Capacidade antioxidante de bebidas aromatizadas: … · contrebalancé par des mécanismes de défense antioxydante. ... Ensuite, une évaluation des paramètres de la capacité antioxydante

Maria de Fátima de Sá Barroso

Mestre em Engenharia do Ambiente

Capacidade antioxidante de bebidas aromatizadas:

águas e chás

junho, 2011

III

Tese de Doutoramento

Capacidade antioxidante de bebidas aromatizadas:

águas e chás

Maria de Fátima de Sá Barroso

Dissertação de candidatura ao grau de Doutor em

Ciências Farmacêuticas – Química Analítica,

apresentada à Faculdade de Farmácia da Universidad e do Porto

Orientação

Professora Doutora Maria Beatriz Prior Pinto Oliveira

Porto

junho, 2011

IV

© Autorizada a reprodução parcial desta dissertação (condicionada à autorização

das editoras das revistas onde os artigos foram pub licados) apenas para efeitos de

investigação, mediante declaração escrita do intere ssado, que a tal se

compromete.

V

A realização deste trabalho foi possível graças à c oncessão de uma Bolsa de

Doutoramento (SFRH/BD/29440/2006) pela Fundação par a a Ciência e a Tecnologia

(FCT), financiada pelo Programa Operacinal Potencia l Humano (POPH) - Quadro de

Referência Estratégico Nacional (QREN) - Tipologia 4.1 - Formação Avançada,

comparticipado pelo Fundo Social Europeu (FSE) e po r Fundos Nacionais do

Ministério da Ciência, Tecnologia e Ensino Superior (MCTES). Em associação à

Bolsa de Doutoramento, a candidata contou ainda com subsídios para a realização

de trabalho no estrangeiro, para deslocamento a con gressos internacionais e para

a execução gráfica desta tese.

VII

Os estudos apresentados nesta dissertação foram re alizados no Laboratório

Grupo de Reacção e Análise Química (GRAQ) do Instit uto Superior de Engenharia

do Porto do Instituto Politécnico do Porto, no Serv iço de Bromatologia da

Faculdade de Farmácia da Universidade do Porto, no Laboratório de Química

Orgânica Física/Química Radicalar do Departamento d e Química da Faculdade de

Ciências e Tecnologia da Universidade Nova de Lisbo a e no Grupo de

Electroanálisis da Faculdade de Química da Universi dade de Oviedo.

IX

AGRADECIMENTOS

.“…Eu não posso brincar contigo, disse a raposa. Não me cativaram ainda. Ah! desculpa, disse o principezinho. Após uma reflexão, acrescentou:

- Que quer dizer "cativar"? ……

- É uma coisa muito esquecida, disse a raposa. Significa "criar laços. - Criar laços?..”

Antoine de Saint-Exupéry em O Principezinho

Tal como Saint-Exupéry também a autora desta dissertação cativou, e deixou-se

cativar pelas inúmeras pessoas com quem se cruzou a nível profissional e pessoal nestes

4 anos de trabalho. Sem o apoio de muitas pessoas teria sido muito difícil chegar à reta

final deste trabalho e atracar num porto seguro. Por isso ficam os meus agradecimentos à:

Fundação para a Ciência e Tecnologia pela concessão de uma bolsa de doutoramento

(SFRH/BD/29440/2006), sem a qual seria impossível realizar este trabalho.

Ao diretor da Faculdade de Farmácia da Universidade do Porto Professor Doutor José

Luís da Costa, ao ex-diretor Professor Doutor José Manuel Sousa Lobo e à Doutora

Isabel Guimarães pelo apoio prestado nas questões burocáticas associadas ao processo

de doutoramento.

À minha orientadora Professora Doutora Beatriz Oliveira por me ter acolhido como

aluna, pela ajuda prestada no decorrer do trabalho, pelo apoio incondicional e pela

amizade que foi crescendo nestes 4 anos, e também à Professora Doutora Cristina

Delerue-Matos, com quem tenho o previlégio de ter criado laços à mais de 10 anos, pelo

seu apoio incondicional, pelos seus conselhos e por me ajudar a crescer a nível pessoal

e científico.

À Professora Teresa Teles, à Professora Sandra Ramos à Engenheira Aurora Silva e

à Engenheira Elisa Soares pelo apoio prestado nas análises mineralógicas e estudos

estatísticos.

Ao Professor João Paulo Noronha, pela simpatia e entusiasmo com que me recebeu

no seu laboratório, pels constante preocupação e permanente disponibilidade e por todo

o apoio científico prestado que permitiu valorizar este trabalho.

X

À Professora Rosa Fireman Dutra e ao Doutor Joilson Jesus, que me receberam no

outro lado do Atlântico de braços abertos, por estarem sempres disponiveis, pelo apoio e

por todo o ensinamento. Ao Professor Lauro Kubota pelo valioso contributo na análise

dos resultados.

À Professora Noemí de-los-Santos Alvarez, e ao Professor Paulino Tuñón Blanco, por

me terem acolhido num magnífico laboratório, por me terem oferecidas condições ótimas

de trabalho, e por todo apoio científico e aprendizagem adquiridos no decorrer do

trabalho.

Aos meus colegas do laboratório por me ouvirem e me apoiarem em inúmeras

situações, e em especial à Doutora Marta Neves pela simpatia e ajuda aquando na

estadia no estrangeiro.

À minha família, país, irmãos, sobrinhos, marido e filhos, mais do que agradecer a

paciência, apoio e conselhos seguem as minhas desculpas pela ausência e pelo apoio

que não foram prestados.

XI

RESUMO

A produção contínua de radicais livres, durante os processos metabólicos, é

compensada pelos mecanismos de defesa antioxidante. Estes visam eliminar ou reduzir

os níveis destes radicais nas células e assim proteger o organismo dos seus efeitos

pejorativos.

Os alimentos e as bebidas são uma boa fonte exógena de antioxidantes. As águas

com sabores, sendo constituídas por aromas naturais, extratos de vegetais (chá) e sumos

de fruta, devem apresentar alguma capacidade antioxidante. Este tipo de refrigerante

(águas com sabores) foi desenvolvido recentemente e, por isso, não se encontrava

caraterizado no que concerne aos fatores com efeito benéfico na saúde humana,

nomeadamente a capacidade antioxidante.

O objetivo principal desta dissertação consistiu em melhorar o conhecimento acerca da

composição química e antioxidante das diversas águas com sabores disponíveis no

mercado português.

Numa primeira fase efetuou-se a caraterização mineralógica destas águas, tendo-se

avaliado um total de 18 minerais: 4 macrominerais (Ca, Mg, K e Na), 3 microminerais (Fe,

Cu e Zn) e 11 elementos vestigiais (Al, As, Cd, Cr, Co, Hg, Mn, Ni, Pb, Se e Si). Os

teores determinados destes minerais estavam dentro dos limites estipulados por lei e

eram variáveis de marca para marca, o que está relacionado com a sua origem.

Seguiu-se a avaliação de parâmetros relacionados com a ação antioxidante, através

de métodos óticos convencionais, nomeadamente, o teor fenólico total, o teor de

flavonoídes total, o poder redutor e a atividade anti-radicalar das águas e dos aromas

usados na sua formulação. Para um melhor conhecimento da composição química dos

aromas foi avaliado o perfil de terpenóides por HS-SPME/GC-MS.

De acordo com os resultados, as amostras de águas e aromas apresentam capacidade

antioxidante e atividade anti-radicar. No entanto, não foram detetados flavonóides. A

análise dos aromas indicou a presença de monoterpenos e sesquiterpenos.

Na sequência do trabalho desenvolveram-se metodologias alternativas para a

quantificação da capacidade antioxidante de águas com sabores. Procedeu-se à

construção de biossensores de ADN, usando bases púricas (adenina ou guanina) ou

cadeias simples de ADN imobilizadas na superfície de elétrodos de carbono vítreo ou de

pasta de carbono, respetivamente.

O princípio de funcionamento destes biossensores baseou-se na avaliação do dano

provocado por radicais livres (hidroxilo, superóxido e sulfato) e da proteção de

antioxidantes (ácido ascórbico, ácido gálico, ácido cafeico, ácido cumárico e resveratrol),

XII

usando-se a voltametria de onda quadrada, voltametria cíclica ou voltametria de impulso

diferencial, como técnicas de deteção

Verificou-se que os radicais livres hidroxilo (OH•), superóxido (O2•-) e sulfato (SO4

•-)

provocam danos oxidativos na adenina e na guanina imobilizada na superfície do

elétrodo de carbono vítreo. Por outro lado, os antioxidantes, ácido ascórbico, ácido gálico,

ácido cafeíco, ácido cumárico e o resveratrol protegem a adenina e a guanina dos danos

provocados pelos radicais.

Com os biossensores de cadeia simples de ADN os radicais livres usados foram o

hidróxilo (OH•) e o superóxido (O2•-). Neste caso o dano oxidativo e a proteção produzida

pelo ácido ascórbico foram avaliados através da medição da corrente eletrocatalítica do

NADH.

Estes biossensores foram usados para determinar a capacidade antioxidante total das

águas aromatizadas, tendo-se verificado capacidade de proteção ao dano causado por

radicais livres.

Palavras-chave : águas com sabores, antioxidante, radicais livres, ADN, biossensores

XIII

ABSTRACT

The continuous production of free radicals during the metabolic process is balanced by

antioxidant defense mechanisms. These mechanisms aim to eliminate or reduce the

levels of these radicals in the cells and thus protect the organism from its deleterious

effects.

Food and beverages are good exogenous sources of antioxidants. Flavoured water,

which has natural flavours, vegetable extracts (like tea) and fruit juices, should have some

antioxidant capability. This type of beverage (flavoured water) is quite recent and so, its

beneficial health factors, namely antioxidant capability, are not yet assessed.

The main goal of this dissertation consisted in improving the knowledge concerning the

chemical and antioxidant composition of the several flavoured waters available in the

Portuguese consumer market.

In a first stage, a minerologic caracterization of the waters was done; a total of 18

mineral were assessed: 4 macrominerals (Ca, Mg, K and Na), 3 microminerals (Fe, Cu

and Zn), and 11 trace elements (Al, As, Cd, Cr, Co, Hg, Mn, Ni, Pb, Se and Si). The levels

of all these minerals were within legal limits but varied from brand to brand, due to its

different sources.

Then, an assessment of antioxidant capability parameters was done. Conventional

optical methods were used, namely, total phenolic content, total flavonoid content,

reducing power and DPPH radical scavenging activity of the waters and flavours used. To

better understand the chemical composition of flavours, the terpenoid profile was

assessed, using HP-SPME/GC-MS.

According to the results, the water samples and flavours presented antioxidant capacity

and radical scavenging activity. However, no flavonoids were found. The analysis of

flavour detected the presence of monoterpenes and sesquiterpenes.

In the work flow, alternative methods were developed to quantify the antioxidant

capacity of flavoured waters. DNA biosensors were assembled, using purine bases

(adenine and guanine) or single strainded DNA, immobilized on the glassy carbon

electrode surface or carbon paste electrode surface, respectively.

The operating principle of these biosensors was based in assessing damage promoted

by free radicals (hydroxyl, superoxide, sulfate) and the protection made by the

antioxidants (ascorbic acid, gallic acid, caffeic acid, coumaric acid and resveratrol), using

square wave voltammetry, cyclic voltammetry and differential pulse voltammetry, as the

detection technique.

It was verified that the free radicals hydroxyl (OH•), superoxide (O2•-) and sulfate (SO4

•-)

induce oxidative damage in adenine and in the guanine immobilized in the electrode

XIV

surface. On the other hand, antioxidants, ascorbic acid, gallic acid, caffeic acid, coumaric

acid and resveratrol protect adenine and guanine from free radical damage.

With DNA-based biosensors, hydroxyl and superoxide were the free radicals used. In

this case, the oxidative damage and the protection granted by ascorbic acid were

assessed through measuring electrocatalytic current of NADH.

These biosensors were used to determine the total antioxidant capability of flavoured

waters, and it was verified that these waters present antioxidant capacity.

Keywords : flavoured waters, antioxidant, free radical, biosensor

XV

RÉSUMÉ

La production continue de radicaux libres au cours du processus métabolique est

contrebalancé par des mécanismes de défense antioxydante. Ces mécanismes visent à

éliminer ou à réduire les niveaux de ces radicaux dans les cellules et donc de protéger

l'organisme contre ses effets délétères.

Les aliments et les boissons sont de bonnes sources d'antioxydants exogènes. Les

eaux aromatisées, qui ont des arômes naturels, des extraits de végétaux (comme le thé)

et des jus de fruits, devraient avoir une certaine capacité antioxydante. Ce type de

boissons (eaux aromatisées) sont assez récente et donc, ses effets bénéfiques pour la

santé, à savoir la capacité antioxydante, ne sont pas encore évalués. L'objectif principal

de cette thèse a consisté à améliorer les connaissances concernant la composition

chimique et antioxydantes de diverses eaux aromatisées disponibles sur le marché des

consommateurs portugais.

En premier, une caractérisation minéralogique des eaux a été faite; un total de 18

minéraux ont été évalués: 4 macrominéraux (Ca, Mg, K et Na), 3 microminéraux (Fe, Cu

et Zn), et 11 éléments résiduels (Al , As, Cd, Cr, Co, Hg, Mn, Ni, Pb, Se et Si). Les

niveaux de tous ces minéraux étaient dans les limites légales, mais variaient d'une

marque à l'autre, en raison de ses différentes sources.

Ensuite, une évaluation des paramètres de la capacité antioxydante a été faite. Les

méthodes classiques optiques ont été utilisés, à savoir, le contenu totale en phénoliques,

le contenu totale en flavonoïdes, le pouvoir réducteur et l'activité antioxydant mesurée par

le radical DPPH des eaux et arômes utilisés. Pour mieux comprendre la composition

chimique des saveurs, le profil de terpénoïdes a été évaluée, en utilisant HP-SPME/GC-

MS.

Selon les résultats, les échantillons d'eaux et de saveurs présentent des capacités

antioxydantes et une activité anti-radicale. Toutefois, aucune flavonoïdes n’ont été

trouvées. L'analyse des aromes détecte la présence de monoterpènes et sesquiterpènes.

Dans le deroulement des travaux, d'autres méthodes ont été développées pour

quantifier la capacité antioxydante des eaux aromatisées. Des biocapteurs d'ADN ont été

assemblés, en utilisant des bases puriques (adénine et la guanine) ou brin simple l'ADN

immobilisé sur la surface de l'électrode de carbone vitreux ou sur la surface de l’électrode

de pâte de carbone, respectivement.

Le principe de fonctionnement de ces biocapteurs a été fondé dans l'évaluation des

dommages provoqués par les radicaux libres (hydroxyle, superoxyde, sulfate) et la

protection faites par les antioxydants (l’acide ascorbique, l’acide gallique, l’acide caféique,

XVI

l'acide coumarique et le resvératrol), en utilisant la voltamétrie à onde carrée, voltamétrie

cyclique et la voltamétrie à impulsion différentielle, ainsi que la technique de détection.

Il a été vérifié que les radicaux libres hydroxyles (OH•), superoxyde (O2• -) et de sulfate

(SO4• -) induisent des lésions oxydatives dans l'adénine et la guanine qui a été

immobilisée dans la surface de l'électrode. D’ une autre part, les antioxydants, l’acide

ascorbique, l’acide gallique, l’acide caféique, l'acide coumarique et le resvératrol

protégent l’adénine et la guanine des dommages provoqués par les radicaux libres. Avec

les biocapteurs basés sur l'ADN, les radicaux libres qui ont été utilisées ont été l’hydroxyle

et le superoxyde. Dans ce cas, les dommages oxydatifs et la protection accordée par

l'acide ascorbique ont été évalués par le mesure de courant électrocatalytique de NADH.

Ces biocapteurs ont été utilisés pour déterminer la capacité antioxydante totale des

eaux aromatisées, et l’on observe que ces eaux présentent des capacités antioxydantes.

Mots-clés : eaux aromatisées, antioxydant, radicaux libres, biocapteur

XVII

RESUMEN

La continua producción de radicales libres en los procesos metabólicos es

contrarrestada por los mecanismos antioxidantes de defensa. El objetivo de dichos

mecanismos es eliminar o reducir los niveles de estos radicales en las células y, por lo

tanto, proteger el organismo de sus efectos nocivos.

Los alimentos y las bebidas son una buena fuente de antioxidantes. Las aguas con

sabores, las cuales contienen aromas naturales, extractos vegetales (como el te) y zumos

de frutas, deberían tener actividad antioxidante. Este tipo de bebidas (aguas de sabores)

es bastante reciente y, por tanto, su efecto beneficioso sobre la salud en relación con su

capacidad antioxidante aún no ha sido estudiada.

El principal objetivo de esta Tesis Doctoral consiste en mejorar el conocimiento

relacionado con la composición química y de antioxidantes de varias agua de sabores

disponibles en el mercado portugués.

En una primera etapa, se llevó a cabo la caracterización mineralógica. Un total de 18

minerales fueron estudiados: 4 macrominerales (Ca, Mg, K y Na), 3 microminerales (Fe,

Cu y Zn) y 11 elementos traza (Al, As, Cd, Cr, Co, Hg, Mn, Ni, Pb, Se y Si). El nivel

encontrado de estos minerales está dentro de los límites legales aunque varía entre

marcas debido a su diferente origen.

Posteriormente, se estudiaron los parámetros relacionados con la capacidad

antioxidante. Para ello se emplearon métodos ópticos convencionales como el contenido

fenólico total, contenido de flavonoides total, el poder reductor y la actividad de

eliminación del radical DPPH. Para comprender mejor la composición química de los

aromas, el perfil de terpenoides fue estudiado por HP-SPME/GC-MS.

Los resultados obtenidos permiten afirmar que las muestras de agua y aromas

presentan capacidad antioxidante y de eliminación de radicales. Sin embargo, no se

encontraron flavonoides. El análisis de aroma detectó la presencia de monoterpenos y

sesquiterpenos.

En el curso del trabajo también se desarrollaron métodos alternativos para cuantificar

la capacidad antioxidante de aguas de sabores. Se prepararon sensores usando las

bases púricas (adenina y guanina) o ADN monocatenario inmovilizado sobre electrodos

de carbono vítreo o de pasta de carbono, respectivamente.

El principio de operación de estos biosensores está basado en la detección del daño

ocasionado por los radicales libres (hidroxilo, superóxido, sulfato) y la protección ejercida

por los antioxidantes (ácido ascórbico, ácido gálico, ácido cafeico, ácido cumárico y

resveratrol) usando voltametría de onda cuadrada, voltametría cíclica y voltametría de

pulso diferencial como técnicas de detección.

XVIII

Se comprobó que los radicales libres hidroxilo(OH•), superóxido (O2•-) y sulfato (SO4

•-)

inducen daño oxidativo en adenina y guanina inmovilizada sobre la superficie electródica.

Por otra parte, los antioxidantes ácido ascórbico, ácido gálico, ácido cafeico, ácido

cumárico y resveratrol protegen a la adenina y a la guanina del daño de los radicales

libres.

Con los biosensores basados en ADN, los radicales libres usados fueron el hidroxilo y

el superóxido. En estos casos, el daño oxidativo y la protección ejercida por el ácido

ascórbico fue estudiada a través de la medida de la corriente electrocatalítica de NADH.

Estos biosensores fueron usados para determinar la capacidad antioxidante total de

aguas con sabores comprobándose que, en efecto, estas aguas presentan capacidad

antioxidante.

Palabras clave : aguas de sabores, antioxidantes, radical libre, biosensor.

XIX

TRABALHOS (PUBLICAÇÕES E COMUNICAÇÕES) DESENVOLVIDO S NO ÂMBITO

DO PROJETO DE DOUTORAMENTO

Publicações de artigos em revistas de circulação in ternacional com arbitragem

científica referenciadas no Journal Citation Reports da ISI Web of Knowledge :

1. Flavoured versus natural waters: Macromineral (Ca, Mg, K, Na) and micromineral (Fe,

Cu, Zn) contents

M. Fátima Barroso, Aurora Silva, Sandra Ramos, M. T. Oliva-Teles, Cristina Delerue-

Matos, M. Goreti F. Sales, M.B.P.P. Oliveira

Food Chemistry, 2009, 116 (2), 580-589

2. Survey of trace elements (Al, As, Cd, Cr, Co, Hg, Mn, Ni, Pb, Se, and Si) in retail

samples of flavoured and bottled waters

M. F. Barroso, S. Ramos, M. T. Oliva-Teles, C. Delerue-Matos, M. G. F. Sales, M. B.

P. P. Oliveira

Food Additives and Contaminants: Part B – Surveillance, 2009, 2 (2), 121-130

3. Flavored waters: Influence of ingredients on antioxidant capacity and terpenoid profile

by HS-SPME/GC-MS

M. Fátima Barroso, J. P. Noronha, Cristina Delerue-Matos, M. B. P. P. Oliveira

Journal of Food and Agricultural Chemistry, 2011, 59 (9), 5062-5072

4. DNA-based biosensor for the electrocatalytic determination of antioxidant capacity in

beverages

M. F. Barroso, N. de-los-Santos-Álvarez, M.J. Lobo-Castañón, A. J. Miranda-Ordieres,

C. Delerue-Matos, M. B. P. P. Oliveira, P. Tuñón-Blanco

Biosensors and Bioelectronics, 2011, 26 (5), 2396-2401

5. Electrochemical DNA-sensor for evaluation of total antioxidant capacity of flavours and

flavoured waters using superoxide radical damage

M. F. Barroso, C. Delerue-Matos, M. B. P. P. Oliveira


XX

6. Electrocatalytic evaluation of DNA damage by superoxide radical for antioxidant

capacity assessment

M. F. Barroso, N. de-los-Santos-Álvarez, M. J. Lobo-Castañón, A. J. Miranda-

Ordieres, C. Delerue-Matos, M. B. P. P. Oliveira, P. Tuñón-Blanco

Journal of Electroanalytical Chemistry, 2011, em publicação

doi:10.1016/j.jelechem.2011.04.022

7. Electrochemical evaluation of total antioxidant capacity of beverages using a purine-

biosensor


Food Chemistry (submetido)

8. Evaluation of total antioxidant capacity of flavoured waters using sulfate radical

damage of purine-based sensors


Electrochimica Acta (submetido)

9. Towards a reliable technology for antioxidant capacity and oxidative damage

evaluation: electrochemical (bio)sensors

M. Fátima Barroso, N. de-los-Santos-Álvarez, C. Delerue-Matos, M. B. P. P. Oliveira

Biosensor and Bioelectronics (submetido)

Publicações de artigos ou resumos alargados em atas de encontros científicos

1. Águas naturais versus aromatizadas: Influência dos ingredientes adicionados na

composição mineral

M. F. Barroso, C. Delerue-Matos, M. G. Sales, M. B. P. P. Oliveira

Atas do 9º Encontro de Química dos Alimentos, 29 Abril-2 Maio, 2009, Angra do

Heroísmo, Açores

Comunicações em poster em encontros científicos internacionais:

1. Evaluation of trace elements in flavoured waters: a case study

M. F. Barroso, M. T. Oliva-Teles, C. Delerue-Matos, M. G. Ferreira Sales, M. B. P. P.

Oliveira

Rapid Methods Europe 2008 for Food and Feed Safety and Quality, P34, 21-23

Janeiro, 2008, Noordwijkerhout, Holanda, 2008

XXI

2. Macrominerals in flavoured waters

M. Fátima Barroso, M.T. Oliva-Teles, Cristina Delerue-Matos, M. Goreti F. Sales,

M. B. P. P. Oliveira

AOAC Europe Section International Workshop e II encontro Nacional de

Bromatologia, Hidrologia e Toxicologia, P3FNFA, 17-18 Abril, 2008, Lisboa, Portugal

3. Antioxidant capacity evaluation in flavour and flavoured waters by total phenolic and

flavonoid contents, reuducing power and DPPH scavenging assays

Maria Fátima Barroso, Cristina Delerue-Matos, João Paulo Noronha, Maria Beatriz

Oliveira, Maria Goreti Sales

1st European Food Congress, P208, 4-9 Novembro, 2008, Ljubljana, Slovenia

4. Investigations on the electrocatalytic assessment of antioxidant capacity using a DNA-

Modified carbon paste electrode

M. F. Barroso, N. de-los-Santos-Álvarez, M. J. Lobo-Castanon, A. J. Miranda-

Ordieres, M. G. Ferreira Sales, M. B. P. P. Oliveira, C. Delerue-Matos

Euro Analysis 2009, P081-B2, 6-10 Setembro, 2009, Innsbruck, Áustria

5. Development of a DNA-modified sensor to evaluate the total antioxidant capacity of

flavoured waters

M. Fátima Barroso, J. Paulo Noronha, M. Goreti F. Sales, Cristina Delerue-Matos, M.

Beatriz P. P. Oliveira

4th International Symposium on Recent Advances in Food Analysis, N-23, 4-6

Novembro, 2009, Prague, Czech Republic

6. Antioxidant capacity of flavoured waters by electrochemical DNA-Biosensor

M. Fátima Barroso, M. Goreti Sales, Cristina Delerue-Matos, M. B. P. P. Oliveira

13th International Conference on Electroanalysis, Pin-40, 20-24 Junho, 2010, Gigon,

Spain

7. DNA damage generated by a sulphate radical and the protective effect of dietary

antioxidants using an electrochemical DNA biosensors

M. Fátima Barroso, M. Goreti Sales, Cristina Delerue-Matos, M. B. P. P. Oliveira

13th International Conference on Electroanalysis, Pin-41, 20-24 Junho, 2010, Gigon,

Spain

XXII

Comunicações em poster em encontros científicos nacionais:

1. Aguas naturais versus aromatizadas: Influência dos ingredientes adicionados na

composição mineral

M. F. Barroso, S. Ramos, C. Delerue-Matos, M. G. Sales, M. B. P. P. Oliveira

9º Encontro de Química dos Alimentos, P42, 29 Abril-2 Maio, 2009, Angra do

Heroísmo, Açores.

XXIII

ÍNDICE

Agradecimentos IX

Resumo XI

Abstract XIII

Résumé XV

Resumen XVII

Trabalhos desenvolvidos no âmbito do projeto de dou toramento XIX

Índice XXIII

Lista de Abreviaturas e Símbolos XXV

I. REVISÃO DO ESTADO DA ARTE 1

Organização e estrutura da dissertação 3

Objetivos 5

Introdução geral 7

Referências 12

CAPÍTULO 1. Métodos eletroquímicos 17

Towards a reliable technology for antioxidant capacity and oxidative damage evaluation: electrochemical (bio)sensors 19

II. INVESTIGAÇÃO E DESENVOLVIMENTO 51

CAPÍTULO 2. Composição mineralógica 53

2.1. Flavoured versus natural waters: Macromineral (Ca, Mg, K, Na) and

micromineral (Fe, Cu, Zn) contents 55

2.2. Survey of trace elements (Al, As, Cd, Cr, Co, Hg, Mn, Ni, Pb, Se, and

Si) in retail samples of flavoured and bottled waters 77

CAPÍTULO 3. Perfil antioxidante – métodos convenc ionais 99

Flavored Waters: Influence of Ingredients on Antioxidant Capacity and

Terpenoid Profile by HS-SPME/GC-MS

101

CAPÍTULO 4. Construção de biossensores de bases púricas

127

4.1. Electrochemical evaluation of total antioxidant capacity of beverages

using a purine biosensor 129

XXIV

4.2. Electrochemical DNA-sensor for evaluation of total antioxidant

capacity of flavours and flavoured waters using superoxide radical

damage

149

4.3. Evaluation of total antioxidant capacity of flavoured waters using

sulfate radical damage of purine-based sensors

167

CAPÍTULO 5. Construção de biossensores de ADN 187

5.1. DNA-based biosensor for the electrocatalytic determination of

antioxidant capacity in beverages

189

5.2. Electrocatalytic evaluation of DNA damage by superoxide radical for

antioxidant capacity

207

CONSIDERAÇÕES FINAIS 225

XXV

LISTA DE ABREVIATURAS E SÍMBOLOS

Na lista apresentada constam termos em português e em inglês, consoante a língua

em que são utilizados ao longo da dissertação.

A• reactive radical

AA ascorbic acid

AAS atomic absorption spectrophotometry

ACu ácido cumárico

AdSV Adsorptive stripping voltammetry

AH phenolic antioxidant

AOC antioxidant capacity

APHA American Public Health Association

AuE gold electrode

BHA tert-butylhydroxy anisole

BHT butylated hydroxytolurene

CA caffeic acid

CF carbon fiber

CMEs chemically modified electrodes

CNTECE carbon nanotube epoxy composite electrode

CPE carbon paste electrode

CV cyclic voltammetry

DA dopamine

dA21 deoxyadenylic acid oligonucleotide

DCTMACl docosyltrimethylammonium chloride

DNA deoxyribonucleic acid

DPPH 2,2-diphenyl-1-picrylhydrazyl

DPV differential pulse voltammetry

dsDNA double strainded DNA

E potential

EFSA European Food Safety Authority

EI electronic impact

Eº´ formal potential

Ep peak potential

EPA Environmental Protection Agency

FAD flavin adenine dinucleotide

XXVI

FAO Food and Agriculture Organisation

FC Folin-Ciocalteu

FDA Food and Drug Administration

FIA flow injection analysis

FNB/IOM Food and Nutrition Board of the Institute of Medicine

FRAP ferric reducing antioxidant power

GA gallic acid

GCE glassy carbon electrode

GC-MS gas chromatography-mass spectrometry

GPES general purpose electrochemical system

H2O2 hydrogen peroxide

HAT hydrogen atom transfer

HOCl hypochorous acid

HOPC highly oriented pyrolytic graphite

HRP horseradish peroxidase

HS-SPME Headspace-solid-phase microextraction

i current

Ia electrocatalytic current in the presence of an antioxidant

ICP–AES inductively coupled plasma-atomic emission spectroscopy

ICP–MS inductively coupled plasma-mass spectrometry

Id electrocatalytic current in the absence of an antioxidant

ip peak current

JECFA Joint Expert Committee on Food Additives

LDL low-density lipoprotein

LOD limit of detection

LOQ limit of quantification

LSV linear sweep voltammetry

LSW linear sweep voltammetry

M metal

m/m massa/massa

MAE microwave-assisted extraction

MWCNs multi-walled carbon nanotubes

NADH nicotinamide adenine dinucleotide disodium salt, reduced form

NiHCF nickel hexacyanoferrate

NO• nitric oxide radical

NPs nanoparticules

XXVII

O2•- Superoxide radical, radical superóxido

OH• hydroxyl radical, radical hidroxilo

ORAC oxygen radical absorbance capacity

PAMAM poly(amidoamine)

PBS phosphate buffer saline

PCL photochemiluminescence

PDMS/DVB polydimethylsiloxane/divinylbenzene

PGE pyrolytic graphite electrode

PGEs pencil graphite electrodes

PPO polyphenol oxidase

PRTC peroxyl radical trapping capacity

PS polystyrene

PTDI provisional tolerable daily intake

PtE platinium electrode

PTWI provisional tolerable weekly intake

RDA recommended dietary allowance

RE relative error

REC recovery

RES resveratrol

RfD reference dose

RNS reactive nitrogen species

RO• alkoxyl radical

ROO• peroxyl radical

ROS reactive oxygen species

RSA radical-scavenging activity

RSD relative standard deviations

SAM self assembled monolayer

SAMe s-adenosyl-L-methionine

SD standard deviation

SET single electron transfer

SFE supercritical fluid extraction

SO4•– sulfate radical, radical sulfato

SOD superoxide dismutase

SPE screen printed electrode

SPE-Au gold-screen printed electrode

SPME solid-phase microextraction

XXVIII

ssDNA single strainded DNA

SSPE concentrated saline sodium phosphate EDTA

SV stripping voltammetry

SWCNs single-walled carbon nanotubes

SWV square wave voltammetry

TAC total antioxidant capacity

TEAC trolox equivalent antioxidant capacity assay

TFC total flavonoid content

TOSC total oxidant scavenging capacity

TPC total phenol content or total phenolic content

TRAP total radical-trapping antioxidant parameter

Trolox 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid

UV / VIS ultraviolet-visible

v/v volume/volume

WHO World Health Organization

XOD xanthine oxidase

Γ surface coverage

~ approximately

1

I.

REVISÃO DO ESTADO DA ARTE

Organização e estrutura da dissertação

3

ORGANIZAÇÃO E ESTRUTURA DA DISSERTAÇÃO

A presente dissertação inclui todos os artigos científicos (6 publicados e 3 submetidos)

resultantes do projeto de doutoramento, pretendendo-se assim que todos os trabalhos

desenvolvidos tivessem já sido objeto de análise crítica por parte de diferentes revisores

internacionais, especialistas na área de investigação em que se inserem, e selecionados

de acordo com os critérios de cada revista em que o respetivo artigo foi publicado.

Todos os artigos estão escritos em inglês. Optou-se por manter a formatação original,

adaptada ao corpo da tese, com a qual os textos foram submetidos ou publicados, de

acordo com as normas específicas de cada revista. Assim, encontram-se variações na

estrutura dos diferentes artigos apresentados ao longo dos capítulos e no modo como

são indicadas as referências. Toda a bibliografia que não está integrada nas publicações,

é apresentada de acordo com a norma de Vancouver aconselhado pelas “normas de

formatação das dissertações de mestrado e teses de doutoramento” da Faculdade de

Farmácia da Universidade do Porto.

Os textos que se apresentam em português foram elaborados de acordo com o novo

acordo ortográfico.

A dissertação está organizada em duas partes: a Parte I corresponde à Revisão do

Estado da Arte ; a Parte II é designada por Investigação e Desenvolvimento , onde se

inclui toda a investigação desenvolvida na componente experimental.

Assim, na Parte I , com uma abordagem mais teórica, faz-se inicialmente uma

introdução geral ao tema “águas com sabores”, com o objetivo de contextualizar o

trabalho desenvolvido, abordando questões como o consumo desta bebida em Portugal e

fazendo referência à legislação nacional sobre sobre águas de consumo. A esta breve

introdução, segue-se a descrição dos objetivos do trabalho. Ainda dentro desta primeira

parte, é apresentado o Capítulo 1 constituído por um artigo, em que se faz a Revisão do

Estado da Arte no que se refere a todos os métodos eletroquímicos utilizados para a

determinação da capacidade antioxidante de diversas matrizes.

A Parte II da dissertação é constituída por 4 capítulos (capítulos 2 a 5). No geral, cada

capítulo é constituído por um ou mais artigos. O Capítulo 2 integra 2 artigos científicos

correspondentes à determinação mineralógica das águas com sabores e respetivas

águas naturais. O primeiro artigo refere-se à determinação de macrominerais (Ca, Mg, K

e Na) e microminerais (Fe, Cu e Zn) enquanto que o segundo aborda a determinação dos

minerais vestigiais (Al, As, Cd, Cr, Co, Hg, Mn, Ni, Pb, Se e Si). O Capítulo 3 (1 artigo)

corresponde à determinação do perfil antioxidante das águas com sabores e seus

aromas através de métodos óticos convencionais. O perfil antioxidante foi feito através da

determinação do teor fenólico total, teor total de flavonóides, poder redutor e atividade

I. Revisão do estado da arte

4

anti-radicalar. O trabalho apresentado neste capítulo foi elaborado em colaboração com o

Laboratório de Cromatografia da Linha de Química Orgânica Física/Química Radicalar do

departamento de Química da FCT-UNL. O Capítulo 4 (3 artigos) e Capítulo 5 (2 artigos)

correspondem ao desenvolvimento das metodologias analíticas alternativas, para a

determinação da capacidade antioxidante, baseadas em técnicas eletroquímicas e na

construção de biossensores. No Capítulo 4 refere-se a construção de biossensores de

adenina e de guanina. Estes biossensores foram sujeitos ao ataque oxidativo induzido

por 3 radicais livres (hidroxilo, superóxido e sulfato) e avalia-se o efeito protetor

promovido por 5 antioxidantes (ácido ascórbico, ácido gálico, ácido cafeico, ácido

cumárico e resveratrol). Cada artigo corresponde a um dos radicais. No Capítulo 5 , os

artigos descrevem a construção de biossensores de cadeia simples de ADN em que se

usaram o radical hidroxilo e o superóxido para a promoção do dano oxidativo. O trabalho

experimental associado a este capítulo foi desenvolvido no Grupo de Electroanálisis da

Faculdade de Química da Universidade de Oviedo.

No final da dissertação apresentam-se as Considerações Finais . Optou-se por esta

designação, uma vez que se foram apresentando conclusões parciais ao longo dos

diferentes capítulos. Nesta medida, o final desta dissertação não é necessariamente um

culminar de um projeto, mas pretende apenas apresentar um conjunto de reflexões sobre

o trabalho realizado, perspetivando projetos e atividades de investigação futuras.

Objetivos

5

OBJETIVOS

Considerando a ausência de estudos científicos na matriz água aromatizada,

considerou-se importante avaliar algumas propriedades desta nova bebida lançada no

mercado. Por isso os objetivos desta dissertação foram:

• Caraterizar quimicamente, no que diz respeito ao teor de macrominerais,

microminerais e minerais vestigiais, as águas com sabores;

• Analisar os aromas adicionados a estas águas por HS-SPME/GC-MS;

• Avaliar o perfil antioxidante destas águas e dos aromas usados na sua formulação,

através da avaliação do teor fenólico total, do teor de flavonóides total, poder redutor e

atividade anti-radicalar usando os métodos óticos convencionais;

• Desenvolver metodologias analíticas alternativas para a determinação da capacidade

antioxidante das águas com sabores:

• Construir biossensores eletroquímicos de cadeia simples de ADN ou de bases

púricas (adenina ou guanina);

• Estudar o dano oxidativo induzido por radicais livres (hidroxilo, superóxido e

sulfato) na superfície modificada do elétrodo.

• Estudar a proteção promovida por antioxidantes na superfície modificada do

elétrodo;

• Avaliar a capacidade antioxidante total das águas com sabores utilizando estes

biossensores.

Introdução geral

7

INTRODUÇÃO GERAL

É do conhecimento geral que a água é indispensável à vida sendo um fator primordial

na qualidade de vida e imprescindível a todos os aspetos da existência humana. A água

representa cerca de 60 a 70 % do peso corporal e é responsável pelo controlo da

temperatura corporal, pela manutenção do volume vascular, por possibilitar as reações

enzimáticas envolvidas na digestão, e pela absorção e metabolismo. Além de transportar

substâncias como nutrientes e oxigénio para as células, promove a excreção de toxinas

através dos rins. Uma hidratação correta contribui para o bem-estar e para a manutenção

da saúde.

A ingestão diária de água deve compensar as perdas de fluidos que ocorrem através

da excreção urinária, transpiração e respiração (1,2). Dependendo do género e da idade

do indivíduo, as quantidades diárias recomendadas para a ingestão de água podem

variar entre 0,7 L (bebés entre os 0-6 meses) 2,7 L (mulheres com idade superior a 19

anos) e 3,7 L (homens com idade superior a 19 anos). As quantidades de água

recomendadas incluem a água presente nos alimentos sólidos e líquidos (como por

exemplo a sopa) e bebidas. De acordo com Ershow and Cantor (1989), 28 % da água

ingerida pelos adultos corresponde à água dos alimentos, 28 % à água potável, enquanto

44 % corresponde a outro tipo de bebidas (3,4).

A qualidade e quantidade de água disponível para consumo humano tem sido uma

preocupação ao longo da história. As principais fontes deste bem essencial são as

massas de água superficial e subterrânea. A população atual, preocupada com a

qualidade da água, tem aumentado o consumo de água engarrafada (5).

Estão disponíveis no mercado vários tipos de água engarrafada: água mineral natural,

água mineral gasosa, água mineral natural gaseificada e água de nascente. A água

mineral natural é uma água de circulação subterrânea, considerada bacteriologicamente

própria, com caraterísticas físico-químicas estáveis na origem, caraterizada por um teor

de substâncias minerais, oligoelementos ou outros constituintes, de que podem

eventualmente resultar efeitos favoráveis à saúde e que se distingue da água de

consumo. Este tipo de água pode conter gás natural (água mineral natural gasosa), ser

reforçada com gás (água mineral natural reforçada com gás carbónico) ou conter

somente gás adicionado (água mineral natural gaseificada). Por outro lado, a água de

nascente é uma água subterrânea, considerada bacteriologicamente própria, com

caraterísticas físico-químicas que a tornam adequada para consumo humano no seu

estado natural (5,6). Estas águas encontram-se regulamentas pelo Decreto-Lei nº 156/98

que define e carateriza as águas minerais naturais e as águas de nascente e estabelece

as regras relativas à exploração, acondicionamento e comercialização (6).


8

No entanto, em determinadas águas minerais naturais, devido à sua origem

hidrogeológica, podem estar presentes, no estado natural, elementos químicos que, a

partir de uma certa concentração, podem representar um risco para a saúde pública (7).

Assim, o decreto-Lei nº 72/2004 estabelece uma lista com os limites máximos de

constituintes que se encontram, naturalmente, nas águas minerais naturais. Os

constituintes presentes nessa lista são o antimónio, arsénio, bário, boro, cádmio, cromo,

cobre, cianeto, fluoretos, chumbo, manganês, mercúrio, níquel, nitratos, nitritos e o

selénio (7).

Embora a água seja muito importante para o ser humano, muitos indivíduos não

consomem água, preferindo ingerir outro tipo de bebidas (sumos, refrigerantes ou

chás/infusões). Em resposta às necessidades atuais do consumidor e conhecedores das

suas preferências, os industriais de bebidas desenvolvem com frequência novos tipos de

bebidas. É o caso das águas com sabores, baseadas na adição de um ou mais aromas

(naturais ou sintéticos) à água natural. Estas águas podem conter, para além do aroma,

sumos de fruta, conservantes, reguladores de acidez e edulcorantes. De acordo com a

Portaria nº 703/96, as águas com sabores podem, legalmente, ter diferentes

designações. No caso de estas águas conterem entre 6 % e 16 % (m/m) de sumo de

fruta, a designação será “refrigerante de sumo de frutos”; e refrigerante aromatizado no

caso desta bebida resultar da diluição de aromatizantes. Se estas bebidas não

contiveram açúcares nem edulcorantes a designação será “água aromatizada” (8). Neste

trabalho optou-se pela designação de águas com sabores.

O consumo de água com sabores é um mercado em ascensão, tendo sido vendidos

em Portugal e no 1º semestre de 2010 cerca de 6,07 milhões de litros (9). No mercado

português existe uma grande oferta destas bebidas. As mais usuais e frequentes são as

águas com aroma de limão, mas também podem ser encontradas bebidas com aroma de

laranja, ananás, pêssego, melão, morango, maçã e goiaba. Há marcas comerciais que

apostaram no desenvolvimento de águas com dois aromas em simultâneo

(laranja/framboesa, pêssego/ananás, maçã/chá, framboesa/ginseng, pêssego/chá,

manga/biloba, melão/hortelã). Sendo estas bebidas relativamente recentes, não havia

publicações científicas abordando o estudo deste tipo de águas quando se iniciou este

trabalho.

Analisando os rótulos das garrafas de águas com sabores das diferentes marcas

comerciais disponíveis no mercado português, verifica-se uma grande diversidade nos

ingredientes adicionados à água. Para além dos aromas (naturais ou sintéticos) algumas

marcas referem a adição de fibras alimentares (ex, dextrina de trigo), sumos de fruta

(concentração máxima de 2,3 %) compostos bioativos (ginseng, L-carnitina, chá verde e

branco, ginkgo biloba), vitaminas (vitamina C e complexos da vitamina B) e ingredientes

Introdução geral

9

que, não tendo uma relação positiva com o bem-estar e saúde do consumidor, são

necessários para garantir a qualidade desejada, tais como, agentes acidificantes (ácido

cítrico e citrato de sódio), edulcorantes (acessulfame-K, aspartamo e sucralose) e

conservantes (sorbato de potássio e benzoato de sódio).

Em termos energéticos, as águas com sabores que contêm edulcorantes apresentam

um valor energético inferior (0,4 a 1,3 kcal/ 100 mL) ao de águas com sabores que não

contêm edulcorantes (9 a 13 kcal/ 100 mL). Comparativamente com os refrigerantes,

néctares e sumos de fruta, as águas com sabores um baixo valor calórico, apresentando

os primeiros um valor calórico que oscila entre os 21 kcal/ 100 mL e os 46 kcal/ 100 mL.

Naturalmente que as águas com sabores não devem substituir a água potável, mas

podem ser uma alternativa interessante aos refrigerantes. Estes possuem na sua

composição ingredientes com efeito negativo na saúde dos consumidores, especialmente

crianças. O consumo moderado de águas com sabores pode ser feito com prazer e sem

grandes preocupações de saúde. No entanto, será importante referir que são mais caras

(20-40 %) do que as águas naturais engarrafadas (2, 10-13).

Considerando que as águas com sabores podem conter, além da sua composição

normal, sumos de fruta e aromas naturais, extraídos de frutas ou vegetais (exemplo do

chá), é esperado que estas águas possam apresentar valores de capacidade antioxidante

superiores.

De acordo com Laguerre e colaboradores (2010), os antioxidantes são substâncias

que, mesmo quando presentes em baixas concentrações, comparativamente com um

substrato oxidante, protegem (por si mesmo ou através dos seus produtos de oxidação) o

substrato dos danos provocados pela oxidação (14). Os alimentos (fruta, vegetais,

legumes e cereais) e bebidas (sumos, chá, café e vinho) são boas fontes externas de

antioxidantes (15-20). Os compostos que vulgarmente conferem capacidade antioxidante

aos alimentos são o ácido ascórbico, os ácidos fenólicos, os tocoferóis, os terpenos,

entre outros (21). Por isso, aumentar a ingestão de antioxidantes na dieta alimentar

(incluindo as águas com sabores) pode ajudar a fortalecer os mecanismos de defesa

antioxidante do organismo humano (21). Conscientes da importância que estes alimentos

e bebidas têm na defesa do organismo, contra os radicais livres produzidos durante o

metabolismo celular, têm sido efetuados muitos estudos para caraterizar o perfil

antioxidante de uma diversidade de alimentos e bebidas, bem como no desenvolvimento

de novos produtos alimentares fortificados com antioxidantes.

Embora não existam estudos (nacionais ou internacionais) publicados sobre a

capacidade antioxidante de águas com sabores (somente existem os trabalhos realizadas

nesta dissertação) alguns grupos portugueses têm-se dedicado ao estudo e valorização,

quer de recursos naturais quer de produtos locais, a fim de desenvolver sistemas


10

económicos, social e ambientalmente sustentáveis. Uma das formas de valorizar esses

produtos é conhecer o seu perfil antioxidante.

Há vários grupos de investigação portugueses que têm contribuído para o

conhecimento do perfil antioxidante destes produtos. Das diversas matrizes estudadas

salienta-se o estudo da capacidade antioxidante de cogumelos (20,22,23), ervas

aromaticas (15), verduras (24), castanha e amêndoa (25), plantas aromáticas (26), frutos

e sumos de fruta (18) e vinho (27). Sendo o vinho um produto muito produzido em

Portugal e valorizado nacional e internacionalmente, alguns grupos de investigação

traçaram o perfil antioxidante de vários tipos de vinhos produzidos no nosso país (17, 28-

31). O café é uma das bebidas mais consumidas em Portugal e nesta área o Laboratório

de Bromatologia da Faculdade de Farmácia da Universidade do Porto tem desenvolvido

um amplo trabalho de investigação incluindo a avaliação da capacidade antioxidante

desta bebida (32,33).

Outro tipo de matrizes, como por exemplo chá, frutos vermelhos (romã, mirtilos e

morangos) e óleos têm sido avaliados por grupos de investigação da Universidade do

Minho (34) e da Universidade do Porto (35,36).

A nível internacional há um grande número de publicações visando a avaliação da

capacidade antioxidante de diferentes tipos de alimentos e produtos processados. Dada a

extensão da bibliografia serão apenas citados alguns artigos mais recentes (37-39).

A metodologia analítica mais vulgarmente usada para quantificar a capacidade

antioxidante de produtos é a espetrofotometria de UV-Vis (40). Estes métodos

convencionais têm a desvantagem de ter tempos de reação e análise longos; as

amostras com cor precisarem de um pré-tratamento; e, sendo o alimento uma matriz

complexa, conter muitos interferentes (40). Considerando ser importante encontrar

alternativas aos métodos convencionais, foram desenvolvidos e construídos, ao longo

deste projeto, biossensores para a quantificação da capacidade antioxidante total de

águas com sabores. A nível da literatura, encontram-se algumas publicações envolvendo

a utilização de biossensores de ADN para a avaliação da capacidade antioxidante de

extratos vegetais e chás (41,42).

Esta dissertação surge no seguimento da investigação científica desenvolvida sobre

sensores e biossensores eletroquímicos no Grupo de Reacção e Análise Química

(GRAQ) do Instituto Superior de Engenharia do Instituto Politécnico do Porto e sobre

antioxidantes realizada no Laboratório de Bromatologia da Faculdade de Farmácia da

Universidade do Porto. Pretende-se assim dar mais um passo num processo cumulativo

em relação ao estado da arte.

Alguns dos trabalhos apresentados foram também desenvolvidos no Laboratório de

Cromatografia da Linha de Química Orgânica Física/Química Radicalar do Departamento

Introdução geral

11

de Química da Faculdade de Ciências e Tecnologia da Universidade Nova de Lisboa sob

supervisão do Professor Doutor João Paulo Noranha e no Laboratório de Electroanálisis

do Departamemto de Química Física e Analítica da Faculdade de Química da

Universidade de Oviedo, sob orientação da Professora Doutora Noemí de los Santos

Álvarez.


12

REFERÊNCIAS

1. Wolf A, Bray GA, Popkin BM. A short history of beverages and how our body treats

them. Obes Rev 2008 Mar; 9 (2):151–64.

2. Daniels MC, Popkinnure BM. Impact of water intake on energy intake and weight

status: A systematic review. Nutr Rev 2010 Dec; 68 (9): 505–21.

3. Ershow AG, Cantor KP. Total Water and Tapwater Intake in the United States:

Population-based estimates of quantities and sources. Bethesda, MD: Life Sciences

Research Office, 1989.

4. Institute of Medicine of the National Academies. Dietary reference intakes for water,

potassium, sodium, chloride, and sulfate. Food and Nutrition Board. The National

Academies Press. Washington, D.C. 2005.

Disponível em http://books.nap.edu/openbook.php?record_id=10925&page=86

[acedido em 12/05/2011]

5. Food and Drug Administration (FDA). Bottled water regulation and the FDA. Center

for Food Safety and Applied Nutrition.

Disponível em http://www.fda.gov/NewsEvents/Testimony/ucm170932.htm [acedido

em 12/05/2011]

6. Ministério da Agricultura, do Desenvolvimento Rural e das Pescas. Decreto-Lei nº

156/98 de 6 de Junho. Diário da República. 1ª Série, nº 131, 2593-2598.

7. Ministério da Agricultura, do Desenvolvimento Rural e das Pescas. Decreto-Lei nº

72/2004 de 25 de Março. Diário da República. 1ª Série, nº 72, 1916-1919.

8. Ministérios da Economia, da Agricultura, do Desenvolvimento Rural e das Pescas,

da Saúde e do Ambiente. Portaria nº 703/96 de 6 de Dezembro. 1ª Série, nº 282,

4387-4887.

9. ANIRSF (Associação Nacional dos Industriais de Refrigerantes e Sumos de Frutos).

Informar nº29. Disponível em http: www.anirsf.pt [acedido em 03/11/2010].

10. Ray GA. Soft drink consumption and obesity: it is all about fructose. Curr Opin Lipidol

2010 Feb; 21 (1): 51-7.

11. Moreno LA, Rodriguez G, Fleta J, Bueno-Lozano M, Lazaro A, Bueno G. Trends of

dietary habits in adolescents. Crit Rev Food Sci Nutr 2010 Jan; 50 (2): 106-12.

12. Lino CM, Pena A. Occurrence of caffeine, saccharin, benzoic acid and sorbic acid in

soft drinks and nectars in Portugal and subsequent exposure assessment. Food

Chem 2010 Jul; 121 (2): 503–8.

13. Pena A, Lino C, Silveira MIN. Survey of caffeine levels in retail beverages in

Portugal. Food Addit Contam Part A-Chem 2005 Feb; 22 (2): 91–6.

Introdução geral

13

14. Laguerre M, Decker EA, Lecomte J, Villeneuve P. Methods for evaluating the

potency and efficacy of antioxidants. Curr Opin Clin Nutr Metab Care 2010 Sep; 13

(5): 518-25.

15. Guimarães R, Barreira JCM, Barros L, Carvalho AM, Ferreira ICFR. Effects of oral

dosage form and storage period on the antioxidant properties of four species used in

traditional herbal medicine. Phytother Res 2010 Apr; 25 (4): 484-92.

16. Alves RC, Costa ASG, Jerez M, Casal S, Sineiro J, Nunez MJ, Oliveira B. Antiradical

activity, phenolics profile, and hydroxymethylfurfural in espresso coffee: Influence of

technological factors. J Agric Food Chem 2010 Dec; 58 (23): 12221-29.

17. Feliciano RP, Bravo MN, Pires MM, Serra AT, Duarte CM, Boas LV, Bronze MR.

Phenolic content and antioxidant activity of Moscatel dessert wines from the Setubal

region in Portugal. Food Anal Meth 2009 Jun; 2 (2): 149-61.

18. Guimarães R, Barros L, Barreira JCM; Sousa MJ, Carvalho AM, Ferreira, ICFR.

Targeting excessive free radicals with peels and juices of citrus fruits: grapefruit,

lemon, lime and orange. Food Chem Toxicol 2010 Jan; 48 (1): 99-106.

19. Assunção M, Santos-Marques MJ, Monteiro R, Azevedo I, Andrade JP, Carvalho F,

Martins MJ. Red Wine Protects against Ethanol-Induced Oxidative Stress in Rat

Liver. J Agric Food Chem 2009 Jul; 57 (14): 6066-73.

20. Vaz JA, Barros L, Martins A, Santos-Buelga C, Vasconcelos MH, Ferreira ICFR.

Chemical composition of wild edible mushrooms and antioxidant properties of their

water soluble polysaccharidic and ethanolic fractions. Food Chem 2011 May; 126 (2):

610-16.

21. Calixto F, Silva L. O consumo de antioxidantes e o balanço energético na dieta de

praticantes de hidroginástica em idade sénior. Motri 2007 Apr; (3) 2: 7-8.

22. Heleno SA, Barros L, Sousa MJ, Martins A, Santos-Buelga C, Ferreira ICFR.

Targeted metabolites analysis in wild Boletus species. LWT-Food Sci Technol 2011

Jul; 44 (6): 1343-8.

23. Reis FS, Ferreira ICFR, Banos L, Martins A. A comparative study of tocopherols

composition and antioxidant properties of in vivo and in vitro ectomycorrhizal fungi.

LWT-Food Sci Technol 2011 May; 44 (4): 820-4.

24. Martins D, Barros L, Carvalho AM, Ferreira ICFR. Nutritional and in vitro antioxidant

properties of edible wild greens in Iberian Peninsula traditional diet. Food Chem 2011

Mar; 125 (2): 488-94.

25. Barreira JCM, Ferreira ICFR, Oliveira MBPP, Pereira JA. Antioxidant Potential of

Chestnut (Castanea sativa L.) and Almond (Prunus dulcis L.) By-products. Food Sci

Technol Int 2010 Jun; 16 (3): 209-16.


14

26. Guimaraes R, Sousa MJ, Ferreira ICFR. Contribution of essential oils and phenolics

to the antioxidant properties of aromatic plants. Ind Crop Prod 2010 Sep; 32 (2): 152-

6.

27. Goncalves B, Falco V, Moutinho-Pereira J, Bacelar E, Correia C. Effects of elevated

CO2 on Grapevine (Vitis vinifera L.): Volatile composition, phenolic content, and in

vitro antioxidant activity of red wine. J Agric Food Chem 2009 Jan; 57 (1): 265-73.

28. Azevedo J, Teixeira N, Oliveira J, Freitas, Mateus N. Effect of sugar acylation on the

antioxidant properties of Vitis vinifera red grape malvidin-3-glucoside. Int J Food Sci

Technol 2011 Feb; 46 (2): 343-49.

29. Goncalves FJ, Jordão AM. Influence of different commercial fining agents on

proanthocyanidin fraction and antioxidant activity of a red wine from baga grapes. J

Int Sci Vigne Vin 2009 Apr-Jun; 43 (2) 111-20.

30. Canas S, Casanova V, Belchior AP. Antioxidant activity and phenolic content of

Portuguese wine aged brandies. J Food Compos Anal 2008 Dec; 21 (8): 626-33.

31. Novak IJaneiro P, Seruga M, Oliveira-Brett AM. Ultrasound extracted flavonoids from

four varieties of Portuguese red grape skins determined by reverse-phase high-

performance liquid chromatography with electrochemical detection. Anal Chim Acta

2008 Dec; 630 (2): 107-15.

32. Alves RC, Casal S, Oliveira MBPP. Tocopherols in coffee brews: Influence of coffee

species, roast degree and brewing procedure. J Food Compos Anal 2010 Dec; 23

(8): 802-8.

33. Alves RC, Almeida IMC, Casal S, Oliveira MBPP. Method development and

validation for isoflavones quantification in coffee. Food Chem 2010 Oct; 122 (3): 914-

9.

34. Lima CF, Fernandes-Ferreira M, Pereira-Wilson C. Drinking of Salvia officinalis tea

increases CCl4-induced hepatotoxicity in mice. Food Chem Toxicol 2007 Mar; 45 (3):

456-64.

35. Faria A, Oliveira J, Neves P, Gameiro P, Santos-Buelga C, de Freitas V, Mateus N.

Antioxidant properties of prepared blueberry (Vaccinium myrtillus) extracts. J Agric

Food Chem 2005 Aug; 53 (17): 6896-902.

36. Costa J, Amaral JS, Mafra I, Oliveira MBPP. Refining of Roundup Ready (R) soya

bean oil: Effect on the fatty acid, phytosterol and tocopherol profiles. Eur J Lipid Sci

Technol 2011 Apr; 113 (4): 528-535.

37. Gollucke, Andrea PB, Recent applications of grape polyphenols in foods, beverages

and supplements. Recent Pat Food Nutr Agric 2010 Jun; 2 (2): 105-9.

38. Pharn-Huy LAN, He H, Phamhuy C. Green tea and health: An overview. J Food Agric

Environ 2008 Jan; 6 (1): 6-13.

Introdução geral

15

39. Ferruzzi MG. The influence of beverage composition on delivery of phenolic

compounds from coffee and tea. Physiol. Behav 2010 Apr; 100 (1): 33-41.

40. Prior RL, Wu X, Schaich K. Standardized methods for the determination of

antioxidants capacity and phenolics in foods and dietary supplements. J Agric Food

Chem 2005 May; 53 (10): 4290-302.

41. Mello LD, Hernandez S, Marrazza G, Mascini M, Kubota LT. Investigations of the

antioxidant properties of plant extracts using a DNA-electrochemical biosensor.

Biosens. Bioelectron 2006 Jan; 21 (7): 1374-82.

42. Ferancova A, Heilerova L, Korgova E, Silhar S, Stepanek I, Labuda J. Anti/pro-

oxidative properties of selected standard chemicals and tea extracts investigated by

DNA-based electrochemical biosensor. Eur Food Res Technol 2004 Sep; 219 (4):

416-20.

17

Capítulo 1

Métodos eletroquímicos

Towards a reliable technology for antioxidant capacity and oxidative damage evaluation:

electrochemical (bio)sensors

M. Fátima Barroso, N. de-los-Santos-Álvarez, C. Delerue-Matos, M. B. P. P. Oliveira

Biosensors and Bioelectronics (submetido)

1. Métodos eletroquímicos

19

Towards a reliable technology for antioxidant capac ity and

oxidative damage evaluation: electrochemical (bio)s ensors

M. Fátima Barrosoa,b, N. de-los-Santos-Álvarezc, C. Delerue-Matosa, M. B. P. P. Oliveirab aREQUIMTE/Instituto Superior de Engenharia do Porto, Dr. Bernardino de Almeida 431,

4200-072 Porto, Portugal bRequimte, Serviço de Bromatologia, Faculdade de Farmácia, Universidade do Porto, R.

Aníbal Cunha n. 164, 4050-047 Porto, Portugal cDepartamento de Química Física y Analítica, Universidad de Oviedo, Julián Clavería 8,

33006 Oviedo, Spain

Abstract

To counteract and prevent the deleterious effect of free radicals the living organisms

have developed complex endogenous and exogenous antioxidant systems. Several

analytical methodologies have been proposed in order to quantify antioxidants in food,

beverages and biological fluids. This paper revises the electroanalytical approaches

developed for the assessment of the total or individual antioxidant capacity. Four

electrochemical sensing approaches have been identified, based on the direct

electrochemical detection of antioxidant at bare or chemically modified electrodes, and

using enzymatic and DNA-based biosensors.

Keywords: Antioxidants; free radicals; electrochemistry; chemically modified electrode;

enzymatic biosensors; DNA-based biosensor.

Disponível online em ww.sciencedirect.com

Bionsensor and Bioelectronics submitted


21

1. Introduction

The field of antioxidants has grown over the past decades invading a large number of

areas that affect food and health such as nutrition, biochemistry, pharmacology,

physiology, food processing and analytical chemistry. In general, researchers have been

focused on the i) study of the antioxidant pathway on free radical scavenging activity to

stop radical chain reactions; ii) study of the antioxidant protective role on proteins, lipids

and DNA against the free radical generated in vivo; iii) evaluation of the antioxidant profile

of natural foodstuffs and beverages because of their increasing appreciation; iv) synthesis

of antioxidants and development of novel artificially antioxidant-enriched food products; v)

study of the antioxidant efficacy on the preservation of foodstuff against deterioration

carried out by free radicals produced by food; vi) development of analytical methods for

the evaluation of the antioxidant content in food, beverages and biological samples.

The definition of antioxidant has been adjusted over the time (Halliwell and Gutteridge,

1990; Krinkly, 1992) and recently Laguerre et al. (2010) defined a biological antioxidant as

a substance that, when present at low concentrations compared to an oxidizable

substrate, protects (by itself and through its oxidation products) that substrate from

oxidation, and ultimately protects the organism from harmful effects of oxidative stress. An

imbalance between the generation of oxidants, either free radical or non-free radicals, and

the antioxidant system is known to cause oxidative stress, which is frequently associated

with many complex diseases (cardiovascular diseases, inflammatory disorders and

cancer) (Laguerre et al., 2010). Oxidation of key biological molecules (e.g. proteins,

carbohydrates, lipids and nucleic acids) is usually the mechanism that triggers these

pathologies along with modulation of gene expression and the inflammatory response

(Laguerre et al., 2007). Fortunately, nature has developed complex endogenous and

exogenous antioxidant systems to counteract and prevent the deleterious effect of

oxidants and minimize the oxidative stress in most living beings. Endogenous antioxidants

include enzymes that have the ability to promote an efficient repair of oxidative damaged

sites on macromolecules such as DNA. In general, hydrophilic antioxidants react with

oxidant compounds in the cell and blood plasma while hydrophobic antioxidants protect

cell membranes from lipid peroxidation (Barroso et al., 2011a). Fig. 1 sumarizes the

antioxidant defence system.


22

Fig. 1. Antioxidant defence system

Foodstuffs constitute an excellent exogenous source of natural antioxidants. Although

further and more conclusive studies must be done to establish the precise correlation

between the intake of antioxidants and the reduction in the oxidative stress level, it seems

reasonable that increasing intake of dietary antioxidants may help to maintain an

adequate antioxidant status and, so, the normal physiological functions of a living system.

Some functional foods, vegetables, fruits, whole-grain cereals, juice, tea, wine are good

sources of exogenous antioxidants (Almajano et al. 2008; Frankel, 2007; Huang et al.

2009; Lee et al. 2009).

Oxidant compounds (radicals and non-radicals) can be generated as a consequence of

normal aerobic metabolism, and are able to induce damage to the cells by reacting with

biomolecules (proteins, lipids, among others) and causing serious lesions on DNA.

Radical species such as nitric oxide (NO•), superoxide (O2•-), hydroxyl (OH•), peroxyl

(ROO•) and alkoxyl (RO•) radicals, are known as reactive oxygen species (ROS) and

reactive nitrogen species (RNS) produced during the mitochondrial respiratory chain

(electron transport) or during the lipid oxidation chain reaction. Hypochlorous acid (HClO)

obtained by the enzymatic system myeloperoxidase/H2O2/Cl- is another free radicals

involved. All of them tend to donate or take another electron to attain stability, which

confers their characteristic high reactivity (Halliwell and Gutteridge, 1999; Rice-Evans C.A.

antioxidant defence system

endogenous exogenous(dietary source)ascorbic acid phenolic acidtocopherolcarotenesterpenes

enzymaticsuperoxide dismutase (SOD)catalaseperoxidasemyeloperoxidaseCofactorCu/Zn SOD (cytoplasm) Mn/SOD (mitochondria)

non-enzymaticglutathionecarnosinehistidineuric acid

repair mechanism

DNA-repair enzymesprevention mechanism

inhibition of radical generation (scavenger activity)stop radical chain reaction

stabilization of biological sites

antioxidant defence system

endogenous exogenous(dietary source)ascorbic acid phenolic acidtocopherolcarotenesterpenes

enzymaticsuperoxide dismutase (SOD)catalaseperoxidasemyeloperoxidaseCofactorCu/Zn SOD (cytoplasm) Mn/SOD (mitochondria)

non-enzymaticglutathionecarnosinehistidineuric acid

repair mechanism

DNA-repair enzymesprevention mechanism

inhibition of radical generation (scavenger activity)stop radical chain reaction

stabilization of biological sites


23

and Burdon, 1994). The group of non-radical compounds contains a large variety of

substances, some of which are extremely reactive. Among these compounds produced in

high concentration in the living cell are the hypochlorous acid (HClO) and the hydrogen

peroxide (H2O2) (Kohen and Nyska, 2002).

Several analytical methods have been developed in order to measure the total or

individual antioxidant capacity of foodstuff, beverages, processed food and biological

samples. The quantification of the individual antioxidants present in samples can be

performed by chromatographic techniques and using several extraction techniques, like

solid-phase microextraction (SPME), supercritical fluid extraction (SFE) and microwave-

assisted extraction (MAE) (Abidi, 2000; El-Agamey et al., 2004; Valls et al., 2009).

However, considering the complexity of the composition of foods and biological samples

separating each antioxidant compound and studying it individually is costly and inefficient,

notwithstanding the possible synergistic interactions among the antioxidant compounds in

a sample. This is the case of the formation of highly active antioxidant compounds from

less active promoters or positive interactions between two antioxidants with different

action pathway. Therefore, it is very appealing for researchers to have methods for the

quick quantification of antioxidant effectiveness (Huang et al., 2005).

The conventional analytical methods used for the quantification of the total antioxidant

capacity can be divided in two major mechanisms, based on the chemical reaction

involved: Hydrogen atom transfer (HAT) and single electron transfer (SET). HAT-based

methods measure the ability of an antioxidant to quench radicals by hydrogen donation,

while SET-based methods measure the ability of an antioxidant to transfer one electron to

reduce any compound, including metals, carbonyls and radicals (Huang et al., 2005; Prior

et al., 2005). SET and HAT mechanisms almost always occur together, with the balance

determined by antioxidant structure and pH (Prior et al., 2005). The assessment of

antioxidant capacity using the HAT reactions mechanisms can be performed by several

techniques, such as the oxygen radical absorbance capacity (ORAC), total radical-

trapping antioxidant parameter (TRAP), total oxidant scavenging capacity (TOSC), β-

carotene bleaching by ROO•, and low-density lipoprotein (LDL) oxidation. On the other

hand, the SET reaction mechanisms can be used in various analytical methods, like ferric

reducing antioxidant power (FRAP), trolox equivalent antioxidant capacity assay (TEAC),

DPPH-based assay (2,2-diphenyl-1-picrylhydrazyl), total phenolic assay by Folin-

Ciocalteu (FC). These methods present some disadvantages such as long analysis time,

pre-treatment of coloured samples, expensive equipment requirements, and correction for

interfering substances. These analytical methods and the reactions behind them have

been very well described previously (Sánchez-Moreno, (2002) and Prior et al., (2005)).


24

The use of electrochemical devices for the assessment of the total antioxidant capacity

in samples is a good alternative to the optical methods. In general, they do not require

sophisticated and expensive equipment, are very sensitive and easy to miniaturize, which

pave the way to portability. In the particular case of antioxidant assessment, some

antioxidant capacity assays are based on electron transfer reactions that occur in food

samples, which can be easily monitored by electrochemical techniques. Additionally,

natural antioxidants included in the diet usually exhibit a non negligible native

electroactivity that can be exploited for detection. In that sense, the electrochemical

detection of the antioxidant is a direct measure of the total reducing power of the

antioxidant compounds present in samples, without the use of reactive species (Blasco et

al., 2007).

This paper highlights the most recent electroanalytical methodologies (since 2007) for

detection of individual antioxidants or the total antioxidant capacity in several kinds of

matrices. Four electrochemical approaches for this purpose are described. The simplest

one is based on the direct electrochemical evaluation of antioxidant on bare electrode

surfaces. The other three strategies rely on modification of the electrode surfaces. A

distinction between chemically and biologically modified electrodes is herein established

mainly because the difference in the principle of measurement. Among the latter group,

enzyme-based and DNA-based sensors are revised separately.

2. Direct electrochemical detection of antioxidants

In the last twenty years, a great number of electrochemical approaches have been

developed for the direct determination of the total or individual antioxidants present in

beverages/food. For the total antioxidant content, cyclic voltammetry (CV), is the most

widely used electrochemical technique although differential pulse voltammetry (DPV) and

square wave voltammetry (SWV) are advantageous in terms of sensitivity for individual

detection.

As reducing agents, antioxidants tend to be easily oxidized at bare electrodes such as

glassy carbon (GCE) (Intarakamhang et al., 2011) and Pt (Pisoschi et al., 2009)

electrodes (Fig. 3A). In general, this technique allows the evaluation of the antioxidant

capacity by measuring the charge associated to the oxidation process of the antioxidants

present in the sample, that is, measuring the area under the corresponding anodic waves.

It is worth noting that this method of estimating the antioxidant capacity assumes that the

reducing power is an absolute measure of the antioxidant capacity, which is open to

discussion (Cheng and Li, 2004). The antioxidant capacity is actually a wider concept

including radical scavenging efficiency, metal chelating capacity and oxidative enzyme


25

inhibition capacity (Huang et al, 2005). In spite of the complexity of food samples that

frequently contain more than one antioxidant compound, the potential at which the

oxidation takes place enables the identification of the type of antioxidant involved

(oxidizable functional group). Additionally, the peak potential (Ep) is an indication of the

reducing power because the higher the potential the more difficult the oxidation is

indicating a less reducing power. Total phenolic content can also be electrochemically

evaluated.

It has been suggested that isoflavonoids are more advantageous for human health than

flavonoids because of their cardiovascular protective properties as well as antioxidant

activity. Consistenly with this statement isoflavonoid genistein presented a lower Ep than

flavonoid apigenin. This higher antioxidant activity was confirmed by TEAC assay (Han et

al., 2009).

As mentioned in the introduction, the antioxidant capacity can be estimated by a large

number of methods. However, the values obtained only reflect the chemical reactivity

under the experimental conditions of the assay. For this reason it is not unexpected that

polyphenolic compounds showed different antioxidant capacities depending on the

spectrophotometric assay used. CV provides helpful information to understand this

behaviour. Two flavonols, kaempferol and morin that differ in a single hydroxyl group

underwent a first oxidation at 4′-OH in their B-ring. However, the Ep of kaemperol was 80

mV less positive than the morin one, which clearly explains its higher antioxidant capacity

when exposed to oxidants unable to oxidize the additional 2’-OH that is only present in

morin. On the contrary, under stronger oxidants conditions, morin exhibited almost twice

antioxidant capacity because two oxidation processes took place (He et al., 2009).

Piljac-Žegarac (2010) used CV to study the electrochemical properties of antioxidants

present in fruit tea infusions as well as to estimate the antioxidant capacity. Three

compounds were identified through their redox processes. The easiest oxidized

compound (Ep=130 mV) was ascribed to the oxidation of the ene-diol of ascorbic acid.

The low magnitude of its current indicated a very low concentration in tea infusions, which

maybe a consequence of its thermal instability in the boiling water used for preparation. A

quasi-reversible redox process (Eº’=395 mV) was attributed to the oxidation of the ortho-

dihydroxy-phenol and gallate group of low formal potential phenolics because the resulting

product is a stable quinone, which was further reduced in the backward scan. The less

oxidizable compound presented an irreversible oxidation process between 670 and 700

mV, which was ascribed to the oxidation of the monophenol group or the meta-diphenols

on the A-ring of flavonoids that led to a phenoxy radical or a phenoxonium ion undergoing

successively secondary reactions. The antioxidant capacity of these fruit teas was

determined by estimating the integrated area under the peak up to 600 mV.


26

Fig. 3 . Principle of operation for the four electrochemical sensing approaches described for

antioxidant quantification

The CV profile of flavonoids (quercetin and its glucosides) in onions was also studied

(Zielinska et al., 2008). The registered cyclic voltammogram showed up to four anodic

waves that were associated to quercetin, quercetin-3,4’-diO-β-glucoside, quercetin-3-O-β-

glucoside and quercetin-4’-O-β-glucoside, from most to less oxidizable compounds. The

data provided by CV were confirmed by HPLC-UV-MS. The quantitative analysis in terms

of total antioxidant capacity was carried out by measuring the total charge (area under the

anodic waves). This value was referred to the charge corresponding to the anodic wave of

increasing concentrations of trolox standards. This method can be considered an

alternative method for the quercetin glucosides quality control without sample

pretreatment. Comparison with other spectrophotometric methods was also discussed.

The CV method was also applied to the evaluation of the antioxidant capacity of dark- and

light-grown buckwheat (Zielinska et al., 2007). The broad anodic wave observed indicated

the presence of several antioxidants such as flavonoids, phenolic acids and water-soluble

vitamins (B1, B6, C).

Native DNA signal

Damaged DNA signal

DNA signal (antioxidant andfree radical)

e-

O2Antioxidant

(Antioxidant)ox

TyrLac

H2O

e-

O2Antioxidant

(Antioxidant)ox

TyrLac

H2O(Med)rd

(Med)ox

e-Antioxidant

(Antioxidant)ox

Direct oxidation on bare electrodes

Modified electrodes

CME

enzymatic

DNA

(A)

(B)

(C)

(D)

(E)

e-

(Med)rd

(Med)oxAntioxidant

(Antioxidant)ox

Native DNA signal

Damaged DNA signal


Native DNA signal

Damaged DNA signal


Native DNA signal

Damaged DNA signal


e-

O2Antioxidant

(Antioxidant)ox

TyrLac

H2O

e-

O2Antioxidant

(Antioxidant)ox

TyrLac

H2O(Med)rd

(Med)ox

e-Antioxidant

(Antioxidant)ox


Modified electrodes

CME

enzymatic

DNA

(A)

(B)

(C)

(D)

(E)

e-

(Med)rd

(Med)oxAntioxidant

(Antioxidant)ox

e-

O2Antioxidant

(Antioxidant)ox

TyrLac

H2O

e-

O2Antioxidant

(Antioxidant)ox

TyrLacTyrLac

H2O

e-

O2Antioxidant

(Antioxidant)ox

TyrLac

H2O(Med)rd

(Med)ox

e-

O2Antioxidant

(Antioxidant)ox

TyrLacTyrLac

H2O(Med)rd

(Med)ox

e-Antioxidant

(Antioxidant)ox

e-Antioxidant

(Antioxidant)ox


Modified electrodes

CME

enzymatic

DNA

(A)

(B)

(C)

(D)

(E)

e-

(Med)rd

(Med)oxAntioxidant

(Antioxidant)ox

e-

(Med)rd

(Med)oxAntioxidant

(Antioxidant)ox


27

(Makhotkina and Kilmartin, 2010) proposed several analysis in wines. It is known that

polyphenol catalyzed the oxidation of sulfur dioxide. Cyclic voltammograms of SO2-

containing wines showed a catalytic current due to the polyphenol-mediated oxidation of

SO2. Addition. of acetaldehyde suppressed this current because actetaldehyde interacts

with SO2 in solution, reducing its availability to be catalytically oxidized. So, the content of

free sulfur dioxide can be estimated. The simplicity and quickness of this approach make

it a promising alternative method to the classic titration procedure. The total phenolic

content and the flavonol level was estimated by measuring the area under the CV up to

1.2 V and the peak current at 1.12 V, respectively. The current intensity of the anodic

wave at about 450 mV allowed the estimation of catechol and galloyl-containing

compounds content. In all cases much better correlations were obtained for white wines

than for red ones. This was attributed to the more complex composition of red wines that

can not be compared to a mixture of a limited number of model compounds as the

standards used for calibration.

Burratti et al., (2008) developed a renewable, low-cost flow injection analysis (FIA)

system composed of a pencil (graphite) working electrode and a adapted PVC Pasteur

pipette as a wall-jet flow cell, for the determination of reducing power and total phenolic

content of several types of tea infusions. These inexpensive electrodes have been

previously used for other purposes and showed excellent electrochemical features such

as high sensitivity, low background current, wide potential window and chemical inertness.

The reducing power and the total phenolic content were quantified by measuring the

oxidation current at +0.5 V (from the most easily oxidized compounds) and +0.8 V (all

polyphenols), respectively. The total phenolic content was correlated to FC phenolic index

and the reducing power to the DPPH assay.

Pisoschi et al. developed a chronobiamperometric method for the detection of natural

juices. The radical DPPH• can interact with antioxidants leading to DPPH. The analytical

signal was the differential current measured at two Pt electrodes polarized at a small

potential difference in the presence of the reversible redox couple DPPH•/DPPH. While

the reduction took place in a Pt electrode the oxidation is monitored in the other one. In

the presence of antioxidants the differential current increased as a consequence of the

consumption of radical (Pisoschi et al., 2009).

All these direct electrochemical methods described are rapid, reliable, easy to carry out

and in general, potentials below 700 mV are applied, which improves the selectivity. This

is consistent with the low oxidation potential commonly found in food and biological

samples informing about their high antioxidant capacity (Blasco et al., 2007). Table 1

summarizes the approaches reported for the direct electrochemical evaluation of the

antioxidant status.

I. R

evis

ão d

o es

tado

da

arte

28

Tab

le 1

. M

ain

anal

ytic

al fe

atur

es o

f dire

ct e

lect

roch

emic

al d

etec

tion

of a

ntio

xida

nts

Ant

ioxi

dant

S

ampl

e V

olta

mm

etric

tech

niqu

e (e

lect

rode

) Li

near

ran

ge

Lim

it of

det

ectio

n R

efer

ence

Asc

orbi

c ac

id

Tab

lets

, fru

it ju

ice,

he

rbal

tea

extr

act

DP

V (

GC

E)

0.1-

8.0

mM

50

µM

In

tara

kam

hang

et a

l., 2

011

Cat

echi

n, q

uerc

etin

, rut

in

- C

V, D

PV

, SW

V (

GC

E)

- -

Med

vido

vić-

Kos

anov

ić e

t al

., 20

10

Isor

ham

netin

T

able

ts

CV

, DP

V (

GC

E)

10 -

100

nM

5.0

nM

Liu

et a

l., 2

008

Ant

ioxi

dant

cap

acity

F

ruit

tea

infu

sion

C

V (

GC

E)

- -

Pilj

ac-Ž

egar

ac e

t al,

2010

T

otal

pol

yphe

nols

F

ree

SO

2 W

hite

and

red

win

e C

V (

GC

E)

- -

Mak

hotk

ina

and

Kilm

artin

20

10

Tot

al o

xida

nt-s

cave

ngin

g ca

paci

ties

B

lood

pla

sma

CV

(G

CE

) -

- K

oren

et a

l., 2

009

Tot

al a

ntio

xida

nt c

apac

ity

Fru

it ju

ice,

sof

t drin

k C

V (

Pt)

-

76.8

nM

P

isos

chi e

t al.,

200

9 2-

Sty

rylc

hrom

ones

(f

lavo

ne)

- C

V (

GC

E)

- -

Gom

es e

t al.,

200

8

Que

rcet

in

Oni

ons

CV

(G

CE

) -

- Z

ieliń

ska

et a

l., 2

008

Toc

ophe

rol

- C

V (

GC

E)

- -

Yao

et a

l., 2

008

Pom

iferin

, Is

opom

iferin

, O

sajin

C

atal

posi

de

- S

WV

(C

PE

) 0.

1-10

0 ng

mL-1

1–12

ng

mL-1

0.

1-1

ng m

L-1

0.1-

1 ng

mL-1

50 p

g m

l-1

800

pg m

l-1

40 p

g m

l-1

10 n

g m

l-1

Dio

pan

e ta

l., 2

008

Pol

yphe

nols

W

ines

(G

CE

) -

- F

ell e

t al.,

200

7 ph

enol

ic a

cids

-

CV

(G

CE

) -

- S

imić

et a

l., 2

007

α-Li

poic

aci

d D

ieta

ry s

uppl

emen

ts

CV

; DP

V; S

WV

2.

5-75

µM

1.

8 µ

M

Cor

dune

anu

et a

l., 2

007


29

3. Chemically modified electrodes for detection of antioxidants

The finding that rational modification of a conductive substrate can lead to electrode

surfaces possessing not only the properties of the substrate but also those of the

immobilized compound has paved the way to the development of chemically modified

electrodes (CMEs). Since then, the field of CMEs is an area of growing interest because

of their convenience for a great variety of fundamental studies and applications.

IUPAC defines a CME as an electrode made of a conducting or semiconducting material

that is coated with a selected monomolecular, multimolecular, ionic, or polymeric film of a

chemical modifier and that by means of faradaic (charge-transfer) reactions or interfacial

potential differences (no net charge transfer) exhibits chemical, electrochemical, and/or

optical properties of the film (Durst et al., 1997). The substrate is the platform on the

modifier is assembled and it is selected among the electrode materials available according

to the desired properties, e.g., mechanical and chemical stability, resistance to fouling etc.

The modifying layer can be fixed at the electrode surface by different techniques: i)

physical adsorption; ii) formation of organized monolayers (self-assembly) iii)

electropolymerization; iv) covalent attachment; v) entrapment in a polymer or inorganic

film; and vi) incorporation into an electrode matrix. The discussion about the preparation

and properties of the different layers is out of the scope of this revision but the general aim

is to obtain enhanced properties such as faster electron transfer, increased current and

elimination of interferences. The detection of antioxidants can also benefit from these

modifications as discussed below.

Ascorbic acid (AA) is electroactive and can be directly detected on bare electrodes.

However, the presence of interfering compounds that oxidize at similar potentials such as

dopamine (DA), is an important drawback. Immobilization of cationic surfactants on the

electrode surface made more difficult the oxidation of positively charged DA because of

electrostatic repulsion. Therefore, the Ep shifted to more positive potentials improving the

peak separation up to 200 mV using docosyltrimethylammonium chloride adsorbed on

GCE (Luo et al., 2010).

Adsorption is the simplest immobilization method for carbon nanotubes. This twenty-

year old material is an excellent promoter of the electron transfer and electrocatalytic

activity, so it is not rare that has been tested for the detection of uncountable molecules

including antioxidants. Not only single-walled carbon nanotubes (SWCNTs) increased the

electrochemical area of the GCE twice but also the electron transfer was accelerated and

the current of catechin increased by a factor of 10 due to their high adsorption capacity.

Consequently, subnM concentrations were detected in combination with DPV, which is

more sensitive than CV (Yang et al., 2009). Other carbon materials such as graphene


30

nanosheets (Du et al., 2010) and acetylene black nanoparticles (Song et al., 2010) were

used for rutin determination taking advantage of their high number of adsorption active

sites. A technique directly based on the enhanced sensitivity associated to accumulation

processes is adsorptive stripping voltammetry (AdSV). Using a lead film plated on GCE,

rutin was accumulated and further oxidized by SWV allowing the detection of subnM

concentrations. The main drawback was the toxicity of the electrode material (Tyszczuk,

2009).

In many cases the molecule immobilized is a redox mediator that shuttles electrons from

the antioxidant to the electrode surface reducing the potential at which the oxidation of the

antioxidant occurs. This is advantageous in order to eliminate interferences. The mode of

action is depicted in Fig. 3B. The oxidized form of the immobilized mediator is able to

oxidize the antioxidant chemically in the electrode-solution interface. The reduced

mediator is then electrochemically oxidized on the electrode surface, which generates

more oxidized mediator available for oxidizing the antioxidant (catalytic cycle). As a result

an increased anodic current is observed (catalytic current) increasing the sensitivity of the

determination.

Films of mediators can be electrogenerated on the electrode substrate. This is the case

of o-aminophenol that was electrografted on GCE and effectively catalyzed the oxidation

of AA in the presence of organic hydroxyacids and sugars commonly found in fruit juices.

The sensor was successfully applied to AA detection in commercial fruit juices (Civit et al.,

2008). Electropolymerization of aspartic acid (Wang et al. 2010) or glutamic acid (Santos

et al., 2007b) also resulted in higher oxidation currents of catechin and rutin, respectively

because of increased adsorption and kinetics, which allowed the analysis of catechin in

several types of teas with good recoveries and rutin in pharmaceutical formulations. Poly

(3-(3-Pyridyl) acrylic acid (Zhang et al., 2007b), polycaffeic acid (Li et al., 2008) and p-

aminobenzene sulfonic acid doped polyaniline (Zhang et al., 2008b) films also promote

the peak separation of AA and DA.

Carbon pastes are easily modified by incorporation of modifiers into the own material

(graphite powder and a binder, e.g. parafilm oil, nujol, silicon). This method of

immobilization is simple, fast and the resulting electrode surface is easily renewable by

polishing or discarding after measurement and immediately replacing by a new one from

the large amount previously prepared.

Ionic liquids can totally or partially replace the classical binders in carbon pastes

because they exhibit high ionic conductivity and a wide potential window. In spite of the

dramatic increase in capacitive current, the concomitant increase in faradaic current

allowed the determination of rutin, although the detectability was highly dependent on the

ionic liquid selected (Sun et al., 2008; Zhang and Zheng, 2008).


31

Cobalt 5-nitrosalophen, a well known mediator of the oxidation of AA, was introduced

into carbon paste along with a cationic surfactant to minimize the interference of DA by

peak separation (about 374 mV) (Shahrokhian and Zare-Mehrjardi, 2007b). The limit of

detection was improved in comparison with the docosyltrimethylammonium chloride GCE

(Luo et al., 2010) probably due to the presence of the mediator. A better peak resolution

(395 mV) and an order of magnitude lower limit of detection were achieved by preparing a

MWCNTs-thionine-nafion carbon composite (Shahrokhian and Zare-Mehrjardi, 2007a).

This material without nafion also exhibited a good discrimination ratio towards other

interfering drugs, a very useful feature for AA determination in human blood plasma

(Shahrokhian and Asadian, 2010).

Micromolar concentrations of rutin cannot be detected on bare CPE. However, after

addition of poly(vinylpyrrolidone) to the carbon matrix, a significant peak was observed.

This is due to the strong adsorption of phenolic compounds on this polymer. In spite of the

high oxidation potential, the CME did not show interferences in pharmaceutical

formulations (Franzoi et al., 2008).

Fatibello-Filho’s group has exploited the catalytic properties of copper immobilized on

polyester resin incorporated into carbon matrix to detect catechin in teas (Freitas and

Fatibello, 2010a), rutin in pharmaceutical preparations (Freitas et al., 2009) and the

synthetic phenolic antioxidants butylated hydroxyanisole (BHA) and butylated

hydroxytoluene (BHT) (Freitas and Fatibello, 2010b). These phenolic compounds have to

be controlled in food samples because it is believed to cause nutrient loss and have

potential toxic effects. In spite of the high detection potential, the method showed good

recoveries in mayonnaise samples and compared well with the most commonly used

HPLC method. BHA was also monitored on films of nickel hexacyanoferrate (NiHCF)

generated on p-phenylendiamine modified carbon paste. The electrocatalytic effect of

NiHCF allowed the detection of BHA at relatively low potentials, 400 mV, in a flow

injection system with excellent reproducibility and stability. The interference of more easily

oxidized water-soluble compounds such as AA was eliminated during the sample

pretreatment but the signal from propyl gallate could not be prevented (Prabakar and

Narayanan, 2010).

Molecular recognition is another strategy to improve the selectivity of the detection by

preconcentration on specific receptors. An example of this approach is the ability of

cyclodextrins to form inclusion complexes with organic and inorganic compounds

depending on the size of their cavity that was applied to the selectively detection of

catechin in different beverages. Slightly higher sensitivities were achieved by SWV than

DPV (El-Hady, 2007; El-Hady and El-Maali, 2008).


32

A completely different approach is based on the correlation between the antioxidant

properties of flavonoids and other phenolic compounds and their ability to promote the

enlargement of AuNPs (Wang et al., 2007) or Au nanoshells (Ma and Qian, 2010). The

Au seeds were immobilized on SAMs of cysteamine or on aminopropyltriethoxysilane-

modified ITO electrodes in the formed of SiO2-Au nanocomposite, respectively. Equimolar

concentrations of ferrycianide/ferrocyanide or ferricyanide alone were used as probes in

solution to test the growth of AuNPs. In the presence of flavonoids the redox process of

the probe clearly diminished indicating that its access to the electrode surface is blocked.

This was attributed to the growth of the Au seeds induced by reducing activity of

flavonoids or phenolic compounds. This resistance to the electron transfer linearly

correlated with the concentration of antioxidant in different ranges, which allowed

comparing the antioxidant capacity of several pure compounds. Since this methodology

cannot distinguish the origin of the antioxidant power in mixtures, it was used for the

assessment of the total antioxidant capacity of herbal extracts (Wang et al., 2007).

Self-assembly of thiolated compounds is a spontaneous process on Au surfaces that has

become very common for immobilization of molecules. Recently, a nickel (II) complex was

prepared on the surface of a mercaptopropanoic acid monolayer. This complex was able

to oxidize catechin on the interface, which was subsequently reduced on the electrode

surface. Results from SWV compared well with those from capillary electrophoresis

(Moccelini et al., 2009). In table 2 the main characteristics of the CMEs applied to

antioxidant detection is summarized.

4. Enzymatic biosensors for antioxidant evaluation

Electrochemical biosensors, a subclass of chemical sensors, combine the excellent

detectability of electrochemical transducers with the high selectivity of biological

recognition elements such as enzymes, proteins, antibodies, nucleic acids, etc. Since the

principle of measurement is different, in this section, biosensors employing enzymes for

antioxidant detection are revised while the DNA-based biosensors are discussed in

section 5.

33

Tab

le 2

. C

hem

ical

ly m

odifi

ed e

lect

rode

dev

ices

for

the

antio

xida

nt m

easu

rem

ent.

Ant

ioxi

dant

S

ampl

e Im

mob

iliza

tion

proc

edur

e (e

lect

rode

) T

echn

ique

Li

near

ran

ge

Lim

it of

det

ectio

n R

efer

ence

Asc

orbi

c ac

id

Inje

ctio

n D

ocos

yltr

imet

hyla

mm

oniu

m c

hlor

ide

(GC

E)

DP

V

0.01

-1.0

mM

4.

0 µ

M

Luo

et a

l, 20

10

Cat

echi

n -

SW

CN

Ts-

CT

AB

(G

CE

) D

PV

0.

372–

2.38

nM

0.

112

nM

Yan

g et

al.,

200

9 R

utin

T

able

ts

Gra

phen

e na

nosh

eets

(G

CE

)

CV

0.

1-10

µM

21

nM

D

u et

al.,

201

0 R

utin

ch

ines

e m

edic

ines

ac

etyl

ene

blac

k na

nopa

rtic

les

(GC

E)

DP

V

20-5

0 µ

g L-1

10

µg

L-1

Son

g et

al.,

201

0 R

utin

T

able

ts

Pb

film

(G

CE

) A

dSV

0.

5 -1

0 nM

0.

25 n

M

Tys

zczu

k, 2

009

AA

F

ruit

juic

es

o-am

inop

heno

l film

(G

CE

) am

pero

met

ry

2-20

µM

0.

86 µ

M

Civ

it et

al.,

200

8 C

atec

hin

Tea

bev

erag

e P

oly-

aspa

rtic

aci

d fil

m (

GC

E)

D

PV

0.

25-3

0 µ

M

72 n

M

Wan

g an

d F

an, 2

010

Rut

in

Pha

rmac

eutic

al

form

ulat

ion

Pol

yglu

tam

ic a

cid

(GC

E)

SW

V

0.7-

10 µ

M

- S

anto

s et

al.

2007

Asc

orbi

c ac

id

- P

oly

(3-(

3-P

yrid

yl)

Acr

ylic

(G

CE

) C

V

10-4

00 µ

M

0.8

µM

Z

hang

, et a

l., 2

007b

A

scor

bic

acid

P

harm

aceu

tical

s P

olyc

affe

ic a

cid

(G

CE

) C

V

0.2-

1.2

mM

9.

0 µ

M

Li e

t al.,

200

8 A

scor

bic

acid

T

able

ts, u

rine

p-am

inob

enze

ne s

ulfo

nic

acid

dop

ed p

olya

nilin

e (G

CE

) D

PV

35

-175

µM

7.

5 µM

Z

hang

et a

l., 2

008b

Rut

in

Tab

lets

io

nic

liqui

d (C

PE

) C

V

0.5-

100

µM

0.

35 µ

M

Sun

et a

l., 2

008

Rut

in

Tab

lets

, urin

e H

ydro

phili

c io

nic

liqui

d (C

PE

) S

WV

0.

04-1

0 µ

M

10 n

M

Zha

ng a

nd Z

heng

, 200

8 A

scor

bic

acid

-

MW

CN

TS

/Naf

ion-

coba

lt(II)

-nitr

osal

ophe

n (C

PE

) D

PV

0.

5-10

0 µ

M

0.1

µM

S

hahr

okhi

an a

nd Z

are-

Meh

rjard

i, 20

07b

Asc

orbi

c ac

id

- T

hion

ine-

nafio

n-M

WC

NT

s (C

PE

)

DP

V

0.1-

80.0

µM

0.

08 µ

M

Sha

hrok

hian

and

Zar

e-M

ehrja

rdi,

2007

a

Asc

orbi

c ac

id

Hum

an b

lood

se

rum

T

hion

ine-

nafio

n-M

WC

NT

s (C

PE

)

DP

V

1-10

0 µ

M

0.8

µM

S

hahr

okhi

ana

and

Asa

dian

a 20

10

Asc

orbi

c ac

id

Pha

rmac

eutic

al

prep

arat

ion

Cob

alt(

II)-n

itros

alop

hen-

tetr

aoct

ylam

mon

ium

bro

mid

e (C

PE

) D

PV

1-

100

µM

0.7

µM

S

hahr

okhi

an a

nd Z

are-

Meh

rjard

i, 20

07c

Rut

in

Pha

rmac

eutic

al

form

ulat

ions

po

ly(v

inyl

pyrr

olid

one

(CP

E)

LSV

0.

39 -

13.0

µM

0.15

µM

F

ranz

oi e

t al.,

200

8

Cat

echi

n T

eas

Cu

(II)

- po

lyes

ter

resi

n (C

PE

)

SW

V

0.09

9-1.

2 µ

M

58 n

M

Fre

itas

and

Fat

ibel

lo-

Filh

o 20

10a

BH

A/B

HT

F

ood

sam

ples

C

u (I

I) p

olye

ster

res

in (

CP

E)

S

WV

0.

34-4

1.0

µM

(bot

h)

72 n

M, 9

3 nM

F

reita

s F

atib

ello

-Filh

o,

2010

b

BH

A

Spi

ked

pota

to c

hips

N

iHC

F-

(gra

phite

com

posi

te)

FIA

, DP

V

1.2-

1070

µM

0.

6 µ

M

Pra

baka

r et

al.

2010

C

atec

hin

T

ea

Bev

erag

es

Inco

rpor

atio

n of

β-

cycl

odex

trin

(C

PE

)

SW

V

SW

V

DP

V

Up

to 7

0 µ

g m

L-1

0.00

1-7.

2 µ

g m

L-1

0.00

2-4.

2 µ

g m

L-1

1.35

µg

mL-1

0.12

ng

mL-1

0.

30 n

g m

L-1

El-H

ady

and

El-M

aali

2008

E

l-Had

y 20

07

34

Ant

ioxi

dant

S

ampl

e Im

mob

iliza

tion

proc

edur

e (e

lect

rode

) T

echn

ique

Li

near

ran

ge

Lim

it of

det

ectio

n R

efer

ence

Que

rcet

in

daiz

eol

puer

arin

Her

bal e

xtra

cts

Au/

cyst

eam

ine/

AuN

Ps

C

V

10-1

00 µ

M

0.1-

10 µ

M

0.5-

1 µ

M

1 nM

10

nM

10

0 nM

Wan

g et

al.,

200

7a

Siri

ngic

aci

d -

Gol

d na

nosh

ells

(IT

O)

CV

5-

100

µM

-

Ma

and

Qia

n, 2

010

Cat

echi

n G

reen

tea

Ni(I

I) c

ompl

ex o

n S

AM

S

WV

3.

31-2

5.3

µM

0.

826

µM

M

occe

lin e

t al.,

200

9 A

scor

bic

acid

F

ood

Film

of b

inuc

lear

Cu

com

plex

(G

CE

) D

PV

5.

0-16

0.0

µM

2.

8 µ

M

Wan

g et

al.,

200

7b

Rut

in

Tab

lets

P

oly(

p-am

inob

enze

ne s

ulfo

nic

acid

) (

GC

E)

DP

V

0.25

- 10

.0 µ

M

0.1 µ

M

Che

n et

al.,

201

0

LSV

- lin

ear

swee

p vo

ltam

met

ry.


35

The enzyme is the most critical component of the enzyme electrode since it provides the

selectivity for the sensor and catalyzes the formation of an electroactive product for

detection. Recovery of enzymes from solution is, in general, very expensive, so their

immobilization on the transductor allows their reusability, which contributes to a reduction

in the analysis cost. This fact along with the feasibility of obtaining enzyme preparations of

high purity at a reasonable price has fuelled the construction of enzyme electrodes. It is

expected that this trend will continue because of their relative simplicity and rapid

analysis. For antioxidant detection oxidase enzymes such as polyphenol oxidase (PPO)

(Tan et al., 2011), tyrosinase (Cortina-Puig et al., 2010; El Kaoutit et al., 2007; Singh,

2011; Wang and Hasebe, 2011), laccase (El Kaoutit et al., 2008, Di fusco et al., 2010;

Ibarra-Escutia et al., 2010; Tan et al. 2009), horseradish peroxidise (HRP) (Santos et al.,

2007), ascorbate oxidase (Chauhan et al., 2010; Pisoschi, et al., 2010; Vig et al., 2010;

Wang et al., 2008) or uricase (Moraes et al., 2007; Zhang et al., 2007) are used. These

redox enzymes are very convenient for electrochemical detection because the product

monitored usually exhibits a redox process at low potentials, where the electrochemical

interferences are minimized.

Immobilization of enzymes on the electrode surface is a critical step for the successful

design of a biosensor. A great variety of approaches have been reported as it is summed

up in table 3. All of them try to obtain the highest enzyme loading without losing activity,

minimizing its leaching from the electrode surface to the bulk solution and inactivation of

the layer by radicals generated in the enzymatic reactions or polymerization of the product

formed as in the case of polyphenols.

Antioxidant detection by means of enzymes mostly relies on the measurement of the

reduction current of the enzymatically oxidized antioxidant as schematically depicted in

Fig. 3C. Tyrosinase and laccase are Cu containing oxidases that catalyses the reduction

of oxygen to water in the presence of phenolic compounds. Most methods for phenolic

compounds cited in table 3 measures the reduction current of the corresponding quinone

formed. The use of a redox mediator along with the enzyme (Fig. 3D) was also proposed

for the detection of catechol on Pt at +0.65 (Akyilmaz et al., 2010). Gallic acid (Di Fusco et

al., 2010), caffeic acid (Cortina-Puig et al., 2010), catechol (Zejli et al., 2008) or rosmarinic

acid (Diaconu et al., 2010) among others have been selected as standards for the

assesment of phenolic content in food and environmental samples. HRP can also catalyze

the conversion of some phenolic compounds. This enzyme along with an adsorbed

mediator (methylene blue) improved the reduction current of catechol (Santos et al.,

2007a).

Measurement of oxygen consumption after enzymatic reaction with Clark electrodes is

another possibility reported for the detection of ascorbic (Pisoschi et al. 2010) or uric acids


36

(Zhang et al., 2007a). A combination of two enzymes, uricase and microperoxidase-11

allowed the detection of uric acid by monitoring the formation of H2O2 (Behera and Raj,

2007). To eliminate interferences, a flow injection system with differential measurement

was reported for AA. An enzyme cell was constructed with ascorbate oxidase immobilized

in a membrane and the electrochemical cell was placed separately. The sample was

injected twice, after the enzyme cell to measure the total oxidation current, and before the

enzyme cell to measure the decreased oxidation current due to the enzymatic depletion of

AA (Vig et al., 2010).

A different strategy is based on the scavenging activity of antioxidant against ROS.

Superoxide radical is enzymatically generated by xanthine oxidase (XOD) in the presence

of (hipo)xanthine. Superoxide dismutase causes the dismutation of the radical forming

H2O2, which can be reoxidized on the electrode surface. However, the high potential

needed, 0.65V at Pt electrodes (Campanella et al., 2009), limits the selectivity and practical

applicability. To solve this drawback, immobilization of cytochrome c that is oxidized by the

radical is advantageous. This compound is further reduced at the electrode surface. The

presence of antioxidants diminished the reduction current of cytochrome c by scavenging

the radical formed, resulting in signal off approaches (Cortina-Puig et al., 2009a; Cortina-

Puig et al., 2009b). Table 3 summarizes the more relevant analytical features of successful

applications of enzymatic biosensors for the evaluation of antioxidant status.

5. DNA biosensors for quantification of antioxidant s

Not only enzymes but also nucleic acids can be immobilized on electrochemical

transducers. DNA layers can act as biomolecular recognition elements for diagnostics of

genetic or infection diseases as well as the detection of pathogens in food and

environmental samples taking advantage of one of the most specific reactions known:

hybridization. Likewise the so-called aptamers (synthetic single stranded oligonucleotides)

can act as high affinity receptors similarly to antibodies for a great variety of ligands

(Miranda-Castro et al., 2009). However, the usefulness of DNA layers is not restricted to

these important applications. On the contrary, they can be the target for the antioxidant

assessment by mimicking the damage caused in vivo by ROS. Nucleobases are the main

targets of ROS leading to oxidation of bases and, ultimately to their release but sugar are

also weak points which may result in strand breaking. From this fact, it seems reasonable

that the protective role of antioxidants at a cellular level could be properly studied by

monitoring the DNA integrity. Several DNA-based electrochemical sensors have been

developed for the measurement of the antioxidant capacity of different compounds.

37

Tab

le 3

. E

nzym

atic

bio

sens

ors

for

antio

xida

nt q

uant

ifica

tion.

A

ntio

xida

nt

S

ampl

e

Enz

yme

Im

mob

iliza

tion

proc

edur

e (e

lect

rode

)

Line

ar r

ange

Det

ectio

n lim

it

refe

renc

e

Phe

nolic

su

bstr

ates

P

lant

s La

ccas

e

Ent

rapm

ent i

n ch

itosa

n-M

WC

NT

s co

mpo

site

on

Au

0.91

- 12

µM

(ro

smar

inic

)

0.23

3 µM

D

iaco

nu e

t al.,

201

0

Pol

yphe

nol

Win

es

Lacc

ase

Cro

sslin

ked

with

pol

ymer

on

SW

CN

T-S

PE

or

MW

CN

T-S

PE

0.

1- 1

8.0

mg

L-1 (

galli

c ac

id)

0.3

mg

L- (ga

llic

acid

) D

i fus

co e

t al.,

201

0

Pol

yphe

nolic

co

mpo

unds

W

ines

La

ccas

e M

embr

ane

on P

t ele

ctro

de

5 -3

5 µM

(ca

ffeic

aci

d)

0.88

µM

(ca

ffeic

ac

id)

Júni

or a

nd R

ebel

o 20

08,

Gil

and

Reb

elo

2010

C

atec

hol

- La

ccas

e C

ossl

inke

d to

chi

tosa

n on

MW

CN

T/ G

CE

0.

1-50

µM

20

nM

T

an e

t al.

2009

C

atec

hin

Gre

en te

a

Hum

an u

rine

La

ccas

e co

vale

ntly

imm

obili

zed

on d

endr

imer

-en

caps

ulat

ed A

uNP

s 0.

1-10

µM

0.

05 µ

M

Rah

man

et a

l., 2

008

Ros

mar

inic

aci

d P

lant

ext

ract

La

ccas

e In

corp

orat

ed to

ioni

c liq

uid

carb

on p

aste

0.

99-6

5.4

µM

0.

188

µM

F

ranz

oi e

t al.,

200

9 P

olyp

heno

ls

Bee

r La

ccas

e C

ossl

inki

ng o

nto

sono

gel-c

arbo

n el

ectr

ode

0.04

-2 µ

M (

caffe

ic a

cid)

0.

04-2

µM

(fe

rulic

aci

d)

0.1-

22 µ

M (

galli

c ac

id)

0.04

-3 µ

M (

cate

chin

) 0.

04-8

µM

(ep

icat

echi

n)

0.06

µM

0.

16 µ

M

0.41

µM

0.

10 µ

M

0.16

µM

El K

aout

it et

al.,

200

8

Caf

feic

aci

d C

atec

hol

Hyd

roqu

inon

e R

esor

cino

l

Tea

infu

sion

s La

ccas

e E

trap

men

t in

a m

embr

ane

on S

PE

0.

5-13

0 µ

M

0.5-

175

µM

1.

1-13

0 µ

M

50-2

50 µ

M

0.52

4 µ

M

0.55

8 µ

M

1.07

1 µ

M

5.43

2 µ

M

Ibar

ra-E

scut

ia e

t al.,

201

0

Tot

al p

heno

lic

cont

ent

Pla

nt e

xtra

cts

Lacc

ase

Ads

orbe

d on

a S

PE

7x

10-7

–1.5

x10-6

M

11.9

9x10

-7M

Li

tesc

u et

al.,

201

0

Phe

nolic

co

mpo

unds

-

Lacc

ase

Phy

sica

l ads

orpt

ion

on c

arbo

n ce

ram

ic

elec

trod

e 0.

1-10

µM

(ca

tech

ol)

0.06

µM

H

aghi

ghi e

t al.,

200

7

Phe

nols

and

po

lyph

enol

s B

eers

and

In

dust

rial

was

tew

ater

s

Tyr

osin

ase

Naf

ion

coat

ed o

n so

noge

l-car

bon

elec

trod

e -

64 n

M c

atec

hol

85 µ

M g

allic

aci

d 1.

25 µ

M c

atec

hin

96 n

M p

heno

l 30

nM

4-c

hlor

o-3-

met

hylp

heno

l

El K

aout

it et

al.,

200

7

Cat

echo

l C

affe

ic a

cid

Chl

orog

enic

ac

id

Spi

ked

river

wat

er

Tyr

osin

ase

Alu

min

a so

l-gel

on

sono

gel-c

arbo

n el

ectr

ode

0.1-

30 µ

M

5-30

µM

5-

30 µ

M

30 n

M

0.62

µM

0.

61 µ

M

Zej

li et

al.

2008

Cat

echo

l W

ater

sam

ples

T

yros

inas

e O

n A

uNP

s en

caps

ulat

ed-d

endr

imer

lin

kded

to a

con

duct

ing

poly

mer

on

GC

E

0.00

5- 1

20 µ

M

2 nM

S

ingh

, 201

1

p-ch

loro

phen

ol

p-cr

esol

, P

heno

l,

Cat

echo

l

- T

yros

inas

e C

oval

ent t

o ac

tivat

ed c

arbo

n fe

lt

2.6

-300

0 nM

2.7

-100

0 nM

0019

-10

µM

2.3-

3000

nM

2.6

nM

2.7

nM

18.7

nM

2.

3 nM

Wan

g an

d H

aseb

e, 2

011

38

Ant

ioxi

dant

Sam

ple

E

nzym

e

Imm

obili

zatio

n pr

oced

ure

(ele

ctro

de)

Li

near

ran

ge

D

etec

tion

limit

re

fere

nce

Cat

echo

l A

queo

us a

nd

orga

nic

med

ia

Tyr

osin

ase

Ent

rapm

ent i

nto

laye

red

doub

le h

ydro

xide

-al

gina

te h

ybrid

nan

ocom

posi

te o

n G

CE

2

nM–3

0 µ

M (

aq)

0.01

-3 µ

M (

orga

nic)

0.

5 nM

0.

01 µ

M

Lópe

z et

al.,

201

0

Cat

echo

l,

Cat

echi

n,

Caf

feic

aci

d

Gal

lic a

cid

Tea

sam

ples

T

yros

inas

e O

n di

azon

ium

-fun

ctio

naliz

ed S

PA

uE

0.1-

22 µ

M

2.8-

29 µ

M

0.3-

83 µ

M

2.5-

65 µ

M

0.1

µM

2.8

µM

0.2

µM

2.1

µM

Cor

tina-

Pui

g et

al.,

201

0

Phe

nolic

co

mpo

unds

-

Tyr

osin

ase

Ads

orpt

ion

on n

anoc

ompo

site

(A

u)

0.01

-7 µ

M

5 µ

M

Lu e

t al.,

201

0

Cat

echo

l -

Tyr

osin

ase

Inco

rpor

atio

n to

CP

E

- 0.

008 µ

M

Ard

uini

et a

l., 2

010

Phe

nolic

co

mpo

unds

M

icro

caps

ules

T

yros

inas

e D

rop-

coat

ing

on e

lect

rosp

inne

d m

embr

ane

GC

E

1–10

0 µ

M (

pyro

cate

chol

) 0.

05 µ

M

Are

cchi

et a

l., 2

010

p-cr

esol

P

heno

l 4-

chlo

roph

enol

- T

yros

inas

e A

dsor

bed

on Z

nO n

anor

od c

lust

ers

at

nano

crys

talli

ne d

iam

ond

elec

trod

es

1-21

0 µ

M

1-19

0 µ

M

1-25

0 µ

M

0.2 µ

M

0.5 µ

M

0.4 µ

M

Zha

o et

al.,

200

9a

p-cr

esol

4-

clor

ophe

nol

phen

ol

- T

yros

inas

e C

oval

ently

on

ZnO

nan

orod

s at

na

nocr

ysta

lline

dia

mon

d el

ectr

ode

1-17

5 µ

M

1-15

0 µ

M

1-15

0 µ

M

0.1

µM

0.

2 µ

M

0.25

0M

Zha

o el

al 2

009b

Phe

nolic

co

mpu

nds

Land

fill l

each

ate

T

yros

inas

e O

n m

etha

cryl

ic-a

cryl

ic m

embr

anes

on

SP

CE

6.

2-54

.2 µ

M

0.13

µM

H

anifa

h et

al.,

200

9

cate

chol

-

Tyr

osin

ase

Inco

rpor

ated

to M

WC

NT

-epo

xy c

ompo

site

U

p to

0.1

5 m

M

0.01

mM

Lo

pez

et a

l., 2

009

Tot

al p

heno

lic

com

poun

ds

Red

win

es

Tyr

osin

ase

Ioni

c liq

uid

mod

ified

MW

CN

Ts

on IT

O

0.01

-0.0

8 m

M

- K

im e

l al.,

200

9

Cat

echo

l In

dust

rial

was

tew

ater

P

olyp

heno

l ox

idas

e

Inco

rpor

ated

to p

olya

nilin

e fil

m o

n P

t

2.5-

140

µM

0.

05 µ

M

Tan

et a

l., 2

011

Pol

yphe

nols

T

ea le

aves

, al

coho

lic b

ever

ages

, w

ater

Pol

yphe

nol

oxid

ase

B

SA

-glu

tara

ldeh

yde

cros

slin

king

on

PV

C

mem

bran

e pl

aced

on

Pt

1.25

–10 µ

M

7.5 µ

M

Cha

wla

et a

l., 2

010

Cat

echo

l -

Pol

yphe

nol

oxid

ase

Ele

ctro

poly

mer

izat

ion

on S

PP

t 5-

100

µM

0.

318

µM

A

kyilm

az e

t al.,

201

0

Phe

nolic

co

mpo

unds

S

pike

d riv

er w

ater

P

olyp

heno

l ox

idas

e C

ross

linki

ng a

fter

entr

apm

ent i

n B

iOx

film

on

GC

E

4 nM

-15 µ

M (

cate

chol

) 1

nM

Sha

n et

al.,

200

9

Asc

orbi

c ac

id

Mul

tivita

min

ef

ferv

esce

nt ta

blet

w

hite

win

es

Asc

orba

te

oxid

ase

Flo

w s

yste

m. O

n m

embr

ane

cell

not

dire

ctly

on

the

elec

trod

e su

rfac

e 25

-400

µM

5

µM

V

ig e

t al.,

201

0

Asc

orbi

c ac

id

Ser

um, f

ruit

juic

e,

vita

min

c ta

blet

s A

scor

bate

ox

idas

e C

oval

ent t

o eg

g sh

ell m

embr

ane

on A

uE

10 –

100

µM

1.0 µ

M

Cha

uhan

et a

l., 2

010

Asc

orbi

c ac

id

Fru

it ju

ice

Asc

orba

te

oxid

ase

On

nylo

n m

embr

ane

with

glu

atar

alde

hyde

on

Cla

rk e

lect

rode

0.

10-0

.55

mM

0.

023

mM

P

isos

chi,

et a

l., 2

010

L-as

corb

ic a

cid

Fru

it ju

ice

Asc

orba

te

oxid

ase

Mic

elle

s on

pol

ysty

rene

coa

ted

GC

E

5 -4

00 µ

M

4 µ

M

Wan

g et

al.,

200

8

39

Ant

ioxi

dant

Sam

ple

E

nzym

e

Imm

obili

zatio

n pr

oced

ure

(ele

ctro

de)

Li

near

ran

ge

D

etec

tion

limit

re

fere

nce

Uric

aci

d H

uman

ser

um

Urin

e sa

mpl

es

Uric

ase

On

eggs

hell

mem

bran

e on

a C

lark

sen

sor

4.0-

640

µM

2.

0 µ

M

Zha

ng e

t al.,

200

7

Uric

aci

d C

linic

al te

sts

Uric

ase

Laye

r by

laye

r on

Pru

ssia

n bl

ue-I

TO

0.

1-0.

6 µ

M

0.15

µM

M

orae

s et

al.,

200

7 U

ric a

cid

- m

icro

pero

xida

se-

11

Uric

ase

MxP

-11

was

cov

alen

tly b

ound

to S

AM

U

ricas

e in

to c

hito

san

on M

xP-1

1 el

ectr

ode

5–15

0 µ

M

2 µ

M

Beh

era

and

Raj

. 200

7

Phe

nolic

co

mpu

nds

- H

RP

O

n m

ethy

lene

blu

e-M

WC

NT

- C

PE

1-

150 µ

M

0.5 µ

M

San

tos

et a

l., 2

007

Ant

ioxi

dant

ca

paci

ty

Gar

lic b

ulbs

O

rang

e ju

ices

X

OD

C

oval

ently

atta

ched

to m

ixed

SA

Ms

Alli

n st

anda

rd

Asc

orbi

c ac

id s

tand

ard

- C

ortin

a-P

uig

et a

l., 2

009,

20

09b

antio

xida

nt

capa

city

P

apay

a fr

uit a

nd

papa

ya-b

ased

food

S

uper

oxid

e di

smut

ase

On

kapp

a-ca

rrag

eena

n ge

l bet

wee

n ce

llulo

se a

ceta

te a

nd a

dya

lisis

mem

bran

e on

Pt

- -

Cam

pane

lla e

t al.,

200

9


40

The simplest immobilization strategy is physical adsorption of double stranded DNA

(dsDNA) (Diopan et al., 2008) or single stranded DNA (ssDNA) (Barroso et al., 2011b;

Barroso et al., 2011c) on CPE. Covalent attachment of dsDNA on PAMAM-encapsulated

Au-Pd/chitosan composite was also proposed (Qian et al., 2010). Guanine or adenine free

bases were also used as oxidant layers adsorbed on GCE (Barroso et al., 2011d; 2011e;

2012f). Although the resulting modified electrodes cannot be considered as DNA-based

biosensors, they are included in this section because the principle of measurement is also

based on the electroactivity of purine bases, the most easily oxidized ones, after radical

attack. The oxidation current of guanine or adenine dramatically decreases in the

presence of radicals when comparing with their native electroactivity. When an antioxidant

is added to the radical-containing solution, the current is partially recovered, which is

attributed to the scavenging activity of these compounds. CV, SWV and DPV were the

electrochemical techniques most widely used for these studies (Fig. 3E).

Since the antioxidant capacity is highly dependent on the source of ROS, that is, a

single antioxidant exhibits different scavenging efficiency against different radicals, the

development of methods based on several source of ROS is needed. Most methods rely

on the generation of hydroxyl radical through a Fenton-type reaction. Recently sulfate

radical (Barroso et al., 2011e) and superoxide radical enzymatically generated (Barroso et

al, 2011 d, Barroso et al. 2011b) were also reported.

An indirect electrocatalytic voltammetric method to assess total antioxidant capacity

using a DNA-modified CPE was developed (Barroso et al., 2011a, b). It was reported that

the electrochemical oxidation of both adenine and guanine homopolynucleotides in neutral

or alkaline conditions led to the formation of a common oxidized product that catalyzed the

oxidation of NADH (de-los-Santos-Alvarez et al., 2007). Therefore, the oxidative lesions

generated after immersion of the DNA-CPE in free radicals (hydroxyl or superoxide

radical) can be indirectly quantified after the electrochemical oxidation of the adenines that

remained unoxidized on the electrode surface. Under the conditions tested both radicals

produce around 80% of damage on DNA. In the presence of antioxidants (ascorbic acid,

gallic acid, caffeic acid and resveratrol) an increase in the electrocatalytic current of NADH

was obtained, which probed the capacity of these compounds as free radical scavengers.

The efficiency varied from 19 to 63% and a different trend was observed depending on the

source of radicals used suggesting a different mode of action against both radicals.

Zhang et al. observed an increase in the electroactivity of DNA after damage because of

the strand breaking that leaves the purine bases more exposed to the electrode surface

facilitating the electron transfer. In addition to this, it was also found that the role of AA

was prooxidant at concentrations below 1.5 mM (increasing damage) but antioxidant at

higher concentrations (Zhang et al., 2008a).


41

Ziyatdinova et al observed that the degraded DNA is less effective in blocking the

electron transfer from anionic redox probes in solution. So, the current recovery due to the

protection of rutin allowed its detection by CV (Ziyatdinova et al., 2008).

Finally, it is worth mentioning that DNA layers were also used as simple modifiers

without being subjected to radical damage, similarly to approaches explained in section 3

(Liu et al., 2011; Wang et al., 2011).

Table 4 shows the main features of the DNA-based biosensors reported for the

evaluation of antioxidant status.

42

Tab

le 4

. D

NA

-bas

ed b

iose

nsor

app

lied

for

the

antio

xida

nt s

tatu

s ev

alua

tion.

Ant

ioxi

dant

Sam

ple

Tar

get l

ayer

Im

mob

iliza

tion

proc

edur

e F

ree

radi

cal

Line

ar r

ange

Det

ectio

n lim

it re

fere

nce

pom

iferin

, is

opom

iferin

, os

ajin

and

ca

talp

osid

e

- ss

DN

A

Ads

orpt

ion

afte

r da

mag

e on

CP

E

Hyd

roxy

l rad

ical

-

- D

iopa

n e

tal.,

200

8

TA

C

Bev

erag

es

dA21

A

dsor

ptio

n on

CP

E

Hyd

roxy

l rad

ical

0.

05-1

.00 µ

M

50 n

M

Bar

roso

et a

l., 2

011c

TA

C

Bev

erag

es

Fla

vour

ed w

ater

s dA

21

Ads

orpt

ion

on C

PE

S

uper

oxid

e ra

dica

l 10

-100

µM

10

µM

B

arro

so e

t al.,

201

1b

TA

C

Bev

erag

es

Fla

vour

ed w

ater

s G

uani

ne

aden

ine

Ads

orpt

ion

on G

CE

S

uper

oxid

e ra

dica

l 1.

00–5

.00

mg

l-1

0.50

–4.0

0 m

g l-1

0.

77 m

g l-1

0.

50 m

g l-1

B

arro

so e

t al.,

201

1e

TA

C

Bev

erag

es

Fla

vour

ed w

ater

s G

uani

ne

aden

ine

Ads

orpt

ion

on G

CE

S

ulfa

te r

adic

al

0.50

– 4

.00

mg

l-1

0.50

– 4

.00

mg

l-1

0.47

mg

l-1

0.50

mg

l-1

Bar

roso

et a

l., 2

011f

TA

C

Bev

erag

es

Fla

vour

ed w

ater

s G

uani

ne

aden

ine

Ads

orpt

ion

on G

CE

H

ydro

xyl r

adic

al

0.50

– 2

.50

mg

l-1

2.00

– 6

.00

mg

l-1

0.29

mg

l-1

0.99

mg

l-1

Bar

roso

et a

l., 2

011

Ser

icin

-

dsD

NA

C

oval

ently

to P

AM

AM

den

drim

er

enca

psul

ated

Au-

PdN

Ps/

GC

E

Hyd

roxy

l rad

ical

-

- Q

ian

et a

l., 2

010

Asc

orbi

c ac

id

Man

nito

l

dsD

NA

A

dsor

ptio

n af

ter

dam

age

on G

CE

H

ydro

xyl r

adic

al

- -

Zha

ng e

t al.,

200

8

Rut

in

Tea

ext

ract

s ds

DN

A

Ads

oprt

ion

on M

WN

T/S

PC

E

Hyd

roxy

l rad

ical

-

- Z

iyat

dino

va e

t al.,

200

8

Uric

aci

d

- ds

DN

A

PA

MA

M-M

WC

NT

-chi

tosa

n A

uE

- 0.

5-10

0 µ

M

0.07

µM

Li

u et

al.,

201

1

baic

alei

n M

edic

inal

tabl

ets

Spi

ked

hum

an b

lood

ds

-DN

A

Lagm

uir-

Blo

dget

t film

-

0.01

-2 µ

M

6.0

nM

Wan

g et

al.,

201

1


43

6. Conclusions

Measurement of the individual antioxidant or total antioxidant capacity can be performed

by several techniques such as spectrophotometry, chromatography and electrochemistry.

Considering the complexity of food composition and the possibility of synergistic

interactions among the antioxidant compounds in the sample, the evaluation of total

antioxidant capacity is desired instead the individual antioxidant measurement.

Electrochemistry is well suited approach for the antioxidant evaluation because

antioxidant mechanism in vitro systems is based on the transference of charge involved in

electron transfer during the redox reactions. The use of electrochemical devices presents

several advantages such as short detection time, small sample volume, high accuracy,

simplicity and lacks interferences from coloured samples avoiding time-consuming pre-

treatments.

Four different electrochemical approaches for the antioxidant evaluation were identified

in this review. Direct electrochemical detection on bare and CMEs, enzymatic biosensors

and DNA-based biosensors were described in detail. Direct electrochemical detection

presents the advantage of simplicity. In order to increase the rate of the electrode reaction

as well the sensitivity and selectivity, chemical modification of electrode surface can be

successfully carried out. The use of enzymes on electrodes for antioxidant sensing

combines the high selectivity of these biomolecules with a significant amplification of the

analytical signal. Besides, enzymatic biosensors are reusable, relatively simple, rapid and

inexpensive. The potential window at which they operate minimize the electrochemical

interferences in complex real samples. Nevertheless, the use of DNA-based biosensor is

preferred because the principle of measurement is closer to the activity of antioxidant in

biological systems. dsDNA, ssDNA or nucleobases immobilized on the electrode are

exposed to radical attack similarly to what occurs within the cell, which may generate

replication errors and subsequent misleading protein synthesis. Therefore, DNA-based

biosensors are considered promising tools for rapid screening of TAC in different kind of

matrices.

Acknowledgements

M. Fátima Barroso is grateful to Fundação para a Ciência e a Tecnologia for a Ph.D.

grant (SFRH/BD/ 29440/2006). N.S.Á thanks to Ministerio de Ciencia y Tecnología for a

Ramón y Cajal contract and for financial support (Project CTQ2008-02429 granted to

“Grupos Consolidados”).


44

References

Abidi, S.L., 2000. J. Chromatogr. A 881 (1-2), 197-216.

Akyilmaz, E., Kozgus, O., Turkmen, H., Cetinkaya, B., 2010. Bioelectrochemistry 78 (2)

135-140.

Almajano, M.P., Carbó, R., Jiménez, J.A.L., Gordon, M,H., 2008. Food Chem. 108 (1), 55-

63.

Arduini, F., Di Giorgio, F., Amine, A., Cataldo, F., Moscone, D., Palleschi, G., 2010. Anal.

Lett. 43 (10-11), 1688-1702.

Arecchi, A., Scampicchio, M., Drusch, S., Mannino, S., 2010. Anal. Chim. Acta 659 (1-2),

133-136.

Barroso, M. F., Delerue-Matos, C., Oliveira, M. B. P. P., (2011d). Biosens. Bioelectron. 26

(9) 3748-3754.

Barroso, M. F., Delerue-Matos, C., Oliveira, M. B. P. P., (2011e) Evaluation of total

antioxidant capacity of flavoured waters using sulfate radical damage of purine-based

sensor. (submitted).

Barroso, M. F., Delerue-Matos, C., Oliveira, M. B. P. P., (2011f) Electrochemical

evaluation of total antioxidant capacity of beverages using a DNA biosensor:

Voltammetric detection of DNA damage with hydroxyl radical and its protection with

ascorbic acid and phenolic acids. Submitted.

Barroso, M.F., de-los-Santos-Alvarez, N., Lobo-Castañón, M.J., Miranda-Ordieres, A.J.,

Delerue-Matos, C., Oliveira, M.B.P.P., Tuñón-Blanco, P., 2011b. In press. J.

Electroanal. Chem. doi:10.1016/j.jelechem.2011.04.022

Barroso, M.F., de-los-Santos-Alvarez, N., Lobo-Castañón, M.J., Miranda-Ordieres, A.J.,

Delerue-Matos, C., Oliveira, M.B.P.P., Tuñón-Blanco, P., 2011c. DNA-based biosensor

for the electrocatalytic determination of antioxidant capacity in beverages Biosens.

Bioelectron. 26 (5), 2396-2401.

Barroso, M.F., Noronha, J.P., Delerue-Matos, C., Oliveira, M. P. P., (2011a). J. Agric.

Food Chem. 59 (9), 5062-5072.

Behera, S., Raj, C.R., 2007. Biosens. Bioelectron. 23(4), 556-561.

Blasco, A.J., Crevillén, A.G., González, M.C., Escarpa, A., 2007. Electroanalysis, 19(22)

2275-2286.

Buratti, S., Scampicchio, M., Giovanelli, G., Saverio Mannino, S., 2008. Talanta 75 (1),

312–316.

Campanella, L., Gatta, T., Gregori, E., Tomassetti, M., 2009. Monatsh. Chem. 140 (8),

965-972.


45

Chauhan, N., Dahiya, T, Priyanka, Pundi, C. S. 2010. J. Mol. Catal. B-Enzym. 67 (1-2),

66-71.

Chawla, S., Narang, J., Pundir, C. S., 2010. Anal. Methods 2 (8), 1106-1111.

Chen, X., Wang, Z., Zhang, F., Zhu, L., Li, Y., Xia, Y. 2010. Chem. Pharm. Bull. 58 (4),

475-478

Cheng, Z.Y., Li, Y.Z., 2004. Redox Rep. 9 (4), 213-217.

Civit, L., Nassef, H.M., Fragoso, A., O'Sullivan, C.K., 2008. J. Agric. Food Chem. 56 (22),

10452-10455.

Corduneanu, O., Garnett, M., Brett, A. M. O., 2007. Anal. Lett. 40 (9), 1763–1778.

Cortina-Puig, M., Munoz-Berbel, X., Calas-Blanchard, C., Marty, J.L., 2009a. Talanta 79

(2), 289-294.

Cortina-Puig, M., Munoz-Berbel, X., Calas-Blanchard, C., Marty, J.L., 2010. Microchim.

Acta 171 (1-2), 187-193.

Cortina-Puig, M., Munoz-Berbel, X., Rouillon, R., Calas-Blanchard, C., Marty, J.L., 2009b.

Bioelectrochem. 76 (1-2), 76-80.

de-los-Santos-Alvarez, N., de-los-Santos-Alvarez, P., Lobo-Castañón, M.J., Lopez, R.,

Miranda-Ordieres, A.J., Tuñón-Blanco, P., 2007. Electrochem. Commun. 9 (8), 1862-

1866.

Di Fusco, M., Tortolini, C., Deriu, D., Mazzei, F., 2010. Talanta 81 (1-2), 235-240.

Diaconu, M., Litescu, S.C., Radu, G.L., 2010. Sensor Actuat. B-Chem. 145 (2), 800-806.

Diopan, V., Babula, P., Shestivska, V., Adam, V., Zemlicka, M., Dvorska, M., Hubalek, J.,

Trnkova, L., Havel, L., Kizek, R., 2008. J. Pharm. Biomed. Anal. 48 (1), 127-133.

Du, H.J., Ye, J.S., Zhang, J.Q., Huang, X.D., Yu, C.Z., 2010. Electroanalysis 22 (20),

2399-2406.

Durst, R.A., Baumner, A.J., Murray, R.W., Buck, R.P., Andrieux, C.P., 1997. Pure Appl.

Chem. 69 (6), 1317-1323.

El Kaoutit, M., Naranjo-Rodriguez, I., Temsamani, K.R., Hernandez-Artiga, M.P., Bellido-

Milla, D., Hidalgo-Hidalgo de Cisneros, J.L., 2008. Food Chem. 110 (4) 1019-1024.

El Kaoutit, M., Naranjo-Rodriguez, I., Temsamani, K.R., Hidalgo-Hidalgo de Cisneros,

J.L., 2007. Biosens. Bioelectron. 22 (12), 2958–2966.

El-Agamey, A., Lowe, M., McGarvey, D. J., Mortensen, A., Philips, D.M., Truscott, T.G.,

Young, A.J., 2004. Arch. Biochem. Biophys. 430 (1), 37-48.

El-Hady, D.A., 2007. Anal. Chim. Acta 593 (82), 178-187.

El-Hady, D., Nagwa El-Maali, N., 2008. Microchim, Acta 161 (1-2), 225–231

El-Hady, D.A., El-Maali, N.A., 2008. Talanta 76 (1), 138-145.

Fell, A.J., Dykes, S.I., Laura, N., Kilmartin, P.A., 2007. Am. J. Enol. Vitic. 58 (4), 443-450.


46

Frankel, E.N., Antioxidants in food and biology facts and fiction. Publisher Oily Press.

Bridgwater, England, vol 20 2007.

Franzoi, A.C., Dupont, J., Spinellia, A., Vieira, L.C., 2009. Talanta 77 (4), 1322-1327.

Franzoi, A.C., Spinelli, A., Vieira, L.C., 2008. J. Pharm. Biomed. Anal. 47 (4-5), 973-977.

Freitas, K.H.G., Fatibello, O., 2010a. Anal Lett 43 (13), 2091-2104.

Freitas, K.H.G., Fatibello, O., 2010b. Talanta 81 (3), 1102-1108.

Freitas, K.H.G., Medeiros, R.A., Fatibello, O., 2009. Anal Lett 42 (6), 881-897.

Gil, D. M. A., Rebelo, M.J.F., 2010. Eur. Food Res. Technol. 231 (2), 303–308.

Gomes, A., Fernandes, E., Garcia, M.B.Q., Silva, A.M.S., Pinto, D.C.G.A., Santos,

C.M.M., Cavaleiro, J.A.S., Lima, J.L.F.C. 2008. Bioorg. Med. Chem. 16 (17), 7939-

7943.

Haghighi, B., Panah, A., Shleev, S.L., 2007. Electroanalysis 19 (9), 907-917.

Halliwell B., Gutteridge J.M., 1999. Free radicals in biology and Medicine, third edition.

Oxoford University Press, Midsomer Norton, Avon, England.

Halliwell, B., Gutteridge, J.M., 1990. Role of free radicals and catalytic metal ions in

human disease: an overview. Methods Enzymol. 186, 1-85.

Han, R.M., Tian, Y.X., Liu, Y., Chen, C.H., Ai, X.C., Zhang, J.P., Skibsted, L.H., 2009. J.

Agric. Food Chem. 57 (9), 3780–3785.

Hanifah, S.A., Heng, L.Y., Ahmad, M., 2009. Anal. Sci.25 (6), 779-784.

He, J-B., Yuan, S-J., Du, J.Q., Hu, X-R., Wang, Y., 2009. Bioelectrochemistry 75 (2), 110–

116.

Huang, Z., Wang, B., Eaves, D.H., Shikany, J.M., Pace, R.D., 2009. Int. J. Food Sci. Nutr.

60 (2), 100-108.

Huang, D., Ou, B., Prior, R.L., 2005. The chemistry behind antioxidant capacity assays. J.

Agr. Food Chem. 53 (6), 1841-1856.

Ibarra-Escutia, P., Gómez, J.J., Calas-Blanchard, C., Marty, J.L., Ramírez-Silva, M. T.,

2010. Talanta 81 (4-5), 1636-1642.

Intarakamhang, S., Leson, C., Schuhmann, W., Schulte, A., 2011. Anal. Chim. Acta 687

(1), 1–6.

Júnior, A.R.S., Rebelo, M.J.F., 2008. Port. Electrochim. Acta 26 (1), 117-124.

Kim, K.-I., Hang, H.-Y., Lee, J.-C., Choi, S.-H., 2009. Sensors 9 (9), 6701-6714.

Kohen, R., Nyska, A., 2002. Toxicol. Pathol. 30 (6), 620-650.

Koren, E., Lipkin, J., Klar, A., Hershkovitz, E., Ginsburg, I., Kohen, R. 2009. J. Inherit.

Metab. Dis. 32 (5), 651–659.

Krinsky, N.I., 1992. Proc. Soc. Exp. Biol. Med. 200 (2), 248-254.

Laguerre, M., Decker, E.A., Lecomte, J., Villeneuve, P., 2010. Curr. Opin. Clin. Nutr.

Metab. Care 13 (5), 518-525.


47

Laguerre, M., Lecomte, J., Villeneuve, P., 2007. Progr.Lipid res. 46 (5), 244-282.

Lee, K.W., Kim, Y.J., Kang, N.J., Kim, J.H., Lee, S.J., Kim, D.O., Lee, C.Y., Lee, H.J.,

2009. Int. J. Food Sci. Nutr. 60 (1), 12-20.

Li, N.B., Ren, W., Luo, H.Q., 2008. J. Solid State Electr. 12 (6), 693-699.

Litescu, S.C., Eremia, S.A.V., Bertoli, A., Pistelli, L., Radu, G.-L., 2010. Anal. Lett. 43 (7-

8), 1089-1099.

Liu, A.-L., Zhanga, S.-B., Chen, W., Huang, L.-Y., Lin, X.-H., Xing-Hua Xia, X.-X., 2008.

Talanta 77 (1), 314–318.

Liu, X.G., Peng, Y.H., Qu, X.J., Ai, S.Y., Han, R.X., Zhu, X.B., 2011. J. Electroanal. Chem.

654 (1-2), 72-78.

Lopez, B.P., Merkoci, A., 2009. Analyst 134 (1), 60-64.

Lopez, M.S.P., Leroux, F., Mousty, C., 2010. Sens. Actuator B-Chem. 150 (1), 36-42.

Lu, L., Zhang, L., Zhang, X., Huan, S., Shen, G., Yu, R., 2010.. Anal. Chim. Acta 665 (2),

146-151.

Luo, L.Q., Li, Q.X., Ding, Y.P., Zhang, Y., Shen, X., 2010. J. Solid State Electr. 14 (7),

1311-1316.

Ma, X.Y., Qian, W.P., 2010. Biosens. Bioelectron. 26 (3), 1049-1055.

Makhotkina, O., Kilmartin, P.A., 2010. Anal. Chim. Acta 668 (2), 155-165.

Medvidović-Kosanović, M., Šeruga, M., Jakobek, M.J., Novak, I., 2010. Croat. Chem. Acta

83 (2), 197–207.

Miranda-Castro, R., de-los-Santos-Álvarez, N., Lobo-Castañón, M.J., Miranda-Ordieres,

A.J., Tuñón-Blanco, P., 2009. Electroanalysis 21 (19), 2077-2090.

Moccelini, S.K., Fernandes, S.C., de Camargo, T.P., Neves, A., Vieira, I.C., 2009. Talanta

78 (3), 1063-1068.

Moraes, M. L., Rodrigues Filho, U. P., OliveiraJr., O. N., Ferreira, M. 2007. J. Solid State

Electr. 11 (11), 1489-1495.

Piljac-Žegarac, J., Valek, L., Tamara Stipčević, T.,Martinez, S., 2010.. Food Chem. 121

(3), 820-825

Pisoschi, A.M., Cheregi, M.C., Danet, A.F., 2009. Molecules 14 (1), 480-493.

Pisoschi, A.M., Negulescu, G.P., Pisoschi, A., 2010. Rev. Chim.-Bucharest 61 (4), 339-

344.

Prabakar, S.J.R., Narayanan, S.S., 2010. Food Chem. 118 (2), 449-455.

Prior, R.L., Wu, X., Schaich, K. 2005. J. Agr. Food Chem. 53 (10), 4290-4302.

Qian, P., Ai, S.Y., Yin, H.S., Li, J.H., 2010. Microchim. Acta 168 (3-4), 347-354.

Rahman, M.A., Noh, H.B., Shim, Y.B., 2008. Anal. Chem. 80 (21), 8020-8027.

Rice-Evans, C.A., Burdon R.H. 1994. New Comprehensive Biochemistry: Free Radical

Damage and its Control, Vol. 28. Elsevier Science, Amerterdam.


48

Sánchez-Moreno, C., 2002. Food Sci. Technol. Int. 8 (3),121-137.

Santos, A.S., Pereira, A.C., Sotomayor, M.D.R.T., Tarley, C.R.T., Duran, N., Kubota, L.T.,

2007a. Electroanalysis 19 (5), 549-554.

Santos, D.P., Bergamini, M.F., Santos, V.A.F.F.M., Furlan, M., Zanoni, M.V.B., 2007b.

Anal. Lett. 40 (16-18), 3430-3442.

Shahrokhian, S., Asadian, E., 2010. Electrochim. Acta 55 (3), 666-672.

Shahrokhian, S., Zare-Mehrjardi, H.R., 2007a. Electrochim. Acta 52 (22), 6310-6317.

Shahrokhian, S., Zare-Mehrjardi, H.R., 2007b. Sensor Actuat. B-Chem. 121 (2), 530-537.

Shahrokhian, S., Zare-Mehrjardi, H.R., 2007c. Electroanalysis 19 (21), 2234 – 2242.

Shan, D., Zhang, J., Xue, H.-G., Zhang, Y.-C., Cosnier, S., Ding,-N., 2009. Biosens.

Bioelectron. 24 (12), 3671-3676.

Simić, A., Manojlović, D., Šegan, D., Todorović, M., 2007. Molecules 12 (10), 2327-2340.

Singh, R. P., 2011. Analyst 136 (6), 1216-1221.

Song, J.C., Yang, J., Zeng, J.F., Tan, J.A., Zhang, L., 2010. Microchim. Acta 171 (3-4),

283-287.

Sun, W., Yang, M.X., Li, Y.Z., Jiang, Q., Liu, S.F., Jiao, K., 2008. J. Pharm. Biomed. Anal.

48 (5), 1326-1331.

Tan, Y., Deng, W., Ge, B., Xie, Q., Huang, J., Yao, S., 2009. Biosens. Bioelectron. 24 (7),

2225–2231.

Tan, Y., Kan, J., Shengqi, L., 2011. Amperometric biosensor for catechol using

electrochemical template process. Sens. Actuators B Chem. 152 (2), 285–291.

Tyszczuk, K., 2009. J. Pharm. Biomed. Anal. 49 (2), 558-561.

Valls, J., Millán, S., Martí, M.P., Borràs, E., Arola, L., 2009. J. Chromatogr. A 1216 (43),

7143-7172.

Vig, A., Igloi, A., Adanyi, N., Gyemant, G., Csutoras, C., Kiss, A., 2010. Bioproc. Biosyst.

Eng. 33 (8), 947-952.

Wang, F., Zhao, F.Y., Zhang, Y.Z., Yang, H.G., Ye, B.X., 2011. Talanta 84 (1), 160-168.

Wang, J., Zhou, N.D., Zhu, Z.Q., Huang, J.Y., Li, G.X., 2007a. Anal. Bioanal. Chem. 388

(5-6), 1199-1205.

Wang, M., Xu, X., Gao, J. 2007b. J. Appl. Electrochem. 37 (6), 705-710.

Wang, X., Watanabe, H., Uchiyama, S., 2008. TALANTA 74 (5) 1681-1685.

Wang, X.G., Fan, Y.J., 2010. Microchim. Acta 169 (1-2), 173-179.

Wang, Y., Hasebe, Y., 2011. Anal. Bioanal. Chem. 399 (3), 1151-1162.

Yang, L.J., Tang, C., Xiong, H.Y., Zhang, X.H., Wang, S.F., 2009. Bioelectrochem. 75 (2),

158-162.

Yao, W.W., Peng, H.M., Webster, R.D., R.D. Gill, P.M.W., 2008. J. Phys. Chem. B, 112

(22), 6847–6855.


49

Zejli, H., de Cisneros, J.L.H.H., Naranjo-Rodriguez, I., Liu, B.H., Temsamani, K.R., Marty,

J.L., 2008. Anal. Chim. Acta 612 (2), 198-203.

Zhang, J.J., Wang, B., Li, Y.F., Jia, W.L., Cui, H., Wang, H.S., 2008a. Electroanalysis

20(15), 1684-1689.

Zhang, L., Zhang, C.H., Lian, J.Y., 2008b. Biosens. Bioelectron. 24 (4), 690-695.

Zhang, Y., Wen, G.M., Zhou, Y.H., Shuang, S.M., Dong, C., Choi, M.M.F., 2007a.

Biosens. Bioelectron. 22(8), 1791-1797.

Zhang, Y., Zheng, J.B., 2008. Talanta 77 (1), 325-330.

Zhang, Y.H., Su, S., Pan, Y., Zhang, L.P., Cai, Y.J., 2007b. Ann. Chim-Rome 97 (8), 665-

674.

Zhao, J., Zhi, F., Zhou, Y. L., Yan, W. B., 2009. Anal. Sci. 25 (9), 1083-1088.

Zhou, A.L., Kikandi, S., Sadik, O.A., 2007. Electrochem. Commun. 9 (9), 2246-2255.

Zielinska, D., Szawara-Nowak, D., Ornatowska, A., Wiczkowski, W., 2007. J. Agric. Food

Chem. 55 (24), 9891–9898.

Zielinska, D., Wiczkowski, W., Piskula, M.K., 2008. J. Agric. Food Chem. 56 (10), 3524–

3531.

Ziyatdinova, G., Galandova, J., Labuda, J., 2008. Int. J. Electrochem. Sc. 3 (3), 223-235.

51

II.

INVESTIGAÇÃO E DESENVOLVIMENTO

53

Capítulo 2

Composição mineralógica

2.1.

Flavoured versus natural waters: Macromineral (Ca, Mg, K, Na) and micromineral (Fe, Cu,

Zn) contents

M. Fátima Barroso, Aurora Silva, Sandra Ramos, M.T. Oliva-Teles, Cristina Delerue-Matos, M.

Goreti F. Sales, M.B.P.P. Oliveira

Food Chemistry, 2009, 116 (2), 580–589

2.2.

Survey of trace elements (Al, As, Cd, Cr, Co, Hg, Mn, Ni, Pb, Se, and Si) in retail samples

of flavoured and bottled waters

M. F. Barroso, S. Ramos, M. T. Oliva-Teles, C. Delerue-Matos, M. G. F. Sales, M. B. P. P.

Oliveira

Food Additives and Contaminants: Part B – Surveillance, 2009, 2 (2), 121-130

2.1. Macrominerais e microminerais

55

Flavoured versus natural waters: Macromineral (Ca, Mg, K, Na)

and micromineral (Fe, Cu, Zn) contents

M. Fátima Barrosoa,b, Aurora Silvaa, Sandra Ramosc, M. T. Oliva-Telesa, Cristina Delerue-

Matosa, M. Goreti F. Salesa, M. B. P.P. Oliveirab aRequimte/Instituto Superior de Engenharia do Porto, Rua Dr. António Bernardino de

Almeida 431, 4200-072, Porto, Portugal. bRequimte/Serviço de Bromatologia, Faculdade de Farmácia, Universidade do Porto, Rua

Aníbal Cunha, 164, 4099-030 Porto, Portugal. cInstituto Superior de Engenharia do Porto, Rua Dr. António Bernardino de Almeida 431,

4200-072, Porto, Portugal.

Abstract

Macro (Ca, Mg, K, Na) and micromineral (Fe, Zn, Cu) composition of 39 waters was

analysed. Determinations were made by atomic flame spectrophotometry for

macrominerals and electrothermic atomization in graphite furnace for microminerals.

Mineral contents of still or sparkling natural waters (without flavours) changed from

brand to brand. Mann–Whitney test was used to search for significant differences between

flavoured and natural waters. For that, the concentration of each mineral was compared to

the presence of flavours, preservatives, acidifying agents, fruit juice and/or sweeteners,

according to the labelled composition.

The statistical study demonstrated that flavoured waters generally have increased

contents of K, Na, Fe and Cu. The added preservatives also led to significant differences

in the mineral composition. Acidifying agents and fruit juice can also be correlated to the

increase of Mg, K, Na, Fe and Cu. Sweeteners do not provide any significant difference in

Ca, Mg, Fe and Zn contents.

Keywords: Macrominerals; Microminerals; Atomic spectrophotometry; Flavoured waters;

Health benefits

Available online at www.sciencedirect.com

Food Chemistry 2009, 116 (2), 580-589


57

1. INTRODUCTION

Water makes up more than two thirds of the human body, and it is the most consumed

drink in the world. To answer to consumer´s preferences, industries have applied several

technical improvements to plain water. Today, a significant part of marketed water is

flavoured. It consists in the addition of flavours, juices, bioactive compounds,

preservatives and/or sweeteners that provide singular tastes and smells appreciated by

consumers.

In the case of flavoured waters, either mineral or spring sources are used, both having

important mineral contents. According to Food and Drug Administration (FDA, 2002),

mineral water arises from a geologically and physically protected underground source,

characterised by constant levels and relative proportions of minerals and trace elements

at the source. Spring water is derived from an underground formation from which water

flows naturally to the surface at an identified location.

Minerals are necessary for human life and play important roles in metabolic functions

(Biziuk & Kuczynska, 2007) such as, maintenance of pH, osmotic pressure, nerve

conductance, muscle contraction, energy production, and in almost all other aspects of

life. Depending on the amounts needed, minerals can be divided into macro (g or mg/day)

and microminerals (few mg or µg/day). Physiologically, the most important macrominerals

are Ca, K, Na and Mg, and the same for Fe, Cu and Zn as microminerals (Silvera &

Rohan, 2007).

Bioavailability of minerals is affected by several factors. Host factors can be defined as

any attribute that can influence the amount of metal exposure, uptake, absorption,

biokinetics and susceptibility of an individual. Such factors include age, gender, size and

weight, nutritional status, genetics and some behaviours (Robson, 2003).

Although minerals are essential to normal health and development, they can become

toxic in higher amounts. Risk assessments of chemical elements show high intakes that

result in toxicity or nutritional problems related to low or no intakes (Goldhaber, 2003). So,

it is important to establish an adequate intake of certain substance to avoid adverse health

effects (Nasreddine & Parent-Massin, 2002). To answer this goal, the Joint Expert

Committee on Food Additives (JECFA) of the Food and Agriculture Organisation of the

United Nations and the World Health Organisation (FAO/WHO, 2000) established

acceptable or tolerable intakes for substances that exhibit thresholds of toxicity.

Provisional Tolerable Daily Intake (PTDI), calculated on a daily basis for certain

substances that do not accumulate in the human body, is the reference value that

indicates the safe level of intake. US Environmental Protection Agency (EPA) has also

carried out the Reference Dose (RfD), being the amount of a daily human exposure

II. Investigação e desenvolvimento

58

(including sensitive subgroups) to a certain compound, without an appreciable risk during

a lifetime (EPA, 1993). RfD is generally expressed in mg/kgbodyweight/day. US Food and

Nutrition Board of the Institute of Medicine (FNB/IOM) set forward the Recommended

Dietary Allowance (RDA) as the average daily intake that meets the nutrient requirements

of nearly all healthy individuals in a particular life stage and gender (Institute of Medicine

from United States, 2007). Table 1 presents the different values established for the

aforementioned minerals.

Table 1 Recommended dietary allowances (RDA), provisional tolerable intakes (PTDI)

and reference dose (RfD) for the studied minerals.

Mineral RDA (mg/day) PTDI (mg/kgbodyWeight/day) RfD (mg/kg/day)

Ca 1000a - -

Mg 420male;a

320female;a - -

K 4700b - -

Na 1500b - -

Fe 8male;a

18female;a 0.8c -

Zn 11male;a

8female;a 1.0d 0.3f

Cu 0.9a 0.5e -

Adult Values (male or female) with 31-50 years old.

- Not available. aInstitute of Medicine from United States (2001). bInstitute of Medicine from United States (2004). cWHO/FAO (1983). dWHO/FAO (1982a). eWHO/FAO (1982b). fEnvironmental Protection Agency of United States, EPA (2005)

Several analytical methods have been developed to determine the mineral contents in

biological, food and environmental samples. The most commonly employed techniques

are described below. Inductively coupled plasma mass spectrometry (ICP–MS) (Liu,

Chen, Yang, Chiang, & Hsu, 2007), inductively coupled plasma atomic emission

spectroscopy (ICP–AES) (Mehra & Baker, 2007), these techniques allow a multielement

analysis; however the equipment used is very expensive and also have high operation

coasts. Atomic absorption spectrophotometry (AAS) with flame or electrothermic

atomisation in graphite furnace (Galani-Nikolakaki,Kallithrakas-Kontos, & Katsanos, 2002;

Tamasi & Cini, 2004). The AAS based technique is robust, well establish, easy to use,

and presents good detection and quantification levels, mg/L and µg/L for flame or graphite


59

furnace technique, respectively. The use of voltammetry (Melucci, Torsi, & Locatelli, 2007)

to quantify metals is an inexpensive and fast technique but normally associated with the

use of mercury electrode considered as toxic and environmental unfriendly.

The present study aims to evaluate the contents of Ca, Mg, Na, K, Cu, Fe and Zn in 39

mineral and spring water samples, with and without flavours. Atomic absorption

spectrophotometry with flame or electrothermic atomisation in graphite furnace was the

implemented methodology. A nutritional and statistic study was carried out to compare

these water kinds.

There are no known reports, or any type of evaluation, concerning the mineral contents

of these flavoured waters. So, the presented research work is crucial to consumer’s

information about the advantages/disadvantages of the consumption of these beverages.

2. Materials and methods

2.1. Reagents and equipment

The water used had ultrapure quality (18.2MΩcm-1) and was obtained from a Millipore

Simplicity 185 system.

All reagents and solvents used were suprapure grade and acquired from Merck, except

CsCl that was from Sigma. Standard solutions of each element (Ca, Mg, Na, K, Cu, Fe

and Zn) were daily prepared by dilution of the corresponding stock solutions (1000 mg/L),

with water and 0.1% (v/v) nitric acid, and stored in polyethylene bottles.

Mg(NO3)2 (0.1%; v/v) was used as matrix modifier for the determination of Fe and Zn

and CsCl (0.1%; v/v) to evaluate Na and K contents.

All glassware and polyethylene vessels were soaked with 10% HNO3, at least overnight,

and then rinsed with ultrapure water prior to use.

Macrominerals were quantified in a Perkin Elmer AAnalystTM 200 spectrophotometer

with an air–acetylene flame. Ca, Mg, Na and K were analysed at wavelengths of 422.7,

285.2, 589.6 and 766.5 nm, respectively.

Microminerals (Fe, Zn and Cu) were quantified in an Analytik Jena Zeenit 650

spectrophotometer with electrothermic atomisation in graphite furnace (wavelengths of

248.4, 213.9 and 324.8 nm, respectively) equipped with an Analytik Jena MPE60

autosampler. Pyrolytically coated graphite tubes with integrated pin platform (Analytik

Jena AG) were used. Specific interferences from the matrix were not observed in all

samples and the Zeeman background correction was sufficient. Hollow cathode lamps


60

were used (Varian). A stream of ultrapure argon at 5.5 bar was used in the electrothermic

determinations, except in the Auto-zero and atomisation step.

2.2. Samples and sample preparation

Thirty-nine water samples (flavoured and the natural ones) corresponding to 10 different

brands (mineral and spring) were collected in several supermarkets in the North of

Portugal. Each brand (still or sparkling) had different flavours and aromas.

Table 2 summarises the labelled information, namely the presence of vitamins,

sweeteners and preservatives.

All samples were acidified with suprapure HNO3 (1 mL/L) and stored in sealed

polyethylene bottles maintained at 4 ºC. The gas of sparkling water was removed by

sonication, before HNO3 treatment or acidification.

2.3. Validation of the methodology

Calibration standards were daily prepared (all samples were determined in triplicate).

The proposed methods were validated by linear range, limit of detection (LOD), limit of

quantification (LOQ), precision and accuracy. LOD and LOQ were defined, respectively,

as three and 10 times the standard deviation of 10 blank signals divided by the slope of

the calibration plot (Miller & Miller, 2000). The precision was investigated considering the

intra-day and inter-day determinations of standard solutions and expressed by relative

standard deviations (RSD). For intra-day evaluation each concentration was assessed by

three measurements, at three times along a working-day. The inter-day precision

measurements were performed over a period of one week. Accuracy and reproducibility

were checked by the recovery (REC), the relative error (RE) and the RSD.

2.4. Data analysis

All results were expressed as mean ± standard deviation. In the statistical analysis, data

were presented as median (1st quartile– 3rd quartile). The significance of the differences

between natural waters with and without natural gas was tested by Mann–Whitney test.

Comparisons between natural water group and the respective flavoured water group were

carried out by Wilcoxon test (dependent samples). All statistical analysis was performed

using the Statistica7 software, p < 0.05 was considered statistically significant.


61

Table 2 Label information in bottled flavoured waters evaluated.

Brand Sample Flavour Composition (mg/L) Other ingredients

Still water A (Mineral) 1 Lemon

2 Mango

3 Strawberry

Energy value 1.3 kcal/100 mL, Na+ (<200), Proteins (<1000); carbohydrate (1000), lipids (<1000)

Fibres (1%), Wheat dextrin (0.1%) citric acid, potassium sorbate, sodium benzoate, sodium citrate, acesulfame-K

4 Natural

Total dissolved solids (47), SiO2

(12.7), Ca2+ (0.75), F- (<0.08), NO3

- (1.7), Na+ (6.9), Mg2+ (1.7), Cl- (9.4), HCO3

- (11.6), pH 5.7

B (Spring)

5

Pineapple /orange

Apple juice concentrate, calcium lactate, citric acid, potassium sorbate, sodium benzoate, acesulfame-K, aspartame

6 Lemon

Apple juice concentrate, citric acid, potassium sorbate, sodium benzoate, vitamins: niacin, pantothenic acid, B6, folic acid, biotin, B12, acesulfame-K, aspartame

7 Natural Total dissolved solids (52), SiO2

(21.5), Na+ (<5.5), Ca2+ (<4.3), HCO3

- (<24), Cl- (<30); pH 5.8-6.9

C (Mineral)

8

Lemon/ magnesium

Energy value 13 kcal/100 mL, proteins (<1000), sugar (23000), Sat. fatty acids (<500), fibres (<1000), Na+ (12), Mg2+ (450)

Fruit juice concentrate, citric acid, potassium sorbate, dimethyl dicarbonate, magnesium carbonate, ginseng, vitamins (mg/100 mL): B3 (2.7), B5 (0.9), B6 (0.3), B8 (0.022), B9

(0.03), B12 (1.5x10-4)

9

Apple/ white tea

Energy value 17 kcal/100 mL, proteins (<1000), Sat. fatty acids (<500), Sugar (43000), fibres (<1000), Na+ (14), Ca2+ (1200)

Fruit juice concentrate, calcium lactate, citric acid, malic acid, potassium sorbate, vitamins (mg/100 mL): B3 (2.7), B5 (0.9), B6 (0.3), B8 (0.022), B9 (0.03) and B12 (1.5x10-4)

10

Pineapple /fibre

Energy value 9 kcal/100 mL, proteins (<1000), sugar (23000), Sat. fatty acid (<1000), fibres (9000), Na+ (14)

Fruit juice concentrate, wheat dextrin (0.9%), citric acid, potassium sorbate, dimethyl dicarbonate, L-carnitine (200 mg/L)

11 Natural

Total dissolved solid (45), SiO2

(18), HCO3- (5.1), Cl- (7.4), NO3

-

(2.1), Ca2+ (0.8), Na+ (5.8), Mg2+ (0.5), pH 5.7

D (Mineral) 12 Apple

13 Orange

/peach 14 Lemon

Energy value 0.9 kcal/100 mL

Citric acid, dimethyl dicarbonate, sodium benzoate, sucralose, acesulfame-K

15 Natural

Conductivity (515 µS/cm), HCO3

- (315), SO4

2- (25), Cl- (10), Ca2+ (83), Mg2+ (24), Na+ (4.7)

Sparkling water

E (Mneral) 16 lemon

Added gas 17 orange/ raspberry

18 peach/ pineapple

19 guava

Energy value 1.4 kcal/100 mL, proteins (<1000), carbohydrates, (<100) lipids (<1000)

Lemon juice, carbon dioxide, citric acid, sodium citrate, potassium sorbate, sodium benzoate, acesulfame-K, aspartame, vitamins (mg/100 mL): B3 (2.7), B12 (0.15)

20 natural Total dissolved solids (47), SiO2

(12.7), Ca2+ (0.75), F- (<0.08),


62

Brand Sample Flavour Composition (mg/L) Other ingredients NO3

- (1.7), Na+ (6.9), Mg2+ (1.7), Cl- (9.4), HCO3

- (11.6)

F (Mineral) 21

lemon/ green tea

Natural gas 22 raspberry/ ginseng

23 peach/ white tea

24 mango/ ginkgo biloba

25 melon/ mint

Energy value 19 kcal/100 mL, proteins (1000), sugar (43000), lipids (<500), Na+ (600)

Lemon, apple and pear juice, fibres (<1000), citric acid

26 natural

Total dissolved solids (3011), SiO2

(62), HCO3- (2125), Cl- (31), NO3

- (0.3), Ca2+ (103), Na+ (622), Mg2+ (28), pH 6.1

G (Mineral) 27 lemon

Citric acid, vitamin C (12 mg/250 mL), potassium sorbate, acesulfame-K, sucralose

Added gas 28 lime

29 apple

30 peach

Energy value 0.4 kcal/100 mL

Citric acid, vitamin C (12 mg/250 mL), potassium sorbate, acesulfame-K, sucralose

31 natural

Total dissolved solids, 180ºC (497), Cl- (78), SO4

2- (22), HCO3-

(373), Na+ (37), Ca2+ (105), Mg2+ (29), pH 5.43

H (mineral) 32 lemon Energy value 4 kcal/100 mL proteins (<1000), carbohydrates (10000), lipids (<1000)

Lemon and apple juice, citric acid, vitamin C (30 mg/100 mL), sodium benzoate, potassium sorbate, aspartame

Natural gas

33 natural

Total dissolved solids (2776), SiO2 (37), HCO3

- (1954), Cl- (34), NO3-

(0.3), Ca2+ (75), Na+ (607), Mg2+ (14), pH 6.1

I (Mineral) 34 lemon Energy value 2 kcal/100 mL, carbohydrates (5000)

Lemon and apple juice, citric acid, aspartame, sodium benzoate

Natural gas 35 green apple

Energy value 4 kcal/100 mL, proteins (500), carbohydrates (9000)

Apple juice, citric acid, sodium benzoate, sucralose

36 strawberry

Energy value 2 kcal/100 mL, carbohydrates (5000)

Apple and strawberry juice, citric acid, sodium benzoate, aspartame

37 natural

CO2 free (2260), Total dissolved solids (3535), SIO2 (26.7), Cl- (145), HCO3

- (2393), Na+ (741), Mg2+ (35.2), Ca2+ (116)

J (Spring) 38 lemon

Energy value 4 kcal/100 mL, proteins (<1000), carbohydrates (<1000), lipids (<1000)

Lemon juice, citric acid, sodium citrate, potassium sorbate, sodium benzoate, aspartame, acesulfame-K

Added gas 39 natural

Total dissolved solids (180ºC) (21), SiO2 (8.5), HCO3

- (5.8), Cl- (3.4), Ca2+ (1.2), Mg2+ (0.3), K+ (0.2), Na+ (2.5)


63

3. Results and discussion

Macro (Ca, Mg, K, Na) and microminerals (Fe, Cu, Zn) are essential, in different

amounts, to normal human development. So, an adequate intake from dietary sources is

very important to avoid deleterious effects on human health and general well-being.

Before the evaluation of the mineral content in the samples it was necessary to

implement and validate the employed methodologies.

3.1. Minerals quantification

Table 3 summarises the data from calibration curves and the performance

characteristics for the seven minerals in study. Linearity ranges were from 0.10 to 5.00

mg/L in macrominerals and from 1.0 to 20.0 µg/L in microminerals. The calculated LOD

values ranged from 4.6 to 30.2 µg/L for macrominerals, and 0.21 to 1.7 µg/L for

microminerals. LOQ values range from 15.2 to 100.2 µg/L and 0.7 to 5.5 µg/L for

macrominerals and microminerals, respectively.

Precision and accuracy values are shown in Table 3. No significant differences were

found between intra-day and inter-day experiments. RSD values ranged from 1.0% to

4.3%, and confirmed the high precision of the method. REC and RE values assessed the

accuracy of the results. RE were always <10.0% and recovery trials ranged from 99% to

110%, confirming the accuracy of the implemented method.

64

Tab

le 3

Cal

ibra

tion

curv

es, l

imit

valu

es, p

reci

sion

and

acc

urac

y ob

tain

ed fo

r th

e m

iner

als

stud

ied.

R

EC

(%

) =

[met

al] fo

und/

[met

al] a

dded

x 1

00; R

E (

%)

= ([

met

al] fo

und

– [m

etal

] add

ed)/

[met

al] a

dded

x 1

00; R

SD

(%

) =

σ/[m

etal

] mea

n fo

und

x 10

0

a)A

vera

ge o

f thr

ee m

easu

rem

ents

thre

e tim

es a

long

a d

ay; b)

RE

C, r

ecov

ery;

c)R

E, r

elat

ive

erro

r; d)

RS

D, r

elat

ive

stan

dard

dev

iatio

n

e)A

vera

ge o

f thr

ee m

easu

rem

ents

ove

r a

wee

k.

Par

amet

ers

Ca

Mg

K

Na

Fe

Zn

Cu

Line

ar c

once

ntra

tion

(µg/

L)

100

0.0-

5000

.0

100

.0–4

00.0

4

00.0

–150

0.0

400

.0–1

500.

0 6

.0-2

0.0

2.5

0-20

.0

1.0

-10.

0

Slo

pe (

Abs

µg-1

L)

5.2

±0.2

(x1

0-5

) 9

.3±0

.5 (

x10

-4)

1.9

±0.8

(x1

0-4

) 2

.7±0

.2 (

x10

-4)

0.0

118±

0.00

02

11.

08±0

.04

0.0

14±0

.001

Inte

rcep

t (A

bs)

0.0

10±0

.006

2

.4±0

.1 (

x10

.2)

1.9

±0.7

(x1

0-3

) -

0.02

±0.0

2 1

.18x

10-2

± 0

.02x

10-2

0

.9±0

.4

0.0

03±0

.002

Cor

rela

tion

coef

ficie

nt (

n =

3)

0.9

97

0.9

95

0.9

97

0.9

92

0.9

994

1.0

0 0

.998

1

LOD

(µ

g/L)

30

.2

4.6

2

7.3

17.1

1.

7 0

.59

0.2

1

LOQ

(µ

g/L)

10

0.2

15.

2 9

1.0

57.1

5.

5 1

.90

0.7

Intr

a-da

y st

udie

sa)

Add

ed (

µg/

L):

foun

d (µ

g/L)

1000

.0

1020

.0

300.

0

305.

0

120

0.0

119

0.2

1000

.0

1090

.0

12.0

12.5

20.0

19.7

5.0

5.6

RE

Cb)

(%

) 10

2 10

1.7

99.

2 10

9.0

104.

2 98

.5

112.

0

RE

c) (%

) 2.

0 1.

7 -

0.8

9.0

4.2

-1.5

12

.0

RS

Dd)

(%

) 4.

0 3.

8 1

.0

2.3

3.4

1.8

4.3

Inte

r-da

y st

udie

se)

Add

ed (

µg/

L)

foun

d (µ

g/L)

1000

.0

1015

.0

300.

0

298.

0

1200

.0

1210

.0

1000

.0

1100

.0

12.0

12.3

20.0

20.4

5.0

5.3

RE

C (

%)

101.

5 99

.3

100.

8 11

0.0

102.

5 10

2.0

106.

0

RE

(%

) 1.

5 -0

.7

0.8

10.0

2.

5 2.

0 5.

7

RS

D (

%)

4.3

2.9

3.7

2.8

1.7

2.5

3.8


65

3.2. Global discussion

Table 4 shows the minerals contents of still and sparkling waters, respectively. The

mineral composition of the natural waters (without flavours) changed from brand to brand.

This was due to their different natural origins, from different geological structures.

According to the values described in the label (Table 2) it is possible to discriminate

three groups in natural waters, attending to the total dissolved solids values:

- one with reduced values ranging from 21 to 47 mg/L corresponding to natural waters

4, 7, 11, 20 and 39 (50% of the samples considering the flavoured ones). This group

included all spring waters studied (samples 7 and 39); sample 39 had added gas. The

other samples included in this group were mineral waters and only sample 20 had added

gas;

- another group included two samples with intermediate values (samples 15 and 31)

with total dissolved solids ranging 500 mg/L. This group corresponds to 25% of the total

samples analysed;

- the third group included mineral waters with natural gas and their total dissolved solids

ranged from 2776 to 3535 mg/L (samples 26, 33 and 37).

From all the evaluated samples, the excessive consumption of the mineral waters,

pertaining to this group, with or without flavours, can contribute to the development of

health problems namely kidneys disorders. Considering the values of Table 1, three

bottles of this water (about 1 L) per day correspond to an ingestion of a 1/3 of the RDA for

Na.

Taking into account the macromineral values presented in the label and the determined

ones (Tables 2 and 4), in general, there are in agreement. All label samples described the

contents of Ca, Mg and Na and only one (sample 39) presented K contents. None of them

expressed micromineral contents.

It is important to stress that the samples with high total dissolved solids (the integrated

measure of the concentrations of common ions, Na, K, Ca, Mg, Cl) have a high

contribution of Na, with contents ranging 600 mg/L. This fact reinforces the care needed in

the consumption of this foodstuff, with special attention to children and adults with renal

disorders. Nevertheless, chloride contents are not very high, decreasing the possible

influence in the development or contribution to hypertension.


66

Table 4 Mineral contents in bottled waters.

Macrominerals, mg/L (%) Microminerals, µg/L (%) Brand Sample No. Ca Mg K Na Fe Zn Cu Still A 1 9.1 ± 0.5 3.4 ± 1.0 117.5 ± 0.1 209.1 ± 0.4 18.9 ± 3.4 24.7 ± 5.7 13.3 ± 8.5 2 8.2 ± 0.3 4.0 ± 0.8 107.5 ± 0.5 190.1 ± 2.1 16.5 ± 11.6 16.3 ± 11.0 12.5 ± 0.6 3 8.4 ± 0.5 4.4 ± 2.1 113.7 ± 0.9 151.0 ± 8.6 42.6 ± 3.9 15.3 ± 3.9 12.3 ± 5.8 4 0.8 ± 3.0 1.5 ± 4.0 1.0 ± 0.6 6.7 ± 0.4 5.1 ± 3.2 11.9 ± 8.4 9.1 ± 7.0

B 5 212.8 ± 0.8 0.2 ± 1.2 65.7 ± 0.7 46.1 ± 10.2 108.4 ± 10.2 8.3 ± 0.7 0.7 ± 5.7 6 3.1 ± 0.7 0.3 ± 2.8 137.4 ± 0.4 57.7 ± 6.9 262.9 ± 4.3 7.8 ± 1.3 2.3 ± 1.8 7 1.5 ± 1.4 0.2 ± 1.7 0.5 ± 0.5 4.2 ± 2.2 2.3 ± 15.2 8.6 ± 0.3 -

C 8 1.6 ± 0.5 28.0 ± 2.6 111.8 ± 0.1 16.6 ± 7.3 83.7 ± 2.0 29.4 ± 2.6 1.2 ± 4.1 9 238.8 ± 6.7 33.2 ± 3.3 105.7 ± 0.2 16.8 ± 2.4 87.0 ± 13.9 28.5 ± 2.8 2.3 ± 2.6 10 2.4 ± 1.4 0.4 ± 0.5 94.3 ± 1.2 10.6 ± 5.3 1 9.5 ± 7.7 30.9 ± 2.8 0.5 ± 0.6 11 0.2 ± 0.3 0.4 ± 1.3 2.8 ± 1.0 8.1 ± 0.6 1.8 ± 14.2 22. 5 ± 4.4 -

D 12 69.7 ± 0.6 12.2 ± 0.5 21.5 ± 0.9 35.4 ± 1.3 2.9 ± 33.6 5.8 ± 17.2 - 13 55.1 ± 0.5 10.7 ± 0.8 17.3 ± 0.8 31.9 ± 1.0 4.1 ± 6.9 6.9 ± 1.7 - 14 53.3 ± 1.4 14.3 ± 1.4 15.9 ± 1.6 31.9 ± 2.3 - 9.2 ± 0.6 - 15 69.9 ± 2.9 20.1 ± 1.6 1.1 ± 1.3 4.8 ± 1.3 - 6.5 ± 21.5 -

Sparkling E 16 1.1 ± 2.9 2.1 ± 2.2 246.9 ± 1.0 619.4 ± 1.1 6.3 ± 14.4 11.1 ± 7.2 9.0 ± 0.9 17 2.3 ± 0.7 2.3 ± 4.2 86.26 ± 3.0 245.2 ± 0.8 13.9 ± 12.7 20.4 ± 13.2 7.9 ± 1.9 18 1.1 ± 1.3 2.2 ± 4.2 92.7 ± 0.6 221.7 ± 2.3 35.4 ± 11.8 12.8 ± 10.9 9.6 ± 2.2 19 1.3 ± 1.6 2.2 ± 3.9 96.2 ± 0.2 290.9 ± 0.8 8.9 ± 1.0 13.4 ± 14.9 9.4 ± 1.2 20 0.2 ± 5.2 1.7 ± 1.7 0.5 ± 0.5 8.2 ± 1.8 83. 6 ± 2.5 31.2 ± 10.6 8.7 ± 0.3

F 21 82.5 ± 0.6 20.4 ± 1.5 36.7 ± 0.5 655.6 ± 1. 9 29.2 ± 7.3 12.2 ± 11.5 - 22 82.5 ± 1.4 19.4 ± 2.7 31.8 ± 2.0 618.2 ± 1.5 122.8 ± 1.5 8.3 ± 16.9 1.5 ± 6.9 23 79.5 ± 0.6 20.2 ± 1.2 41.9 ± 1.1 641.6 ± 2.4 9.4 ± 6.4 12.4 ± 25.8 1.4 ± 3.5 24 77.8 ± 1.6 21.3 ± 1.3 25.8 ± 1.1 496.6 ± 2.3 196.8 ± 11.7 10.3 ± 18.4 5.1 ± 3.2 25 89.6 ± 1.2 21.4 ± 2.6 41.8 ± 0.2 667.3 ± 1.8 27.7 ± 3.2 65.9 ± 2.1 1.5 ± 3.1 26 87.8 ± 0.6 20.0 ± 4.0 42.7 ± 1.5 560.1 ± 3.6 4.5 ± 11.5 4.8 ± 25.0 -

G 27 110.8 ± 1.7 15.0 ± 0.5 122.4 ± 1.8 56.9 ± 1.2 20.3 ± 6.4 15.4 ± 4.5 - 28 108.4 ± 0.2 14.4 ± 1.1 6.2 ± 0.5 43.4 ± 0.7 49.2 ± 4.4 6.8 ± 1.5 0.7 ± 2.7 29 112.1 ± 0.4 23.8 ± 0.9 88.7 ± 2.3 45.7 ± 1.4 67.0 ± 4.0 27. 6 ± 15.6 2.0 ± 4.3 30 98.8 ± 1.5 25.9 ± 5.5 78.6 ± 0.2 47.1 ± 1.8 41.0 ± 0.5 5.6 ± 3.6 - 31 108.0 ± 1.1 33.7 ± 7.3 2.2 ± 0.5 43.1 ± 2.4 58.0 ± 4.3 2.8 ± 0.1 -

H 32 58.2 ± 0.3 10.8 ± 12.3 37.1 ± 1.0 603.0 ± 0.9 35.1 ± 5.6 11.4 ± 14.0 -

33 81.6 ± 1.8 10.9 ± 2.0 24.8 ± 2.2 596.8 ± 0.5 25.8 ± 2.2 11.2 ± 21.4 -

I 34 123.0 ± 1.6 28.7 ± 21.7 36.2 ± 4.5 671.3 ± 0.1 39.7 ± 2.5 3.0 ± 0.3 - 35 118.6 ± 0.2 30.1 ± 11.4 39.6 ± 0.4 593.2 ± 1.9 104.4 ± 0.8 6.4 ± 10.9 - 36 127.4 ± 1.1 29.2 ± 3.2 34.5 ± 1.3 555.1 ± 1.3 36.4 ± 2.8 8.3 ± 1.1 0.9 ± 5.3 37 123.0 ± 1.5 24.9 ± 5.2 38.3 ± 0.6 535.1 ± 1.2 8.7 ± 1.2 3.5 ± 0.8 - J 38 3.6 ± 3.6 1.3 ± 1.2 69.2 ± 0.8 359.7 ± 1.4 10.9 ± 2.1 21.7 ± 3.0 7.2 ± 0.7 39 1.1 ± 1.4 0.2 ± 2.6 0.37 ± 1.0 3.7 ± 1.6 20 .7 ± 6.3 24.9 ± 1.4 1.0 ± 4.6

- Not detected.

In what concerns microminerals, several natural water samples have important levels of

Fe; it is the case of samples 20 and 31 (83.6 and 58.0 µg/L) as well as samples 33 and 39

(25.8 and 20.7 µg/L). Samples 4 and 20 presented 9 µg/L of Cu and sample 39 had 1

µg/L. Only these three natural water samples presented detectable Cu values.


67

From Table 2, it was possible to obtain more information, namely about the added

ingredients in the flavoured waters. Amongst them can be cited:

- fibres, that are listed in 11 samples from brands A, C and F;

- fruit juices or concentrates in about 50% of the samples. Only flavoured brands A, D

and G do not refer the addition of this type of ingredient;

- vitamins: 11 samples refer the presence of vitamins of B complex (seven samples) and

C (four samples). According to the label, the added amounts are very different in several

brands, some of them only refer its presence;

- other bioactive compounds, namely ginseng, L-carnitine, white and green tea and

ginkgo biloba; they are present in some flavoured waters from different brands.

Inevitably, these waters also need other ingredients, without positive relation with well-

being and health, but necessary to assure the desired quality for the producer and

consumers, and the safety of the product, such as, acidifying agents, sweeteners and

preservatives.

About 50% of the flavoured samples contained sweeteners as ingredients. There are

samples with only one (acesulfame-K, sucralose or aspartame) and with blends of two

sweeteners (acesulfame- K and aspartame; acesulfame-K and sucralose). The most used

were acesulfame-K (present in 14 samples) and aspartame in 10 samples. It is interesting

to note that, in general, the samples from the same brand have the same sweetener, with

exception of brand I that use different sweeteners for different flavours.

Brands C and F do not use sweeteners, providing more energetic products (9–13 and

19 kcal/100 mL), respectively. In the case of sweetened samples its energy value ranged

from 0.4 to 4 kcal/100 mL.

In what concerns to preservatives and the information contained in the label, each

sample can contain one (potassium sorbate or sodium benzoate) or two preservatives

(potassium sorbate and sodium benzoate; potassium sorbate and dimethyl dicarbonate;

sodium benzoate and dimethyl dicarbonate) simultaneously.

Flavoured waters would not ideally replace natural water, but can be an interesting

alternative to soft drinks which are products with more ingredients with negative influence

in health, namely obesity and generally in children health. Its moderate consumption can

be made with pleasure and without major concerns. It is also important to refer that

flavoured waters are more expensive (20–40%) than natural ones.

Confirming the referred above, except for Fe (p = 0.215), there were statistical

differences in mineral levels between natural waters (with and without natural gas or

added gas) (p < 0.05 for Ca, Mg, K, Na, Zn and Cu).


68

After the presented discussion it is consensual that the minerals contents in flavoured

waters are higher than in natural ones. One justification for that is the use of several

ingredients in the salt form.

3.2.1. Flavour factor analysis

Considering only the flavour factor, the contents of K, Na, Fe and Cu are higher in the

flavoured waters than in the natural ones. For Mg and Ca median concentrations were

slightly higher in the natural waters. However, the difference observed between the

median concentration of the two groups (natural and flavoured) are statistically significant

only for K, Na, Cu (p < 0.001) and Fe (p = 0.045). Fig. 1 shows the median concentration

of the minerals studied in flavoured and natural waters.

Natural water Flavoured water0

20

40

60

80

100

120

Cal

cium

(m

g/L

)

Natural water Flavoured water

0

2

4

6

8

10

12

14

16

18

20

22

24

Mag

nesi

um (

mg/

L)


0

20

40

60

80

100

120

Pota

ssiu

m (

mg/

L)


0

100

200

300

400

500

600

700

Sodi

um (

mg/

L)

Natural water Flavoured water0

10

20

30

40

50

60

70

Iron

(m

cg/L

)


0

1

2

3

4

5

6

7

8

Cop

per

(mcg

/L)


0

2

4

6

8

10

12

14

16

18

20

22

24

Zin

c (m

cg/L

)

Fig. 1. Concentrations of Ca (p = 0.767), Mg (p = 0.456), K (p < 0.001), Na (p < 0.001), Fe (p =

0.045), Cu (p < 0.001) and Zn (p = 0.114) in flavoured waters and correspondent natural water.

The bars represent the median and the whisker represents the inter-quartile range.

3.3. Individual mineral composition

Calcium is the most important macromineral, with amounts ranging from 0.2 to 213

mg/L. Analysing Table 4 it was possible to verify that Ca contents are higher in sparkling


69

waters (except in brands E and J, that have added gas) than in still waters. In the former

group included the samples with the highest content in total dissolved solids and the one

with the lowest contents (brand J). In flavoured still waters, Ca levels increased, except in

brand D. This brand only has preservatives, sweeteners and acidifying agents.

Samples 5 and 9 presented the highest Ca contents (140 and 600 times higher than the

corresponding natural water). In this case, the addition of calcium lactate (Table 2) as

acidifying agent can justify the increased contents. Nevertheless the detected values are

lower than the claimed in the label (Ca 1200).

In sparkling waters Ca levels were kept fairly constant in all samples. As an exception,

in brand H, the flavoured water had lower contents than the natural one. This situation

could not be explained with the available information in Table 2.

Magnesium is a cofactor in almost all phosphorylation reactions involving ATP and is an

indirect antioxidant, being important for the control of the pro-oxidant and antioxidant

status (Lukaski, 2004). Its concentration ranged from 0.2 to 33.2 and 0.2 to 33.7 mg/L in

still and sparkling waters, respectively. Sample 9 had the highest Mg contents (when

compared with the corresponding natural sample). This increment was higher than the

determined in sample 8 with magnesium carbonate incorporation. According to Table 2

this sample should contain 450 mg/L. This claim does not correspond to the actual water

content, and therefore, the consumer is misled when looking for a good source of Mg. Mg

contents have little variation in the water samples evaluated. There are samples with an

increase in the contents and in others a decrease occurs.

Potassium is the most abundant positively charged electrolyte inside cells, being very

important for the muscle contractility, including cardiac muscle (European Food Safety

Authority (EFSA), 2006).

K concentration ranged from 0.5 to 137.4 mg/L and 0.4 to 246.9 mg/L in still and

sparkling waters, respectively. In general, water samples (still and sparkling) presented an

increment in K levels in the flavoured waters that can be explained by the addition of

ingredients in a K salt form, which is the case of some preservatives.

Sodium is the major extracellular electrolyte with functions in nerve conduction, active

transport and formation of the mineral apatite of the bone (WHO/FAO, 2003).

Table 4 points out high Na contents in some water samples, as referred above. In

flavoured waters, both still and sparkling, a significant increase in the Na contents was

verified, comparing to the natural ones. One of the major influential factors is the addition

of sodium benzoate and sodium citrate as preservatives.

Iron is essential for the haemoglobin (oxygen transport), myoglobin, fatty acid, DNA and

neurotransmitters synthesis, in peroxide conversions, in purine metabolism and in the

nitric oxide production (Lukaski, 2004).


70

Fe contents of samples 14 and 15 (Table 4) were lower than the LOD value. Samples 4,

7, 11, 12, 13 and 26 (Table 4) presented levels between LOD and LOQ values.

In the other samples, Fe contents ranged from 16.5 to 262.9 µg/L and 6.3 to 196.8 µg/L

in still and sparkling water, respectively. In samples with added gas, iron contents are

lower than in the natural ones, which is an interesting factor.

Zinc is essential to enzymes function, acting as catalyst or stabilizing protein structure

(Silvera & Rohan, 2007; Zuliani, Kralj, Stibilj, & Milačič, 2005). Zn in excess competes with

the absorption of Cu and Fe. Zn was detected in all samples, in levels ranging from 5.8 to

30.9 µg/L and 2.8 to 65.9 µg/L in still and sparkling waters, respectively. The behaviour

presented is not constant, with increased levels in some samples and decreased in

others, comparing flavoured and natural waters. Sample 25 (melon/mint) presented the

highest contents of Zn and sample 10 (pineapple/fibre) the second higher content. A

possible justification can be the flavour used in the case of sample 25 (brand F). In brand

C all samples had similar contents regardless of flavour contribution.

Copper is an essential cofactor for a variety of enzymes and, like Zn and Fe, is involved

in the regulation of the expression of the genes for the metal-binding proteins (Zuliani et

al., 2005). Deficient intakes can promote breast cancer and cardiovascular diseases. Cu

was not detected in all samples of brands D and H, neither in some natural water

(samples 7, 11, 26, 31 and 37). As shown in Table 4, the presence of flavours can

increase Cu levels in water samples. In brand I, sample 36 (strawberry flavour) is the only

one with a detectable Cu level. In brand G only samples 28 and 29 (lime and apple) had a

detectable Cu level.

3.4. Effects of same labelled compounds in mineral composition

Table 5 shows the results of the statistical analysis considering the following factors:

preservatives, acidifying agents, fruit juice and sweeteners. These ingredients are added

to natural waters and this study aimed to verify its influence in the contents of the macro

and microminerals evaluated. Their influence in each mineral will be appreciated

individually.


71

Table 5 Statistical analysis of minerals content results obtained.

Natural water Flavoured water p Ca (mg/L) Preservatives Potassium sorbate+sodium benzoate (n=12) Potassium sorbate (n=7) Sodium benzoate (n=6) Acidifying agents Citric acid+sodium citrate (n=9) Citric acid+natural flavour (n=8) Citric acid (n=12) Fruit juice Yes (n=19) No (n=10) Sweeteners Yes (n=20) No (n=9)

0.22 (0.22-0.83) 108.00 (0.99-110.25) 96.45 (69.90-123.00) 0.83 (0.22-0.95) 87.80 (87.80-123.00) 69.90 (0.52-108.00) 1.47 (0.22-87.80) 69.90 (0.83-108.00) 35.68 (0.83-108.00) 87.80 (0.21-87.80)

2.25 (1.12-8.17) 103.55 (4.41-116.98) 94.17 (54.62-124.09) 3.07 (1.19-8.31) 86.03 (80.24-121.90) 84.25 (53.73-111.76) 77.75 (2.25-118.60) 62.40 (8.93-108.95) 54.17 (3.21-111.76) 82.45 (40.06-98.98)

0.018* 0.722 0.138 0.008* 0.063 0.754 0.845 0.959 0.494 0.678

Mg (mg/L) Preservatives Potassium sorbate+sodium benzoate (n=12) Potassium sorbate (n=7) Sodium benzoate (n=6) Acidifying agents Citric acid+sodium citrate (n=9) Citric acid+natural flavour (n=8) Citric acid (n=12) Fruit juice Yes (n=19) No (n=10) Sweeteners Yes (n=20) No (n=9)

1.72 (1.52-1.72) 24.87 (0.65-33.70) 22.48 (20.09-24.87) 1.52 (0.88-1.72) 19.99 (19.99-28.87) 20.09 (0.36-33.70) 1.72 (0.36-19.99) 20.09 (1.52-33.70) 6.32 (1.52-24.87) 19.99 (0.36-19.99)

2.20 (2.06-3.37) 19.41 (6.03-28.52) 21.50 (11.82-29.43) 2.20 (1.69-3.08) 21.35 (20.25-29.09) 14.33 (10.69-25.35) 19.41 (2.06-27.95) 13.25 (4.32-17.20) 7.54 (2.17-21.61) 20.42 (16.89-24.69)

0.019* 0.875 0.345 0.007* 0.036* 0.272 0.001* 0.028* 0.881 0.173

K (mg/L) Preservatives Potassium sorbate+sodium benzoate (n=12) Potassium sorbate (n=7) Sodium benzoate (n=6) Acidifying agents Citric acid+sodium citrate (n=9) Citric acid+natural flavour (n=8) Citric acid (n=12) Fruit juice Yes (n=19) No (n=10) Sweeteners Yes (n=20) No (n=9)

0.45 (0.45-1.03) 2.51 (2.21-34.96) 19.72 (1.10-38.34) 0.45 (0.45-1.03) 42.74 (38.34-42.74) 2.21 (1.10-2.81) 2.81 (0.45-42.70) 1.10 (1.03-2.21) 1.07 (0.45-2.21) 42.74 (2.81-42.74)

96.18 (86.26-117.50) 83.65 (36.46-113.24) 28.04 (16.98-37.04) 107.47 (89.47-127.45) 36.45 (32.45-41.24) 72.11 (18.39-102.82) 65.65 (36.70-96.18) 83.65 (16.98-114.68) 82.42 (36.42-112.17) 41.79 (28.77-99.99)

0.018* 0.008* 0.249 0.008* 0.030* 0.002* 0.014* 0.005* <0.001* 0.594

Na (mg/L)

0.017* 0.002* 0.029* 0.007* 0.050* 0.002*

Preservatives Potassium sorbate+sodium benzoate (n=12) Potassium sorbate (n=7) Sodium benzoate (n=6) Acidifying agents Citric acid+sodium citrate (n=9) Citric acid+natural flavour (n=8) Citric acid (n=12) Fruit juice Yes (n=19) No (n=10) Sweeteners Yes (n=20) No (n=9)

8.20 (6.70-8.20) 43.10 (8.10-535.10) 269.95 (4.80-535.10) 6.70 (5.45-8.20) 560.10 (535.10-560.10) 8.10 (4.80-43.10) 8.20 (8.10-560.10) 6.70 (4.80-43.10) 8.20 (4.80-43.10) 560.10 (8.1-560.10)

245.20 (209.00-359.70) 57.30 (43.98-583.68 295.25 (31.90-612.73) 221.70 (170.55-325.30) 629.90 (564.63-664.38) 39.40 (20.58-46.85) 496.60 (57.70-619.40) 46.40 (34.53-160.76) 199.60 (46.35-506.25) 496.60 (16.7-648.60)

0.001* 0.005* <0.001* 0.051


72

Natural water Flavoured water p Fe (µg/L)

Preservatives Potassium sorbate+sodium benzoate (n=12) Potassium sorbate (n=7) Sodium benzoate (n=6) Acidifying agents Citric acid+sodium citrate (n=9) Citric acid+natural flavour (n=8) Citric acid (n=12) Fruit juice Yes (n=19) No (n=10) Sweeteners Yes (n=20) No (n=9)

83.60 (5.10-83.60) 8.60 (3.00-58.00) 4.70 (0.81-8.60) 20.70 (5.10-83.60) 4.50 (4.50-8.60) 2.05 (1.05-58.00) 4.50 (2.30-25.80) 5.10 (0.80-58.00) 8.60 (3.00-58.00) 4.50 (1.80-4.50)

13.90 (8.90-18.90) 41.80 (35.43-79.45) 20.25 (2.30-55.85) 16.50 (9.90-39.00) 38.05 (28.00-118.18) 38.05 (7.95-79.45) 35.40 (13.90-104.30) 19.60 (3.83-44.25) 27.70 (9.40-42.20) 49.2 (23.55-104.90)

0.128 0.050* 0.046* 0.594 0.012* 0.158 0.044* 0.646 0.601 0.015*

Zn (µg/L) Preservatives Potassium sorbate+sodium benzoate (n=12) Potassium sorbate (n=7) Sodium benzoate (n=6) Acidifying agents Citric acid+sodium citrate (n=9) Citric acid+natural flavour (n=8) Citric acid (n=12) Fruit juice Yes (n=19) No (n=10) Sweeteners Yes (n=20) No (n=9)

31.20 (11.90-31.20) 7.35 (2.80-11.73) 6.65 (5.10-8.23) 24.90 (11.90-31.20) 4.80 (4.80-6.00) 7.75 (2.80-19.68) 8.60 (4.80-24.90) 6.35 (2.80-11.90) 8.90 (5.95-21.65) 4.80 (4.80-22.50)

16.30 (13.80-21.70) 9.85 (6.50-24.55) 6.65 (5.00-8.79) 15.30 (11.95-21.05) 9.30 (12.35-6.88) 10.30 (6.83-28.28) 12.20 (8.30-21.70) 12.25 (6.55-18.40 ) 11.25 (7.13-16.08) 12.40 (23.55-30.15)

0.176 0.015* 0.991 0.260 0.043* 0.015* 0.796 0.018* 0.925 0.008*

Cu (µg/L) Preservatives Potassium sorbate+sodium benzoate (n=12) Potassium sorbate (n=7) Sodium benzoate (n=6) Acidifying agents Citric acid+sodium citrate (n=9) Citric acid+natural flavour (n=8) Citric acid (n=12) Fruit juice Yes (n=19) No (n=10) Sweeteners Yes (n=20) No (n=9)

8.65 (8.65-9.08) 0.10 (0.07-0.15) 0.08 (0.06-0.10) 8.65 (4.85-8.08) 6.10 (0.06-0.10) 0.10 (0.10-0.15) 0.10 (0.10-1.04) 0.15 (0.10-9.08) 0.15 (0.10-8.65) 0.10 (0.10-0.10)

9.35 (7.95-12.45) 0.61 (0.10-1.80) 0.10 (0.09-0.29) 9.35 (7.56-12.35) 1.13 (0.10-1.51) 0.33 (0.10-1.07) 1.46 (0.50-7.17) 0.41 (0.10-12.30) 1.43 (0.10-9.27) 1.39 (0.61-1.93)

0.051 0.018* 0.285 0.018* 0.026* 0.028* 0.001* 0.063 0.005* 0.012*

*Statistically significant p<0.05. Data are presented was median (1st quartile–3rd quartile).

3.4.1. Calcium

Taking into account the preservatives added, it was observed that the blend of

potassium sorbate and sodium benzoate lead to significant statistical differences (p =

0.018). For the remaining preservatives the differences observed are not significant (p >

0.05).

In what concerns acidifying agents, only the waters with, simultaneously, citric acid and

sodium citrate presented statistically significant differences from its natural corresponding


73

water (p = 0.008). The addition of fruit juice (p = 0.845) as well as sweeteners does not

influence Ca concentration.

3.4.2. Magnesium

In the case of Mg the statistical study showed that the addition of preservatives

(potassium sorbate and sodium benzoate) (p = 0.019) and acidifying agents increased

significantly Mg concentration. Regarding fruit juice, the statistical values are p = 0.001

and 0.028 for flavoured waters with and without fruit juice, respectively.

No other factor affected Mg concentration in a significant way.

3.4.3. Potassium

The addition of preservatives, only one (potassium sorbate) or in combination

(potassium sorbate and sodium benzoate) as well as all acidifying agents, increased

significantly K levels (p = 0.008 and 0.018, respectively) (Table 5).

Regarding the presence or absence of juice, there is also a significant statistical

difference in K concentration between flavoured and natural waters. The addition of

sweeteners also increased significantly K concentration (p < 0.001).

3.4.4. Sodium

Statistical analysis showed that the influence of numerous factors led to significant

differences in the results. This can be seen in Table 5.

3.4.5. Iron

The presence of potassium sorbate or sodium benzoate, as preservatives, induced

significant differences in Fe contents, as well as the presence of citric acid and natural

flavours. Also, significant differences, in Fe content, can be verified amongst waters

without fruit juice or sweeteners.

3.4.6. Zinc

The statistical analysis for the influence in Zn concentration showed that the major

influence came from potassium sorbate and citric acid (Table 5). As referred for Fe,


74

significant differences amongst waters without fruit juice and sweeteners could also be

noticed.

3.4.7. Copper

Regarding the influence of the considered factors in Cu concentration, the ones that

caused statistical significant differences were: potassium sorbate (in the preservative

group), all acidifying agents and the presence of fruit juice. The presence of sweeteners

influenced the levels of this micromineral in all samples.

4. Conclusion

This study leads to conclude that flavoured waters can be an adequate alternative to

consumers that do not like natural water. The different ingredients added to natural waters

hardly influence its mineral composition. All consumers are advised to read the label

content, in order to avoid some health problems that can occur with some mineral waters

and some specific groups of consumers. Also, flavoured waters could represent

advantages due to the presence of certain minerals, some vitamins, antioxidants and

bioactive compounds. Some preservatives, acidifying agents and sweeteners are not

hazardous if consumed with moderation.

Acknowledgement

M. Fatima Barroso is grateful to Fundacao para a Ciencia e a Tecnologia for the Ph.D.

Grant (SFRH/BD/29440/2006).


75

References

Biziuk, M., & Kuczynska, J. (2007). Mineral Components in Foods. Taylor & Francis, cap 1.

Environmental Protection Agency of United States (EPA). (1993). Reference Dose (RfD):

Description and Use in Health Risk Assessments Background Document 1A. Integrated

Risk Information System (IRIS). Available from: <http://www.epa.gov/iris/rfd.htm>

Environmental Protection Agency of United States (EPA). (2005). Zinc and compounds,

Integrated Risk Information System (IRIS). Available from:

<http://www.epa.gov/IRIS/subst/0426.htm>.

European Food Safety Authority (EFSA). (2006). Tolerable upper intake levels for

vitamins and minerals. Scientific Committee on Food, Scientific Panel on Dietetic

Products, Nutrition and Allergies. Available from:

<http://www.efsa.europa.eu/EFSA/efsa_locale-1178620753812_1178633962601.htm>.

FAO/WHO. (2000). Codex committee on nutrition and foods for special dietary uses. In

22nd Session of JOINT FAO/WHO food standards programme. Berlin, Germany.

Available from: <http://www.codexalimentarius.net/>.

Food and Drug Administration (FDA). (2002). Bottled Water Regulation and the FDA.

Center for Food Safety and Applied Nutrition. Available from:

<http://www.cfsan.fda.gov/~dms/botwatr.html>.

Galani-Nikolakaki, S., Kallithrakas-Kontos, N., & Katsanos, A. A. (2002). Trace element

analysis of Cretan wines and wine products. Science of the Total Environment, 285,

155-163.

Goldhaber, S. B. (2003). Trace element risk assessment: essentiality vs. toxicity.

Regulatory Toxicology and Pharmacology, 38, 232-242.

Institute of Medicine from United States. (2001). Dietary Reference Intakes (DRIs):

Elements, Food and Nutrition Board, Institute of Medicine, National Academies.

Available from: <http://www.iom.edu/Object.File/Master/7/294/Webtableminerals.pdf>.

Institute of Medicine from United States. (2004). Dietary Reference Intakes: Electrolytes

and Water, Food and Nutrition Board, Institute of Medicine, National Academies.

Available from:

<http://www.iom.edu/Object.File/Master/20/004/Electrolytes%20Table%20for%20web.p

df>.

Institute of Medicine from United States. (2007). DRI values definitions. Food and Nutrition

Board, Institute of Medicine, National Academies. Available from:

<http://www.iom.edu/CMS/3788/4574/45105.aspx>.


76

Liu, C. L., Chen, Y. S., Yang, J. H., Chiang, B. H., & Hsu, C. K. (2007). Trace element

water improves the antioxidant activity of buckwheat (Fagopyrum esculentum Moench)

Sprouts. Journal of Agricultural and Food Chemistry, 55, 8934-8940.

Lukaski, H. C. (2004). Vitamin and Mineral Status: Effects on physical performance.

Nutrition, 20, 632-644.

Mehra, A., & Baker, C.L. (2007). Leaching and bioavailability of aluminium, copper and

manganese from tea (Camellia sinensis). Food Chemistry, 100, 1456-1463.

Melucci, D., Torsi, G., & Locatelli, C. (2007). Analytical procedure for the simultaneous

voltammetric determination of trace metals in food and environmental matrices. Critical

comparison with atomic absorption spectroscopic measurements. Annali di Chimica, 97,

141-151.

Miller, J. N., & Miller, J. C. (2000). Statistics and chemometrics for analytical chemistry.

Pearson Education, London.

Nasreddine, L., Parent-Massin, D. (2002). Food contamination by metals and pesticides in

the European Union. Should we Worry?. Toxicology Letters, 127, 29-41.

Robson, M. (2003). Methodologies for assessing exposures to metals: human host

factors. Ecotoxicology and Environmental Safety, 56, 104-109.

Silvera, S. A. N., & Rohan, T. E. (2007). Trace elements and cancer risk: a review of the

epidemiologic evidence. Cancer Causes Control, 18, 7-27.

Tamasi, G., & Cini, R. (2004). Heavy metals in drinking waters from Mount Amiata

(Tuscany, Italy). Possible risks from arsenic for public health in the Province of Siena.

Science of the Total Environment, 327, 41-51.

World Health Organization (WHO). (1982a). In 26th meeting of joint FAO/WHO expert

committee on food additives, WHO food additives series no. 17, Geneva. Available

from: <http://www.inchem.org/documents/jecfa/jecmono/v17je33.htm>.

World Health Organization (WHO). (1982b). In 26th meeting of joint FAO/WHO expert

committee on food additives, WHO food additives series no. 17, Geneva. Available

from: <http://www.inchem.org/documents/jecfa/jecmono/v17je31.htm>.

World Health Organization (WHO). (1983). Evaluation of certain food additives and

contaminants. 26th report of the joint FAO/WHO expert committee on food additives,

WHO food additives series no. 696, Geneva. Available from:

<http://www.inchem.org/documents/jecfa/jecmono/v18je18.htm>.

WHO/FAO. (2003). Diet, nutrition and the prevention of chronic diseases. WHO Technical

Report Series 916. Geneva.

Zuliani, T., Kralj, B. L., Stibilj, V., Milačič, R. (2005). Minerals and trace elements in food

commonly consumed in Slovenia. Italian Journal of Food Science, 17(2), 155-166.

2.2. Minerais vestigiais

77

Survey of trace elements (Al, As, Cd, Cr, Co, Hg, M n, Ni, Pb, Se,

and Si) in retail samples of flavoured and bottled waters

M.F. Barrosoab, S. Ramosc, M.T. Oliva-Telesa, C. Delerue-Matosa, M.G.F. Salesa and

M.B.P.P. Oliveirab aRequimte/Instituto Superior de Engenharia do Porto, Rua Dr. António Bernardino de

Almeida 431, 4200-072, Porto, Portugal. bRequimte, Serviço de Bromatologia, Faculdade de Farmácia, Universidade do Porto,

Rua Aníbal Cunha, 164, P-4099-030, Porto, Portugal. cInstituto Superior de Engenharia do Porto, Rua Dr. António Bernardino de Almeida 431,

4200-072, Porto, Portugal.

Abstract

Concentrations of eleven trace elements (Al, As, Cd, Cr, Co, Hg, Mn, Ni, Pb, Se, and

Si) were measured in 39 (natural and flavoured) water samples. Determinations were

performed using graphite furnace electrothermetry for almost all elements (Al, As, Cd, Cr,

Co, Mn, Ni, Pb, and Si). For Se determination hydride generation was used, and cold

vapour generation for Hg. These techniques were coupled to atomic absorption

spectrophotometry. The trace element content of still or sparkling natural waters changed

from brand to brand. Significant differences between natural still and natural sparkling

waters (p < 0.001) were only apparent for Mn. The Mann–Whitney U-test was used to

search for significant differences between flavoured and natural waters. The concentration

of each element was compared with the presence of flavours, preservatives, acidifying

agents, fruit juice and/or sweeteners, according to the labelled composition. It was shown

that flavoured waters generally increase the trace element content. The addition of

preservatives and acidifying regulators had a significant influence on Mn, Co, As and Si

contents (p < 0.05). Fruit juice can also be correlated to the increase of Co and As.

Sweeteners did not provide any significant difference in Mn, Co, Se and Si content.

Keywords : health significance; metals analysis – atomic absorption spectrometry (AAS);

method validation; statistical analysis; heavy metals – arsenic; heavy metals – cadmium;

heavy metals – mercury; metals – nutritional; beverages; drinking water; water

Available online at www.inforworld.com

Food additives and Contaminants

2009, (2) 2, 121-130


79

Introduction

Trace elements are required in small quantities (µg kg-1 body weight) in the human diet

to maintain normal physiological functions (Goldhaber 2003; Silvera and Rohan 2007).

They are found naturally in the environment, and human exposure is derived from a

variety of sources, including the air, food and drinking water (Silvera and Rohan 2007).

Some trace elements have no known beneficial biological function in humans, but

generally these elements have an impact on human health in many ways (Santos et al.

2004). In several cases, slightly high exposure may be harmful to human health (Ysart et

al. 2000). Some trace elements, such as mercury (Hg), lead (Pb), arsenic (As), cadmium

(Cd), aluminium (Al), chromium (Cr) and nickel (Ni), produce a variety of toxic effects

when intake exceeds the established limit (World Health Organization (WHO) 1996;

Berdanier 1998; Chou and De Rosa 2003; Aziz et al. 2006; Diawara et al. 2006; Mehra

and Baker 2007; Barton 2008; Jorhem et al. 2008; Mor and Ceylan 2008). On the other

hand, some nutritional problems are related to low or no intake of some elements, like

selenium (Se), cobalt (Co) and manganese (Mn) (Magnuson et al. 1997; Barceloux 1999;

Institute of Medicine (IOM) of National Academies 2001; Goldhaber 2003; Alinnor 2005).

Almost all trace elements can be found in bones and teeth (Berdanier 1998).

The intake of trace elements by consumers depends on their content in foodstuffs, the

consumer’s drink/beverage eating habits, and of course the kind of drinking water. The

risks to health of the presence of certain trace elements in water can be assessed by

comparing estimates of dietary exposures with the provisional tolerable weekly intake

(PTWI) recommended by the Joint Expert Committee on Food Additives (JECFA) of the

Food and Agricultural Organization of the United Nations and the World Health

Organization (FAO/WHO). PTWI values range from 0.005 to 0.025 mg kg-1 body weight

week-1 for As, Cd, Hg and Pb. Consumers can also use the recommended dietary

allowance (RDA) from the US Food and Nutrition Board of the Institute of Medicine

(FNB/IOM). The RDA ranges between 25 and 1800 µg day-1 for the trace elements Cr, Se

and Mn (IOM 2001; WHO 2008).

The results presented in this paper focus on the determination of eleven trace elements

(Al, As, Pb, Cd, Co, Cr, Hg, Mn, Ni, Se and Si) in 39 mineral or spring bottled water

samples (with and without flavours) from the market in Portugal. Trace element

concentrations in water depend on soil characteristics, such as organic matter content,

pH, clay mineralogy, and physical and chemical forms in which they are dispersed, which

can affect the bioavailability of these elements (Santos et al. 2004). Flavoured water

consists of water with the addition of flavours, juices, bioactive compounds, preservatives

and/or sweeteners that provide a characteristic taste and odour appreciated by


80

consumers. Mineral and spring waters with important mineral contents are used in the

case of flavoured waters. According to the US Food Drug and Administration (USFDA)

(2002), mineral water arises from a geologically and physically protected underground

source, characterized by constant levels and relative proportions of minerals and trace

elements at the source. Spring water is derived from an underground formation from

which water flows naturally to the surface at an identified location.

In this study atomic absorption spectrometry (AAS) was used with electrothermic

atomization in graphite furnace and hydride generation for the quantification of the

selected trace elements. A nutritional and statistical study was carried out to compare

these different kinds of water. The knowledge of trace element concentrations in waters

can provide important information on the impact of the use of these beverages. Until now,

in this respect no published studies were found.

Material and methods

Apparatus

For the quantification of almost all trace elements (Al, As, Pb, Cd, Co, Cr, Mn, Ni and Si)

an Analytik Jena Zeenit 650 spetrophotometer was used with electrothermic atomization

in a graphite furnace equipped with an Analytik Jena MPE60 autosampler. Pyrolytically

coated graphite tubes with integrated pin platform were used. Specific interferences from

the matrix were not observed in all samples and the Zeeman background correction was

sufficient. Hollow cathode lamps were used (Varian). A stream of ultrapure Ar at 5.5 bar

was used in the electrothermic determinations.

Hydride generation atomic absorption spectroscopy (HS60; Analityk Jena) was used for

Se and Hg quantifications. In order to determine the total content of Se, a preliminary

reduction of Se(VI) to Se(IV) is needed because only Se(IV) is hydride active (Magnuson

et al. 1997). The efficiency of the reduction depends on the temperature, reduction time

and HCl concentration. The resulting Se hydride was drawn off under inert atmosphere

(Ar) into a quartz T-tube at 960ºC and the determinations were performed at 196.0 nm. Hg

was determined by cold vapour generation at room temperature at 253.7 nm.

The optimization of analytical conditions for the quantification of trace elements was

based on the Standard Methods for the Examination of Water and Wastewater (American

Public Health Association (APHA), 1995).


81

Reagents

The water used was ultrapure quality (18.2MΩ cm-1) and obtained from a Millipore

Simplicity 185 system.

All chemicals used for the analytical determinations were supra-pure grade and

acquired from Merck. Standard solutions of each element (Al, As, Pb, Cd, Co, Cr, Hg, Mn,

Ni, Se and Si) were prepared daily by dilution of the stock solutions (1000 mg l-1) with

water and 0.1% (v/v) nitric acid and stored in polyethylene bottles.

All glassware and polyethylene vessels were soaked with 10% HNO3 at least overnight

and then rinsed with ultrapure water before use.

Samples collection and pre-treatment

Thirty-nine water samples corresponding to ten different brands (mineral and spring)

were collected in several supermarkets in Portugal. Each brand (still or sparkling) had

different flavours and aromas. Table 1 summarizes the nutrient information on the labels,

namely the presence of vitamins, sweeteners and preservatives.

All samples were acidified with supra-pure HNO3 (1 ml l-1) and stored in sealed

polyethylene bottles maintained at 4ºC. In the sparkling waters, gas was removed by

sonication before HNO3 conservation and storage.

Matrix modifiers were used in order to eliminate the interference of the water sample

matrix. Mg(NO3)2 (0.1%; v/v) was used for the analysis of Al, Cr and Co. For the As

quantification Pd(NO3)2 (0.1%; v/v) was used NH4H2PO4 (0.1%; v/v) was used for Cd and

Pb analyses. A mixture of Mg(NO3)2 and Pd(NO3)2 was utilized for Si determinations.

For the evaluation of Se, water samples were acidified with HCl 6 mol l-1 and heated at

about 100ºC during 45 min. After this treatment the samples could be subjected to hydride

generation that was performed with a 0.3% sodium borohydride (NaBH4) in 0.1% NaOH

and acidic medium (HCl 3%).

To analyse Hg, all water samples were digested with HNO3 and HCl followed by

reduction to elementary mercury vapour by SnCl2 (Akagi et al. 2000; Ohno et al. 2007).


82

Table 1. Label information in bottled flavoured waters evaluated.

Brand Sample Flavour Other ingredients Still water

A Mineral 1 Lemon 2 Mango

3 Strawberry

preservatives (potassium sorbate, sodium benzoate) acidity regulators ( acid citric, sodium citrate), sweeteners

4 Natural

B Spring 5 Pineapple /orange Apple juice concentrate, preservatives (potassium sorbate, sodium benzoate) acidity regulators (citric acid) and sweeteners

6 Lemon

Apple juice concentrate, preservatives (potassium sorbate, sodium benzoate) acidity regulators (citric acid), sweeteners, added vitamins

7 Natural

C Mineral 8 Lemon/magnesium

9 Apple/ white tea

10 Pineapple/fibre

Fruit juice concentrate, preservative (potassium sorbate), acidity regulator (citric acid), added vitamins

11 Natural

D Mineral 12 Apple

13 Orange/peach

14 Lemon

Preservatives (sodium benzoate), acidity regulator (citric acid), sweeteners

15 Natural

Sparkling water

E Mineral Added gas

16

Lemon

17 Orange/raspberry

18 Peach/pineapple

19 Guava

Lemon juice, preservatives (potassium sorbate, sodium benzoate), acidity regulator (citric acid, sodium citrate) and sweeteners

20 Natural

F Mineral Natural gas

21

Lemon/green tea

22 Raspberry/ginseng

23 Peach/white tea

24 Mango/ginkgo beloba

25 Melon/mint

Lemon, apple and pear juice, acidity regulator (citric acid)

26 Natural


83

Brand Sample Flavour Other ingredients

G Mineral Added gas

27

Lemon

Preservatives (potassium sorbate), regulator acidity (citric acid) added vitamin, sweeteners

28 Lime

29 Apple

30 Peach

Preservatives (potassium sorbate), regulator acidity (citric acid) added vitamin, sweeteners

31 Natural

H Mineral Natural gas

32

Lemon

Lemon and apple juice, preservatives (sodium benzoate, potassium sorbate) acidity regulator (citric acid), added vitamins, sweeteners

33 Natural

I Mineral Natural gás

34

Lemon

Lemon and apple juice, preservative (sodium benzoate), acidity regulator (citric acid), sweeteners

35

Green Apple

Apple juice, preservative (sodium benzoate), acidity regulator (citric acid), sweeteners

36

Strawberry

Apple and strawberry juice, preservative (sodium benzoate), acidity regulator (citric acid), sweeteners

37 Natural

J Spring Added gas

38

Lemon

Lemon juice, preservative (sodium benzoate, potassium sorbate), acidity regulator (citric acid, sodium citrate), sweeteners

39 Natural

Note: Natural waters have no added ingredients.

Analytical quality assurance

Triplicate determinations were made on all samples. Validation of the proposed methods

was evaluated by linearity range, limit of detection (LOD), limit of quantification (LOQ),

precision and accuracy. The LOD is defined as 3 σ s-1 and the LOQ 10 σ s-1, where σ is

the standard deviation of the blank signal (n = 20) and s is the slope of the calibration plot

(Miller and Miller 2000). The precision of the proposed methods was investigated by intra-

and inter-day determinations of standard solutions and expressed by relative standard

deviations (RSD). For intra-day studies, each concentration was assessed by performing

three repeated measurements three times during a working day. The inter-day

measurements studies were performed over 1 week. Accuracy and reproducibility of

methods were checked by recoveries.

Statistical analysis

All results are expressed as mean ± standard deviation (SD). In the statistical analysis

data were presented as median (first quartile – third quartile). The significance of the


84

differences between natural still and natural sparkling waters was tested by a Mann–

Whitney U-test. Comparisons between natural water group and the respective flavoured

water group were carried out by a Wilcoxon test (dependent samples). All graphical and

inferential statistical analysis was performed using the Statistica7 software. p ≤ 0.05 was

considered as being statistically significant.

Results and discussion

Trace elements quantification

Table 2 summarizes the data obtained for calibration curves and the performance

characteristics of the methods optimized for all elements studied. The linearity range was

from 0.20 to 200.0 µg l-1.The calculated LOD values ranged from 0.005 to 3.70 µg l-1 and

LOQ values ranged from 0.018 to 12.34 µg l-1. No significant differences were found

between intra- and inter-day experiments. RSD values ranged from 0.5% to 9.6%,

showing a high degree of precision. Recoveries were determined and all values ranged

from 99% to 110%, showing good accuracy.

Table 2. linear range, limit values and precision obtained for the trace elements studied.

Notes: a)Average of three measurements, three times along a day. b)RSD: relative standard deviation.c)Average of three

measurements over a week.

Label information

Figure 1 shows the median concentration of the trace elements analysed in flavoured

and natural waters. The mineral composition of natural waters (without flavours) depends

of the brand, and the different natural waters origins, namely different geological

structures. Labelling only indicated the concentrations of macrominerals (Ca2+, Na+, Mg2+,

K+) and some anions (NO3-, Cl-, HCO3

-). No information is given concerning trace element

Parameters Al As Cd Cr Co Hg Mn Ni Pb Se Si

Linear concentration

(µg l-1)

16.00-

40.00

1.00–

10.00

0.20–

1.00

1.00–

10.00

2.00–

10.00

0.10–

10.00

1.00–

5.00

2.00–

10.00

2.00–

10.00

0.200–

1.00

20.00–

200.00

LOD ( µg l-1) 0.73 0.11 0.005 0.04 0.11 0.02 0.13 0.07 0.49 0.04 3.70

LOQ ( µg l-1) 2.45 0.40 0.018 0.14 0.35 0.05 0.45 0.23 1.63 0.15 12.34

Intra-day studiesa)

RSDb) (%) 9.60 3.94 0.56 2.02 3.66 0.48 2.63 2.61 3.28 2.41 2.41

Inter-day studiesc)

RSDb) (%) 3.80 1.24 4.05 2.99 2.50 5.20 1.87 3.76 4.27 5.64 5.64


85

contents, with the exception of some natural waters that reported Si contents. Si values

determined in this work were similar to those described in the labels.

Natural water Flavoured water0,0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

Hg

(ug

l-1)

p = 0.310


20,0

40,0

60,0

80,0

100,0

120,0

140,0

160,0

Mn

(ug

l-1)

p = 0.458


0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

Hg

(ug

l-1)

p = 0.310


20,0

40,0

60,0

80,0

100,0

120,0

140,0

160,0

Mn

(ug

l-1)

p = 0.458


20,0

40,0

60,0

80,0

100,0

120,0

140,0

160,0

Al (

ug l-1

)

p = 0.024


0,2

0,4

0,6

0,8

1,0

1,2

1,4

1,6

1,8

2,0

As

(ug

l-1)

p = 0.001


20,0

40,0

60,0

80,0

100,0

120,0

140,0

160,0

Al (

ug l-1

)

p = 0.024


0,2

0,4

0,6

0,8

1,0

1,2

1,4

1,6

1,8

2,0

As

(ug

l-1)

p = 0.001


0,2

0,4

0,6

0,8

1,0

1,2

1,4

1,6

1,8

2,0

Cr

(ug

l-1)

p = 0.709


0,2

0,4

0,6

0,8

1,0

1,2

1,4

1,6

Co

(ug

l-1)

p = 0.064


0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

Hg

(ug

l-1)

p = 0.310


20,0

40,0

60,0

80,0

100,0

120,0

140,0

160,0

Mn

(ug

l-1)

p = 0.458


0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

Hg

(ug

l-1)

p = 0.310


20,0

40,0

60,0

80,0

100,0

120,0

140,0

160,0

Mn

(ug

l-1)

p = 0.458


86

Figure 1. Al, As, Cr,Co, Hg, Mn, Ni, Pb, Se and Si concentrations in flavoured and correspondent natural

eaters. Bars represent the median and whiskers represent the inter-quartile range.

Several natural water samples showed important levels of Al, being the case for

samples 33 and 37 (441 and 158 µg l-1). Ni, Pb and Se were only detected in one (sample

20), three (samples 4, 20 and 39) and three (samples 20, 21 and 31) natural water

samples, respectively. Hg was detected in 50% of natural waters (in five samples). Cd

was not detected in all natural waters.

Except for manganese, there were no statistical differences in trace element

concentrations between natural still and natural sparkling waters (p < 0.001). From the

bottle labels (Table 1) it was possible to obtain more information, namely about the added


0,1

0,2

0,3

0,4

0,5

Ni (

ug l-1

)

p = 0.197


0,2

0,4

0,6

0,8

1,0

1,2

1,4

Pb (

ug l-1

)

p = 0.225


0,1

0,2

0,3

0,4

0,5

Ni (

ug l-1

)

p = 0.197


0,2

0,4

0,6

0,8

1,0

1,2

1,4

Pb (

ug l-1

)

p = 0.225


0,05

0,10

0,15

0,20

0,25

0,30

0,35

Se (

ug l-1

)

p = 0.237


10,0

20,0

30,0

40,0

50,0

Si (

ug l-1

)

p = 0.299


0,05

0,10

0,15

0,20

0,25

0,30

0,35

Se (

ug l-1

)

p = 0.237


10,0

20,0

30,0

40,0

50,0

Si (

ug l-1

)

p = 0.299


87

ingredients to the flavoured waters. Among them can be cited fibres, fruit juices and

vitamins.

Inevitably, these waters also need ingredients, other than those added to provide health

and well-being claims, but necessary to assure the quality desired for the producer and

consumers and safety of the product such as acidifying agents, sweeteners and

preservatives.

About 50% of the flavoured water samples contained sweeteners as ingredients. There

were samples with only one (acesulfame-K, sucralose or aspartame) and with blends of

two sweeteners (acesulfame-K and aspartame; acesulfame-K and sucralose). The most

used was acesulfame-K (present in 14 samples) and aspartame (ten samples). It is

interesting to note that, in general, the samples from the same brand have the same

sweetener, with the exception of band I.

Concerning the preservatives, each sample contained one (potassium sorbate or

sodium benzoate) or two (potassium sorbate and sodium benzoate). Flavoured waters

would not ideally replace natural water, but can be an interesting alternative to soft drinks.

Flavour factor analysis

Considering only the flavour factor, the contents on Cr, Co and Al were higher in the

flavoured waters than in the natural ones. For Mn, As, Pb, Se, and Si the median

concentrations were slightly higher in the natural waters. Ni and Hg when present had

similar concentrations in both groups (Figure 1). However, the differences observed

between the median concentration of the two groups (natural and flavoured) were

statistically significant only for As (p = 0.001) and Al (p = 0.024).

Considering that Cd was not detected in any natural water samples, it was not possible

to perform any statistical analysis.

Discussion of individual trace element composition

Aluminium (Al) was detected in all 39 water samples. Its concentrations ranged from 5.8

to 143.5 µg l-1 (still waters) and from 7.5 to 441 µg l-1 (sparkling waters). In general,

flavoured waters contain Al concentration higher than natural ones, probably due to the

addition of the indicated ingredients. Nookabkaew et al. (2006) and Mehra and Baker

(2007) detected Al contents in tea samples ranging from 1878 to 16 520 µg l-1 and from

458 to 1307 mg kg-1, respectively. Theses values are much higher than those found in

flavoured waters. A high Al content in tea samples is not surprising, because the tea plant

is known to be an Al accumulator (Mehra and Baker 2007). Another paper reported Al


88

concentrations in drinking waters ranging from 3.9 to 16.4 µg l-1 (Tamasi and Cini 2004),

values that are similar to those found in two samples analysed (samples 15 and 20). The

other samples had values several fold higher.

As occurs naturally in soil and in many kinds of rock, especially in minerals that contain

lead and copper (Chou and De Rosa 2003) and it is a natural constituent of water. No As

was detected in 20 samples analysed (five still and 15 sparkling). Some brands (D, F, H

and I) have no As detectable content in all samples analysed (natural and flavoured). In

other brands some flavoured samples have lost the As content detected in the

corresponding natural water. Six samples had values between the LOD and the LOQ and

13 samples had As levels between 0.45 and 2.63 µg l-1.

Drinking water is one of the most important sources of As exposure. There are many

reports of chronic arsenism resulting from drinking water containing high levels of As in

endemic areas worldwide (Chou and De Rosa 2003). Analysis of tea samples indicated

As levels ranging from 0.2 to 1.5 µg l-1 (Nookabkaew et al. 2006). These values are in

agreement with those found in the flavoured waters studied.

Cd is widely distributed in the environment being a contaminant of soil, air, water, plants,

and food supplies. Unless contamination has occurred, the levels of Cd in most foods are

normally very low. In this work only four water samples contained Cd (all flavoured) with

values around 0.1 µg l-1. Similar results were found in drinking water and teas (Tamasi

and Cini 2004; Nookabkaew et al. 2006).

Cr is often found in soil and groundwater of abandoned industrial plants. In the waters

studied no Cr was found in five samples (four still and one sparkling). Only one sample

(number 12) was between the LOD and the LOQ values. Cr values ranged from 0.13 to

5.42 µg l-1. The higher values of Cr were found in sparkling flavoured waters, samples 19

(5.42 µg l-1) and 39 (2.13 µg l-1). Cr levels increased in all still flavoured waters, probably

due to the addition of flavours and other ingredients. Almost all flavoured sparkling brand

samples had a lower Cr concentration than the natural waters. Brand H showed the same

Cr value for the two samples analysed and only in brand B the Cr content increased in

waters with flavour. Comparing these values with those obtained by Nookabkaew et al.

(2006), where Cr ranged from 3 to 14 µg l-1, it is easy to verify that flavoured waters had

lower values of Cr.

Co was not detected in ten samples (three still, seven sparkling) and the values from

nine samples were between the LOD and the LOQ values. The other samples showed a

Co concentration ranging from 0.17 to 2.2 µg l-1. In all still flavoured waters the Co

contents increased. In sparkling flavoured waters some samples had increased levels,

and in others a decrease occurred without any obvious explanation. These results are


89

similar to those obtained by other authors (Tamasi and Cini 2004; Nookabkaew et al.

2006) in drinking waters.

The persistence in the environment, bioaccumulation, transport in the aquatic chain and

presence in a variety of foods make Hg among the most dangerous of all metals in the

human food chain. The widespread use of Hg and its derivatives in industry and

agriculture (now banned in most countries) had resulted in serious environmental

pollution. Only six of 24 sparkling samples contained Hg (from 0.06 to 0.3 µg l-1). Still

waters had Hg levels higher than sparkling ones. Two natural waters had contents of 0.8

and 0.67 µg l-1. Similar values were also reported in other studies with herbal tea and

drinking water (Alinnor 2005; Nookabkaew et al. 2006).

Mn occurs in the environment and is a natural water component. In the samples studied

Mn was not detected in six still water samples. The other samples had Mn levels ranging

from 0.28 to 1.22 µg l-1 in still water and from 0.77 to 236 µg l-1 in sparkling waters. It was

shown that sparkling natural water with natural gas had a higher Mn content (about 100

time higher) than still/added gas natural water. In general, the addition of flavours and

other ingredients increased the Mn concentration. This behaviour may be due to different

physiological properties or structures of flavours/aromas, levels of phytochelating

phenolics and other mineral-binding components present in the final product. Similar

behaviour can be found with respect to other trace elements. Several tea samples had

higher values of Mn than the flavoured waters studied. In tea samples, Mn concentration

ranges from 483 to 2766 µg l-1 (Nookabkaew et al. 2006; Mehra and Baker 2007).

Ni is used in many industrial and consumer products, including stainless steel, magnets,

coinage, and special alloys. In the present study only one still water sample had a

detectable content of Ni. In sparking water samples only brand E contained Ni in all

samples. In the other brands only one or two samples had a detectable Ni content. The

values are in agreement with those in Tamasi and Cini (2004).

Pb has probably the longest history of environmental contamination and toxicity to

humans. Its presence in the human food chain continues to be a great health problem

worldwide. Pb values ranged from 0.79 to 2.22 µg l-1 and from 0.55 to 1.28 µg l-1 in still

and sparkling waters, respectively. It was detected in 50% of the analysed samples.

Generally, the addition of flavours and aromas increased Pb content. The values found

are similar to those obtained in drinking waters and herbal teas (Soylak et al. 2002;

Tamasi and Cini 2004; Nookabkaew et al. 2006; Obiri 2007).

Se is frequently found in combination with Pb, Cu, Hg, and Ag in the environment. Se

was not found in eleven still and ten sparkling waters. Se content values are low (from

0.05 to 0.44 µg l-1). These values are similar to others obtained by other authors in tea

samples (Nookabkaew et al. 2006).


90

Si is the second most abundant element on Earth and is usually found in the form of

silicon dioxide (also known as silica) and silicate. Si was found in all water samples. Si

concentration values ranged from 4.8 to 541.1 µg l-1. The highest Si values were found in

sparkling waters. Considering still waters, only in brand D did Si concentration levels

increase in flavoured water samples. In other brands, Si levels decreased. In sparkling

waters, Si contents decreased in all flavoured waters compared with the values presented

in natural ones. Sripanyakorn et al. (2004) reported Si levels ranging from 9.6 to 22.5 µg l-1

in several kinds of beers. Dejneka and Łukasiak (2003) analysed fruit juice where the Si

concentration ranged from 3.66 to 14.38 µg l-1. These values are similar to the majority of

the samples analysed, but there was one sample with 541.1 mg l-1. According with the

label information this sample had raspberry/ginseng flavour with addition of lemon, apple

and pear juice, fibres and citric acid.

Effects of some labelled compounds in mineral compo sition

Considering the following factors – preservatives, acidifying agents, fruit juice and

sweeteners – a statistical study was made. These ingredients are added to natural waters

and this study aimed to verify their influence on the contents of the trace elements

evaluated. Their influence in each element will be appreciated individually.

Aluminium

Statistical analysis showed that almost all ingredients added to the water do not

influence Al content (p > 0.05). Only significant differences could be found in flavoured

waters without sweeteners (p < 0.05).

Arsenic

Taking into account the added preservatives, it was observed that potassium sorbate

and the blend of potassium sorbate and sodium benzoate led to significant differences (p

< 0.05). Concerning acidifying agents, only the sample with, simultaneously, citric acid

and sodium citrate had significant differences compared with its natural corresponding

water (p = 0.043). Regarding fruit juice, the statistical values were p = 0.018 and 0.028 for

flavoured waters with and without fruit juice, respectively. The addition of sweeteners (p =

0.008) also seemed to influence As concentration.


91

Chromium

In the case of Cr, the statistical study showed that the addition of acidifying regulators

(citric acid and natural flavour) (p = 0.012) increased Cr concentrations significantly.

Regarding the presence or absence of sweeteners, only flavoured waters without

sweeteners increased Cr contents significantly (p = 0.046). No other factor affected Cr

concentration in a significant way.

Cobalt

Regarding the influence of the considered factors on Co concentration, those that

caused significant differences were potassium sorbate and sodium benzoate (in the

preservatives group; p = 0.028); and citric acid and sodium citrate (in acidifying regulators;

p = 0.017). The presence of fruit juice also influenced the levels of this trace element in all

samples (p = 0.019).

Manganese

The blend of potassium sorbate and sodium benzoate as preservatives led to significant

differences in Mn contents, as well as the presence of citric acid and sodium citrate (p =

0.017 for both). Also, a significant difference was verified in flavoured waters without fruit

juice (p = 0.018).

Silicon

The addition of preservatives (potassium sorbate and sodium benzoate) (p = 0.008) and

acidifying regulators (citric acid and sodium citrate) (p = 0.008) increased Si levels

significantly. Regarding the presence of fruit juice or sweeteners, these groups do not

influence Si concentration (p = 0.070 and 0.765, respectively).

Cadmium, mercury, nickel, lead and selenium

Taking into account that only four water samples contained Cd, it was not possible to

perform the statistical analysis for this trace element.

For Hg, Ni, Pb and Se, it was difficult to carry out the statistical analysis because these

elements had several missing values. However, it was shown that the addition of

ingredients in flavoured waters does not affect Hg, Ni, Pb and Se concentrations.


92

Conclusion

The different ingredients added to natural waters (flavours, preservatives, acidifying

regulators, fruit juice and sweeteners) usually influence the concentrations of trace

element. It was verified that preservatives provide a significant contribution to the contents

of As, Co, Mn and Si. The acidifying regulators also produce differences in the

concentrations of As, Cr, Co, Mn and Si. Fruit juice and sweeteners can be correlated to

the increase of Co and As levels.

Although there was an increase in concentrations of trace elements by the added

ingredients, it did not impact on human health because levels never exceeded the

established values recommended by the WHO and the IOM. It is important to note that Cd

was not detected in natural waters and that 50% of the analysed samples had detectable

levels of Hg and Pb. Mn levels were the only element where there were differences in

natural still and sparkling waters (100 times higher).

Acknowledgement

M. Fátima Barroso is grateful to the Fundação para a Ciência e a Tecnologia for a PhD

grant (Grant Number SFRH/BD/29440/2006).

93

Sup

lem

enta

r in

form

atio

n (D

atas

ete)

Tab

le 3

. Min

eral

con

tent

ana

lyse

d in

mar

ket b

ottle

d w

ater

s.

sam

ple

Al

As

Cd

Cr

Co

Hg

Mn

Ni

Pb

Se

Si

nº

µg

l-1 (

%)

mg

l-1 (

%)

1 77

.80

± 3.

00

0.26

± 3

.00

0.17

± 4

.41

0.48

± 1

.80

1.38

± 1

2.64

0.

38 ±

2.9

1 1.

08 ±

8.2

6 -

1.17

± 1

6.63

-

9.3

± 0.

8

2 14

3.53

± 2

.00

0.17

± 1

0.00

-

1.92

± 9

.30

1.48

± 9

.83

0.13

± 1

0.09

0.

90 ±

5.5

1 0.

08 ±

20.

00

1.18

± 7

.81

- 9.

9 ±

8.3

3 48

.12

± 1.

91

- 0.

05 ±

2.0

9 0.

60 ±

11.

24

1.49

± 2

.18

0.12

± 7

.99

0.52

± 2

.81

- 1.

23 ±

13.

83

- 10

.9 ±

3.2

4 11

1.11

± 1

0.90

0.

38 ±

19.

40

- 0.

26 ±

7.5

0 1.

26 ±

10.

20

0.67

± 9

.37

0.28

± 1

2.74

-

1.17

± 6

.10

- 14

.0 ±

1.5

5 11

2.29

± 9

.2

0.74

± 1

3.09

-

0.44

± 0

.76

- -

1.22

± 2

3.88

-

2.22

± 1

2.01

-

17.7

± 8

.1

6 20

.43

± 2.

3

1.05

± 6

.66

- 0.

64 ±

2.8

8 0.

54 ±

13.

00

0.06

± 6

.54

- -

- 0.

22 ±

5.0

8 24

.4 ±

3.9

7 41

.97

± 2.

68

2.14

± 1

.87

- -

0.14

± 8

.30

0.08

± 1

.57

- -

- -

25.7

± 2

.4

8 74

.67

± 9.

1

0.64

± 1

7.36

-

0.28

± 1

2.88

0.

60 ±

28.

20

0.05

± 4

.14

1.46

± 1

2.09

-

1.48

± 4

.45

0.24

± 1

5.46

17

.5 ±

8.4

9 97

.06

± 2.

30

0.45

± 1

0.94

-

0.32

± 4

.23

0.95

± 8

.71

0.07

± 1

7.97

1.

37 ±

11.

85

- 0.

79 ±

10.

12

- 15

.8 ±

5.6

10

87.2

4 ±

4.77

1.

09 ±

2.8

5 -

0.31

± 4

.29

0.94

± 6

2.7

5 -

0.85

± 2

.23

-

- -

17.0

± 4

.8

11

34.0

4 ±

4.60

1.

75 ±

4.2

7 -

- 0.

57 ±

20.

62

0.80

± 5

.15

0.82

± 1

4.69

-

- -

49.7

± 7

.9

12

51.7

9 ±

5.80

-

- 0.

13 ±

18.

17

- -

- -

0.79

± 9

.99

0.20

± 4

.11

7.4

± 1.

7

13

43.0

7 ±

6.0

- -

0.25

± 1

2.13

0.

29 ±

4.0

0 -

- -

0.89

± 1

5.61

0.

12 ±

11.

63

10.9

± 6

.0

14

57.8

0 ±

6.70

- -

0.81

± 2

3.67

-

0.34

± 5

.97

- -

- 10

.0 ±

4.9

15

5.80

± 2

.80

-

- -

- -

- -

- -

5.2

± 7.

4

16

21.0

0 ±

15.2

0

- -

0.50

± 5

.83

1.65

± 2

7.84

0.

30 ±

5.4

0 7.

40 ±

0.8

1 0.

37 ±

2.6

2

0.64

± 1

8.04

-

10.6

± 1

.3

17

51.8

0 ±

1.97

-

- 0.

30 ±

10.

19

1.59

± 5

.77

0.26

± 0

.48

18.9

1 ±

1.32

0.

35 ±

5.8

9

- -

12.3

± 0

.1

18

66.0

3 ±

1.30

-

- 0.

33 ±

2.8

9 1.

78 ±

14.

08

- 19

.11

± 5.

15

1.05

± 2

.58

0.78

± 1

6.37

-

16.7

± 2

.3

19

8.90

± 3

.05

0.

24 ±

7.9

5 -

5.42

± 8

.37

2.20

± 5

.82

0.

13 ±

6.2

0 12

.19

± 1.

07

0.27

± 6

.96

0.87

± 7

.56

- 11

.8 ±

3.8

20

7.50

± 8

.00

0.

49 ±

16.

63

- 0.

24 ±

4.3

0 1.

60 ±

10.

70

0.12

± 0

.45

7.45

± 1

0.90

0.

31 ±

6.0

9 0.

94 ±

8.9

8 0.

42 ±

11.

58

18.1

± 9

.6

94

sam

ple

Al

As

Cd

Cr

Co

Hg

Mn

Ni

Pb

Se

Si

nº

µg

l-1 (

%)

mg

l-1 (

%)

21

104.

91 ±

1.9

3

- -

0.33

± 1

0.19

0.

59 ±

12.

11

- 2

23.6

7 ±

1.53

-

- 0.

28 ±

17.

87

46.5

± 3

.5

22

221.

42 ±

1.4

0

- -

0.61

± 9

.13

0.48

± 1

0.05

-

118.

72 ±

1.8

9

0.15

± 8

.15

1.28

± 8

.99

0.05

± 9

.03

541.

1 ±

9.1

23

119.

22 ±

5.1

0

- -

0.75

± 5

.21

- -

176.

46 ±

1.3

7

0.08

± 1

5.00

0.

87 ±

15.

10

0.10

± 1

3.72

22

1.5

± 6.

0

24

323.

96 ±

1.1

0

- -

0.91

± 5

.09

- 0.

06 ±

21.

93

184.

86 ±

0.7

0 -

- -

44.2

± 0

.5

25

402.

22 ±

6.3

-

- 0.

35 ±

8.3

6 -

- 21

6.30

± 1

.52

- -

- 32

.0 ±

4.6

26

56.5

1 ±

1.80

-

- 1.

71 ±

2.1

3 0.

54 ±

6.0

0 -

185.

62 ±

2.5

0 -

- 0.

30 ±

3.0

5 79

.6 ±

5.8

27

68.1

0 ±

14.3

1

0.54

± 4

4.84

-

0.22

± 4

.66

0.32

± 3

.25

- 1.

83 ±

0.6

2

- 1.

22 ±

2.5

3 0.

44 ±

9.4

6 30

.0 ±

6.0

28

28.5

4 ±

0.20

0.

85 ±

3.4

9 -

- 0.

31 ±

1.2

4 -

3.58

± 2

.86

- -

0.12

± 1

6.57

27

.0 ±

9.2

29

253.

34 ±

1.8

0.

31 ±

12.

96

0.04

± 5

.98

0.34

± 6

.95

0.17

± 3

.89

- 4.

09 ±

3.3

1 0.

39 ±

2.6

6 1.

59 ±

3.4

3 0.

08 ±

16.

46

25.8

± 3

.9

30

95.8

2 ±

0.70

0.

35 ±

18.

32

- 0.

16 ±

18.

44

0.20

± 4

.05

- 1.

68 ±

7.1

1 -

0.55

± 3

.93

0.14

± 6

.09

26.2

± 8

.6

31

63.1

7 ±

0.70

0.

94 ±

0.9

3 -

0.24

± 1

3.49

0.

39 ±

5.0

0 -

1.50

± 3

.98

- -

0.17

± 7

.03

21.4

± 5

.0

32

221.

43 ±

4.7

0 -

- 0.

23 ±

8.0

5 -

- 22

9.40

± 0

.89

- -

- 47

.3 ±

10.

9

33

441.

26 ±

2.8

0

- -

0.22

± 1

.82

0.30

± 5

.96

- 23

6.6

2 ±

0.40

-

- -

35.5

± 5

.5

34

343.

66 ±

0.6

0

- -

0.33

± 4

.09

- -

79.7

1 ±

1.24

-

- 0.

03 ±

8.5

4 26

.5 ±

3.8

35

50.0

1 ±

0.50

-

- 0.

49 ±

1.6

1 0.

45 ±

6.0

5 -

58.9

0 ±

0.91

0.

12 ±

7.5

6 0.

85 ±

3.2

2 0.

13 ±

9.1

4 23

.9 ±

1.7

36

58.9

7 ±

0.70

-

- 0.

43 ±

8.8

7 0.

25 ±

5.0

6 -

72.9

1 ±

1.73

-

0.78

± 7

.29

0.03

± 1

2.27

25

.1 ±

8.5

37

158.

00 ±

4.5

0

- -

0.63

± 4

.71

0.26

± 3.

86

- 86

.55

± 0.

89

- -

- 25

.6 ±

1.7

3

38

106.

72 ±

8.7

0

1.10

± 8

.79

0.07

± 3

.36

0.82

± 4

.17

- -

1.55

± 0

.38

0.53

± 5

.43

- 0.

11 ±

20.

80

4.8

± 0.

8

39

99.0

6 ±

1.10

2.

63 ±

0.7

0 -.

2.

13 ±

1.1

7 -

0.1

± 4

.27

0.77

± 7

.20

-

0.88

± 1

1.89

-

9.2

± 0.

7


95

References

Akagi H, Castillo ES, Cortes-Maramba N, Francisco-Rivera T, Timbang TD. 2000. Health

assessment for mercury exposure among schoolchildren residing near a gold

processing and refining plant in Apokon, Tagum, Davao del Norte, Philippines. Sci Total

Environ. 259:31-43.

Alinnor IJ. 2005. Assessment of elemental contaminants in water and fish samples from

Aba river. Environ Monit Assess. 102:15-25.

American Public Health Association (APHA). 1995. Standard methods for the examination

of water and wastewater. 20th ed. Washington, DC: American Public Health

Association.

Aziz SN, Boyle KJ, Rahman M. 2006. Knowledge of arsenic in drinking-water: risks and

avoidance in Matlab, Bangladesh. J Hlth Pop Nutr. 24:327-335.

Barceloux DG. 1999. Manganese. Clin Toxicol. 37:293-307.

Barton H, 2008. Predicted intake of trace elements and minerals via household drinking

water by 6-year-old children from Kraków, Poland. Part 3: Aluminium. Food addit

Contam. 25(5):588-603.

Berdanier CD. 1998. Advanced Nutrition Micronutrients. Boca Raton (FL): CRC Press

LLC.

Chou CHSJ, De Rosa CT. 2003. Case studies-Arsenic. Int J Hyg Environ Hlth. 206:381-

386.

Dejneka W, Łukasiak J. 2003. Determination of total and bioavailable silicon in selected

foodstuffs. Food Contr. 14:193-196.

Diawara MM, Litt JS, Unis D, Alfonso N, Martinez L, Crock JG, Smith DB, Carselha J.

2006. Arsenic, cadmium, lead, and mercury in surface soils, Pueblo, Colorado:

Implications for population health risk. Environ Geochem Hlth. 28:297-315.

Goldhaber SB. 2003. Trace element risk assessment: essentiality vs. toxicity. Regul

Toxicol Pharmacol. 38:232-242.

Institute of Medicine (IOM) of National Academies. 2001. Dietary Reference Intakes for

vitamin A, Vitamin K, arsenic, boron, chromium, copper, iodine, iron, manganese,

molybdenum, nickel, silicon, vanadium and zinc. [cited 27 August 2008] Available from:

http://www.iom.edu/object.file/Master/54/395/DRIs.Elements.pdf/

Jorhem L, Åstrand C, Sundström B, Baxter M, Stokes P, Lewis J, Grawé KP. 2008.

Elements in rice from the Swedish market: 1. Cadmium, lead and arsenic (total and

inorganic). Food Addit Contam. 25(3): 284–292

Magnuson ML, Creed JT, Brockhoff CA.1997. Speciation of selenium and arsenic

compounds by capillary electrophoresis with hydrodynamically modified electroosmotic


96

flow and on-line reduction of selenium (VI) to selenium (IV) with hydride generation

inductively coupled plasma mass spectrometric detection. Analyst. 122:1057-1061.

Mehra A, Baker CL. 2007. Leaching and bioavailability of aluminium, copper and

manganese from tea (Camellia sinensis). Food Chem. 100:1456-1463.

Miller JN, Miller JC. 2000. Statistics and chemometrics for analytical chemistry. London

(UK): Pearson Education.

Mor F, Ceylan C. 2008. Cadmium and lead contamination in vegetables collected from

industrial, traffic and rural areas in Bursa Province, Turkey. Food addit contam. 25:611-

615.

Nookabkaew, Rangkadilok N, Satayavivad J. 2006. Determination of trace elements in

herbal tea products and their infusions consumed in Thailand. J Agricult Food Chem.

54:6939-6944.

Obiri S. 2007. Determination of heavy metals in water from boreholes in Dumasi in the

Wassa West district of western region of Republic of Ghana. Environ Monit Assess.

130:455-463.

Ohno T, Sakamoto M, Kurosawa T, Dakeishi M, Iwata T, Murata M. 2007. Total mercury

levels in hair, toenail, and urine among women free from occupational exposure and

their relations to renal tubular function. Environ Res. 103:191-197.

Santos EE, Lauria DC, Silveira CLP. 2004. Assssment of daily intake of trace elements

due to comsumption of foodstuffs by adults inhabitants of Rio de Janeiro city. Sci Total

Environ. 327:69-79.

Silvera SAN, Rohan TE. 2007. Trace elements and cancer risk: a review of the

epidemiologic evidence. Canc Cause Contr. 18:7-27.

Soylak M, Aydin FA, Saracoglu S, Elci L, Dogan M. 2002. Chemical analysis of drinking

water samples from Yozgat, Turkey. Pol J Environ Stud. 11:151-156.

Sripanyakorn S, Jugdaohsingh R, Elliot H, Walker C, Mehta P, Shoukru S, Thompson PH,

Powell JJ. 2004. The silicon content of beer and its bioavailability in healthy volunteers.

Br J Nutr. 91:403-409.

Tamasi G, Cini R. 2004. Heavy metals in drinking waters from Mount Amiata (Tuscany,

Italy). Possible risks from arsenic from public health in the Province of Siena. Sci Total

Environ. 327:41-51.

US Food Drug and Administration (USFDA). 2002. Bottled Water Regulation and the FDA.

Washington DC: USFDA, Center for Food Safety and Applied Nutrition. Available from:

http://www.cfsan.fda.gov/~dms/botwatr.html

World Health Organization (WHO). 1996. Trace elements in human nutrition and health.

Geneva (Switzerland): WHO.


97

World Health Organization (WHO). 2008. homepage website available from: :

http://www.inchem.org/

Ysart G, Miller P, Croasdale M, Crews H, Robb P, Baxter M, L’Argy C, Harrison N. 2000.

1997 UK Total Diet Study – dietary exposures to aluminium, arsenic, cadmium,

chromium, copper, lead, mercury, nickel, selenium, tin and zinc. Food Addit Contam.

17:775-786.

99

Capítulo 3

Perfil antioxidante – métodos convencionais

Flavored waters: Influence of ingredients on antioxidant capacity and terpenoid profile by

HS-SPME/GC-MS

M. Fátima Barroso, J. P. Noronha, Cristina Delerue-Matos, M. B. P. P. Oliveira

Journal of Food and Agricultural Chemistry, 2011, 59 (9), 5062 - 5072

3. Perfil antioxidante

101

Flavored Waters: Influence of Ingredients on Antiox idant

Capacity and Terpenoid Profile by HS-SPME/GC-MS

M. Fátima Barroso,†,§ J. P. Noronha,*,# Cristina Delerue-Matos,§ and M. B. P. P. Oliveira†

†Requimte/Faculdade de Farmácia, Universidade do Porto, Rua Aníbal Cunha 164, 4099-

030 Porto, Portugal §Requimte/Instituto Superior de Engenharia do Porto, Rua Dr. António Bernardino de

Almeida 431, 4200-072 Porto, Portugal #Requimte/CQFB, Departamento de Química, Faculdade de Ciências e Tecnologia,

Universidade Nova de Lisboa, 2829-516 Caparica, Portugal

Abstract

The antioxidant profiles of 39 water samples (29 flavored waters based on 10 natural

waters) and 6 flavors used in their formulation (furnished by producers) were determined.

Total phenol and flavonoid contents, reducing power, and DPPH radical scavenging

activity were the optical techniques implemented and included in the referred profile.

Flavor extracts were analyzed by HS-SPME/GC-MS to obtain the qualitative and

quantitative profiles of the volatile fraction of essential oils. Results pointed out a higher

reducing power (0.14-11.8 mg of gallic acid/L) and radical scavenging activity (0.29-211.5

mg Trolox/L) of flavored waters compared with the corresponding natural ones, an

interesting fact concerning human health. Bioactive compounds, such as polyphenols,

were present in all samples (0.5-359 mg of gallic acid/L), whereas flavonoids were not

present either in flavored waters or in flavors. The major components of flavor extracts

were monoterpenes, such as citral, α-limonene, carveol, and α-terpineol.

KEYWORDS: total antioxidant capacity, flavored water, essential oils, total phenols and

flavonoids contents, radical scavenging activity, reducing power, HS-SPME/GC-MS

Available online at pubs.acs.org

Journal of Agricultural and Food Chemistry

2011, 59 (9), 5062-5072


103

INTRODUCTION

Reactive oxygen species (ROS) are continuously produced in all living beings,

especially in higher organisms, as a result of normal cellular metabolism, phagocytises,

inflammation, and exogenous factors such as ionizing radiations and xenobiotics.1 ROS

can induce cell damage by reacting with biomolecules (proteins, lipids) and cause serious

lesions in the DNA molecule,2 such as strand breaks, DNA-protein cross-linking, and

base-free sites.3 The mammalian body has certain endogenous antioxidant defense

mechanisms to combat and reduce oxidative damage such as enzymatic systems, and

exogenous antioxidant systems, such as as vitamins, minerals, and proteins. Antioxidants,

which can inhibit or delay the oxidation of a substrate in a chain reaction, therefore,

appear to be very important in the prevention of many diseases.4 Foodstuffs constitute an

excellent exogenous source of natural antioxidants. It is known that vegetables, fruits,

whole-grain, and some beverages (tea, juice, wine) are rich in antioxidants and bioactive

compounds. Examples of antioxidants present in food are vitamins (particularly C and E),

phenolic compounds (flavonoids, catechins, flavones, flavonols, anthocyanins), and

carotenoids including β-carotene.5 A healthy diet should provide an adequate and

continuous supply of these antioxidants. Other antioxidants, such as ubiquinol and thiol

compounds, produced in small amounts by the organism, can be obtained in higher

amounts by dietary supplements.6 Consequently, interest is increasing in new effective

natural antioxidants as well as in the chemical and biochemical characterization of

foodstuffs and beverages to evaluate them with regard to their antioxidant profiles.

To answer consumers’ preferences, the food industry has applied several technical

improvements to plain water. Today, a significant part of commercialized water is in

flavored formulation. Flavors, juices, bioactive compounds, preservatives, and/or

sweeteners are added to water, providing a product with singular tastes and smells

appreciated by consumers.

Flavors (or essential oils) from fruits contain 85-99% of volatile and 1-15% of nonvolatile

compounds. Volatile constituents are a mixture of monoterpenes and sesquiterpenes,7

being flavonoids present in the nonvolatile fraction.8 Terpenes and flavonoids present

antioxidant and antiradical properties9 and can be transferred to water samples if

flavors/aromas extracts are used. Therefore, drinking this type of beverage can improve

the daily intake of antioxidants, contributing to the exogenous protective system. However,

there are no reports concerning the antioxidant properties of these waters, although their

macro- and micromineral compositions are known.10,11 These properties will be a new

source of information for consumer’s about the advantages/disadvantages on the

consumption of these beverages.


104

Antioxidant capacity determination is not an easy task to perform. Several factors

(substrates, conditions, analytical methods, and concentrations) can affect the estimated

values, and it is difficult to measure each antioxidant component separately and/or the

interactions among different antioxidant components in the samples.4 Total antioxidant

capacity measures can be classified in two groups: assays based on the inhibition of

human low density lipoprotein oxidation or those based on oxygen free radical scavenging

ability. Current In vitro methods for antioxidant efficacy evaluation have as a basic

principle the oxidation inhibition of a suitable substrate. After oxidation of the substrate,

under standard conditions, the extent of the reaction is determined at a fixed time point or

over the range that is characteristic of the generated free radical.3 UV-vis

spectrophotometric, chemiluminescence, fluorometric,4 and chromatographicmethods12

can be used to do that.

In the present study, four optical methods were applied to evaluate the antioxidant

profile of 39 mineral and spring, natural and flavored water samples, and 6 flavors/aromas

used in their formulation. This was carried out by means of the total phenol content (TPC),

total flavonoid content (TFC), reducing power, and 2,2-diphenyl-1-picrylhydrazyl (DPPH)

radical-scavenging activity (RSA). The volatile fractions of the flavor extracts were isolated

by headspace solid-phase microextraction (SPME) and analyzed by gas chromatography-

mass spectrometry (GC-MS).

MATERIALS AND METHODS

Chemicals . Gallic acid, (-)-epicatechin, and 6-hydroxy-2,5,7,8-tetramethylchroman-2-

carboxylic acid (Trolox, a water-soluble analogue of vitamin E) standards were from

Sigma-Aldrich or Fluka. Folin-Ciocalteu reagent and DPPH were obtained from Sigma-

Aldrich. All of these chemicals, of the highest quality available (95-99%), were used

without purification. Other compounds of analytical grade, such as sodium carbonate,

sodium nitrite, aluminum chloride, sodium hydroxide, ethanol, and sodium acetate (0.1

mol/L, pH 4.3), were from Merck. All solvents used were of HPLC grade. Standard

antioxidant solutions were prepared daily and stored in the dark at 4 ºC Cwhen not in use.

Water used was ultrapure (18.2MΩ/cm), obtained from a Millipore Simplicity 185 system.

For spectrophotometric measurements a Shimadzu 160-A spectrophotometer was used.

Sample Preparation . Mineral water arises from a geologically and physically protected

underground source, characterized by constant levels and relative proportions of minerals

and trace elements at the source. Spring water derives from an underground formation

from which water flows naturally to the surface at an identified location.


105

Thirty-nine water samples, corresponding to 10 different brands, acquired in

supermarkets in northern Portugal and stored in the dark at 4 ºC were analyzed. Each

brand (still or sparkling, mineral or spring water) had different flavors and aromas. Natural

waters of each brand were used as control. Sonication was used to eliminate gas from

sparkling water samples.

Table 1 summarizes the nutrient information on the labels, taking into account its

different composition in gas, flavor, vitamins, preservatives, acidifying regulators, and

sweeteners.

Six flavors or concentrate extracts (lime, tangerine, strawberry, lemon, apple, and

gooseberry) used in the formulation of some water brands, and provided by producers,

were also analyzed. As expected, these flavors had no description about its chemical

composition.

TPC Determination . TPC values of flavors and flavored waters were determined by a

colorimetric assay based on procedures described by Singleton and Rossi13 with some

modification. Folin-Ciocalteu reagent and the reduced phenols produced a stable blue

product at the end of reaction. The reaction mixture (20 µL of sample, 1.58 mL of ultrapure

water, and 100 µL of Folin-Ciocalteu reagent) was sonicated for 30 s. After this, it was

added to 300 µL of 7% Na2CO3, and the mixture was incubated for 10 min at 50 ºC.

Factor dilutions of 10 times on the mother standard antioxidant gallic acid (GA) were

carried out to obtain a calibration curve ranging from 0 to 5.00 mg of GA/L of water.

Quantifications were carried out in triplicate, and the absorbance was measured at 760

nm.

TFC Determination . TFC was determined by a colorimetric assay based on the

formation of flavonoid-aluminum compound.14 One milliliter of flavored water was mixed

with 4 mL of ultrapure water and 300 µL of 5% NaNO2 solution. After 5 min, 300 µL of

10% AlCl3 solution was added. After 6 min, 2 mL of 1 mol/L NaOH and 2.4 mL of

ultrapure water were added. The solution was mixed well, and the absorbance of a pink

color was read at 510 nm. (-)-Epicatechin was used to plot the standard curve ranging

from 0 to 66.26 mg/L, and the results of TFC were expressed as milligrams of epicatechin

per liter of water. All measurements were carried out in triplicate.

10

6

Tab

le 1

. Lab

el In

form

atio

n in

the

Eva

luat

ed B

ottle

d F

lavo

red

Wat

ers

bran

d sa

mpl

e fla

vor

juic

e vi

tam

in

pres

erva

tives

ac

idify

ing

regu

lato

rs

swee

tene

rs

othe

r in

gred

ient

s

Stil

l Wat

er A

1 le

mon

pota

ssiu

m s

orba

te,

sodi

um b

enzo

ate

citr

ic a

cid,

sodi

um c

itrat

e

aces

ulfa

me-

K

fibre

s (1

%),

whe

at d

extr

in (

0.1%

) m

iner

al

2 m

ango

po

tass

ium

sor

bate

,

sodi

um b

enzo

ate

citr

ic a

cid,

sod

ium

citr

ate

aces

ulfa

me-

K

fibre

s (1

%),

whe

at d

extr

in (

0.1%

)

3

stra

wbe

rry

pota

ssiu

m s

orba

te,

sodi

um b

enzo

ate

citr

ic a

cid,

sodi

um c

itrat

e

aces

ulfa

me-

K

fibre

s (1

%),

whe

at d

extr

in (

0.1%

)

4

natu

ral

B

5 pi

neap

ple/

oran

ge

appl

e

pota

ssiu

m s

orba

te,

sodi

um b

enzo

ate

citr

ic a

cid

aces

ulfa

me-

K,

aspa

rtam

e

calc

ium

lact

ate

spr

ing

6 le

mon

ap

ple

niac

in, p

anto

then

ic a

cid,

B6,

folic

aci

d, b

iotin

, B12

pota

ssiu

m s

orba

te,

sodi

um b

enzo

ate

citr

ic a

cid

aces

ulfa

me-

K,

aspa

rtam

e

7

natu

ral

C

8 le

mon

/mag

nesi

um

frui

t (m

g/10

0 m

L): B

3 (2

.7),

B5

(0.9

),

B6

(0.3

), B

8 (0

.022

), B

9 (0

.03)

,

B12

(1.

5x10

-4)

pota

ssiu

m s

orba

te,

dim

ethy

l dic

arbo

nate

citr

ic a

cid

mag

nesi

um c

arbo

nate

,

gin

seng

min

eral

9

appl

e/w

hite

tea

frui

t (m

g/10

0 m

L): B

3 (2

.7),

B5

(0.9

),

B6

(0.3

), B

8 (0

.022

), B

9 (0

.03)

B12

(1.5

x10-4

)

pota

ssiu

m s

orba

te

citr

ic a

cid

m

alic

aci

d,

calc

ium

lact

ate

10

pi

neap

ple/

fibre

fr

uit

po

tass

ium

sor

bate

,

dim

ethy

l dic

arbo

nate

citr

ic a

cid

w

heat

dex

trin

(0.

9%)

L-ca

rniti

ne (

200

mg/

L)

11

na

tura

l

D

12

appl

e

di

met

hyl d

icar

bona

te,

sodi

um b

enzo

ate

citr

ic a

cid

sucr

alos

e,

aces

ulfa

me-

K

min

eral

13

or

ange

/pea

ch

dim

ethy

l dic

arbo

nate

, so

dium

ben

zoat

e ci

tric

aci

d su

cral

ose,

aces

ulfa

me-

K

10

7

bran

d sa

mpl

e fla

vor

juic

e vi

tam

in

pres

erva

tives

ac

idify

ing

regu

lato

rs

swee

tene

rs

othe

r in

gred

ient

s

14

le

mon

dim

ethy

l dic

arbo

nate

, so

dium

ben

zoat

e ci

tric

aci

d

sucr

alos

e,

aces

ulfa

me-

K

15

na

tura

l

E

16

lem

on

lem

on

(mg/

100

mL)

: B3

(2.7

), B

12 (

0.15

) po

tass

ium

sor

bate

,

sodi

um b

enzo

ate

citr

ic a

cid,

sodi

um c

itrat

e

aces

ulfa

me-

K,

aspa

rtam

e ca

rbon

dio

xide

min

eral

17

or

ange

/ras

pber

ry

oran

ge, r

aspb

erry

(m

g/10

0 m

L): B

3 (2

.7),

B12

(0.

15)

pota

ssiu

m s

orba

te,

sodi

um b

enzo

ate

citr

ic a

cid,

sodi

um c

itrat

e

aces

ulfa

me-

K,

aspa

rtam

e ca

rbon

dio

xide

add

ed g

as

18

peac

h/pi

neap

ple

peac

h, p

inea

pple

(m

g/10

0 m

L): B

3 (2

.7),

B12

(0.

15)

pota

ssiu

m s

orba

te,

sodi

um b

enzo

ate

citr

ic a

cid,

sodi

um c

itrat

e

aces

ulfa

me-

K,

aspa

rtam

e ca

rbon

dio

xide

19

gu

ava/

lime

guav

a/lim

e (m

g/10

0 m

L): B

3 (2

.7),

B12

(0.

15)

pota

ssiu

m s

orba

te,

sodi

um b

enzo

ate

citr

ic a

cid,

sod

ium

citr

ate

aces

ulfa

me-

K,

aspa

rtam

e ca

rbon

dio

xide

20

natu

ral

F

21

lem

on/g

reen

tea

frui

t, le

mon

, app

le

citr

ic a

cid

gr

een

tea

min

eral

22

ra

spbe

rry/

gins

eng

rasp

berr

y, a

pple

, pea

r

ci

tric

aci

d

gins

eng

nat

ural

gas

23

pe

ach/

whi

te te

a fr

uit,

peac

h, a

pple

, pe

ar

citr

ic a

cid

w

hite

tea

24

m

ango

/gin

kgo

belo

ba

man

go, a

pple

, pea

r

ci

tric

aci

d

Gin

kgo

bilo

ba

25

m

elon

/min

t fr

uit,

mel

on, a

pple

, pe

ar

citr

ic a

cid

m

int

26

na

tura

l

G

27

lem

on

C

(12

mg/

250

mL)

po

tass

ium

sor

bate

ci

tric

aci

d ac

esul

fam

- K

, su

cral

ose

min

eral

28

lim

e

add

ed g

as

29

appl

e

C (

12 m

g/25

0 m

L)

pota

ssiu

m s

orba

te

citr

ic a

cid

aces

ulfa

me-

K,

sucr

alos

e

30

pe

ach

C

(12

mg/

250

mL)

po

tass

ium

sor

bate

ci

tric

aci

d ac

esul

fam

e-K

, su

cral

ose

31

natu

ral

H

32

lem

on

lem

on, a

pple

C

(30

mg/

100

mL)

so

dium

ben

zoat

e,

pota

ssiu

m s

orba

te

citr

ic a

cid

aspa

rtam

e

min

eral

n

atur

al g

as

33

na

tura

l

I 34

le

mon

le

mon

, app

le

so

dium

ben

zoat

e ci

tric

aci

d as

part

ame

10

8

bran

d sa

mpl

e fla

vor

juic

e vi

tam

in

pres

erva

tives

ac

idify

ing

regu

lato

rs

swee

tene

rs

othe

r in

gred

ient

s

min

eral

35

gr

een

appl

e ap

ple

so

dium

ben

zoat

e ci

tric

aci

d su

cral

ose

nat

ural

gas

36

st

raw

berr

y ap

ple,

str

awbe

rry

so

dium

ben

zoat

e ci

tric

aci

d as

part

ame

37

natu

ral

J

38

lem

on

lem

on

pota

ssiu

m s

orba

te,

sodi

um b

enzo

ate

citr

ic a

cid

so

dium

citr

ate

aspa

rtam

e,

aces

ulfa

me-

K

spr

ing

add

ed g

as

39

natu

ral


109

Reducing Power Assay . Reducing power was determined according to the method of

Oyaizu.15 One milliliter of sample was mixed with 2.5 mL of 0.2 mol/L sodium phosphate

buffer (pH 6.6) and 2.5 mL of 1% potassium ferricyanide. This mixture was incubated for

20 min at 50 ºC, and then 2.5 mL of 10% trichloroacetic acid (w/v) was added and

centrifuged at 1000 rpm for 10 min. The upper layer of the solution (2.5 mL) was mixed

with distilled water (2.5 mL) and 0.5 mL of 0.1% ferric chloride, and the absorbance was

measured at 700 nm. The calibration curve was prepared with GA solutions ranging from

0 to 19.6 mg/L, and the results are given as milligrams of GA per liter of water.

DPPH Radical Scavenging Activity . RSA of samples against the stable nitrogen

radical DPPH• was determined spectrophotometrically at 517 nm.16 DPPH• free radical is

reduced to the corresponding hydrazine when it reacts with hydrogen donors, such as an

antioxidant. In this technique, samples (200 µL) were mixed with 2.80 mL of 1.86x10-4

mol/L ethanolic solution of DPPH•. The mixture, vigorously shaken, was left to stand for 15

min in the dark (until stable absorption values). Lower absorbance values of the reactive

mixture indicated higher free radical scavenging activity. The calibration curve was

prepared with Trolox solutions ranging from 0 to 19.6 mg/L, and the results are given as

milligrams of Trolox per liter of water.

Validation of the Optical Methodologies . Calibration standards were daily prepared,

and all samples were determined in triplicate. The methods were validated by linear

range, limit of detection (LOD), limit of quantification (LOQ), precision, and accuracy. LOD

and LOQ were defined, respectively, as 3 and 10 times the standard deviation of 10 blank

signals divided by the slope of the calibration plot.17 Precision was calculated by intraday

and interday determinations of standard solutions and expressed by relative standard

deviations (RSD). For intraday evaluation, each concentration was assessed by five

measurements, three times during a working day. The interday precision measurements

were made over 1 week. Accuracy and reproducibility were checked by recovery (REC),

relative error (RE), and RSD. All results were expressed as the mean ( standard deviation.

Flavor/Fragrance Extraction by Headspace SPME and D etection by GC-MS .

Extraction of fragrances was carried out by SPME using a 65 µm

polydimethylsiloxane/divinylbenzene (PDMS/DVB) fiber (Supelco, Inc., Bellefonte, PA), for

all experiments. This fiber was selected according to the best results for the extraction of

fruit volatiles.18,19 Fibers were conditioned for 30 min at 250 ºC before use.

For each extraction 2 g of flavor and 0.5 g of NaCl (to inhibit enzymatic reactions and to

favor the transfer of the analytes from the aqueous solution to the headspace) or 50 µL of

extract were transferred into a 10 mL Teflon-lined septum cap vial equipped with a

Tefloncoated magnetic bar. To favor the transfer of the analytes from the aqueous

solution to the headspace, the solution was stirred (200 rpm) at 70 ºC. The PDMS/DVB


110

fiber was used to extract the nonpolar volatile compounds (in the headspace). The fiber

was exposed to the sample headspace for 20 min at 70 ºC. The fiber was then removed

and introduced into the injector port of the GC-MS for desorption at 250 ºC for 3 min, in

the splitless mode.

The separation and detection of the analytes was achieved using a GC-MS system

(Agilent Technologies, USA) with a GC 6850 coupled to a 595C VL MSD mass selective

detector, with a silica capillary column (30 m x 0.32 mm i.d.; df, 0.25 µm) covered with 5%

phenyl/95% dimethylpolysiloxane (DB-5 ms, Agilent-J&W Scientific), kept at 30 ºC for 3

min, and then ramped to 300 ºC at 8 ºC/min and held at the final temperature for 4 min.

The splitless injection (3 min) was achieved with an injector temperature at 250 ºC.

Helium was the carrier gas used at flow of 1.0 mL/min. Ion source, quadrupole, and

transference line were kept at 230, 150, and 280 ºC, respectively. MS spectra were

obtained by electronic impact (EI) at 70 eV and collected at the rate of 1 scan/s over an

m/z range of 35-400, and using MSD ChemStation E.02.00493 software (Agilent

Technologies, USA). Identification of the individual components was performed by

comparing their mass spectra with the standards and spectral libraries of GC-MS (NIST

98 and Wiley 275), enabling the detection of some minor components and identification of

compounds that arise from incompletely resolved chromatographic peaks.

For each compound, quantitation was performed by measuring the corresponding peak

area of the total ion chromatogram and expressed as relative (percent) areas by

normalization.

RESULTS AND DISCUSSION

Descriptive Statistics . Table 1 represents the labeled nutrient information in flavored

waters. About 38% of water samples are still and 62% sparkling (11 water samples with

added gas). Labels indicate the presence of several compounds added for technological

purposes, with biological activity (flavors, juice fruit, and vitamins). Inevitably, these waters

also need other ingredients, without positive relationship with well-being and health, but

necessary to ensure the quality desired for producers and consumers and for the safety of

the product. This is the case of preservatives, acidifying regulators, and sweeteners.

Twelve different flavors were present in flavored waters: lemon (10 samples); mango,

strawberry, lime, and raspberry (2 samples each); pineapple, apple, and orange (3

samples each); peach (4 samples); guava, melon, and green apple (1 sample each).

Lemon is the predominant flavor, present in all water brands (A-J; 10 samples).

Seventeen flavored water samples had only one flavor, and 12 samples had a


111

combination of two flavors. About 50% of the samples have fruit juices or concentrates.

Only flavored brands A, D, and G do not report the addition of this type of ingredient.

Eleven samples, according to the label, have in their composition vitamins of the B

complex (7 samples) and C (4 samples). It is important to remember that vitamin C is an

antioxidant with protection capacity against oxidative stress, being also a cofactor in

several vital enzymatic reactions. Other bioactive compounds (ginseng, L-carnitine, white

and green tea, and Ginkgo biloba) are present in some samples from different brands.

Green tea contains numerous components with antioxidant activity, such as polyphenols

(catechins, epicatechin, epigallocatechin) and vitamins.20 Ginseng is an herbal medicine

with antioxidant and anti-inflammatory activities and G. biloba is rich in phenolic and

flavonoid compounds.

Forty-nine percent of samples contain sweeteners. There are water samples with only

one (acesulfame-K, sucralose, or aspartame) ad with two sweeteners in association

(acesulfame-K and aspartame; acesulfame-K and sucralose). The most used was

acesulfame-K (present in 14 samples), followed by aspartame (10 samples). It is

interesting to note that, in general, the samples from the same brand have the same

sweetener, the exception being brand I that uses different sweeteners for different flavors.

Brands C and F do not have sweeteners, providing more energetic products, of 9-13 and

19 kcal/100 mL, respectively (sweetened samples ranged from 0.4 to 4 kcal/100 mL).

Each sample contains a single preservative (potassium sorbate or sodium benzoate) or

the association of two (potassium sorbate and sodium benzoate; potassium sorbate and

dimethyl dicarbonate; sodium benzoate and dimethyl dicarbonate). From this discussion,

different behaviors and antioxidant values among the samples in the study are expected.

Method Validation . Table 2 presents the results obtained in the validation procedures

of the applied methodologies (TPC, TFC, reducing power, DPPH RSA). Linearity ranges

from 0 to 5.0 mg of GA/L in TPC, from 0 to 66.2 mg of epicatechin/L in TFC, and from 0 to

19.6 mg of GA/L and Trolox/L in reducing power and DPPH RSA methods, respectively.

LOD values ranged from 5.43x10-3 (reducing power) to 1.00x10-1 (TFC) mg of standard

antioxidant/L, and LOQ values ranged from 1.81x10-2 to 3.33x10-1 mg of standard

antioxidant/L.

Precision and accuracy values are also shown in Table 2. RSD values ranged from 2.1

(intraday studies) to 7.4 (interday studies) and confirmed the high precision of the

methods. REC and RE values assessed the accuracy of the results. RE were always

<11.0%, and recovery trials ranged from 93 to 111%, confirming the accuracy of the

implemented methods.


112

Table 2. Calibration Curves, Limit Values, Precisio n, and Accuracy Obtained in the

Determination of Antioxidant Activity Assays

Determination of TPC and TFC . Recently, bottled flavored waters have become

popular, and the consumption of flavored waters is globally increasing, including in

Portugal. In the first half of 2010, 6.08 million liters of this kind of water was consumed by

the Portuguese population. Considering the emergent market of this kind of beverage, it is

important to deepen the knowledge of the antioxidant capacity of these beverages. The

method used to determine TPC has been extensively applied in plants and beverages.

Phenolic and flavonoid compounds, correlated with antioxidant activity, seem to have an

important role in stabilizing lipid oxidation. Generally, the antioxidant mechanism of

phenolic compounds is inactivating lipid free radicals and preventing decompositionof

hydroperoxides into free radicals. This is the case of fruits and beverages in relation to

their phenolic compounds.5 Therefore, in this research TPC and TFC were evaluated in 6

flavors used in flavored water formulation and in 39 water samples commercialized in

northern Portugal. However, TPC determination should always be considered as an

parameters

TPC

(mg Gallic acid/L)

TFC

(mg Epicatechin/L)

reducing Power

(mg Gallic acid/L)

DPPH scavenging activity

(mg Trolox/L)

linear concentration (µg/L) 0 - 5.0 0 - 66.2 0 - 19.6 0 - 19.6

slope (Abs mg/L) 7.34 ± 0.10 (x10-2) 3.52 ± 0.03 (x10-2) 2.74 ± 0.07 (x10-1) -6.76 ± 0.2 (x10-2)

intercept (Abs) -9.24 ± 0.40 (x10-4) 1.67 ± 0.80 (x10-2) -2.65 ± 0.5 (x10-2) 1.30 ± 0.02

correlation coefficient (n = 5) 0.999 0.999 0.998 0.997

LOD (mg standard/L) 3.22x10-2 1.00x10-1 5.43x10-3 2.84x10-2

LOQ (mg standard/L) 1.07x10-1 3.33x10-1 1.81x10-2 9.48x10-2

Intra-day studiesa

added (µg/L)

found (µg/L)

5.0

4.8

6.0

5.8

10.0

11.1

10.0

9.3

RECb (%) 95.0 96.7 111.0 93.0

REc (%) - 5.0 -3.3 11.0 -7.0

RSDd (%) 3.2 4.6 2.1 6.9

Inter-day studiese

added (µg/L)

found (µg/L)

5.0

5.2

6.0

5.6

10.0

9.5

10.0

9.8

REC (%) 104.0 93.3 95.0 98.0

RE (%) 4.0 -6.7 -5.0 -2.0

RSD (%) 4.0 5.8 6.3 7.4

aAverage of three measurements, three times during a day. bREC, recovery. cRE, relative error. dRSD, relative standard deviation. eAverage of five measurements over a week.


113

indicative value instead of an accurate measure of phenolic compounds. This method

should be aware of possible interferences (reducing sugars and some amino acids) that

can overestimate evaluated amounts. On the other hand, it is difficult to measure all

phenolic molecules individually. Nevertheless, this assay is a simple, sensitive, and

precise technique.4 Table 3 presents TPC and TFC values obtained in samples.

With regard to TPC, only natural waters (without added ingredients) and two samples of

flavored waters (13 and 14) present “not detected” values. As an exception, sample 37

(natural water) presents trace TPC levels (0.07 mg of GA/L). TPC values ranged from 8.5

(gooseberry) to 380.2 (lemon) mg of GA/L in flavors and from 0.29 (sample 27) to 284 mg

of GA/L (sample 29) in flavored waters. Comparing flavor TPC values, the highest

contents are from citrus fruits such as tangerine, lime, and lemon, respectively, 117, 359,

and 380 mg of GA/L. These values are similar to those obtained by other authors in juices

of citrus fruits,5 but less than those found in other studies with beverages containing milk

and fruits of the same kind used in this work.21 Gooseberry, in contrast to what was

expected, presented the lowest levels. This flavor is different from Indian gooseberry,

described by Mayachiew and Devahastin,22 as rich in TPC (290 mg of GA/g extract).

Calixto and Goñi23 reported the TPC of beverages (coffee, tea, and red wine) and fruits.

The TPC values were higher than the TPC values obtained in this work and ranged from

76 mg of GA/100 mL in beverages to 538 mg of GA/100 g in dry fruit.

With regard to flavored waters and their TPC contents, the lowest value (0.29 mg of

GA/L, lemon flavor, sample 27) and the highest value (284 mg of GA/L, apple flavor,

sample 29) were determined in samples from the same brand (G). This information can be

important for consumers because they generally correlate brands with similar behaviors.

In this case, flavored waters from the same brand can be distinct. According to Table 3

and the values presented, brand G is unique, having significant differences.

The addition of bioactive compounds such as tea (samples 9, 21, and 23), ginseng

(sample 22), and G. biloba (sample 24) seems to increase TPC contents of the flavored

waters. These samples presented values ranging from 28.1 to 39.7 mg of GA/L, the

highest ones excluding samples 29 and 30. It is important to remember that according to

the label information, these samples (29 and 30) have vitamin C as an added ingredient

(12 mg/ 250 mL), a compound related with antioxidant properties. Nevertheless, the

presence of the vitamin, referred to on the label, does not always imply high TPC levels.

This is the case for samples 27 and 32, with added vitamin, which have very different TPC

values lower than those previously mentioned. Sometimes, the expectation that samples

with bioactive compounds have a dual phenolic protective effect is not true.


114

Table 3. TPC, Reducing Power, and DPPH RSA Determin ed in Flavors and Flavored

and Natural Waters

brand sample TPC (mg GA/L) reducing power (mg GA/L) DPPH (mg trolox/L)

Flavours tangerine 116.70 ± 1.50 10.21 ± 0.09 51.21 ± 0.35 lime 359.30 ± 0.60 11.03 ± 0.07 78.51 ± 0.23

strawberry 15.34 ± 0.10 11.12 ± 0.04 213.53 ± 2.50

lemon 380.20 ± 0.09 10.71± 0.08 38.90 ± 0.42

gooseberry 8.53 ± 0.04 11.80 ± 0.04 211.53 ± 4.60

apple 37.15 ± 0.03 10.62 ± 0.03 54.43 ± 0.29

A 1, lemon 4.68 ± 0.05 2.61 ± 0.11 14.71 ± 0.04 2, mango 7.72 ± 0.10 3.68 ± 0.09 13.16 ± 0.05 3, strawberry 9.26 ± 0.03 3.91 ± 0.08 12.27 ± 0.04 4, natural nda nd 0.76 ± 0.03

B 5, pineapple/orange 18.30 ± 0.09 6.01 ± 0.42 13.49 ± 0.32

6 lemon 17.62 ± 0.020 5.45± 0.14 16.26 ± 0.15

7, natural nd nd 0.62 ± 0.03

C 8, lemon/magnesium 24.44 ± 0.20 8.48 ± 0.20 15.71 ± 0.09

9, apple/white tea 28.10 ± 0.03 8.20 ± 0.15 16.49 ± 0.03

10, pineapple/fibre 11.40 ± 0.05 4.56 ± 0.08 8.05 ± 0.45

11, natural nd nd 0.89 ± 0.04

D 12, apple 0.54 ±0.03 3.31 ± 0.03 46.55 ± 0.45

13, orange/peach nd 2.79 ± 0.06 44.11± 0.07

14, lemon nd 3.07 ± 0.07 44.56 ± 0.04

15, natural nd nd 0.41 ± 0.02

E 16, lemon 2.24 ± 0.11 0.28 ± 0.05 12.38 ± 0.25

17, orange/raspberry 6.18 ± 0.05 1.07 ± 0.02 16.49 ± 0.05

18, peach/pineapple 1.51 ± 0.02 0.14 ± 0.03 15.27 ± 0.10

19, guava/lime 8.57 ± 0.03 5.45 ± 0.04 14.71± 0.04

20, natural nd nd 0.76 ± 0.02

F 21, lemon/green tea 39.70 ± 0.10 9.64 ± 0.03 41.45 ± 0.27

22, raspberry/ginseng 37.90 ± 0.08 13.78 ± 0.05 48.66 ± 0.32

23, peach/white tea 29.20 ± 0.15 8.29 ± 0.02 42.45 ± 0.04

24, mango/ginkgo beloba 36.50 ± 0.02 10.52 ± 0.04 45.89 ± 0.37

25, melon/mint 19.70 ± 0.04 10.11 ± 0.04 41.56 ± 0.05

26, natural nd nd 0.29 ± 0.05

G 27, lemon 0.29 ± 0.02 nd 38.23 ± 0.06

28, lime 1.75 ± 0.04 nd 38.79 ± 0.17

29, apple 284.0 ± 2.3 154.04 ± 0.26 268.89 ± 2.45

30, peach 147.0 ± 1.3 nd 133.87 ± 1.35

31, natural nd nd 0.42 ± 0.06

H 32, lemon 7.23 ± 0.05 3.77 ± 0.03 44.78 ± 0.48

33, natural nd nd 0.31 ± 0.05

I 34, lemon 4.31 ± 0.03 5.91 ± 0.04 43.67 ± 0.28

35, green apple 4.92 ± 0.06 4.00 ± 0.07 54.21 ± 0.03

36, strawberry 5.89 ± 0.03 6.10 ± 0.38 42.67 ± 0.06

37, natural 0.07 ± 0.03 nd 0.27 ± 0.13

J 38, lemon 1.88 ± 0.02 4.33 ± 0.29 41.89 ± 0.04

39, natural nd nd 0.38 ± 0.08 and, not detected


115

Samples from brand D presented the lowest TPC values (from not detected to 0.54 mg

of GA/L). According to the label information, these samples have only flavors in its

formulation. It is possible to speculate whether this is a synthetic substance without the

complexity of vegetable/fruit extracts, namely, without phenolic compounds. Another

approach can be the use of trace amounts without influence in values of the parameters in

appreciation. All brands have water flavored with lemon. TPC values have extreme

discrepancies, ranging from not detected (sample 14) to 17.62 mg of GA/L (sample 6) and

higher in association with magnesium (24.44 mg ofGA/L, sample 8) or green tea (39.70

mg of GA/L, sample 21). It is also interesting to verify that waters flavored with lemon

generally presented the lowest TPC values compared with other flavors, the exception

being the examples referred to above with magnesium and green tea. This is especially

important taking into account that lemon flavor was the richest in TPC. The use of more

diluted extract, due to its strong taste, can be a possible explanation for the obtained

results.

In the case of lime flavor, the second most rich in TPC, is only present in sample 28.

However, this flavor, being slightly poorer in TPC compared with lemon, is present at

levels 6-fold higher in flavored waters from the same brand.

With regard to flavors and their TPC contents, it is important to note that TPC values

from red fruits (strawberry and gooseberry) were the lowest. Taking into account its

antioxidant power, it can be speculated that the mechanism does not involve phenolic

compounds. However, when used as ingredients, they provide water samples the highest

TPC values compared with other samples of the same brand. This is the case of sample 3

in brand A and sample 36 in brand I.

From the studied samples, labels do not reveal the presence of gooseberry. Two

samples (17 and 22) have raspberry as an ingredient, being the second richest

considering all samples of the brands.

Unfortunately, the labels of the samples evaluated do not declare tangerine flavor in

their compositions. Probably it is used in combination with other flavors, in minimal

amounts, and not declared in the final list of ingredients. The same explanation can be

proposed for gooseberry flavor.

With regard to TFC, all flavors and flavored water samples had no flavonoids, in

detectable amounts, in their composition. These results are consistent with those obtained

by Tabard and collaborators.24 Using the same optical technique used in this work, these

authors did not find flavonoids in apple, grape, or vegetable juices. However, those

authors found a high TFC level in red wine. On the other hand, flavors (essential oils)

contain about 1-15% of nonvolatile components when flavonoids are included.8 Therefore,


116

flavonoids are present in small or not detected amounts in flavors. When quantified by a

colorimetric method, and after dilution in water, it is difficult to detect them.

Reducing Power Assay . In reducing power determination, the yellow color of the

solution changes to various shades of green and blue, depending on the compounds

present in the solution. The presence of antioxidants causes the reduction of the Fe3+

/ferricyanide complex to the ferrous form.

Table 3 presents the values of reducing power from flavors and flavored waters. As

expected, flavors presented higher values than flavored waters due to their higher

concentrations of bioactive compounds. Nevertheless, samples from brand F present

similar values, and one sample from brand G (sample 29) presents values 15-fold higher

than those in flavors.

With regard to flavors, reducing powers are very similar (from 10.2 to 11.8 mg of GA/L).

No correlation among the different evaluated parameters was verified. As referred to

above, flavors presented very different TPC values. It is interesting to remember that the

lowest TPC value (gooseberry) corresponds to the highest value in reducing power

determination. Some studies indicated a high reducing power activity in wild fruits.25

All natural waters and three flavored water samples from the same brand (27, 28, and

30) presented values of this parameter below the LOD. It should be noted that samples 27

and 30 have vitamin C as an added ingredient and sample 30 presented the second

highest content in TPC. Inversely, sample 29, with apple flavor, had the highest value in

TPC and reducing capacity and had also vitamin C as an added ingredient. The highest

reducing power values were obtained (like in TPC) in flavored waters with bioactive

compounds (tea, ginseng, and G. biloba) ranging from 8.3 to 13.8 mg of GA/L.

It is verified that samples from the same brand had similar values, except for brand C

(sample 10, without addition of bioactive compounds) and brand E (sample 19 with a

value 5-fold higher than the other samples). This behavior occurred also in TPC values,

sample 19 being also the richest in these compounds.

From a general point of view and except for brand F (with values similar to flavors) and

brand G, as referred to above, all brands can be grouped into two sets: A, D, E, H, and J

with lower values ranging from 3 to 4 mg/L of GA/L; and B, C, and I with relatively higher

values near 6-7mg of GA/L.

DPPH RSA. DPPH RSA is a technique based on the reduction of the DPPH radical in

the presence of a hydrogen-donating antioxidant. A DPPH solution, freshly prepared,

exhibits a deep purple color with maximum absorption at 517 nm. This color disappears in

the presence of an antioxidant, because antioxidant molecules can quench DPPH free

radicals and convert them into a colorless product. Hence, the more rapidly the

absorbance decreases, the more potent is the antioxidant. Table 2 presents RSA values


117

obtained with water samples and flavors. According to the previously discussed

parameters, flavors presented, in general, higher RSA values than flavored waters, except

some samples of brand G (samples 29 and 30) with higher values than some flavors.

Flavor RSA values ranged from 39 (lemon) to 214 (strawberry) mg of Trolox/L. The

highest RSA values were determined in strawberry and gooseberry flavors (214 and 212

mg of Trolox/L), which presented the lowest values of TPC (15 and 9 mg of GA/L). Choi

and collaborators26 reported the RSAs of 34 kinds of citrus essential oils and their

components by HPLC, showing that all essential oils have scavenging effects on DPPH

ranging from 5.4 to 172 mg of Trolox equiv/mL.

As with other parameters (TPC and reducing power), flavored waters with bioactive

compounds (tea, ginseng, and G. biloba) have increased RSA values, demonstrating the

dual effect of radical scavenging of these bioactive compounds. Further global

comparisons are difficult to establish due to the fact that different standards are used in

the several analytical methods described.

GC-MS Analysis of Flavors/Fragances . Six flavors were evaluated with regard to

antioxidant activity, but only citrus flavors (lime, lemon, and tangerine) were analyzed by

GC-MS. Flavors (essential oils) are volatile and complex natural mixtures characterized by

a strong odor, which can contain about 20-60 components at quite different

concentrations. Terpenes and terpenoids constituted the main group of compounds with

other aromatic and aliphatic constituents, all characterized by low molecular weight.7

Volatile compound profiles were obtained by HS-SPME using a PDMS/DVB fiber and

analyzed by GC-MS. Figure 1 shows the chromatograms of citrus flavors (lime, lemon,

and tangerine). The characterization of individual components was performed with mass

spectrometry (MS). Qualitative and quantitative composition of the citrus flavors (lime,

lemon, and tangerine), obtained by comparison of mass spectra data, and library data are

listed in Tables 4-6. A total of 28 terpenes were identified: 22 monoterpenes and 6

sesquiterpenes. Terpenes are a combination of several 5-carbon-base (C5) units called

isoprenes. The main terpenes are monoterpenes (C10) and sesquiterpenes (C15). A

terpene containing oxygen is called a terpenoid. The monoterpenes identified and present

in the flavors can be classified as (i) acyclic (β-myrcene) (ii) monocyclic (α-limonene, γ-

terpinene, o-cymene; β-cymene); (iii) bicyclic (6-isopropylidene-1-

methylbicyclo[3.1.0]hexane, (-)-β-pinene, 3-carene); (iv) terpenoid alcohol acyclic (cis-

geraniol); (v) terpenoid alcohol monocyclic (1,6-dihydrocarveol, 1-terpinen-4-ol, α-

terpineol, L-isopulegol, carveol, cis-p-menth-2,8-dienol, cis-β-terpineol); (vi) terpenoid

aldehyde (geranial, p-mentha-1,8-dien-7-al); (vii) terpenoid ester (linalyl butyrate); (viii)

terpenoid ether (1,4-cineole, 1,8-cineole); and (ix) terpenoid phenol (carvacrol).


118

Figure 1. GC-MS chromatograms of flavor extracts: (a) lime; (b) lemon; (c) tangerine.

a)

b)

c)

11

9

Tab

le 4

. C

hem

ical

Com

posi

tion

of th

e V

olat

ile F

ract

ion

of L

ime

Fla

vor

peak

re

tent

ion

time

(min

) co

mpo

und

MW

m

/z

rela

tive

cont

enta (

%)

Mon

oter

pene

s

1 9

.53

(-)-β

-Pin

ene

13

6 93

; 41;

91

2.1

7 2

10.4

0 1,

4-C

ineo

l 15

4 11

1; 4

3; 7

1 1

.06

3 10

.60

o-cy

men

e 13

4 11

9; 1

34

0.6

7 4

10.6

9 α

-lim

onen

e 13

6 68

; 67;

93

7.4

5 5

10.7

4 eu

caly

ptol

(1,

8-C

ineo

l) 15

4 43

; 81;

108

1

.87

6 11

.33

γ-t

erpi

nene

13

6 93

; 91

1.4

3 7

12.1

7 lin

alyl

but

yrat

e 22

4 93

; 43;

41

1.2

7 8

12.8

7 3-

care

ne

136

93; 9

1; 7

9 0

.95

9 13

.07

p-m

enth

-8-e

n-2-

ol (

1,6-

dihy

droc

arve

ol)

154

93; 1

07; 1

21; 1

36

0.8

6 10

13

.65

n-oc

tana

l dim

ethy

l ace

tal

174

75; 7

1; 4

1 0

.38

11

13.7

0 1-

terp

inen

-4-o

l 15

4 71

; 93;

111

1

.90

12

13.9

8 α

-ter

pine

ol

154

59; 9

3; 1

21

19.3

4 13

14

.09

6-is

opro

pylid

ene-

1-m

ethy

l-bic

yclo

[3.1

.0]h

exan

e 13

6 12

1; 9

3; 1

36

2.4

3 14

14

.19

L-is

opul

egol

15

4 41

; 67;

69;

81;

55

1.0

6 15

14

.71

gera

nyl i

sova

lera

te

238

85; 4

3; 5

7; 4

1; 6

9 0

.52

16

14.8

7 ca

rveo

l (p-

men

tha-

1,8-

dien

-6-o

l) 15

2 11

9; 9

1; 1

34

17.1

3 17

15

.40

citr

al (

gera

nial

) 15

2 69

; 41;

84

27.1

2 18

15

.49

cis-

p-m

enth

-2,8

-die

nol

152

91; 1

34; 4

3; 1

19; 1

34

0.4

7 19

16

.94

cis-

gera

niol

15

4 93

; 41;

91

6.7

4 20

17

.25

β-m

yrce

ne

136

93; 4

1; 9

1 4

.19

Ses

quite

rpen

es

21

17.9

6 β

-car

yoph

ylle

ne

204

41; 6

9; 9

3; 1

33; 7

9 0

.30

22

18.1

5 α

-ber

gam

oten

e 20

4 93

; 41;

119

; 91

0.3

8 23

19

.27

β-b

isab

olen

e 20

4 69

; 41;

93

0.3

1 a R

elat

ive

cont

ent w

as c

alcu

late

d fr

om a

rea

ratio

.

12

0

Tab

le 5

. C

hem

ical

com

posi

tion

of th

e vo

latil

e fr

actio

n of

lem

on fl

avou

r.

peak

re

tent

ion

time

(min

) co

mpo

und

MW

m

/z

rela

tive

cont

enta

(%)

Mon

oter

pene

s 1

10.4

0 ci

s-β

-Ter

pine

ol

154

43; 7

1 2

.55

2 10

.69

α-L

imon

ene

(p-M

enth

a-1,

8-di

ene)

13

6 68

; 67;

93

8.5

7 3

10.7

5 eu

capl

ypto

l (1,

8-C

ineo

l) 15

4 43

; 81;

108

2

.95

4 13

.96

α-T

erpi

neol

(p-

Men

th-1

,8-d

ien-

6-ol

) 15

4 59

; 93;

121

2

.92

5 14

.85

carv

eol (

p-M

enth

a-1,

8-di

en-6

-ol)

152

119;

91;

134

27

.09

6 15

.37

citr

al (

Ger

ania

l) 15

2 69

; 41;

84

44.3

2 S

esqu

iterp

enes

7

17.9

5 β

-Car

yoph

ylle

n 20

4 41

; 69;

93;

133

; 79

11.5

9 a R

elat

ive

cont

ent w

as c

alcu

late

d fr

om a

rea

ratio

.

Tab

le 6

. C

hem

ical

com

posi

tion

of th

e vo

latil

e fr

actio

n of

tang

erin

e fla

vour

. pe

ak

rete

ntio

n tim

e (m

in)

com

poun

d M

W

m/z

a re

lativ

e co

nten

t (%

) M

onot

erpe

nes

1 10

.61

β-c

ymen

e 13

4 11

9; 9

1 0

.89

2 10

.69

α-li

mon

ene

(p-m

enth

a-1,

8-di

ene)

13

6 68

; 67;

93

11.7

8 3

11.3

3 γ-

terp

inen

e (p

-men

tha-

1,4-

dien

e)

136

93; 9

1 3

.76

4 12

.17

linal

yl b

utyr

ate

224

93; 4

3; 4

1 1

.01

5 13

.70

4-te

rpin

eol (

p-m

enth

-1-e

n-4-

ol; 1

-ter

pene

n-4-

ol)

154

71; 9

3; 1

11

4.7

4 6

13.9

5 α

-ter

pine

ol (

p-m

enth

-1-e

n-8-

ol)

154

59; 9

3; 1

21

11.3

8 7

14.1

9 n-

deca

nal

156

41; 4

3; 5

7 1

.01

8 15

.49

p-m

enth

a-1,

8-di

en-7

-al (

(-)-

peril

lald

ehyd

e)

150

68; 7

9 1

.22

9 15

.73

carv

acro

l (p-

cym

en-2

-ol)

150

135;

150

2

.29

10

17.6

7 n-

dode

cana

l 18

4 41

; 57;

55

0.7

1 11

17

.76

met

hyla

min

oben

zoat

e 16

5 16

5; 1

05

48.7

5 S

esqu

iterp

enes

12

17

.96

β-c

aryo

phyl

len

204

41; 6

9; 9

3; 1

33; 7

9 3

.84

13

19.1

4 α

-sel

inen

e 20

4 10

8; 2

04; 9

3 1

.54

14

19.2

2 α

-far

nese

ne

204

41; 9

3 6

.65

15

19.5

3 β

-cad

inen

e 20

4 16

1; 2

04; 1

34

0.4

4 a R

elat

ive

cont

ent w

as c

alcu

late

d fr

om a

rea

ratio

.


121

The sesquiterpenes are classified as (x) acyclic (α-farnesene); (xi) monocyclic (β-

bisabolene); and (xii) bicyclic (β-caryophyllene, α-bergamotene, α-selenene, and β-

cadinene).

For the lime flavor the mass spectral data revealed that monoterpenes represented

>99% of the volatile fraction. Citral (27.12%) was the major ingredient followed by α-

terpineol (19.34%), carveol (17.13%), and α-Limonene (7.45%). The sesquiterpenes β-

bisabolene (0.31%), β-caryophyllene (0.30%), and α-bergamotene (0.38%) were in minor

quantities.

In lemon flavor, the volatile fraction extracted was represented by 88.41% of

monoterpenes and 11.59% of sesquiterpenes. The major ingredients were the terpenes

citral (44.32%), carveol (27.09%), and α-limonene (8.57%) and the sesquiterpene β-

caryophyllene (11.59%). With regard to the tangerine flavor, methyl aminobonzoate

(48.75%), α-limonene (11.78%) and α-terpineol (11.38%) were the major compounds

followed by the sesquiterpene α-farnesene (6.65%). With regard to the tangerine flavor, α-

limonene (11.78%) and α-terpineol (11.38%) were the major compounds followed by the

sesquiterpene α-farnesene (6.65%).

By comparison of the obtained results with those of the literature, several analogies can

be pointed out. According to studies carried out by several authors, the essential oil

obtained from citrus fruit (orange, lemon, bergamot, grapefruit) had a similar composition

to that described in this study, considering only the analysis of the most volatile fraction of

the essence.9,27-29 Some authors reported that the major ingredients present in essential

oils from citrus fruit (orange and lemon) is limonene9,28 followed by α- and β-pinenes and

γ-terpinene.28 However, Caccioni and collaborators27 reported that lemon oil collected in

February showed the highest content of oxygenated compounds, two geraniol-geranial

and nerol-neral couples being the main compounds. Thus, the analysis and extraction of

the compounds in flavors can change in quality and quantity with seasonal variation,

ripeness, soil composition, and geographical region.7,8 Almost authors agree that

monoterpenes make up 97% of the citrus oil composition, with alcohol, aldehydes, and

esters being the lowest percentage components ranging from 1.8 to 2.2%.29 Flavonoids

are another group of components that are present in citrus flavors, making up the

nonvolatile part of the oils.8 Indeed, antimicrobial, antifungal, antioxidant, and radical

scavenging properties have been reported for flavors (essential oils) and fruits.9 Di Vaio

and collaborators9 reported that the peel ethanol extract from lemon presented antioxidant

activity and high radical scavenging power, suggesting that lemon essential oils and their

related flavor components may contribute to preventing oxidation in foods and inhibit lipid

oxidation. Other studies reported by Crowell30 revealed that terpenoids such as carveol


122

and limonene present in plant essential oils are effective in treating breast, liver, and/or

other cancers.

Funding Sources

This research was supported by a Ph.D. grant from FCT (Fundação para a Ciência e

Tecnologia - SFRH/BD/29440/2006).

ACKNOWLEDGMENT

We acknowledge the Frize (Portuguese farmers) for providing flavor samples.


123

REFERENCES

(1) Dixit Y.; Kar A. Antioxidant activity of some vegetable peels determined in vitro by

inducing liver lipid peroxidation. Food Res. Int. 2009, 42, 1351-1354.

(2) Chen, T.-S.; Liou, S.-Y.; Wu, H.-C.; Tsai, F.-J.; Tsai, C.-H.; Huang, C.-Y; Chang, Y.-

L. New analytical method for investigating the antioxidant Power of food extracts on the

basis of their electron-donating ability: Comparison to the ferric reducing/antioxidant

power (FRAP) assay. J. Agric. Food Chem. 2010, 58, 8477–8480.

(3) Sánchez-Moreno, C. Review: Methods used to evaluate the free radical scavenging

activity in foods and biological systems. Food Sci. Technol. Int. 2002, 8, 121-137.

(4) Prior, R. L.; Wu, X.; Schaich, K. Standardized Methods for the determination of

antioxidant capacity and phenolic in foods and dietary supplements. J. Agric. Food Chem.

2005, 53, 4290-4302.

(5) Rapisarda, P.; Fabroni, S.; Peterek, S.; Russo, G.; Mock, H. P. Juice of new citrus

hybrids (Citrus clementina Hort. Ex Tan.C. Sinensis L. Osbeck) as a source of natural

antioxidants. Food Chem. 2009, 117, 212-218.

(6) Lichtenthäler, R.; Marx, F. Total Oxidant Scavenging Capacities of Common

European Fruit and Vegetable Juices. J. Agric. Food Chem. 2005, 53, 103-110.

(7) Bakkali, F.; Averbeck, S.; Averbeck, D.; Adaomar, M. Biological effects of essentials

oils – A review. Food Chem. Toxicol. 2008, 46, 446-475.

(8) Fisher, K.; Phillips, C. Potential antimicrobial uses of essential oils in food: is citrus

the answer?. Trends Food. Sci. Tech. 2008, 19, 156-164.

(9) Di Vaio, C.; Graziani, G.; Gaspari, A.; Scaglione, G.; Nocerino, S.; Ritieni, A.

Essential oils content and antioxidant properties of peel ethanol extract in 18 lemon

cultivars. Sci. Hortic. 2010, 126, 50-55.

(10) Barroso, M. F.; Silva, A.; Ramos, S.; Oliva-Teles, M. T.; Delerue-Matos, C.; Sales,

M. G. F.; Oliveira, M. B. P. P. Flavoured versus natural waters: Macromineral (Ca, Mg, K,

Na) and micromineral (Fe, Cu, Zn) contents. Food Chem. 2009, 116, 580-589.

(11) Barroso, M. F.; Ramos, S.; Oliva-Teles, M. T.; Delerue-Matos, C.; Sales, M. G. F.;

Oliveira, M. B. P. P. A survey of trace elements (Al, As, Cd, Cr, Co, Hg, Mn, Ni, Pb, Se

and Si) in retail samples of flavoured and bottled waters. Food Addit. Contam. Part B-

Surveill. 2009, 2, 121–130.

(12) Haghi, G.; Hatami, A. Simultaneous quantification of flavonoids and phenolic acids

in plant materials by a newly developed isocratic high-performance liquid chromatography

approach. J. Agri. Food chem. 2010, 58, 10812-10816.


124

(13) Singleton, V. L.; Rossi, J. A. Colorimetric of total phenolics with phosphomolybdic-

phosphotungstic acid reagents. Am. J. Enol. Vitic. 1965, 16, 144-158.

(14) Jia, Z.; Tang, M.; Wu, J. The determination of flavonoid contents in mulberry and

their scavenging effects on superoxide radicals. Food Chem. 1999, 64, 555-559.

(15) Oyaizu, M. Studies on products of browning reactions: Antioxidative activities of

products of browning reaction prepared from glucosamine. Jpn. J. Nutr. 1986, 44, 307-

315.

(16) Hatano, T.; Kagawa, H.; Yashura, T.; Okuda, T. Two new flavonoids and other

constituents in locorice root: Their relative astringency and radical scavenging effects.

Chem. Pharm. Bull. 1988, 36, 2090-2097.

(17) Miller, J. N.; Miller, J. C. Statistics and chemometrics for analytical chemistry.

London: Pearson Education, 2000.

(18) Pawliszyn, J. Applications of solid-phase microextraction. Cambridge: Royal

Society of Chemistry, 1999.

(19) Harmon, A. D. Solid-phase microextraction for the analysis of aromas and flavors.

In: MARSILI, R. (Ed.) Flavor, fragrance, and odor analysis. New York: Marcel Dekker,

2002, p. 75-106.

(20) Wang, Y.; Ho, C.-T. Polyphenolic Chemistry of Tea and Coffee: A Century of

Progress. J. Agri. Food Chem. 2009, 57, 8109-8114.

(21) Zulueta, A.; Esteve, M. J.; Frasquet, I.; Frígola, A. Vitamin C, vitamin A, phenolic

compounds and total antioxidant capacity of new fruit juice and skim milk mixture

beverages marketed in Spain. Food Chem., 2007, 103, 1365-1374.

(22) Mayachiew, P.; Devahastin, S. (2008) Antimicrobial and antioxidant activities of

Indian gooseberry and galangal extracts. LWT-Food Sci. Technol. 2008, 41, 1153-1159.

(23) Saura-Calixto, F.; Goñi, I. Antioxidant capacity of the Spanish Mediterranean diet.

Food Chem. 2006, 94, 442–447.

(24) Tabart, J.; Kever, C.; Pincemail, J.; Defraigne, J. O.; Dommes, J. Evaluation of

spectrophotometric methods for antioxidant compound measurement in relation to total

antioxidant capacity in beverages. Food Chem. 2010, 120, 607-614.

(25) Ashwell, R.; Ndhlala, A. R.; Chitindingu, K.; Mupure, C.; Murenje, T.; Ndhlala, F.;

Benhura, M.A.; Muchuweti, M.; Antioxidant properties of methanolic extracts from

Diospyros mespiliformis (jackal berry), Flacourtia indica (Batoka plum), Uapaca kirkiana

(wild loquat) and Ziziphus mauritiana (yellow berry) fruits. Int. J. Food Sci. Technol. 2008,

43, 284–288.

(26) Choi, H.S; Song, H.S.; Ukeda, H.; Masayoshi Sawamura, M. Radical-Scavenging

Activities of Citrus Essential Oils and Their Components: Detection Using 1,1-Diphenyl-2-

picrylhydrazyl. J. Agric. Food Chem. 2000, 48, 4156-4161.


125

(27) Caccioni, D. R. L.; Guizzardi, M.; Biondi, D. M.; Renda, A., Ruberto, G. 1998)

Relationsship betwwen volatile components of citrus essencial oils and antimicrobial

action on Penicillium digitatum and Penicillium italicum. Int. J. Food Microbiol. 1998, 43,

73-79.

(28) Michaelakis, A.; Papachristos, D.; Kimbaris, A.; Koliopoulos, G.; Giatropoulos, A.;

Polissiou, M. G. Citrus essential oils and four enantiometric pinenes against Culex pipiens

(Diptera: Culicidae). Parasitol. Res. 2009, 105, 769-773.

(29) Moufida S.; Marzouk, B. Biochemical characterization of blood Orange, sweet

orange, lemon, bergamot and bitter orange. Phytochemistry 2003, 62, 1283-1289.

(30) Crowell, P.L. Monoterpenes in breast cancer chemoprevention. Breast Cancer Res.

Treat. 1997, 46, 191-197.

127

Capítulo 4

Construção de biossensores de bases púricas

4.1.

Electrochemical evaluation of total antioxidant capacity of beverages using a purine-biosensor


Food Chemistry (submetido)

4.2.

Electrochemical DNA-sensor for evaluation of total antioxidant capacity of flavours and flavoured

waters using superoxide radical damage



4.3.

Evaluation of total antioxidant capacity of flavoured waters using sulfate radical damage of purine-

based sensors


Electrochimica Acta (submetido)

4.1. Biossensores de bases púricas – radical hidroxilo

129

Electrochemical evaluation of total antioxidant cap acity of

beverages using a purine-biosensor

M. Fátima Barroso1,2, C. Delerue-Matos1, M. B. P. P. Oliveira2 1REQUIMTE/Instituto Superior de Engenharia do Porto.

Dr. Bernardino de Almeida 431, 4200-072 Porto. Portugal 2Requimte, Serviço de Bromatologia, Faculdade de Farmácia, Universidade do Porto.

R. Aníbal Cunha n.º 164, 4050-047 Porto. Portugal

Abstract

In this paper, it was evaluated the total antioxidant capacity (TAC) of beverages using

an electrochemical biosensor. The biosensor consisted on the purine base (guanine or

adenine) electro-immobilization on a glassy carbon electrode surface (GCE). Purine base

damage was induced by the hydroxyl radical generated by a Fenton-type reaction. Five

antioxidants were applied to counteract the deleterious effects of the hydroxyl radical. The

antioxidants used were ascorbic acid, gallic acid, caffeic acid, coumaric acid and

resveratrol. These antioxidants have the ability to scavenger the hydroxyl radical and

protect the guanine and adenine immobilized on the GCE surface. The interaction carried

out between the purine-base immobilized and the free radical in the absence and

presence of antioxidants was evaluated by means of changes in the guanine and adenine

anodic peak obtained by square wave voltammetry (SWV). The results demonstrated that

the purine-biosensors are suitable for rapid assessment of TAC in beverages.

Keywords: Purine bases; Total antioxidant capacity (TAC); Ascorbic acid; Phenolic acid;

Reactive oxygen species (ROS); Hydroxyl radical (OH•); Biosensor


Food Chemistry submitted


131

1. Introduction

In recent years, the interest for DNA-based diagnostic tests has been growing. The

development of systems allowing DNA detection is motivated by applications in many

fields: DNA diagnostics, fast detection of biological warfare agents and forensic

applications. Detection of genetic mutations at the molecular level opens up the possibility

of performing reliable diagnostics even before any symptom of a disease appears

(Sassolas, Leca-Bouvier, & Blum, 2008).

Reactive oxygen species (ROS) produced in living organisms by normal metabolism

and by exogenous sources such as carcinogenic compounds and ionizing radiations

induce oxidative DNA damage producing a variety of modifications at DNA level including

base and sugar lesions, strand breaks, DNA-protein cross-link and base-free sites

(Dizdaroglu, Jaruga, Birincioglu, & Rodriguez, 2002; Mello, Hernandez, Marrazza,

Mascini, & Kubota, 2006; Vertuani, Angusti, & Manfredini, 2004). However, the

mammalian cells have developed a complex endogenous defence system to repair the

damaged DNA through specific enzymes such as superoxide dismutase, catalase,

peroxidase, myeloperoxidase, that are involved in the base excision repair (Cadet, Douki,

Gasparutto, & Ravanat, 2003). Beyond this endogenous system, the living organisms also

use exogenous antioxidant compounds. An antioxidant is any substance that when

present at low concentration compared to those of an oxidizable substrate significantly

delays, inhibits or prevents oxidation of that substrate, in a chain reaction, therefore,

appears to be very important in the prevention of many diseases (Frankel, 2007; Halliwell,

Gutteridge, & Cross, 1992; Mello & Kubotta 2007). Antioxidants may delay or inhibit the

chain initiation, propagation and termination by reaction with a peroxyl radical (ROO•) or

alkoxyl radical (RO•) resulting in a lesser reactive radical (A•). In the inhibited oxidation,

termination occurs through the reaction of ROO• and RO• with a chain-breaking phenolic

antioxidant (AH), by interrupting the chain reaction by hydrogen transfer to produce a

phenoxy radical (A•) (Eq. 1 and Eq. 2) that is too stable to continue the chain by reaction.

The antioxidant radical can either react again with the ROO• (Eq. 3) and RO• (Eq. 4) to

form a stable peroxide or hydroxyl or react with another antioxidant radical to form a dimer

(Eq. 5) (Frankel, 2007).


132

(5) A -A A A

(4) ROA A RO

(3) ROOA A ROO

(2) A ROH AH RO

(1) A ROOH AH ROO

→+

→+

→++→+

+→+

••

••

••

••

••

Increasing intake of dietary antioxidant may help to maintain an adequate antioxidant

status and, therefore, the normal physiological functions of a living system. Some

functional foods, vegetables, fruits, whole-grain cereals, wine and infusions are good

sources of exogenous antioxidants (Ignat, Volf & Popa, 2011). These foodstuff and

beverages include in its composition exogenous antioxidants such as vitamins (A, E, C),

phenolic compounds (gallic acid, caffeic acid, ferulic acid, p-coumaric acid, sinapic acid),

flavonoids (quertecin, rutin), minerals (selenium, zinc) or proteins (transferrin,

ceruloplasmin, albumin).

Ascorbic acid is a γ-lactone synthesized by plants and many animals (except primates).

This powerful exogenous antioxidant is a water-soluble vitamin, and plays a key role in the

protection against biological oxidation processes. Indeed, ascorbic acid is a good

scavenger of free radicals acting as a reducing agent by donation of a one electron

producing the semi-dehydroascorbate radical. It justifies its association to protection

against cancer agents by the prevention of formation of carcinogens precursors’

compound (Lee, Davis, Rettmer, & Lable, 1988; Mello & Kubotta, 2007; Smiroff, 2000).

Phenolic compounds (originated from vegetables) alsi present antioxidant activity. In

general, the antioxidant activity of the phenolics-derived compounds is determined by its

ideal chemical structure in terms of some properties such as free-radical scavengers or

chain breakers agents. It also, the fact of the resulting antioxidant-derived radical, namely

phenoxy radical is relatively stable due to the resonance delocalization and lack of

suitable sites for attack by molecular oxygen. The last property, the transition metal-

chelating potential, in special iron and copper supports the role of polyphenols as

preventive antioxidants in terms of inhibiting transition metal-catalysed free radical

formation (Soobrattee, Neergheen, Ramma, Aruoma, & Bahorun, 2005; Thavasi, Leong,

& Bettens, 2006).

Several methods have been proposed for the evaluation of the total antioxidant capacity

(TAC) in biological and food samples. These methodologies are based on UV-vis

spectrometry, chemiluminescence, fluorimetry (Sanchez-Moreno, 2002), chromatography

(Jaitz, Siegl, Eder, Rak, Abranko, Koellensperger, & Hann, 2010) and electrochemistry

techniques (Piljac-Žegarac, Valek, Stipčević, & Martinez, 2010).


133

Electrochemical DNA-based have been developed in order to assess the antioxidant

capacity (Mello, & Kubota, 2007). These biosensors were based on the ds-DNA (double-

stranded DNA) (Mello, Hernandez, Marrazza, Mascini, & Kubota, & 2006), dA21

(deoxyadenylic acid) (Barroso, de-los-Santos-Álvarez, Lobo-Castañón, Miranda-Ordieres,

Delerue-Matos, Oliveira, & Tuñón-Blanco, 2011a) immobilization on the electrode surface,

as oxidation target and a Fenton-type reaction were used for (hydroxyl) OH• generation

(Eq. 6). Hydroxyl radicals interact with DNA bases inducing damage.

)6( OHOH Fe OH Fe 322

2 •−++ ++→+

In this work, the TAC of flavoured waters was evaluated using a purine-based

biosensor. This purine-based biosensor consisted on the electro-deposition of purine

bases (guanine and adenine) on a glassy carbon electrode (GCE) surface. The biosensor

was damaged by the hydroxyl radical according the procedure of Kamel and collaborators

(Kamel, Moreira, Delerue-Matos, & Sales, 2008). The influence of five antioxidants on the

scavenger free radical activity was studied. The antioxidants used were ascorbic acid, and

the following phenolic acids, gallic acid, caffeic acid, coumaric acid and resveratrol

(polyphenol). The protective effect of these five antioxidants on the purine bases was

observed. Square wave voltammetry (SWV) was the electroanalytical technique used to

relate the extent of oxidative damage carried out by the hydroxyl radical and the protective

role made by antioxidants.

2. Material and methods

2.1. Chemicals

Guanine (G-0381), adenine (A-8626), iron (II) sulphate heptahydrate, hydrogen peroxide

(100 % w/v), gallic acid, resveratrol were purchased from Sigma. Caffeic acid was from

Fluka, L(+) ascorbic acid and p-coumaric were acquired from Riedel-de-Haën. Other

chemicals were Merck pro-analysis grade and were used as received. (1 g L-1) Guanine

stock solution was prepared by dissolving an amount of this solid in 0.1 mol L-1 of NaOH

and dilution in phosphate buffered saline (PBS) at pH 7.4. Stock solution of 1 g L-1 of

adenine was prepared in PBS pH 7.4 and stored at +4 ºC.

Working standard solution (ascorbic acid, gallic acid, caffeic acid, coumaric acid) were

prepared daily and immediately before measurements by dissolving an amount of the

solid standard in water until the desired concentration. In order to dissolve the resveratrol


134

antioxidant, an amount of this compound was dissolved in ethanol and then diluted with

water until to the desired concentration.

Hydroxyl radical was generated by mixing Fe2+:EDTA:H2O2 (0.20 mmol L-1: 0.40 mmol L-1

:8 mmol L-1) in the molar ratio of 1:2:40. Mello and collaborators (Mello, Hernandez,

Marrazza, Mascini, & Kubota, 2006) reported that when an excess of hydrogen peroxide

is added in the reaction a high DNA damage is obtained. EDTA was added for solubility

reasons. All solutions were prepared with water purified with a Direct-Q (Millipore) system.

2.2. Apparatus

Square wave voltammetry (SWV) was performed with an Autolab PSTAT 10

potentiostat controlled by GPES software (EcoChemie, The Netherlands). A conventional

three electrode cell was used, which includes glassy carbon electrode (GCE) (0.07 cm2)

as working electrode, a glassy carbon counter electrode and a Ag|AgCl|KClsat reference

electrode to which all potentials are referred. GCE was mechanically polished using a

polishing kit (Metrohm 6.2802.010) first with γ-Al2O3 (0.015 µm) until a shining surface

was obtained and after with only water. After this step the GCE was treated by applying a

fixed potential of +1.7 V for 30 s in PBS pH 4.8. This initial conditioning step improves the

resolution of the analytical signal because the application of high potentials in acidic

medium increases the hydrophilic properties of the electrode surface through the

introduction of oxygenated functionalities (Mello, Hernandez, Marrazza, Mascini, &

Kubota, 2006; Rice, Galus, & Adams, 1983).

2.3. Voltammetric procedure

Unless otherwise mentioned, all experiments consisted of three steps: i) guanine or

adenine electro-immobilization on the purine-based GCE, ii) damage of purine bases by

the immersion of DNA-GCE on the hydroxyl radical, and study the effect of the presence

of antioxidants in the reactive system; iii) detection and measurement of the peak current

of adenine or guanine in a PBS at pH 7.4.

Purine base (adenine or guanine) immobilization was performed by the application of an

adsorptive accumulation step. For that, the activated GCE was immersed in PBS pH 4.8

containing 10 mg L-1 of adenine or 3 mg L-1 of guanine and it was applied a positive

potential of +0.4 V for 180 s, after this the electrode was washed with water (Scheme 1).

For the purine bases biosensor preparation procedure (cleaning and immobilization step)

it was used the conditions optimized in previous works (Kamel, Moreira, Delerue-Matos, &

Sales, 2008; Mello, Hernandez, Marrazza, Mascini, & Kubota, 2006).


135

Scheme 1. Electroimmobilization of the purine base on the CGE surface procedure and the SWV signal of the purine-based sensor in PBS (PH 4.8): a) blank signal (maximum peak current); after b) immersion in hydroxyl radical; c) immersion in hydroxyl radical with an antioxidant (ascorbic acid).

electroimmobilization step

detection stepSWV technique(frequency= 50 Hzstep potential = 4.12 mV amplitude= 0.09 V)

blank signal

immersionin hydroxyl radical

immersion in hydroxyl radicalwith an antioxidant (ascorbic acid)

N

N

NH2

N

HN

activated GCE surface+++++++

E / V15.0

25.0

35.0

45.0

0.50 0.70 0.90 1.10

i / µ

A

Ep = 0.82 V

E / V15.0

25.0

35.0

45.0

0.50 0.70 0.90 1.10

i / µ

A

E / V15.0

25.0

35.0

45.0

0.50 0.70 0.90 1.10

i / µ

A

a)

b)

c)


+ 0.4 V 180 s

N

N

NH2

N

HN

adenineadenine



blank signal



N

N

NH2

N

HN


N

N

NH2

N

HN



E / V15.0

25.0

35.0

45.0

0.50 0.70 0.90 1.10

i / µ

A

Ep = 0.82 V

E / V15.0

25.0

35.0

45.0

0.50 0.70 0.90 1.10

i / µ

A

Ep = 0.82 V

E / V15.0

25.0

35.0

45.0

0.50 0.70 0.90 1.10

i / µ

A

E / V15.0

25.0

35.0

45.0

0.50 0.70 0.90 1.10

i / µ

A

E / V15.0

25.0

35.0

45.0

0.50 0.70 0.90 1.10

i / µ

A

E / V15.0

25.0

35.0

45.0

0.50 0.70 0.90 1.10

i / µ

A

E / V15.0

25.0

35.0

45.0

0.50 0.70 0.90 1.10

i / µ

A

E / V15.0

25.0

35.0

45.0

0.50 0.70 0.90 1.10

i / µ

A

a)

b)

c)


+ 0.4 V 180 s

N

N

NH2

N

HN

adenineadenine


+ 0.4 V 180 s


+ 0.4 V 180 s

N

N

NH2

N

HN

adenineadenine


+ 0.4 V 180 s

NH

NNH2N

HN

O

guanine


blank signal

25.0

35.0

45.0

55.0

0.20 0.40 0.6 0.80E / V

i / µ

A

Ep=0.55 V

25.0

35.0

45.0

55.0

0.20 0.40 0.60 0.80E / V

i / µ

A

25.0

35.0

45.0

55.0

0.20 0.40 0.60 0.80E / V

i / µ

A

HN N

N

NH2

HN

O





guanine


+ 0.4 V 180 s

NH

NNH2N

HN

O

guanine


+ 0.4 V 180 s


+ 0.4 V 180 s

NH

NNH2N

HN

O

guanine


blank signal

25.0

35.0

45.0

55.0

0.20 0.40 0.6 0.80E / V

i / µ

A

Ep=0.55 V

25.0

35.0

45.0

55.0

0.20 0.40 0.6 0.80E / V

i / µ

A

Ep=0.55 V

25.0

35.0

45.0

55.0

0.20 0.40 0.60 0.80E / V

i / µ

A

25.0

35.0

45.0

55.0

0.20 0.40 0.60 0.80E / V

i / µ

A

25.0

35.0

45.0

55.0

0.20 0.40 0.60 0.80E / V

i / µ

A25.0

35.0

45.0

55.0

0.20 0.40 0.60 0.80E / V

i / µ

A

HN N

N

NH2

HN

O


HN N

N

NH2

HN

O






guanine

2

1

a)

b)

c)


136

Purine base damage was carried out by immersing the biosensor in a freshly prepared

Fenton solution in the absence or in the presence of antioxidant in PBS pH 7.4. After a

fixed period of reaction time, the purine-based biosensor was rinsed with water and

immediately immersed in PBS (pH 4.8) to carry out the SWV between +0.2 V and +1.4 V.

(frequency 50 Hz, step potential 4.12 mV and amplitude 0.09 V). The peak current of

guanine and adenine obtained was used as a detection signal. For the electrochemical

studies it was considered that the maximum signal current obtained was for the purine

base signal without damage neither antioxidant effect (Scheme 1).

2.4. Samples

Thirty-nine water samples corresponding to 10 different brands were purchased in

several supermarkets in the North of Portugal and stored in the dark at +4 ºC. Each brand

(still or sparkling, mineral or spring water) had different flavours and aromas. The natural

water of each brand was also used as control. Sonication (30 min) was used to eliminate

gas from the sparkling water samples. The labels on the water bottles indicate the nutrient

information, namely the presence of fruit juice, vitamins, sweeteners and preservatives.

Six liquid flavours used in the formulation of some water brands, provided by a

producer, were also analysed. The flavours used corresponded to different fruit aromas,

such as lime, tangerine, strawberry, lemon, apple and gooseberry. These flavours had no

description about their chemical or aroma composition, but were known to be present in

the flavoured waters used in this study.

2.5. TAC measurement on beverages

The purine-based biosensor was applied to the determination of TAC on flavour and

flavoured waters. For the measurement of TAC in beverages, a volume of the flavoured

water or flavour were diluted in PBS to a final volume of 500 µl. Then, the purine-based

GCE was immersed in the solution and a freshly prepared hydroxyl radical was added for

120 s. After this period of time the biosensor was washed and immersed in PBS buffer to

measured the oxidation peak current of guanine and adenine. Ascorbic acid, gallic acid,

caffeic acid, coumaric acid and resveratrol were the working standard antioxidants used to

study the protective effect made by the antioxidant on the free-radical scavenging and to

carry out the linear calibrations studies. Measurements were made at least three times

and all results were expressed as mean ± standard deviation.


137


Previous studies reported in the literature indicate the oxidative damage of dsDNA

(Mello, Hernandez, Marrazza, Mascini, & Kubota, 2006), dA21 (Barroso, de-los-Santos-

Álvarez, Lobo-Castañón, Miranda-Ordieres, Delerue-Matos, Oliveira, & Tuñón-Blanco,

2011) and purine bases (Kamel, Moreira, Delerue-Matos, & Sales, 2008) induced by the

hydroxyl radical generated by the Fenton solution. Hydroxyl radical (OH•) is one of the

most reactive radical species that induce lesions in DNA. This ROS cause cell injury when

is generated in excess or the cellular antioxidant defence is impaired. When hydroxyl

radical is generated adjacent to DNA, it attacks both deoxyribose sugar and the purine

and pyrimidine bases resulting intermediates radicals, which are the immediate precursors

for DNA base damage (Jaruga & Dizdaroglu, 1996).

In order to study the protective effect promoted by antioxidants on the deactivation of

the hydroxyl radical and consequently protect the purine bases from the oxidative

damage, the purine-based biosensor was placed in a PBS pH 4.8 in presence of an

antioxidant and hydroxyl radical during 120 s. Next the biosensor was rinsed with water

and a SWV was made from +0.2 V to +1.4 V. Fig. 1 shows the performance of the purine-

based biosensor in the presence of antioxidants (0.5 mg L-1 of ascorbic acid, gallic acid,

caffeic acid, coumaric acid and resveratrol) and the hydroxyl radical.

Fig. 1. Effect of the antioxidants presence on the signal of guanine and adenine immobilized on the GCE: blank purine base signal (guanine 3 mg L-1 and adenine 10 mg L-1); after immersion in a hydroxyl radical (Fe2+: EDTA: H2O2; 0.1 mmol L-1: 0.2 mmol L-1: 4.0 mmol L-1 during 120 s); immersion in hydroxyl radical with five different antioxidants (0.50 mg L-1).

100value) expected (maximum done wasdamage no whenmeasured pi

t)antioxidan an of presence the in (or radical hydroxyl withdamage base purine after measured pi (%) signal base purine ×=

0

20

40

60

80

100

120

blank hydroxyl radical ascorbic acidhydroxyl radical

gallic acidhydroxyl radical

caffeic acidhydroxyl radical

coumaric acidhydroxyl radical

Resveratrolhydroxyl radical

guanine

adenine

pu

rin

eba

se s

ign

al (

%)

0

20

40

60

80

100

120

blank hydroxyl radical ascorbic acidhydroxyl radical

gallic acidhydroxyl radical

caffeic acidhydroxyl radical

coumaric acidhydroxyl radical

Resveratrolhydroxyl radical

guanine

adenine

pu

rin

eba

se s

ign

al (

%)


138

To perform this electrochemical study all current peaks were compared with the signal

current obtained with the non damaged adenine and guanine bases (blank signal). Purine

bases of DNA measured in SWV presented two oxidation peaks at around +0.55 V and

+0.82 V corresponding, respectively, to guanine and adenine oxidation peak (Scheme 1a

and 1b). Hydroxyl radical had the ability to produce 61.4% and 55.2% of damage in

guanine and adenine base, respectively (Fig. 1). Other free radicals had also the capacity

to induce oxidative damage on the purine bases. It was verified that and superoxide

radical produce from about 64% of damage on the guanine-based biosensor (Barroso,

Delerue-Matos, & Oliveira, 2011a) and 85% on the dA21 (Barroso, de-los-Santos-Álvarez,

Lobo-Castañón, Miranda-Ordieres, Delerue-Matos, Oliveira, & Tuñón-Blanco, 2011b)

while the sulfate radical produced 61% of damage on guanine-based biosensor (Barroso,

Delerue-Matos, & Oliveira, 2011b).

When it was added an antioxidant (0.5 mg L-1) in the reactive system a less decrease of

the anodic current of guanine and adenine was recorded. It was observed a protective

effect on the purine base carried out by the antioxidants ranged from 47 to 79%. Using the

guanine-biosensor the lowest values were found for caffeic and coumaric acid, 47.6% and

49.1%, respectively. The highest values was obtained for resveratrol (74.6 %) followed by

gallic acid (72.0%) and ascorbic acid (62.8%). Using the adenine-biosensor the protective

effective of the antioxidants ranged from 60 to 79%. The highest values was observed for

the resveratrol antioxidant (79.1%) followed by gallic acid (77.7%) and caffeic acid

(73.6%). The lowest values were found for ascorbic acid (60.4%) and coumaric acid

(61.9%). Using a DNA-based biosensor, ascorbic acid (0.5 µmol L-1) presented a

protective role of 58 % against the hydroxyl radical, and a concentration of 10 µmol L-1 of

ascorbic acid presented a protective role of 53.8 % against the superoxide radical

(Barroso, de-los-Santos-Álvarez, Lobo-Castañón, Miranda-Ordieres, Delerue-Matos,

Oliveira, & Tuñón-Blanco, 2011a; Barroso, de-los-Santos-Álvarez, Lobo-Castañón,

Miranda-Ordieres, Delerue-Matos, Oliveira, & Tuñón-Blanco, 2011b).

The protection action mode of antioxidants may involve multiple mechanisms,

depending on the source material and, possible presence of synergists and antagonists.

In general, the antioxidant activity of ascorbic acid and phenolics-derived compounds is

related to reducing properties as hydrogen or electron-donating agents, which is

determined to its reduction potential (Buettner, 1993; Mello & Kubota, 2007; et al., 2007;

Rice-Evans, 2001).


139

3.1. Optimization of the experimental conditions

In order to evaluate the TAC on beverages, some parameters concerning the damaging

reaction (iron concentration and reaction time between hydroxyl radical) at a fixed time

reaction were implemented in order to achieve the maximum purine base of DNA effect,

but without a complete damage (non-zero ip). The level of purine bases damage was

evaluated as function of the variation of the concentration of Fe2+ keeping constant the

molar ratio Fe2+: EDTA: H2O2 used (1:2:40) (Mello, Hernandez, Marrazza, Mascini, &

Kubota, 2006). Fe2+ concentration was studied between 5.0 µmol L-1 to 1.0 mmol L-1. A

range of 19% to 60% decrease in the ip of guanine and adenine immobilized on the GCE

surface was observed over the Fe2+ concentration studied. When it was used the adenine-

biosensor, a 52% decrease on the ip was recorded when the Fe2+ was increased from 50

µmol L-1 to 0.2 mmol L-1. At Fe2+ concentrations higher than 0.2 mmol L-1 the peak current

remained essentially unchanged so, this concentration was chosen for the next

experiments. At a guanine-biosensor, the increase of Fe2+ concentration promoted a

decrease of 20% to 58% in the ip. At Fe2+ concentration higher than 0.15 mmol L-1 ip was

achieved to remains unchanged, so this value was used for the next experiments.

Reaction time between the hydroxyl radical and the DNA bases immobilized on the GCE

surface depends on the half-life time of the generated free radical, so this parameter is an

important feature to optimize.

In this study the incubation time were ranged from 0 to 120 s. A 62% and a 53%

decrease on the ip of guanine and adenine, respectively was observed after an incubation

time of 120 s. Fig. 2 shows the correlation between the damage on the purine bases

measured (correlated with the anodic peak current) and the incubation time. The

incubation time of 120 s was chosen for both purine-based biosensors for all experiments.

Fig. 2. Influence on the peak current on the biosensor with the incubation time a) 10 mg L-1 adenine base; b) 3 mg L-1 guanine base.

Time (s)

4.0

6.0

8.0

10.0

12.0

14.0

0 20 40 60 80 100 120

a) guanine

b) adenine

i p(µ

A)

Time (s)

4.0

6.0

8.0

10.0

12.0

14.0

0 20 40 60 80 100 120

a) guanine

b) adenine

i p(µ

A)


140

3.2. Determination of TAC

Beverages, such as juice and infusions are an excellent source of exogenous

antioxidants. The total phenolic (TPC) reducing power and DPPH radical scavenging

activity of these flavoured waters were determined using conventional optical methods.

The polyphenols compounds were present in all flavoured water samples (0.5 to 359 mg

of gallic acid L-1). The highest TPC levels were from citrus fruits (tangerine, lime and

lemon) and from waters with bioactive compounds, like tea, gingeng and Gingko biloba.

The reducing power values were ranged from (0.14 to 11.8 mg gallic acid L-1) and DPPH

radical scavenging activity (0.29-211.5 mg trolox L-1) (Barroso, Noronha, Delerue-Matos,

& Oliveira, 2011).

For the evaluation of the TAC of flavoured waters it was used the five antioxidants

referred before. These antioxidants can be found in fruit, grapes, wine and teas. As

expected, the anodic peak current of guanine and adenine immobilised on the GCE

surface increased when the concentration of the antioxidant increased. The analytical

parameters obtained in linearity studies between antioxidants concentration and peak

current of purine-based biosensor are presented in Table 1.

Table 1. Analytical feature obtained for the 5 antioxidants standards. Parameters Ascorbic acid Gallic acid Caffeic acid Coumaric acid Resveratrol

Guanine-GCE

Linear range (mg L-1) 0.50–2.50 0.10–0.50 0.40–0.80 0.31–0.73 0.10–0.50

Slope (µA mg-1L) 2.82 9.33 8.76 9.20 11.8

Intercept (µA) 1.88 4.31 1.27 1.69 3.76

Correlation coefficient (n=5) 0.990 0.986 0.992 0.990 0.986

RSD (%) (mg L-1 ) 3.43 (2.00) 4.87 (0.30) 2.58 (0.50) 4.63 (0.50) 3.25 (0.30)

LOD (mg L-1) 0.29 0.09 0.06 0.08 0.07

Adenine-GCE

Linear range (mg L-1) 2.00–6.00 0.11–0.44 0.10–0.50 0.10–1.00 0.10–0.50

Slope (µA mg-1L) 0.40 7.38 11.9 3.81 8.78

Intercept (µA) 5.08 5.30 2.08 4.35 3.81


RSD (%) (mg L-1) 2.45 (3.00) 5.35 (0.30) 4.86 (0.30) 7.56 (0.50) 6.35 (0.30)

LOD (mg L-1) 0.99 0.08 0.07 0.27 0.10

RSD (%) = σ/[antioxidant]mean found x 100


141

Some authors reported the study of dsDNA (Liu, Roussel, Lagger, Tacchini, & Girault,

2005; Korbut, Buckova, Labuda, & Grundler, 2003; Mello, Hernandez, Marrazza, Mascini,

& Kubota, 2006), or ssDNA (Barroso, de-los-Santos-Álvarez, Lobo-Castañón, Miranda-

Ordieres, Delerue-Matos, Oliveira, & Tuñón-Blanco, 2011a) or purine bases (Kamel,

Moreira, Delerue-Matos, & Sales, 2008) damage induced by hydroxyl radical, generated

by the fenton system (Mello, Hernandez, Marrazza, Mascini, & Kubota, 2006) or UV

radical (Liu, Roussel, Lagger, Tacchini, & Girault, 2005) and its protection with the

ascorbic acid (Kamel, Moreira, Delerue-Matos, & Sales, 2008) gallic acid (Liu, Roussel,

Lagger, Tacchini, & Girault, 2005) and flavonoids (Korbut, Buckova, Labuda, & Grundler,

2003). Zhang et al. (2008) reported the study of DNA damage induced by Fenton system

on a GCE and its protection with the antioxidant ascorbic acid. Ascorbic acid promoted

protective effect on the DNA in a norrow concentration range (from 1.5 to 2.5 mmol l−1)

(Zhang, Wang, Li, Jia, Cui, & Wang, 2008). Nobushi and Uchikura, (2010) reported the

protective effects on the DNA by applying ascorbic acid as a scavenging antioxidant.

Enzyme-modified electrodes using ascorbate oxidase and peroxidase enzymes for the

detection of ascorbic acid showed linear ranges in the submM level (Mello and Kubota,

2007).

The purine-based biosensors were applied to the evaluation of TAC of flavours and

flavoured waters. Table 2 shows the TAC values expressed in mg L-1 of ascorbic acid,

gallic acid, caffeic acid, coumaric and resveratrol. It was verified that all flavours and

flavoured waters presented antioxidant capacity. Like it was expected the natural waters

not presented antioxidant capacity. Flavours presented the highest TAC values, as

demonstrated by results in Table 2. Indeed, flavours are fruit extract and have in its

composition several concentrated antioxidant compounds, so higher TAC values were

expected.

Using the guanine and adenine-biosensor the higher TAC values were found with the

antioxidant standard ascorbic acid. When it was used the guanine-based biosensor, the

flavour that presented the highest TAC value was apple, followed by tangerine,

strawberry, lemon, gooseberry and lime. At the adenine-GCE lime was the flavour that

presented the highest TAC value followed by lemon, apple, tangerine, strawberry and

gooseberry.

When the guanine-biosensor was applied to the quantification of TAC in flavoured

waters, TAC values ranged from 2.98 to 40.32 mg L-1; 0.11 to 2.27 mg L-1; 1.50 to 4.34 mg

L-1, 1.18 to 3.88 mg L-1, 0.10 to 2.05 mg L-1 with the antioxidant ascorbic acid, gallic acid,

caffeic acid, coumaric acid and resveratrol, respectively. Using ascorbic acid as standard

antioxidant, lemon flavoured waters presented the highest TAC values on all brands

(except in brand I).

14

2 Tab

le 2

. T

AC

val

ues

obta

ined

for

the

flavo

urs

and

flavo

ured

wat

ers

usin

g a

guan

ine-

base

d bi

osen

sor

and

aden

ine-

base

d bi

osen

sor.

Gua

nine

-bio

sens

or

Ade

nine

-bio

sens

or

Asc

orbi

c ac

id

Gal

lic a

cid

Caf

feic

aci

d C

oum

aric

aci

d R

esve

ratr

ol

Asc

orbi

c ac

id

Gal

lic a

cid

Caf

feic

aci

d C

oum

aric

aci

d R

esve

ratr

ol

Bra

nd

Sam

ple

m

g L-1

Lem

on

120.

23 ±

11.

48

10.3

0 ±

3.47

45

.67

± 3.

69

38.9

2 ±

3.52

12

.81

± 2.

74

526.

45 ±

35.

62

32.5

2 ±

1.92

3

0.42

± 1

.19

35.4

3 ±3

.71

21.5

3 ±

1.62

Tan

gerin

e 17

3.56

± 2

7.61

26

.41

±8.3

4 62

.84

± 8.

89

55.

27 ±

8.4

6 25

.55

± 6.

60

375.

31 ±

25.

45

16.9

4 ±

2.9

3 25

.38

± 2.

57

19.6

9 ±

2.75

14

.69

± 3.

87

App

le

185.

64 ±

5.7

7 30

.06

± 1.

74

66.7

2 ±

1.86

58

.97

± 1

.77

28.4

3 ±

1.38

47

6.83

± 1

6.69

16

.30

± 1.

97

28.7

6 ±

1.22

30

.26

± 3.

82

19.2

8 ±

1.66

Str

awbe

rry

126.

26 ±

19.

58

12.1

2 ±

5.91

47

.61

± 6.

30

40.7

7 ±

6.00

14

.24

± 4.

68

309.

82 ±

7.8

7 20

.87

± 2.

68

23.1

9 ±

1.66

12

.86

±1.8

7 11

.73

± 2.

26

Goo

sebe

rry

100.

82 ±

10.

03

9.18

± 1

.03

39.1

6 ±

3.22

32

.72

± 3.

07

7.97

± 2

.40

211.

59 ±

17.

82

15.5

8 ±

0.9

6 19

.92

± 0.

59

10.5

0 ±

0.06

10

.71

± 0.

81

Fla

vour

Lim

e 10

2.62

± 1

1.33

10

.97

± 2.

42

40.0

0 ±

3.65

33

.52

± 3

.47

8.59

± 2

.71

571.

91 ±

13.

45

14.6

4 ±

2.29

31

.94

± 4.

52

40.1

7 ±

10.1

2 23

.58

± 6.

13

A

1 Le

mon

12

.27

± 0.

69

2.27

± 0

.29

4.34

± 0

.23

3.88

± 0

.22

2.05

± 0

.17

18.3

3 ±

2.59

1.

38 ±

0.1

6 1.

26 ±

0.0

9 4.

06 ±

0.3

1 0.

71 ±

0.1

2

2 M

ango

9.

53 ±

0.7

0 1.

44 ±

0.0

9 3.

45 ±

0.1

5 3.

04 ±

0.0

4 1.

40 ±

0.0

5 7.

42 ±

3.6

1 0.

51 ±

0.0

2 0.

87 ±

0.0

2 2.

36 ±

0.0

4 0.

21 ±

0.0

6

3 S

traw

berr

y 3.

46 ±

0.4

1 -

1.50

± 0

.13

1.18

± 0

.13

- 7.

21 ±

0.7

6 0.

43 ±

005

0.

93 ±

0.0

3 2.

21 ±

0.0

7 0.

20 ±

0.0

3

4 N

atur

al

- -

- -

- -

- -

- -

B

5 P

inea

pple

/ora

nge

5.42

± 0

.52

0.20

± 0

.02

2.13

± 0

.10

1.78

± 0

.13

0.42

± 0

.09

37.5

4 ±

1.97

0.

18 ±

0.0

6 1.

92 ±

0.0

7 1.

73 ±

0.1

1 1.

57 ±

0.0

9

6

Lem

on

5.13

± 0

.28

0.11

± 0

.08

2.04

± 0

.09

1.69

± 0

.08

0.35

± 0

.07

34.0

6 ±

5.69

0.

09 ±

0.0

3 1.

72 ±

0.1

9 1.

56 ±

0.0

5 1.

41 ±

0.2

6

7

Nat

ural

-

- -

- -

- -

- -

-

C

8 Le

mon

/Mag

nesi

um

8.26

± 0

.28

1.06

± 0

.06

3.04

± 0

.05

2.65

± 0

.09

1.09

± 0

.06

22.6

6 ±

1.19

0.

11 ±

0.0

5

1.49

± 0

.04

1.59

± 0

.10

0.90

± 0

.05

9

App

le/w

hite

tea

6.99

± 0

.14

0.67

± 0

.04

2.63

± 0

.04

2.26

± 0

.04

0.79

± 0

.03

4.16

± 0

.67

0.41

± 0

.05

0.86

± 0

.02

2.17

± 0

.02

0.60

± 0

.03

10

Pin

eapp

le/fi

bre

6.41

± 0

.15

0.50

± 0

.13

2.44

± 0

.17

2.08

± 0

.13

0.65

± 0

.02

13.7

1 ±

5.52

0.

21 ±

0.0

2 1.

30 ±

0.1

8 1.

78 ±

0.0

4 0.

49 ±

0.0

5

11

Nat

ural

-

-

-

-

-

-

-

-

-

-

D

12 A

pple

4.

64 ±

0.6

9 -

1.88

± 0

.22

1.54

± 0

.21

0.23

± 0

.06

5.43

± 0

.96

0.

84 ±

0.0

7 2.

25 ±

0.1

6 0.

12 ±

0.0

2

13

Ora

nge/

peac

h 4.

59 ±

0.3

5 -

1.86

± 0

.11

1.52

± 0

.11

0.22

± 0

.08

3.02

± 0

.09

- 0.

75 ±

0.0

8 1.

52 ±

0.0

5 -

14

Lem

on

4.74

± 0

.28

- 1.

91 ±

0.0

8 1.

57 ±

0.0

4 0.

25 ±

0.0

5 3.

34 ±

0.0

8 -

0.84

± 0

.03

1.81

± 0

.08

-

15

Nat

ural

-

-

-

-

-

-

-

-

-

-

E

16 L

emon

7.

57 ±

0.2

9 0.

85 ±

0.0

6 2.

82 ±

0.1

9 2.

44 ±

0.3

2 0.

93 ±

0.0

6 19

.14

± 2.

41

1.37

± 0

.13

1.29

± 0

.08

4.02

± 0

.26

0.74

± 0

.11

17

Ora

nge/

rasp

berr

y 2.

98 ±

0.4

1 -

1.50

± 0

.14

1.18

± 0

.13

- 4.

23 ±

0.2

5 0.

52 ±

0.1

0 0.

88 ±

0.0

4 2.

39 ±

0.2

0 0.

70 ±

0.0

6

18

Pea

ch/p

inea

pple

3.

86 ±

0.1

4 -

1.63

± 0

.04

1.30

± 0

.05

- 11

.78

± 6.

15

0.34

± 0

.05

1.25

± 0

.21

2.02

± 0

.10

0.41

± 0

.02

19

Gua

va/li

me

3.66

± 0

.70

- 1.

56 ±

0.0

7 1.

24 ±

0.3

2

- 3.

96 ±

0.1

8 0.

11 ±

0.0

5 1.

01 ±

0.2

4 1.

59 ±

0.1

0 -

20

Nat

ural

-

-

-

-

-

-

-

-

-

-

F

21 L

emon

/gre

en te

a 9.

83 ±

0.2

8 1.

53 ±

0.0

4 3.

55 ±

0.9

1 3.

13 ±

0.09

1.

47 ±

0.0

7 15

.94

± 0.

86

0.34

± 0

.05

1.

26 ±

0.0

3 2.

03 ±

0.1

4 0.

60 ±

0.0

4

22

Ras

pber

ry/g

inse

ng

8.90

± 0

.35

1.25

± 0

.10

3.25

± 0

.11

2.84

± 0

.10

1.25

± 0

.08

27.3

6 ±

1.28

0.

28 ±

0.0

8 1.

59 ±

0.0

4 1.

92 ±

0.1

5 1.

11 ±

0.0

6

14

3

Gua

nine

-bio

sens

or

Ade

nine

-bio

sens

or

Asc

orbi

c ac

id

Gal

lic a

cid

Caf

feic

aci

d C

oum

aric

aci

d R

esve

ratr

ol

Asc

orbi

c ac

id

Gal

lic a

cid

Caf

feic

aci

d C

oum

aric

aci

d R

esve

ratr

ol

Bra

nd

Sam

ple

m

g L-1

23

Pea

ch/w

hite

tea

6.89

± 0

.28

0.64

± 0

.03

2.60

± 0

.09

2.23

± 0

.08

0.77

± 0

.06

28.4

9 ±

1.16

0.

39 ±

0.0

4 1.

63 ±

0.0

4 2.

14 ±

0.1

5 1.

16 ±

0.0

5

24

Man

go/g

inkg

o be

loba

7.

48 ±

0.0

8 0.

82 ±

0.0

5 2.

79 ±

0.0

5 2.

41 ±

0.0

8 0.

91 ±

0.0

2 19

.33

±1.9

7 0.

49 ±

0.0

2 1.

40 ±

0.0

7 2.

32 ±

0.0

4 0.

75 ±

0.0

9

25

Mel

on/m

int

8.58

± 0

.38

1.16

± 0

.11

3.15

± 0

.03

2.7

5 ±

0.12

1.

17 ±

0.1

1 24

.89

± 1.

97

0.41

± 0

.05

1.4

9 ±

0.07

2.

17 ±

0.0

7 1.

00 ±

0.0

8

26

Nat

ural

-

-

-

-

-

-

-

-

-

-

G

27 L

emon

7.

47 ±

0.2

0 0.

82 ±

0.0

6 2.

79 ±

0.0

6 2.

40 ±

0.0

6 0.

91 ±

0.0

5 29

.77

± 0.

96

2.02

± 0

.21

1.73

± 0

.03

5.29

± 0

.41

1.22

± 0

.04

28

Lim

e 4.

11 ±

0.0

6 -

1.71

± 0

.02

1.37

± 0

.02

0.10

± 0

.01

5.42

± 1

.98

- 0.

94 ±

0.0

7 0.

76 ±

0.0

3 0.

12 ±

0.0

8

29

App

le

7.48

± 0

.28

0.82

± 0

.09

2.79

± 0

.10

2.41

± 0

.08

0.91

± 0

.07

45.0

5 ±

2.95

1.

20 ±

0.2

1 2.

28 ±

0.0

9 3.

70 ±

0.0

5 1.

91 ±

0.1

3

30

Pea

ch

9.14

± 0

.57

1.33

± 0

.13

3.33

± 0

.15

2.92

± 0

.13

1.31

± 0

.10

14.2

5 ±

2.65

0.

79 ±

0.1

6 1.

12±

0.08

2.

90 ±

0.3

1 0.

51 ±

0.0

2

31

Nat

ural

-

-

-

-

-

-

-

-

-

-

H

32 L

emon

6.

38 ±

0.3

9 0.

49 ±

0.0

7 2.

44 ±

0.0

7 2.

07 ±

0.1

2 0.

65 ±

0.0

9 18

.43

± 1.

27

0.09

± 0

.01

1.35

± 0

.04

1.54

± 0

.03

0.71

± 0

.06

33

Nat

ural

-

- -

- -

- -

- -

-

I 34

Lem

on

34.6

2 ±

1.97

0.

63 ±

0.0

3 1.

91 ±

0.0

7 2.

59 ±

0.5

6

1.44

± 0

.09

34.6

2 ±

1.97

0.

63 ±

0.0

3 1.

91 ±

0.0

7 2.

59 ±

0.5

6

1.44

± 0

.09

35

Gre

en A

pple

37

.40

± 1.

69

0.92

± 0

.14

2.00

± 0

.07

3.

15 ±

0.2

6 1.

57 ±

0.0

2 37

.40

± 1.

69

0.92

± 0

.14

2.0

0 ±

0.07

3.

15 ±

0.2

6 1.

57 ±

0.0

2

36

Str

awbe

rry

40.3

2 ±

2.78

0.

14 ±

0.0

3 2.

07 ±

0.0

3 -

1.

70 ±

0.0

4 40

.32

± 2.

78

0.14

± 0

.03

2.07

± 0

.03

-

1.

70 ±

0.0

4

37

Nat

ural

-

-

-

-

-

-

-

-

-

-

J 38

Lem

on

5.71

± 0

.09

0.29

± 0

.04

2.22

± 0

.09

1.86

± 0

.05

0.48

± 0

.04

36.3

5 ±

2.47

0.

69 ±

0.1

3 1.

86 ±

0.1

5 2.

90 ±

0.2

5 1.

52 ±

0.1

1

39

Nat

ural

-

- -

- -

- -

- -

- -N

ot d

etet

ed.


144

The highest TAC values was found in brand I (sample 34 to 36) followed by brand F

(sample 21 to sample to 25) brand G (sample 27 to 30) and sample C (sample 8 to 10).

Using the gallic acid as a standard antioxidant some flavoured waters not presented

antioxidant activity such as brand D (sample 12 to 14) brand E (sample 17 to 19) sample

3 (brand A), and sample 28 (brand G). The lowest TAC value was from the brand B and

the highest was from brand F. With the caffeic acid standard antioxidant all flavoured

waters presented antioxidant capacity. The highest TAC values were from brand F

following brand A and brand C. Using the coumaric acid and the resveratrol as standard

antioxidant some samples not presented antioxidant activity, such as, sample 36 (brand I)

with coumaric acid and sample 3 (brand A), sample 17, sample 18 and sample 19 (brand

E) with the resveratrol antioxidant. TAC values obtained with the four antioxidant, gallic

acid, caffeic acid, coumaric acid and resveratrol are narrower than the values obtained

with the standard ascorbic acid antioxidant. Theses differences obtained between the

ascorbic acid and the others antioxidants can be elucidated by the fact that ascorbic acid

presented a larger linear range (0.50 to 2.50 mg L-1).

When it was used the adenine-GCE, the highest TAC contents were found with the

ascorbic acid antioxidant. With this antioxidant, TAC values ranged between 3.02 to 45.05

mg L-1 of ascorbic acid. The highest TAC values were obtained in brand I, followed from

brand J, brand B, brand F and brand G. The lowest TAC value was obtained in brand D.

When it was use the gallic acid antioxidant, some flavoured waters not presented

antioxidant activity, such as brand D (sample 12 to 14), and sample 28 (brand G). TAC

values ranged from 0.09 to 2.02 mg L-1 of gallic acid. The highest TAC value was found on

brand G (samples 27 and 29) following brand E and brand A. The lowest TAC value was

found in brand H followed by brand B, brand C and brand F. With the caffeic acid

antioxidant all flavoured waters presented antioxidant activity and the TAC values ranged

from 0.75 to 2.28 mg L-1. When it was used the coumaric acid and the resveratrol as

standard antioxidant some flavoured waters not presented antioxidant activity, such as

sample 36 with the coumaric acid antioxidant and sample 13, sample 14 and sample 19

with the resveratrol antioxidant. TAC values ranged between 0.76 to 5.29 mg L-1 and 0.12

to 1.57 mg L-1 when it was used coumaric acid and resveratrol respectively. Like it was

happened with the guanine-biosensor, larger TAC values were obtained with the ascorbic

acid and the the other four antioxidants presented a narrow TAC range.

Analysing results from Table 2 it is possibly to confirm that the purine bases immobilized

on GCE can be used for the quantification of TAC in beverages, however using the

adenine-GCE and ascorbic acid as antioxidant standard it was obtained the highest TAC

values.

4.1. Biossensores de bases púricas – radical hidróxilo

145

4. Conclusion

A guanine-biosensor and adenine-biosensor for the TAC quantification of beverages was

used. The electroanalytical technique is based on the interaction of adenine or guanine

immobilized on the GCE surface with the hydroxyl radical. The hydroxyl radical had the

capacity to damage the purine base. Five antioxidants (ascorbic acid, gallic acid, caffeic

acid, coumaric acid and resveratrol) were tested as hydroxyl radicals scangers exihiting

efficiencies ranging from 47 to 79 %. The protective effect on the DNA bases performed

by the presence of these antioxidants allowed the evaluation of TAC in food samples.

Ascorbic acid presented the highest TAC values and seems to be the most sensitive

standard antioxidant. The purine-based biosensor developed is disposable, and requires a

very easy, rapid, reproducible preparation and also the advantage to combine with

portable equipment.

Acknowledgements


grant (SFRH/BD/ 29440/2006). The authors thank Frize for providing flavours samples.


146

References

Barroso, M. F., Delerue-Matos, C., & Oliveira, M. B. P. P. (2011a). Electrochemical DNA-

sensor for evaluation of total antioxidant capacity of flavours and flavoured waters

using superoxide radical damage. Biosensors & Bioelectronics, 2011, 26(9), 3748-

3754.

Barroso, M. F., Delerue-Matos, C., & Oliveira, M. B. P. P. (2011b) Evaluation of total

antioxidant capacity of flavoured waters using sulfate radical damage of purine-based

sensor. (submitted).

Barroso, M. F., de-los-Santos-Álvarez, N., Lobo-Castañón ,M. J., Miranda-Ordieres, A. J.,

Delerue-Matos, C., Oliveira, M. B. P. P., & Tuñón-Blanco, P. (2011a). DNA-based

biosensor for the electrocatalytic determination of antioxidant capacity in beverages

Biosensors & Bioelectronics, 26(5), 2396-2401.

Barroso, M. F., de-los-Santos-Álvarez, N., Lobo-Castañón ,M. J., Miranda-Ordieres, A.

J., Delerue-Matos, C., Oliveira, M. B. P. P., & Tuñón-Blanco, P. (2011b).

Electrocatalytic evaluation of DNA damage by superoxide Radical for antioxidant

capacity assessment. Journal of Electroanalytical Chemistry, In press

doi:10.1016/j.jelechem.2011.04.022

Barroso, M.F., Noronha, J.P., Delerue-Matos, C., & Oliveira, M. P. P. (2011). Flavored

Waters: Influence of Ingredients on Antioxidant Capacity and Terpenoid Profile by HS-

SPME/GC-MS. Journal of Agricultural and Food Chemistry, 2011, 59(9), 5062–5072.

Buettner, G.R. (1993). The pecking order of free-radicals and antioxidants - lipid-

peroxidation, alpha-tocopherol, and ascorbate. Archives of Biochemistry and

Biophysics, 300(2), 535-543.

Cadet, J., Douki, T., Gasparutto, D., & Ravanat, J-L. (2003). Oxidative damage to DNA:

formation, measurement and biochemical features. Mutation Research - Fundamental

and Molecular Mechanisms of Mutagenesis Mutation Research, 531(1-2), 5-23.

Dizdaroglu, M., Jaruga, P. Birincioglu, M., & Rodriguez, H. (2002). Free radical-induced

damage to DNA: mechanisms and measurement. Free Radical Biology and Medicine,

32(11), 1102-1115.

Frankel, E. N. (2007). Antioxidants in food and biology facts and fiction. Publisher Oily

Press. Bridgwater, England, vol 20.

Halliwell, B., Gutteridge, J. M., & Cross, C. E. (1992). Free radicals, antioxidants, and

human disease: Where are we now. Journal of Laboratory and Clinical Medicine,

119(6), 598-620.

Ignat, I., Volf, I., & Popa, V. I. (2011). A critical review of methods for characterisation of

polyphenolic compounds in fruits and vegetables, Food Chemistry, 126, 1821-1835.

4.1. Biossensores de bases púricas – radical hidróxilo

147

Jaitz, L., Siegl, K., Eder, R., Rak, G., Abranko, L., Koellensperger, G., & Hann, S. (2010).

LC–MS/MS analysis of phenols for classification of red wine according to geographic

origin, grape variety and vintage. Food Chemistry, 122, 366–372.

Jaruga, P., Dizdaroglu, M. (1996). Repair of products of oxidative DNA base damage in

human cells. Nucleic Acids Research, 24(8), 1389-1394.

Kamel, A. H., Moreira, F. T. C., Delerue-Matos, C., & Sales, M. G. F. (2008).

Electrochemical determination of antioxidant capacities in flavored waters by guanine

and adenine biosensors. Biosensors and Bioelectronics, 24(4), 591-599.

Korbut, O., Buckova, M., Labuda, J., & Grundler, P. (2003). Voltammetric detection of

antioxidative properties of flavonoids using electrically heated DNA modified carbon

paste electrode. Sensors, 3(1)1-10.

Lee, W., Davis, K. A., Rettmer, R. L., & Lable, R. F. (1988). Ascorbic-acid status -

biochemical and clinical considerations. American Journal of Clinical Nutrition 48(2)

289-290.

Liu, J., Roussel, C., Lagger, G., Tacchini, P., & Girault, H. H. (2005). Antioxidant sensors

based on DNA-modified electrodes. Analytical Chemistry, 77(23), 7687-7694.

Mello, L. D., Hernandez, S., Marrazza, G., Mascini, M., & Kubota, L.T. (2006).

Investigations of the antioxidant properties of plant extracts using a DNA-

electrochemical biosensor. Biosensors & Bioelectronics, 21(7), 1374-1382.

Mello, L. D., & Kubota, L. T. (2007). Biosensors as a tool for the antioxidant status

evaluation. Talanta, 72(2), 335-348.

Nobushi, Y., Uchikura, K. (2010). Selective Detection of Hydroxyl Radical Scavenging

Capacity Based on Electrogenerated Chemiluminescence Detection Using Tris(2,2′-

bipyridine)ruthenium(III) by Flow Injection Analysis. Chemical & Pharmaceutical

Bulletin58 (1), 117–120.

Piljac-Žegarac, J., Valek, L., Stipčević, T., Martinez, S. (2010). Electrochemical

determination of antioxidant capacity of fruit tea infusions. Food Chemistry, 121, 820–

825.

Rice, M. E., Galus, Z., & Adams, R. N. (1983). Graphite paste electrodes. Effects of paste

composition and surface states on electron-tranfer rates. Journal of Electroanalytical

Chemistry, 143(1-2), 89-102.

Rice-Evans, C. (2001). Flavonoid antioxidants. Current Medicinal Chemistry, 8(7), 797-

807.

Sanchez-Moreno, C. (2002). Review: Methods used to evaluate the free radical

scavenging activity in foods and biological systems. Food Science and Technology

International, 8(3), 121-137.


148

Sassolas, A., Leca-Bouvier, B. D., & Blum, L. J. (2008). DNA biosensors and microarrays.

Chemical Reviews, 108(1), 109-139.

Smirnoff, N. (2000). Ascorbic acid: metabolism and functions of a multi-facetted molecule.

Current Opinion in Plant Biology, 3(3), 229-235.

Soobrattee, M. A., Neergheen, U. S., Ramma, A. L., Aruoma, O. I., & Bahorun, T. (2005).

Phenolics as potential antioxidant therapeutic agents: Mechanism and actions.

Mutation Research-Fundamental and Molecular Mechanisms of Mutagenesis, 579(1-

2), 200-213.

Thavasi, V., Leong, L. P., & Bettens, R. P. A. (2006). Investigation of the influence of

hydroxy groups on the radical scavenging ability of polyphenols. Journal of Physical

Chemistry A, 110(14), 4918-4923.

Vertuani, S., Angusti, A., & Manfredini, S. (2004). The antioxidant and pro-antioxidants: an

overview. Current Pharmaceutical Design, 10(14), 1677-1694

Zhang, J. J., Wang, B., Li, Y. F., Jia, W. L., Cui, H., & Wang, H. S. (2008). Electrochemical

Study on DNA Damage Based on the Direct Oxidation of 8-Hydroxydeoxyguanosine at

an Electrochemically Modified Glassy Carbon Electrode. Electroanalysis 20(15), 1684–

1689.

4.2. Biossensores de bases púricas - radical superóxido

149

Electrochemical DNA-sensor for evaluation of total antioxidant

capacity of flavours and flavoured waters using sup eroxide

radical damage

M. F. Barroso1,2, C. Delerue-Matos1, M. B. P. P. Oliveira2 1REQUIMTE, Instituto Superior de Engenharia do Porto. R. Dr. Bernardino de Almeida

431, 4200-072 Porto, Portugal 2Requimte, Serviço de Bromatologia, Faculdade de Farmácia, Universidade do Porto. R.

Aníbal Cunha n.º 164, 4050-047 Porto, Portugal

Abstract

In this paper, a biosensor based on a glassy carbon electrode (GCE) was used for the

evaluation of the total antioxidant capacity (TAC) of flavours and flavoured waters. This

biosensor was constructed by immobilising purine bases, guanine and adenine, on a

GCE. Square wave voltammetry (SWV) was selected forthe development of this

methodology. Damage caused by the reactive oxygen species (ROS), superoxide radical

(O2• −), generated by the xanthine/xanthine oxidase (XOD) system on the DNA-biosensor

was evaluated. DNA-biosensor encountered with oxidative lesion when it was in contact

with the O2• −. There was less oxidative damage when reactive antioxidants were added.

The antioxidants used in this work were ascorbic acid, gallic acid, caffeic acid, coumaric

acid and resveratrol. These antioxidants are capable of scavenging the superoxide radical

and therefore protect the purine bases immobilized on the GCE surface. The results

demonstrated that the DNA-based biosensor is suitable for the rapid assess of TAC in

beverages.

Keywords: DNA biosensor; Total antioxidant capacity (TAC); Ascorbic acid; Phenolic

acid; Reactive oxygen species (ROS); Superoxide radical (O2• -).


Biosensors & Bioelectronics 2011, 26 (9), 3748-3754


151

1. Introduction

Recently, bottled flavoured waters are becoming popular, and the consumption of

flavoured waters is globally increasing including Portugal. In the first half of 2010, 6.08

million L of this kind of water were consumed by the Portuguese population (ANIRSF,

2010). Flavoured waters produced from mineral and spring waters consist of the addition

of flavours, juices and sugar or sweeteners that provide water with a particular taste and

aroma appreciated by consumers. Considering that flavours/aromas are fruit extracts, and

fruits are good sources of exogenous antioxidants, it is expected that the use of this fruit

extracts in beverages can introduce antioxidants to the water (Barroso et al., 2009, 2011).

Antioxidant defence mechanisms include the use of enzymes, vitamins, phenolic

compounds, minerals or proteins. Consequently, increasing intake of dietary antioxidants

may help maintain an adequate antioxidant status and, therefore, sustain normal

physiological functions of a living system. Antioxidants are very important in the

mammalian body because they have the ability to combat and reduce oxidative damage

caused by reactive oxygen species (ROS) (Halliwell et al., 1992). ROS are continuously

produced in all living beings as a result of normal cellular metabolism (Benherlal and

Arumughan, 2008).

The superoxide anion radical (O2• −) is the most abundant radical in biological systems

resulting from the univalent reduction of oxygen (Ge and Lisdat, 2002). This radical

species is enzymatically produced by xanthine oxidase (XOD). XOD is a mettalloenzyme

that catalyses the oxidation of hypoxanthine and xanthine to form O2• − that is generated

during the respiratory burst of phagocytic cells such as neutrophils (Gobi and Mizutani,

2000; Laranjinha, 2009).

Several analytical methods have been proposed for the quantification of the total

antioxidant capacity (TAC) in biological and food samples. These methodologies are

based on UV-vis spectrometry, chemiluminescence, fluorimetry, electrochemistry and

chromatography techniques (Sanchez-Moreno, 2002).

Recently, several electrochemical methods based on enzymatic biosensors have been

developed for the determination of superoxide radical and TAC. These biosensors are

based on the immobilization of Cytochrome c (this enzyme acts as an oxidant of

superoxide radical) or on the immobilization of the enzyme superoxide dismutase (SOD;

this enzyme has a protective scavenging function against the superoxide radical), on the

electrode surface (gold, platinium, glass, carbon paste or screen printed electrode (SPE),

SPE-Au) (Ge and Lisdat, 2002; Emregül, 2005). In this type of protein immobilised

biosensor, an electrochemical signal was found to be proportional to the superoxide


152

concentration generated in aqueous solution by the xanthine and xanthine oxidase (Eq.

(1)).

(1) O 2H aciduric O OH xanthine -2

XOD 22

•+ ++ →++

For the immobilization of enzymes on an electrode surface, some strategies have been

demonstrated. The immobilization of the enzyme can be carried out via short-chain thiol

modified gold electrodes, long-chain thiol (mercaptoundecanoic acid), mixed-thiol, long-

chain mixed thiol (mercaptoundecanoic acid/mercaptoundecanol) and hemin modified

electrode (McNeil et al., 1995; Gobi and Mizutani, 2000; Ignatov et al., 2002). However,

the performance of many of these types of devices is interfered by hydrogen peroxide,

uric acid and some communication interference between the protein and the electrode

(Chen et al 2000; Beissenhirtz et al., 2004; Endo et al, 2002; Campanella et al 2004;

Emregül, 2005). The protective effect of antioxidants at a cellular level could only be

achieved by monitoring the DNA integrity (Barroso et al., 2011). For this purpose,

electrochemical DNA-based biosensors have been developed in order to assess the TAC

of foodstuff (Mello et al., 2006; Barroso et al., 2011). In many studies (Fojta et al., 2000;

Mello et al., 2006), the oxidative damage of double stranded DNA or of the nucleobases

(guanine or adenine) by the hydroxyl radical was evaluated. The oxidative damage

produces a significant decrease in the current intensity on the strand scission of DNA or

on the decreasing oxidation current after damage of the nucleobases (Liu et al., 2005;

2006; Mello et al., 2006; Qian et al., 2010). In this work, a DNA-sensor was used in order

to evaluate TAC in bottled flavoured waters. This DNA-biosensor consisted of

electrochemically deposited purine base (adenine or guanine) on a glassy carbon

electrode (GCE). All DNA bases (purine and pyrimidine) can be used for the

electrochemistry study. However, purine bases (adenine and guanine) are more sensitive

for detection and present lower potential peaks than the pyrimidine bases (+1.3 V for

thymine and +1.5 V for cytosine). Considering that purine bases have peaks more well-

defined and larger than those of the pyrimidines (Brett and Matysik, 1997), the purine

bases were used in this study. In experiments evaluating the oxidative damage of the

purine bases, the biosensor was firstly immersed in an aqueous superoxide radical

solution that was generated in the enzymatic reaction between XOD and xanthine (Eq.

(1)). Then, the decrease of the oxidation current of guanine and adenine recorded in

square wave voltammetry (SWV) was used to relate the extent of oxidative damage. The

influence/protection of five antioxidants, such as, ascorbic acid, gallic acid, caffeic acid,

reverastrol and p-coumaric acid was studied.


153


2.1. Chemicals

Guanine, adenine, xanthine oxidase (XOD, X1875) xanthine, gallic acid, resveratrol

were purchased from Sigma. Caffeic acid was from Fluka, L(+) ascorbic acid and

reveratrol was acquired from Riedeil-de-Haën. Other chemicals were Merck pro-analysis

grade and were used as received. Guanine stock solution (1 g L-1) was prepared by

dissolving an amount of this solid in 0.1 mol L-1 of NaOH and diluting in pH 7.4 phosphate

buffered saline (PBS). Stock solution of 1g L-1 of adenine was prepared in PBS pH 7.4

and stored at +4 ºC. For all voltammetric measurements, pH 4.8 PBS was used as the

supporting electrolyte. Superoxide radical was generated by adding XOD (0.0015 U mL-1)

to oxygen-satured PBS (pH 7.4) containing xanthine (10 µmol L-1). All solutions were

prepared with water purified with a Direct-Q (Millipore) system.

2.2. Instrumentation

SWV was performed with an Autolab PSTAT 10 potentiostat controlled by GPES

software (EcoChemie, The Netherlands). A conventional three electrode cell was used,

which includes a GCE (0.07 cm2) as working electrode, a glassy carbon counter electrode

and a Ag|AgCl|KClsat reference electrode to which all potentials were referred. The GCE

was mechanically polished using a polishing kit (Metrohm 6.2802.010) first with γ-Al2O3

(0.015 µm) until a shining surface was obtained and then rinsed with water. After this step

the GCE was treated by applying a fixed potential of +1.7 V for 30 s in PBS pH 4.8. This

initial conditioning step improves the resolution of the analytical signal because the

application of high potentials in acidic medium increases the hydrophilic properties of the

electrode surface through the introduction of oxygenated functionalities (Rice et al., 1983;

Mello et al., 2006).

2.3. Assay procedure

Unless otherwise mentioned, all experiments consisted of three steps: (i) Guanine or

adenine electro-immobilization on the GCE, (ii) damage of purine bases by the immersion

of DNA-GCE in the XOD/xanthine solution, and study of the effect of the presence of

antioxidants in the system, and (iii) detection and measurement of the peak current of

adenine or guanine in a PBS at pH 7.4.


154

Purine bases (adenine or guanine) immobilization was performed by the application of

an adsorptive accumulation step. For that, the activated GCE was immersed in PBS pH

4.8 containing 10 mg L-1 of adenine or 3 mg L-1 of guanine and a potential of +0.4 V was

applied for 180 s. The electrode was next rinsed with water. A reported procedure

(Marrazza et al., 1999; Chiti et al., 2001; Mello et al., 2006) for cleaning and

immobilization step was adopted in this work. DNA damage was carried out by immersing

the biosensor in a freshly prepared XOD/xanthine mixture in the absence or in the

presence of antioxidant in PBS pH 7.4 for a fixed period of reaction time. Next, the

biosensor was immersed in pH 4.8 PBS. SWV was then conducted between +0.2 V and

+1.4 V and the oxidation peak current of guanine and adenine obtained was used as a

detection signal. For the electrochemical studies it was considered that the maximum

signal current obtained were for the purine base electrochemical signal without damage

neither antioxidant effect.

2.4. Samples


several supermarkets in the North of Portugal and stored in the dark at +4 ºC. Each brand


water of each brand was also used as control. Sonication was used to eliminate gas from

the sparkling water samples. The labels on the water bottles indicate the nutrient

information, namely the presence of fruit juice, vitamins, sweeteners and preservatives

(Barroso et al., 2009).

Six liquid flavours used in the formulation of some water brands, provided by a

producer, were also analysed. The flavours used corresponded to different fruit aromas,

including lime, tangerine, strawberry, lemon, apple and gooseberry. These flavours had

no description about their chemical or aroma composition, but were known to be present

in the flavoured waters used in this study.

2.5. TAC measurement on beverages


flavoured waters. For the measurement of TAC in beverages, 100 µL of the flavoured

water or 5 µL of flavour were diluted in PBS to a final volume of 500 µL. Then, the DNA-

GCE was immersed in the solution and a freshly prepared superoxide radical was added.

After 120 s, the biosensor was rinsed and immersed in PBS buffer before SWV of guanine

and adenine was carried out.


155


The ease of oxidation of purine bases in DNA depends, predominantly, on the

secondary structure of the polynucleoside. Owing to the flexibility and better accessibility,

nucleobases in a ssDNA are readily oxidised than in a dsDNA, leading to a higher

oxidation current at an electrode surface (de-los-Santos-Álvarez et al., 2002). SWV was

used to observe the electrochemical response of the oxidation of guanine and adenine

immobilised on a GCE. Fig. 1 (curve a in (i) and (ii)) shows the anodic peak of guanine

and adenine bases. The less positive peak potential (+0.55 V) corresponds to the

oxidation of guanine, while the peak at more positive potential (+0.82 V) corresponds to

the electrooxidation of adenine. This results are in agreement with +0.55 V for guanine

and +0.82 V for adenine reported in the literature (Brett et al., 1994; Brett and Matysik,

1997), which focussed on the dependence of the oxidation peak of purine bases on pH,

buffer and ionic strength.

Fig. 1. SVW obtained in PBS pH 4.8: (i) guanine-biosensor and (ii) adenine-biosensor: after: (a) total oxidation of guanine and adenine signal (maximum peak current), (b) immersion of the biosensor in a superoxide radical solution and (c) immersion in superoxide radical solution with ascorbic acid.

Damage of DNA is the major endogenous type of pathogenesis that induces a variety of

diseases including cancer. ROS induced oxidative lesion in the DNA will cause

modifications at the DNA. Superoxide radical generated in situ by XOD can mediate the

direct strand scission of DNA and this can be attributed to hydrogen atom abstraction of

C5’ of the deoxyribose (Burrows and Muller, 1998). In order to verify if O2•- radicals

generated by xanthine/XOD reaction are able to damage purine base immobilized on the

GCE, the DNA-GCE was placed in a freshly prepared solution of xanthine/XOD in PBS pH

7.4 for 5 min. Next, the biosensor was rinsed with water and SWV at this biosensor was

repeated. A 61.4% and a 64.5% decrease in the anodic peak current (ip) of guanine and

adenine, respectively was observed after the biosensor was immersed on the superoxide

40.0

50.0

60.0

70.0

0.50 0.70 0.90 1.10 1.30

E (V) vs. AgCl/Ag

i / µ

A

a

b

c

40.0

50.0

60.0

70.0

0.50 0.70 0.90 1.10 1.30

E (V) vs. AgCl/Ag

i / µ

A

40.0

50.0

60.0

70.0

0.50 0.70 0.90 1.10 1.30

E (V) vs. AgCl/Ag

i / µ

A

a

b

c

10.0

20.0

30.0

40.0

50.0

0.10 0.30 0.50 0.70 0.90

E (V) vs. AgCl/Ag

i / µ

A

a

b

c

10.0

20.0

30.0

40.0

50.0

0.10 0.30 0.50 0.70 0.90

E (V) vs. AgCl/Ag

i / µ

A

10.0

20.0

30.0

40.0

50.0

0.10 0.30 0.50 0.70 0.90

10.0

20.0

30.0

40.0

50.0

0.10 0.30 0.50 0.70 0.90

E (V) vs. AgCl/Ag

i / µ

A

a

b

c

i) ii)


156

radical solution (curve b in Fig. 1i and ii). This decrease in the peak current was used to

infer damage of the DNA bases after being oxidised by the O2• −radicals. According to the

literature (de-los-Santos-Álvarez et al., 2007; Freidman and Heller, 2004), guanine base is

the most easily oxidized of the nucleic acid base, yielding 8-oxoguanine (8-oxoG) and the

tautomer 8-hydroxyguanine. However, a common diimine structure was produced when

guanine and adenine were electrochemically oxidised at neutral or alkaline solution. As

shown by curve c in Fig. 1i and Fig. 1ii, when an antioxidant, in this case ascorbic acid,

was added to the superoxide radical solution a 43.86% and a 50.11% increase of ip of

guanine and adenine, respectively, compared to curve b of Fig. 1i and Fig. 1ii. Indeed, this

is indicative that the DNA was protected by the antioxidant presents in the solution.

Antioxidants are well-known to exhibit a protective effect with a scavenging effect of ROS

preventing DNA damage. Consequently, the number of lesions diminishes, yielding a

larger number of adenine and guanine for electrochemical oxidation (Barroso et al., 2011).

Indeed, ascorbic acid is considered a good scavenger of free radicals produced during the

metabolic pathways of detoxification. Ignatov et al. (2002) reported the development of a

methodology for the electrochemical detection of antioxidants based on a superoxide

radical measurement with a cytochrome c modified electrode. In this study the authors

have used several antioxidants such as ascorbic acid (standard antioxidant) and sub-

groups of the phenolic acid (flavanols, flavanones, isoflavones, flavones and flavonols).

The antioxidants used by these authors presented scavenger capacity of the superoxide

radical. Considering the good correlation between antioxidant concentration and the

protective effect on the DNA, an analytical procedure to evaluate TAC was developed.


To measure the TAC of beverages, some parameters concerning the damage on the

purine base immobilized on the GCE (xanthine and XOD concentration, reaction time

between superoxide radical and the target molecule) were implemented in order to

achieve the maximum DNA effect, but without a complete damage (non-zero ip). XOD

concentration was studied between 0.0015 and 0.1 U mL-1. A range of 25%-66%

decrease in the ip of guanine and adenine was observed over the XOD concentration

studied. This is indicative of the effectiveness of XOD on the generation of the superoxide

radical. At an adenine-biosensor, a 62% decrease in ip was observed when the XOD

concentration was increased from 0.0015 U mL-1 to 0.07 U mL-1. At higher XOD

concentration, ip was observed to remains essentially unchanged. Considering that the

lowest XOD concentration was 0.0015 U mL-1, this XOD concentration was used for the

next optimisation steps for the adenine biosensor. Similar results were obtained with the


157

guanine-biosensor. The increase of XOD concentration on the reactive system generates

high damage on the DNA as indicated by a decrease of 22 %-60 % in ip. At XOD

concentration higher than 0.008 U mL-1, ip was observed to remains similar, so this value

was used for the next experiments.

For both purine-based biosensors (guanine and adenine) xanthine concentration was

ranged between 10 and 800 µmol L-1. With the increase of xanthine concentration a

decrease between 57% and 66% in the ip of guanine and adenine was observed,

however, the decrease of the ip in all range of xanthine concentration studied was very

similar and remained essentially unchanged. Therefore, to be more cost effective, the

lowest xanthine concentration of 10 µmol L-1 was used in the next optimisation step.

Reaction time between the superoxide radical and the purine bases immobilized on the

GCE depends of the half-life time of the generated ROS, so this parameter is an important

feature to optimize. In this study the incubation time were ranged from 0 to 120 s. Fig. 2

shows the correlation between the damage on the purine base produced by the

superoxide radical (correlated with the ip values) and the incubation time. A more than

50% decrease in the ip was observed with an increase of the reaction time from 0 to 120 s.

However, there was no complete damage of DNA as indicated by the non-zero ip results

shown in the Fig. 2. The lower ip obtained at the adenine biosensor than the guanine

biosensor indicates more damage at the former. However, de-los-Santos Álvarez et al.

(2007) reported more damage of guanine than adenine at a pyrolic graphite electrode in a

neutral and alkaline aqueous solutions. The incubation time of 120 s was chosen for all

experiments.

Fig. 2. Influence on the peak current on the biosensor with the incubation time (a) 10 mg L−1 adenine base, (b)

3 mg L−1 guanine base.

a) adenine

b) guanine

0.0

5.0

10.0

15.0

0 20 40 60 80 100 120

t (sec)

i p/µ

A

a) adenine

b) guanine

0.0

5.0

10.0

15.0

0 20 40 60 80 100 120

t (sec)

i p/µ

A


158

3.2. Determination of TAC

Foodstuff constitutes an excellent source of exogenous antioxidants to counteract the

alteration of lipids in cellular membranes, protein, enzymes, carbohydrates and DNA

promoted by ROS. Antioxidants, such as, ascorbic acid, and phenol-derived compounds

are natural components of fruits and beverages (tea and wine). For the evaluation of the

TAC of flavoured waters, five antioxidants including ascorbic acid, gallic acid, caffeic acid,

coumaric acid and resveratrol were used. Ascorbic acid is a water-soluble vitamin, is

considered a powerfull antioxidant and plays a key role in the protection against biological

oxidation processes participating in many metabolic reactions (Mello and Kubotta, 2007).

Gallic, caffeic and coumaric acid are phenolic acids with a large protective action.

Phenolic acids include several groups such as the hydroxybenzoic acid (gallic acid) and

the hydroxycinnamic acid (caffeic and coumaric acid). In general, the antioxidant activity

of the phenolic-derived compounds is determined by some properties, such as, free-

radical scanvengers (Thavasi, et al., 2006). Resveratrol is a polyphenolic natural product,

derived stilbene that exists in various foods and beverages, has attracted increasing

attention over the past decade because of its multiple beneficial properties, including

chemopreventive and antitumor activities (Fulda, 2010). Linearity studies between the five

antioxidants and ip of guanine and adenine oxidation were carried out. Fig. 3i and ii shows

the SWV of electrochemical current obtained after immersing the purine-biosensor on the

superoxide radical containing increasing concentration of ascorbic acid. As expected, the

oxidation current of guanine and adenine increased when the concentration of ascorbic

acid increased. Similar voltammograms were obtained when the other antioxidants

(resveratrol, gallic, caffeic and coumaric acid) were used with the both DNA-biosensors

(guanine and adenine).

Table 1 presents a summary of analytical parameters of the guanine and adenine

biosensors obtained after being immersed in the respective five antioxidants used. Among

them, ascorbic acid showed the widest linear range from 1.00 to 5.00 mg L-1 at guanine-

GCE and 0.50-4.00 mg L-1 at adenine-GCE. The other antioxidants presented a narrow

linear range, 0.10-1.00 mg L-1 of gallic acid or caffeic acid, 0.10-0.50 mg L-1 of resveratrol

and 0.50-1.00 mg L-1 of coumaric acid when the guanine biosensor was used. For the

adenine biosensor, the linear range was from 0.10 to 0.50 mg L-1 for the antioxidants

caffeic acid, coumaric acid and resveratrol, and from 0.50 to 0.90 mg L-1 for gallic acid.

RSD values were below 10% confirmed the high precision of the methods.


159

Fig. 3. SWV obtained after immersion of (i) guanine-biosensor in superoxide radical containing a standard

solution of ascorbic acid: (a) 1.00, (b) 2.00, (c) 3.00, (d) 4.00 and (e) 5.00 mg L−1 and (ii) adenine-biosensor in

superoxide radical containing a standard solution of ascorbic acid: (a) 0.50, (b) 1.00, (c) 2.00, (d) 3.00 and (e)

4.00 mg L−1. Inset: relationship between ip and ascorbic acid concentration.


Guanine-GCE

Linear range (mg L-1) 1.00–5.00 0.10–1.00 0.1–1.00 0.50–1.00 0.10–0.50

Slope (A mg-1L) 1.05x10-6 5.38x10-6 5.23x10-6 7.33x10-6 1.27x10-5 Intercept (A) 4.11x10-6 4.66x10-6 4.25x10-6 2.09x10-6 1.92x10-6


RSD (%) (mg L-1 ) 3.43 (2.00) 2.36 (0.30) 2.96 (0.50) 1.05 (0.70) 3.86 (0.20)

LOD 0.77 0.10 0.10 0.08 0.06

Adenine-GCE Linear range (mg L-1) 0.50–4.00 0.50–0.90 0.10–0.50 0.10–0.50 0.10–0.50

Slope (A mg-1L) 5.02x10-7 9.40x10-6 1.30x10-5 6.49x10-6 1.11x10-5

Intercept (A) 4.26x10-6 8.00x10-8 1.74x10-6 2.99x10-6 3.02x10-6


RSD (%) 1.00 (2.00) 2.11 (0.70) 4.00 (0.30) 4.93 (0.20) 6.43 (0.20) LOD 0.50 0.06 0.05 0.02 0.10

Table 2 shows the TAC values expressed in mg L-1 of ascorbic acid, gallic acid, caffeic

acid, coumaric acid and resveratrol. All flavours and flavoured waters were observed to

show antioxidant capacity; except the natural waters. Flavours that showed the highest

TAC values are fruit extracts that contain several concentrated antioxidant compounds.

Using the adenine and guanine GCE the highest TAC values were found with the ascorbic

acid standard. At the adenine biosensor apple, fallowed by lemon, gooseberry strawberry,

tangerine and lime were the flavours that showed the highest TAC values. At the guanine-

10.0

20.0

30.0

40.0

50.0

0.30 0.40 0.50 0.60 0.70 0.80

E (V) AgCl/Ag

i/ µA

10.0

20.0

30.0

40.0

50.0

0.30 0.40 0.50 0.60 0.70 0.80

E (V) AgCl/Ag

i/ µA

i)

[AA] mg/L

I p/(µ

A

50.0

60.0

70.0

0.60 0.70 0.80 0.90 1.00 1.10 1.20

E (V) AgCl/Ag

i/ µ

A

i p/µ

A

a

e

ii)

[AA] mg/L

a

e

ii)

[AA] mg/L

a

e

4.0

6.0

8.0

10.0

1.0 2.0 3.0 4.0 5.04.0

6.0

8.0

10.0

1.0 2.0 3.0 4.0 5.0

4.0

5.0

6.0

7.0

0.0 1.0 2.0 3.0 4.04.0

5.0

6.0

7.0

0.0 1.0 2.0 3.0 4.0


160

biosensor tangerine showed the highest TAC value, fallowed by strawberry, apple, lime,

lemon and gooseberry.

When the adenine-biosensor was applied to the analysis of flavoured waters, brand G

showed the highest TAC values (sample 27 and 30), maybe because this brand had in its

composition vitamin C (sample 28 has no vitamin and the TAC value was lower than the

other samples from the same brand). Brand A also presented higher TAC values and the

other commercial brands presented TAC values ranging between 0.33 mg L-1 and 7.31

mg L-1 with the standard ascorbic acid. Using the antioxidant ascorbic acid the lowest TAC

value was obtained from the Brand D (sample 12-14) and sample 35. Analysing TAC

results obtained using the water brands (brand A, B, C, D, E, F and I) it was verified that

the TAC values obtained within the same brand were similar, hence, the Adenine-GCE

might not discriminate the different flavours present in same brand. Using the gallic acid

standard the TAC values ranged from 37 to 57 mg L-1 for the flavours and 0.34-3.37 mg L-1

for the flavoured waters. The lowest TAC values were obtained in brand D (samples 12-

14) and the highest TAC contents were from brand A (samples 1-3). With the caffeic acid

antioxidant the TAC ranged from 13 to 27 mg L-1 and 0.72-1.74 mg L-1 in flavours and

flavoured waters respectively. Similar results were obtained with the other standard

antioxidants, coumaric acid and resveratrol. TAC values obtained with the ascorbic acid

were larger than the other four antioxidants (gallic acid, caffeic acid, coumaric acid and

resveratrol) that presented a narrow TAC levels. Theses differences obtained between the

ascorbic acid and the other antioxidants can be elucidated because the ascorbic acid is a

powerful antioxidant and in this study presented a larger linear range.

A similar behaviour was observed with the guanine-GCE and using the standard

ascorbic acid, brand G (samples 27) presented also the highest TAC values fallowed by

brand F, brand A and brand H. TAC values ranged between 0.68 and 18.7 mg L-1

equivalents of ascorbic acid. It was verified that TAC results obtained within the same

brand were similar (analogous to that at the adenine biosensor) with the exception of

brand C. Considering that sample 9 (from brand C) had two added ingredient; apple and

white tea a higher TAC value was expected compared with the other samples of brand C.

For other antioxidants, the TAC values ranged from 0.34 mg L-1 to 3.15 mg L-1 and 0.41

mg L-1-3.20 mg L-1 or between 0.01 mg L-1 and 4.71 mg L-1 and from 0.33 mg L-1 to 2.19

mg L-1 for the gallic acid, caffeic acid, coumaric acid and resveratrol, respectively. Larger

TAC values were obtained with the ascorbic acid antioxidant and the other four

antioxidants presented a narrow TAC range, a similar behaviour was obtained with the

adenine-GCE.

16

1

Tab

le 2

. T

AC

val

ues

obta

ined

for

the

flavo

urs

and

flavo

ured

wat

ers

usin

g a

guan

ine-

GC

E a

nd a

deni

ne-G

CE

(m

g L-1

) A

deni

ne-G

CE

G

uani

ne-G

CE

Bra

nd

Sam

ple

Asc

orbi

c ac

id

Gal

lic a

cid

Caf

feic

aci

d C

oum

aric

aci

d R

esve

ratr

ol

Asc

orbi

c ac

id

Gal

lic a

cid

Caf

feic

aci

d C

oum

aric

aci

d R

esve

ratr

ol

Fla

vour

Le

mon

16

9.52

± 1

1.20

55

.22

± 3.

90

25.9

3 ±

2.82

32

.68

± 5.

64

25.

53 ±

2.9

3

93.1

4 ±

19.9

3 20

.96

± 3.

89

16.0

2 ±

4.00

47

.36

± 3.

31

17.0

1 ±

1.89

T

ange

rine

131.

6 ±

1.16

39

.14

± 0.

06

14.3

0 ±

0.04

9.

39 ±

0.0

9

13.4

3 ±

0.05

22

0 ±

26.4

7 32

.71

± 9.

07

41.4

9 ±

9.33

68

.40

± 7.

71

29.0

1 ±

4.40

A

pple

20

2.69

± 5

5.08

56

.99

± 2.

94

27.2

1 ±

2.13

35

.25

± 4.

26

26.8

6 ±

2.21

17

7.05

± 3

.10

24.3

3 ±

0.60

32

.87

± 0.

62

61.2

8 ±

0.51

24

.95

± 0.

29

S

traw

berr

y 16

3.75

± 4

.23

37.4

3 ±

3.43

13

.06

± 2.

48

6.90

± 0

.97

12.1

4 ±

2.58

18

6 ±

29.0

3 9.

86 ±

1.6

6 2.

30 ±

5.8

3

32.2

2 ±

4.81

8.

37 ±

2.7

5

G

oose

berr

y 16

9.42

± 5

9.30

55

.22

± 3.

17

25.9

3 ±

2.29

32

.67

± 4.

59

25.5

2 ±

2.38

74

.81

± 7.

18

4.38

± 0

.89

12

.34

± 1.

29

44.3

2 ±

1.81

15

.27

± 6.

73

Li

me

126.

00 ±

6.7

7 39

.44

± 4.

10

14.5

2 ±

2.96

9.

82 ±

0.3

2 13

.66

±3.0

8 13

3.52

± 3

4.21

15

.84

± 2.

68

24.1

3 ±

2.8

7 54

.06

± 5.

67

20.8

3 ±

3.24

A

1 Le

mon

13

.91

± 2.

78

3.23

± 0

.11

1.64

± 0

.08

2.32

± 0

.16

1.63

± 0

.08

15.0

5 ±

2.50

2.

43 ±

0.0

9 2.

13 ±

0.5

7 4.

09 ±

0.4

1 1.

83 ±

0.2

4

2 M

ango

13

.31

± 4.

43

3.16

± 0

.04

1.59

± 0

.03

2.

21 ±

0.5

1.

58 ±

0.0

3 9.

98 ±

5.3

5 1.

44 ±

0.0

4 0.

88 ±

0.3

2 3.

25 ±

0.8

9 1.

35 ±

0.5

1

3 S

traw

berr

y 16

.08

± 2.

06

3.33

± 0

.12

1.71

± 0

.09

2.4

6 ±

0.18

1.

71 ±

0.0

9 6.

58 ±

1.7

8 0.

77 ±

0.0

5 0.

62 ±

0.0

4 2.

69 ±

0.3

0 1.

03 ±

0.1

7

4 N

atur

al

- -

- -

- -

- -

0.01

± 0

.04

-

B

5 P

inea

pple

/ora

nge

1.03

± 0

.03

2.53

± 0

.20

1.13

± 0

.14

1.30

± 0

.28

1.10

± 0

.15

8.00

± 0

.48

1.05

± 0

.09

0.

94 ±

0.0

6 2.

92 ±

0.0

8 1.

17 ±

0.0

5

6

Lem

on

1.29

± 0

.06

2.39

± 0

.08

1.03

± 0

.06

1.09

± 0

.11

1.00

± 0

.06

4.39

± 0

.13

0.34

± 0

.03

0.45

± 0

.05

2.32

± 0

.02

0.82

± 0

.01

7

Nat

ural

-

- -

- -

- -

- 0.

05 ±

0.0

4 -

C

8 Le

mon

/Mag

nesi

um

4.22

± 0

.26

2.38

± 0

.06

1.02

± 0

.05

1.07

± 0

.09

0.99

± 0

.05

5.80

± 2

.02

0.48

± 0

.03

3.20

± 0

.27

1.

62 ±

0.3

4 0.

42 ±

0.0

9

9

App

le/w

hite

tea

3.69

± 0

.99

2.52

± 0

.25

1.12

± 0

.18

1.28

± 0

.36

1.09

± 0

.19

14.6

9 ±

5.21

2.

36 ±

0.0

2

2.54

± 1

.4

4.03

± 0

.86

1.80

± 0

.05

10

Pin

eapp

le/fi

bre

1.70

± 0

.25

2.45

± 0

.40

1.07

± 0

.29

1.18

± 0

.57

1.04

± 0

.30

4.86

± 0

.53

0.44

± 0

.10

1.

54 ±

0.5

3 2.

40 ±

0.0

9 0.

87 ±

0.0

5

11

Nat

ural

-

- -

- -

- -

0.

05 ±

0.0

3 -

D

12 A

pple

0.

33 ±

0.0

7 0.

77 ±

0.5

2 1.

30 ±

0.3

8 1.

65 ±

0.7

6 1.

28 ±

0.3

9 0.

95 ±

0.0

6 0.

67 ±

0.0

2 0.

70 ±

0.0

3 0.

60 ±

0.0

4 0.

15 ±

0.0

3

13

Ora

nge/

peac

h 0.

82 ±

0.0

4 0.

38 ±

0.2

1 -

1.09

± 0

.31

0.99

± 0

.16

0.68

± 0

.03

0.38

± 0

.07

0.41

± 0

.05

0.80

± 0

.02

0.33

± 0

.05

14

Lem

on

0.48

± 0

.09

0.34

± 0

.11

0.99

± 0

.08

1.02

± 0

.15

0.96

± 0

.08

0.81

± 0

.06

0.62

± 0

.09

0.60

± 0

.51

0.75

± 0

.08

0.96

± 0

.04

15

Nat

ural

-

- -

- -

- -

- -

-

E

16 L

emon

6.

12 ±

0.0

4 1.

96 ±

0.4

5 0.

72 ±

0.3

3 0.

47 ±

0.6

5 0.

67 ±

0.3

4 10

.71

± 0.

61

1.58

± 0

.09

1.76

± 0

.48

3.37

±0.

93

1.42

± 0

.53

17

Ora

nge/

rasp

berr

y 7.

31 ±

0.9

5 2.

68 ±

0.6

6 1.

24 ±

0.4

8 1.

52 ±

0.0

6 1.

22 ±

0.5

0 14

.09

± 2.

63

2.24

± 0

.51

1.91

± 0

.58

3.93

± 0

.44

1.74

± 0

.25

18

Pea

ch/p

inea

pple

6.

55 ±

0.7

7 2.

72 ±

0.2

8 1.

26 ±

0.2

0 1.

57 ±

0.2

1 1.

24 ±

0.2

1 14

.20

± 3.

42

2.26

± 0

.27

2.34

± 0

.22

3.95

± 0

.57

1.75

± 0

.32

19

Gua

va/li

me

6.56

± 0

.57

2.45

± 0

.12

1.07

± 0

.09

1.1

8 ±

0.18

1.

04 ±

0.0

9 6.

11±

0.40

1.

70 ±

0.0

3 1.

07 ±

0.7

9

1.58

± 3

.16

1.17

± 0

.04

20

Nat

ural

-

- -

- -

- -

- -

-

F

21 L

emon

/gre

en te

a 4.

14 ±

0.2

1 2.

28 ±

0.0

2 0.

95 ±

0.0

1 0.

94 ±

0.0

3 0.

91 ±

0.0

1 11

.10

± 1.

21

1.66

± 0

.21

1.82

± 0

.44

3.44

± 0

.86

1.46

± 0

.04

22

Ras

pber

ry/g

inse

ng

3.58

± 0

.26

2.39

± 0

.12

1.03

± 0

.09

1.10

± 0

.18

1.00

± 0

.09

12.0

1 ±

0.24

1.

83 ±

0.0

5 1.

67 ±

0.8

1 3.

59 ±

0.0

4 1.

55 ±

0.0

2

23

Pea

ch/w

hite

tea

2.66

± 0

.18

2.03

± 0

.09

0.77

± 0.

06

0.58

± 0

.13

0.73

± 0

.07

12.4

2 ±

0.47

1.

91 ±

0.4

2

1.61

± 0

.63

3.66

± 0

.36

1.58

± 0

.21

24

Man

go/g

inkg

o be

loba

1.

62 ±

0.3

0 2.

22 ±

0.1

2 0.

91 ±

0.0

9 0.

85 ±

0.1

8 1.

87 ±

0.0

9 15

.92

± 1.

51

2.60

± 0

.29

2.40

± 0

.64

4.24

± 0

.25

1.92

± 0

.04

25

Mel

on/m

int

2.10

± 0

.05

2.19

± 0

.15

0.88

± 0

.11

0.8

0 ±

0.21

1.

84 ±

0.1

1

8.17

± 1

.39

1.08

± 0

.06

1.2

6 ±

0.02

2.

95 ±

0.7

3 1.

18 ±

0.0

2

16

2

Ade

nine

-GC

E

Gua

nine

-GC

E

Bra

nd

Sam

ple

Asc

orbi

c ac

id

Gal

lic a

cid

Caf

feic

aci

d C

oum

aric

aci

d R

esve

ratr

ol

Asc

orbi

c ac

id

Gal

lic a

cid

Caf

feic

aci

d C

oum

aric

aci

d R

esve

ratr

ol

26

Nat

ural

-

- -

- -

- -

- -

-

G

27 L

emon

19

.28

± 1.

44

3.37

± 0

.24

1.74

± 0

.18

2.52

± 0

.35

1.74

± 0

.18

18.7

7 ±

1.61

3.

15 ±

0.3

1 3.

29 ±

0.8

1 4.

71 ±

0.2

7 2.

19 ±

0.o

5

28

Lim

e 5.

01 ±

0.2

0 2.

76 ±

0.0

8 1.

29 ±

0.0

6 1.

63 ±

0.1

2 1.

27 ±

0.0

6 10

.92

± 1.

41

1.62

± 0

.28

1.38

± 0

.68

3.41

± 0

.23

1.44

± 0

.13

29

App

le

11.2

3 ±

0.96

3.

17 ±

0.2

6

1.59

± 0

.19

2.22

± 0

.38

1.58

± 0

.20

15.9

8 ±

0.88

2.

61 ±

0.1

7 2.

46 ±

0.7

0 4.

25 ±

0.1

5 1.

92 ±

0.0

8

30

Pea

ch

21.2

5 ±

0.96

2.

45 ±

0.1

3 1.

07 ±

0.0

9 1.

18 ±

0.1

9 1.

04 ±

0.1

0 14

.87

± 0.

61

0.44

± 0

.51

0.67

± 0

.77

2.40

± 0

.43

0.87

± 0

.01

31

Nat

ural

-

- -

- -

- -

- -

0.04

± 0

.02

H

32 L

emon

4.

03 ±

0.4

0 2.

48 ±

0.4

4 1.

09 ±

0.3

2 1.

23 ±

0.6

3 1.

06 ±

0.3

3 8.

92 ±

0.7

5 1.

23 ±

0.5

4 0.

92 ±

0.4

9 3.

08 ±

0.4

6 1.

25 ±

0.0

6

33

Nat

ural

-

- -

- -

- -

- -

-

I 34

Lem

on

3.91

± 0

.73

2.55

± 0

.20

1.15

± 0

.14

1.33

± 0

.29

1.12

± 0

.15

9.48

± 0

.91

1.34

± 0

.18

1.

16 ±

0.0

9 3.

17 ±

0.1

5 1.

31 ±

0.0

9

35

Gre

en A

pple

0.

37 ±

0.2

1 2.

61 ±

0.3

9 1.

19 ±

0.2

8 1.

42 ±

0.5

6 1.

16 ±

0.2

9 14

.48

± 0.

57

2.31

± 0

.11

2.15

± 0

.07

4.00

± 0

.09

1.78

± 0

.05

36

Str

awbe

rry

1.25

± 0

.05

2.05

± 0

.11

0.78

± 0

.08

0.6

1 ±

0.16

0.

74 ±

0.0

8 15

.19

± 1.

09

2.45

± 0

.21

2.4

2 ±

0.25

4.

11 ±

0.1

8 1.

85 ±

0.1

0

37

Nat

ural

-

- -

- -

- -

- -

-

J 38

Lem

on

2.10

± 0

.21

2.06

± 0

.03

0.

79 ±

0.0

2 0.

62 ±

0.0

4 0.

75 ±

0.0

2 10

.49

± 0.

23

1.54

± 0

.04

1.41

± 0

.05

3.34

± 0

.04

1.40

± 0

.02

39

Nat

ural

-

- -

-

-

-

-


163

By analysing the results in Table 2, the applications of adenine and guanine-immobilised

GCEs to the evaluation of TAC in beverages were demonstrated. Standards off all

antioxidants were available for use in the TAC determination in this study. Among them,

we recommend ascorbic acid should be used as a common standard in the determination

of TAC of foodstuff and beverages as it exhibited the widest linear calibration range at

both the guanine and adenine biosensors.

4. Conclusion

Adenine and guanine-immobilised GCEs for the evaluation of TAC in beverages was

developed. The methodology is based on the interaction of adenine or guanine with the

superoxide radical generated by the xanthine/xanthine oxidase system. Five standard

antioxidants (ascorbic acid, gallic acid, caffeic acid, coumaric acid and resveratrol) were

used in order to protect adenine and guanine base. Ascorbic acid presented the highest

TAC values and seems to be the most sensitive standard capable to discriminate the

several ingredients added to the waters.

The biosensors described in this study have some advantages over the conventional

methodologies such as a shorter detection time, a smaller sample volume, higher

accuracy and a high simplicity. In addition, coloured samples can be directly used for the

measurement without pretreatment. The use of these biosensors is closer to biological

systems, with a nucleotide being damaged by free radical.

Acknowledgements


grant (SFRH/BD/ 29440/2006). The authors thank Frize for providing flavours samples.


164

References

ANIRSF, 2010. Associação Nacional dos industriais de refrigerantes e sumos de frutos,

2010. Informar nº29. From the website www.anirsf.pt (acessed 03.11.10)

Barroso, M.F., de-los-Santos-Álvarez, N., Lobo-Castañón ,M.J., Miranda-Ordieres, A.J.,

Delerue-Matos. C., Oliveira, M.B.P.P., Tuñón-Blanco, P., 2011. Biosens. Bioelectron. 26

(5) 2396-2401.

Barroso, M.F., Silva, A., Ramos, S., Oliva-Teles, M.T., Delerue-Matos, C., Sales, M.G.F.,

Oliveira, M.B.P.P., 2009. Food Chem. 116 (2), 580-589.

Beissenhirtz, M.K., Scheller F.W., Lisdat F., 2004. Anal. Chem. 76 (16), 4665-4671.

Benherlal, P.S., Arumughan, C., 2008. Mutat. Res.-Fundam. Mol. Mech. Mutagent. 648

(1-2), 1-8.

Brett, A.M.O., Matysik, F-M., 1997. J. Electroanal. Chem. 429, 95-99.

Brett C.M.A., Brett, A.M.O., Serrano, S.H.P., 1994. J. Electroanal. Chem. 366 (1-2), 225-

231.

Burrows, C.J., Muller, J.G., 1998. Chem. Rev. 98 (3), 1109-1151.

Campanella, L., Bonanni, A., Finotii, E., Tomassetti, M., 2004. Biosensors. Bioelectron. 19

(7), 641-651.

Chen, J., Wollenberger, U., Lisdat, F., Ge B., Scheller F.W., 2000. Sens. Actuators B-

Chem. 70 (1-3), 115-120.

Chiti, G., Marrazza, G., Mascini, M. 2001. Anal. Chim. Acta 427 (2), 155-164.

de-los-Santos-Alvarez, N., de-los-Santos-Alvarez, P., Lobo-Castanon, M.J., Lopez, R.,

Miranda-Ordieres, A.J., Tunon-Blanco, P., 2007. Electrochem. Commun. 9 (8), 1862–

1866.

de-los-Santos-Álvarez, P., Lobo-Castañón, M.J., Miranda-Ordieres, A.J., Tuñón-Blanco,

P., 2002. Anal. Chem. 74 (14), 3342-3347.

Emregül, E., 2005. Anal. Bioanal. Chem. 383 (6), 947-954.

Endo, K., Miyasaka, T., Mochizuki, S., Aoyagi, S., Himi, N., Asahara, H., Tsujioka, K.,

Sakai, K., 2002. Sensor Actuat. B-Chem. 83 (1-3), 30-34.

Fojta, M., Kubicarova, T., Palecek, E., 2000. Biosensors. Bioelectron. 15 (3-4) 107-115.

Freidman, K.A., Heller., A., 2004. J. Am. Chem. Soc. 126 (8), 2368-2371.

Fulda, S., 2010. Drug Discov. Today 15 (17-18), 757-765.

Ge, B., Lisdat, F., 2002. Anal. Chim. Acta 454 (1), 53-64.

Gobi, K.V., Mizutani, F., 2000. J. Electroanal. Chem. 484 (2), 172-181.

Halliwell, B., Gutteridge, J.M., Cross, C.E., 1992. J. Lab. Clin. Med. 119 (6), 598–620.

Ignatov, S., Shishniashvili, D., Ge, B., Scheller, F.W., Lisbat, F., 2002. Biosens.

Bioelectron. 17 (3), 191-199.


165

Laranjinha, J., 2009. Oxidative Stress: From the 1980’s to Resent Update in Oxidative

Stress, Inflammation and Angiogenesis in the Metabolic Syndrome, First ed. Springer.

Liu, J.F., Roussel, C., Lagger, G., Tacchini, P., Girault, H.H., 2005. Anal. Chem. 77 (23),

7687-7694.

Liu, J.F., Su, B., Lagger, G., Tacchini, P., Girault, H.H., 2006. Anal. Chem. 78 (19), 6879-

6884.

Marrazza, G., Chianella, I., Mascini, M. 1999. Anal. Chim. Acta 387 (3), 297-307.

McNeil, C.J. Dale Athey, D. Ho, W.O., 1995. Biosens. Bioelectron.10 (1-2) 75-83.

Mello, L.D., Hernandez, S., Marrazza, G., Mascini, M., Kubota, L.T., 2006. Biosens.

Bioelectron. 21 (7), 1374-1382.

Mello, L.D., Kubota, L.T., 2007. Talanta 72 (2), 335-348.

Qian, P., Ai, S.Y., Yin, H.S., Li, J.H., 2010. Microchim. Acta 168 (3-4), 347-354.

Rice, M.E., Galus, Z., Adams, R.N., 1983. J. Electroanal. Chem. 143 (1-2), 89-102.

Sanchez-Moreno, C., 2002. Food Sci. Technol. Int. 8 (3), 121-137.

Thavasi, V., Leong, L.P., Bettens, R.P.A., 2006. J. Phys.Chem. A 110 (14), 4918-4923.

.

4.3. Biossensores de bases púricas - radical sulfato

167

Evaluation of total antioxidant capacity of flavour ed waters using

sulfate radical damage of purine-based sensors

M.F. Barrosoa,b, C. Delerue-Matosa,, M. B. P. P. Oliveirab aREQUIMTE/Instituto Superior de Engenharia do Porto R. Dr. Bernardino de Almeida 431,

4200-072 Porto. Portugal bRequimte, Serviço de Bromatologia, Faculdade de Farmácia, Universidade do Porto.

R. Aníbal Cunha n.º 164, 4050-047 Porto. Portugal

Abstract

In this study, a method for the electrochemical quantification of the total antioxidant

capacity (TAC) in beverages was developed. This method is based on the oxidative

damage of the purine bases, adenine or guanine, immobilized on glassy carbon electrode

(GCE) surface. The oxidative lesions on the DNA bases were promoted by the sulfate

radical generated by the persulfate/iron (II) system. The presence of antioxidants on the

reactive system promoted the protection of the DNA bases immobilized on the GCE by

scavenging the sulfate radical. Square wave voltammatry (SWV) was the electrochemical

technique used to perform this study. Five antioxidants (ascorbic acid, gallic acid, caffeic

acid, coumaric acid and resveratrol) were used to study its efficiencies on the scavenger

sulfate radical and consequently protect the purine bases immobilized on the GCE. The

results demonstrated that the purine-based biosensor is suitable for rapid assessment of

TAC in flavours and flavoured waters.

Keywords: Purine-based biosensor, Total antioxidant capacity (TAC), Ascorbic acid,

Phhenolic acids, Sulfate radical (SO4• −).


Electrochimica Acta submitted


169

1. Introduction

Free radical generation is directly related with oxidation in food and biological systems. It

is well know that free radicals have the facility to promote alterations on the DNA inducing

oxidative damage and causing several diseases in humans.

Sulfate radical (SO4•–) is a free radical that can also damage DNA. Some studies indicate

that sulfate radical can interact with the nucleic acids producing 8-oxoGuo and induce

modifications on the DNA, such as strand breaks and DNA-protein cross-link. Guanine

base would be the preferential target to suffer these oxidative lesions due to its low

oxidation potential and to its ability to bind to transition metal ions that can catalyze

oxidative processes [1-3]. Some authors reported the oxidation and damage on the

adenine and adenosine produced by the sulfate radical generated by photooxidation of

peroxydisulfate and quantified by UV-Vis method [4,5], while Luke and collaborators [6]

indicated the deprotonation of the nitrogen at acidic medium from the 6-methyl uracil and

5,6-dimethyl uracil induced by the referred radical.

Sulfate radical can be generated by several ways, such as via scission of peroxide bond

by radiolytic, photolytic or thermal activation of the persulfate anion [7,8], or formed via

electron transfer by transition-metal activation of persulfate and via transition-metal

catalysis [9,10]. To counteract and prevent the deleterious effect of free radicals the living

organisms have developed complex endogenous and exogenous antioxidant systems.

The exogenous antioxidant system can be provided by functional foods, vegetables, fruit,

and beverages. These matrices are rich in several antioxidants, like, vitamins (A, E, C, β-

carotene), phenolic compounds, minerals (selenium, zinc) or proteins [11].

Several methods have been proposed for the quantification of total antioxidant capacity

(TAC) in food and biological system. These methodologies are based in photometric,

fluorimetric and chromatographic techniques [12]. Recently, several electrochemical

devices have been developed in order to measure the antioxidants in several types of

matrixes [13-15], however the protective effect of antioxidants at a cellular level could only

be achieved by monitoring the DNA integrity [16]. For this purpose, several

electrochemical DNA-based biosensors have been developed. As far as we know, all the

methodologies cited in the literature reported only the use of the hydroxyl radical

generated by the Fenton system or by the UV irradiation for the evaluation of the oxidative

damage on double stranded of DNA or on the nucleobases (guanine or adenine). In these

works the protective effect on the nucleic acid, produced by the free radical scavenger,

such as vitamins and phenolic acids were also studied [17-22].

In this paper, a purine-based biosensor had been developed in order to assess TAC in

flavoured waters. This biosensor consisted in the electro-deposition of a purine base


170

(adenine or guanine) on a glassy carbon electrode (GCE). To evaluate the oxidative

lesions in the DNA bases, and for the first time, it was used the sulfate radical generated

by the persulfate activated with Fe2+ Eq. (1). For that the purine-based biosensor was

immersed in a freshly solution of sulfate radical for a fixed time period. Depending on the

pH values, the sulfate radical can interconvert in hydroxyl radical according to the

following emical reactions Eq. (2) and Eq. (3). It should be noted that in a general

persulfate oxidation system (acid condition, low [OH-] concentration) reactions of sulfate

radical with water and OH- can be negligible [23,24].

Fe2+ + S2O82- → Fe3+ + SO4

2- + SO4•– (1)

SO4•– + H2O → HO• + HSO4

- (2)

SO4•– + HO- → HO• + SO4

2- (3)

The ability of some antioxidants to scavenge oxidizing free radicals and protect the

integrity of purine bases has also studied. The antioxidants used were, ascorbic acid, and

phenolic acids, such as the hydroxybenzoic acid (gallic acid), hydroxycinnamic acid

(caffeic acid and coumaric acid) and the stilbene (resveratrol). All the electrochemical

studies were performed using the square wave voltammetry (SWV) technique. The

increase of the oxidative current of the purine-based biosensor recorded in SWV observed

when it was added to the reactive system antioxidants allows to the development of a

rapid and alternative electrochemical method to evaluate TAC in beverages, namely

flavoured waters.

2. Experimental

2.1. Chemical reagents

Guanine, adenine, iron (II) sulfate heptahydrate, potassium persulfate, gallic acid, were

purchased from Sigma. Caffeic acid was from Fluka, L(+) ascorbic acid and resveratrol

was acquired from (Riedeil-de-Haën). Other chemicals were Merck pro-analysis grade

and were used as received. Guanine stock solution (1 g L-1) was prepared by dissolving

an amount of this solid in 0.1 mol L-1 of NaOH and diluting it in phosphate buffer (PBS 0.2

mol L-1) at pH 7.4. Stock solution of 1 g L-1 of adenine were prepared in PBS pH 7.4 and

stored at 4ºC.

Working standard solution (ascorbic acid, gallic acid, caffeic acid, coumaric acid) were

prepared daily and immediately before measurements by dissolving an amount of the

solid standard in water until the desired concentration. In order to dissolve the resveratrol


171

antioxidant, an amount of this compound was dissolved in ethanol and then diluted with

water until to the desired concentration.

Considering that free radicals have a very short-lived time, the sulfate radical generation

was prepared immediately before each assay by mixing Fe2+:EDTA: S2O82- (1x10-5: 2x10-5:

2.0x10-5 mol L-1). According Liang et al. [25] if the Fe2+ is in excess (Eq. 4) can destroy the

radical and produce the sulfate ions, so, in this study it was used a molar excess of

persulfate anions. All solutions were prepared with water purified with a Direct-Q

(Millipore) system.

Fe2+ + SO4•– → Fe3+ + SO4

2- (4)

2.2. Apparatus

Square wave voltammetry (SWV) was performed with an Autolab PGSTAT 10

potentiostat controlled by GPES software (EcoChemie, The Netherlands). A conventional

three electrode cell was used, which includes glassy carbon electrode (GCE) (0.07 cm2)

as working electrode, a glassy carbon counter electrode and a Ag|AgCl|KClsat reference

electrode to which all potentials are referred. GCE was mechanically polished using a

polishing kit (Metrohm 6.2802.010) first with γ-Al2O3 (0.015 µm) until a shining surface

was obtained and after with only water. After this step the GCE was treated by applying a

fixed potential of +1.7 V for 30 s in PBS pH 4.8. After this treatment a thin blue film can be

observed on the activated GCE surface. This procedure was made after each experiment.

This initial conditioning step improves the resolution of the analytical signal because the

application of high potentials in acidic medium increases the hydrophilic properties of the

electrode surface through the introduction of oxygenated functionalities [18,26].


Unless otherwise mentioned, most experiments consisted of three steps: i) Guanine or

adenine electro-immobilization on the GCE, ii) damage of DNA bases by the immersion of

purine-GCE on the sulfate radical solution, and study of the effect of the presence of

antioxidants in the system; iii) detection and measurement of the peak height of adenine

or guanine in a PBS at pH 4.8.

Purine bases (adenine or guanine) immobilization was performed by the application of

an adsorptive accumulation step. For that, the activated GCE was immersed in 0.2 mol L-1

PBS pH 4.8 containing 10 mg L-1 of adenine or 3 mg L-1 of guanine and it was applied a

positive potential of +0.4 V for 180 s on permanent stirring, after this the electrode was


172

washed with water. When this procedure was made the purine base was subsequently

immobilized onto activated GCE surface by adsorptive accumulation (controlled-potential

at +0.4 V) involving the application of positive electrode potential to achieve electrostatic

binding of negatively charged purine base. The purine-based biosensor (cleaning and

immobilization step) and the general analytical procedure (buffer composition, pH, ionic

strength) were optimized in previous works [18,27,28].

Purine base damage was carried out by immersing the biosensor in a freshly prepared

sulfate radical in the absence or the presence of antioxidant in PBS pH 4.8 for a fixed

period of reaction time. Damage to purine base layer was made through diffusion of the

radicals to the surface of the transducer. To study the effect of the antioxidant on the free-

radical scavenging, the standard antioxidant (ascorbic acid, caffeic acid, gallic acid,

coumaric acid and resveratrol) were added to the cleavage mixture. Protective effects of

antioxidants in flavours and flavoured waters were done by replacing the standard

antioxidant by samples.

For detection, the damaged purine-based biosensor was immersed in PBS solution pH

4.8. The peak height of the purine base was obtained by sweeping the potential between

+0.2 V and +1.4 V using the SWV technique with a frequency of 50 Hz, step potential of

4.12 mV and amplitude of 0.09 V. For these electrochemical studies it was considered

that the maximum intensity current obtained was for the purine base signal without

damage neither antioxidant effect.

2.4. Samples


several supermarkets in the North of Portugal and stored in the dark at +4ºC. Each brand


water of each brand was also used as control. Sonication was used to eliminate gas from

the sparkling water samples. The labels on the water bottles indicate the nutrient

information, namely the presence of fruit juice, vitamins, sweeteners and preservatives

[29,30].

Six flavours in liquid state used in the formulation of some water brands, provided by a

producer, were also analysed. The flavours used corresponding to different fruit aromas,

such as lime, tangerine, strawberry, lemon, apple and gooseberry. These flavours had no

description about their chemical or aroma composition, only knowing that they are present

in the flavoured waters used in this study.


173

2.5. TAC measurement in flavoured waters


flavoured waters. For the measurement of TAC in beverages, 100 µL of the flavoured

water or 5 µL of flavour were diluted in PBS to a final volume of 500 µL. Then, the purine-

GCE was immersed in the solution and a freshly prepared sulfate radical was added for

120 s and 60 s for guanine-GCE and adenine-GCE respectively. After this period of time

the biosensor was washed and immersed in PBS buffer pH 4.8 to measured the oxidation

current of guanine and adenine. Ascorbic acid, gallic acid, caffeic acid, coumaric acid and

resveratrol were the working standard antioxidants used, to study the protective effect

made by the antioxidant on the free-radical scavenging, and to carry out the linear

calibrations studies.


To develop the purine-based biosensor for the quantification of TAC in beverages, the

first important analytical parameter optimized was the concentration of the purine base

immobilized on the GCE. As can be seen in Fig. 1, a saturation behaviour on the

electrochemical signal of the biosensor was observed for a concentration higher than 3

mg L-1 and 10 mg L-1 for guanine and adenine, respectively. Considering that with these

concentrations it was found the better analytical condition for the immobilization of the

purine bases on the GCE these concentration were chosen for further experiments.

In order to verify if the sulfate radical generated by the persulfate/iron (II) system have

the ability to induce oxidative lesion on the purine base (guanine and adenine)

immobilized on the GCE, the purine-based biosensor was placed in a freshly prepared

SO4•− in PBS pH 4.8 for 1 minute. After this process the purine-based biosensor was

washed with water and a SWV was made from +0.2 to +1.4 V. It was observed that the

sulfate radical induce damage on the purine bases, indeed when the purine bases

immobilized on the GCE interact with the sulfate radical a decrease on the oxidation peak

of guanine and adenine (Fig. 2) was observed.

Fig. 2 shows the performance of the purine-based biosensor in presence of the sulfate

radical and of ascorbic acid antioxidant on radical scavenging. Sulfate radicals had the

ability to produce 63% and 61% of damage in guanine and adenine base, respectively.

When ascorbic acid was added to the reactive system an increase of the anodic current of

guanine and adenine was registered using SWV and a protective effect of 69% and 73%

was observed in guanine-GCE and adenine-GCE, respectively. These results confirm the

ability of the antioxidant to deactivate the sulfate radical and consequently protect the


174

purine bases from the oxidative damage. To perform this electrochemical study all current

peaks were compared with the signal current obtained with the non damaged adenine and

guanine bases (blank signal).

Fig. 1. Square wave voltammogram of a) guanine (3 mg L-1) and b) adenine (10 mg L-1); Influence of the

concentration of the purine base immobilized on the electrochemical current: a) guanine; b) adenine.

Conditions: potential range from to, frequency = 50 Hz, step potential= 4.12 mV and amplitude= 0.09 V.

20.0

30.0

40.0

50.0

60.0

0.20 0.40 0.60 0.80

E(V) vs. AgCl/Ag

i/ µA

20.0

30.0

40.0

50.0

60.0

0.20 0.40 0.60 0.8020.0

30.0

40.0

50.0

60.0

0.20 0.40 0.60 0.8020.0

30.0

40.0

50.0

60.0

0.20 0.40 0.60 0.80

E(V) vs. AgCl/Ag

i/ µA

45.0

50.0

55.0

60.0

65.0

70.0

0.60 0.80 1.00 1.20

E(V) vs. AgCl/Ag

i/ µA

45.0

50.0

55.0

60.0

65.0

70.0

0.60 0.80 1.00 1.20

E(V) vs. AgCl/Ag

i/ µA

4.0

8.0

12.0

16.0

20.0

0.0 2.0 4.0 6.0 8.0 10.0 12.0

Purine base concentration / mg L -1

i p/ µ

A

a)b)

4.0

8.0

12.0

16.0

20.0

0.0 2.0 4.0 6.0 8.0 10.0 12.0

Purine base concentration / mg L -1

i p/ µ

A

a)b)

guanine a)

adenine b)


175

Fig. 2. Signal of the immobilized purine base on the GCE: blank purine base signal (guanine 3 mg L-1 and

adenine 10 mg L-1); after immersion in a sulfate radical (Fe2+ 2.0x10-5 mol L-1; S2O82- 4.0x10-5 mol L-1 for 120

s); immersion in sulfate radical with ascorbic acid antioxidant (4.0 mg L-1).

Purine bases of DNA measured in SWV presented two oxidation peaks at around +0.55

V and +0.82 V corresponding, respectively, to guanine and adenine oxidation peak (Fig.

1). With the analysis of the undamaged purin a maximum current peak was registered

because all adenine and guanine were available for electro-oxidation. But when the

biosensor interact with the sulfate radical, the radical attacks the purine base and promote

the oxidative damage of some amount of the guanine or adenine. So, after this process

the peak current decrease because only the undamaged DNA bases were available for

electro-oxidation in voltammetric techniques. With the addition of antioxidants to the

reactive system the number of lesions diminishes yielding a larger number of adenine or

guanine available for electrochemical oxidation. So, an increase of the peak current

occurs confirming the protective effect on the purine bases made by the antioxidants.

Swaraga and colaborators [4,5] reported the photooxidation of adenosine and adenine

induced by the sulfate radical generated by the photolysis of peroxydisulfate and the

protection of this DNA bases promoted by the antioxidant caffeic acid. Scheme 1 shows a

protective mechanism promoted by caffeic acid on the DNA bases against the sulfate

radical. In this purposed mechanism [5] caffeic acid act as sensitizers and as an efficient

scanvenger of SO4•− by receiving the free electron from the sulfate radical. Considering

the good correlation between antioxidant concentration and the protective effect on the

purine bases of DNA, an analytical procedure to evaluate TAC was developed.

guanine

guanine

guanine

adenine

adenine

adenine

0

25

50

75

100

125

blank sulphate radical ascorbic acid

DN

A s

igna

l / %

guanine

guanine

guanine

adenine

adenine

adenine

0

25

50

75

100

125

blank sulphate radical ascorbic acid

DN

A s

igna

l / %

100value) expected (maximum done wasdamage no whenmeasured i

radical sulfate withdamage base purine after measured i (%) signalDNA

p

p ×=


176

Scheme 1 . Mechanism of adenine protection produced by caffeic acid against the sulfate radical (adapted

from Swaraga and collaborator [5]).


In order to evaluate the TAC on beverages, some parameters concerning the damage

on the purine base immobilized on the GCE surface (persulfate and Fe2+ concentration,

and reaction time between sulfate radical and the target molecule) were optimized in order

to achieve the maximum DNA bases effect, but without a complete damage.

Persulfate (S2O82-) concentration was studied between 2.0x10-5 and 1.0x10-4 mol L-1. It

was observed a decrease on the peak current (ip) of guanine and adenine immobilized on

the GCE in all range of persulfate concentration studied. This is indicative of the powerful

effect of persulfate on the generation of the sulfate radical. However considering that the

persulfate anion caused some electronics problems on the GCE surface, for the next

optimization step, it was chosen the persulfate concentration of 2.0x10-5 and 4.0x10-5 mol

N

N

NH2

N

HN

+ SO4·-

N

N

NH2

N

HN

+ SO42- +

(adenine)

(caffeic acid radical)

·N

N

NH2

NH

N

H + + SO42-(damage)

(protection)

OH

CH=CH-COOH

O·CH=CH-COOH

(caffeic acid)

OH

OH

N

N

NH2

N

HN

+ SO4·-

N

N

NH2

N

HN

+ SO42- +

(adenine)

(caffeic acid radical)

·N

N

NH2

NH

N

H +

·N

N

NH2

NH

N

H + + SO42-(damage)

(protection)

OH

CH=CH-COOH

O·

OH

CH=CH-COOH

O·CH=CH-COOH

(caffeic acid)

OH

OH

CH=CH-COOH

(caffeic acid)

OH

OH

(caffeic acid)

OH

OH


177

L-1 for the guanine and adenine biosensor, respectively. For the both purine-based

biosensors (guanine and adenine) Fe2+ was ranged from 1.0x10-5 and 1x10-4 mol L-1. With

the increases of Fe2+ concentration a decrease of the ip of guanine and adenine was

recorded on SWV. However, the decrease of the ip in all Fe2+ concentration range studied

remain very similar, so, for the next optimization step it was chosen a Fe2+ concentration

of 1.0x10-5 and 2.0x10-5 mol L-1 for guanine and adenine biosensor, respectively. Using

these optimized concentrations of Fe2+ and persulfate it was obtained a molar ratio S2O82-

/Fe2+ of 2 for both purine-based biosensor.

Reaction time between the sulfate radical and the bases immobilized on the GCE

surface depends on the half-life time on the generated free radical, so this parameter is an

important feature to optimize. In this study the incubation time were ranged from 0 to 120

s. Fig. 3 shows the correlation between the damage on the purine base produced by the

sulfate radical (interrelated with the ip values) and the incubation time. A decrease on the

ip was observed with the increase of the reaction time, however it was not observed a

complete damage of the purine bases. When it was used the adenine-biosensor it was

observed the lower ip indicating that adenine is more damaged than the guanine-

biosensor, but 2 minutes after, the damage was similar in the two purine bases. These

results are agreed with those obtained by other authors [31].The incubation time of 120 s

and 60 s was chosen for guanine-GCE and adenine-GCE respectively for all experiments.

Fig. 3. Influence on the peak current of the biosensor after immersion in a sulfate radical with the incubation

time a) 10 mg L-1 of adenine base; b) 3 mg L-1 of guanine base.

3.2. Determination of TAC in flavoured waters

Beverages (such as juice) are an excellent source of exogenous antioxidants. So,

drinking this type of foodstuff can help the human body to provide an adequate and

continuous supply of antioxidants. Antioxidants, such as, ascorbic acid and phenol-

0.00

4.00

8.00

12.00

0 20 40 60 80 100 120

Time / s

i p/ µ

A

a) guanine

b) adenine

0.00

4.00

8.00

12.00

0 20 40 60 80 100 120

Time / s

i p/ µ

A

0.00

4.00

8.00

12.00

0 20 40 60 80 100 1200.00

4.00

8.00

12.00

0 20 40 60 80 100 120

Time / s

i p/ µ

A

a) guanine

b) adenine


178

derived compounds are natural components of fruits. Flavoured waters consist of the

addition of flavours/aromas, juices and considering that these added ingredients are fruit

extracts, they contain natural antioxidants, transferring them to the bottled water. So,

drinking this type of water can increase the daily intake of natural exogenous antioxidants

and may contribute to the protective system against free radicals. The consumption of

flavoured water is increasing over the world namely in Portugal where it was consumed

around 6.08 million L only in the first half of 2010 [32]. Phenolic compounds are correlated

with antioxidant activity and seem to have an important role in stabilizing lipid oxidation

[33]. The total phenolic contents (TPC) of this kind of waters was evaluated (data not

showed). It was observed that all flavoured waters presented phenolic compounds in its

composition. Like it was expected, natural waters did not have TPC. The highest TPC

levels were from citrus fruits flavours (tangerine, lime and lemon) and from waters with

bioactive compounds, like, tea, ginseng and gingko biloba [34]. For the evaluation of the

TAC of flavoured waters it was used five antioxidants compounds: ascorbic acid, gallic

acid, caffeic acid, coumaric acid and resveratrol. Ascorbic acid is a water-soluble vitamin,

considerated a powerfull antioxidant and plays a key role in the protection against

biological oxidation processes participating in many metabolic reactions [15]. Gallic,

caffeic and coumaric acids are phenolic acids with a large protective action. The phenolic

acids, can included several groups, such as the hydroxybenzoic acid (gallic acid) and the

hydroxycinnamic acid (caffeic and coumaric acid). In general, the antioxidant activity of

the phenolic-derived compounds is determined by some properties, such as free-radical

scavengers [35]. Resveratrol is a stilbene, can be found in grapes and wine and have

multiple beneficial properties, including chemopreventive and antitumor activities [36].

Linearity studies between the five antioxidants and ip of guanine and adenine oxidation

was carried out. Fig. 4 (i and ii) shows the SWV of electrochemical current obtained after

immersion the purine-biosensor on the sulfate radical containing increasing concentration

of gallic acid or coumaric acid. Like it was expected the oxidation current of DNA-bases

increased when the concentration of the antioxidant also increase. Similar

voltammograms were obtained when it was used the other antioxidants (ascorbic acid,

resveratrol, gallic, caffeic and coumaric acid) with the both purine-based biosensors

(guanine and adenine).


179

Fig. 4. SWV obtained after immersion of i) guanine-biosensor in sulfate radical containing a standard solution

of gallic acid: a) 0.20; b) 0.30; c) 0.40; d) 0.50 and e) 0.60 mg L-1. Inset: relationship between ip and gallic acid

concentration and a SWV of a flavoured water sample and ii) Adenine-biosensor in sulfate radical containing a

standard solution of coumaric acid: a) 0.50; b) 0.60; c) 0.70; d) 0.80 and e) 1.00 mg L-1. Inset: relationship

between ip and coumaric acid concentration and a SWV of a flavoured water sample.

Table 1 presents the several analytical parameters obtained for the five antioxidants

standard used on the protection of the purine base (adenine or guanine) immobilized on

the GCE against the sulfate radical. Ascorbic acid presents the larger linear range ranging

from 0.50 to 4.00 mg L-1 in both purine-based biosensors. The other antioxidants

presented a narrow linear range, from 0.20 to 0.60 mg L-1 of gallic acid, 0.40 to 1.20 of

caffeic acid, 0.50 to 0.90 mg L-1 of coumaric acid and 0.10 to 0.50 mg L-1 of resveratrol

when it was used the guanine-GCE. For the adenine-GCE the linear ranges were from

20.0

40.0

60.0

0.10 0.30 0.50 0.70 0.90

4.0

6.0

8.0

10.0

0.20 0.40 0.60

i)

Blank signal

Sulfate radical

E(V) vs. AgCl/Ag

30.0

40.0

50.0

0.3 0.5 0.7i/µ

A

Flavoured water sample

i/ µ

A

E(V) vs. AgCl/Ag

i p/µ

A

gallic acid (mg L-1)

20.0

40.0

60.0

0.10 0.30 0.50 0.70 0.90

4.0

6.0

8.0

10.0

0.20 0.40 0.60

i)

Blank signal

Sulfate radical

E(V) vs. AgCl/Ag

30.0

40.0

50.0

0.3 0.5 0.7i/µ

A


i/ µ

A

E(V) vs. AgCl/Ag

i p/µ

A

gallic acid (mg L-1)

E(V) vs. AgCl/Ag

60.0

0.50 0.70 0.90 1.10 1.30

80.0

4.0

6.0

8.0

10.0

0.50 0.60 0.70 0.80 0.90 1.00

ii)

Blank signal

Sulfate radical

40.0

50.0

60.0

70.0

0.6 0.8 1.0 1.2

i/ µ

A

E(V) vs. AgCl/Ag


i/ µ

A

coumaric acid (mg L-1)

i p/µ

A

E(V) vs. AgCl/Ag

60.0

0.50 0.70 0.90 1.10 1.30

80.0

4.0

6.0

8.0

10.0

0.50 0.60 0.70 0.80 0.90 1.00

ii)

Blank signal

Sulfate radical

40.0

50.0

60.0

70.0

0.6 0.8 1.0 1.2

i/ µ

A

E(V) vs. AgCl/Ag


i/ µ

A

coumaric acid (mg L-1)

i p/µ

A


180

0.10 to 0.50 mg L-1 for the antioxidants gallic acid, 0.30 to 0.80 mg L-1 of caffeic acid, 0.50

to 1.00 mg L-1 of coumaric acid and 0.10 to 0.60 mg L-1 of resveratrol. RSD values were

below 10% confirmed the high precision of the methods.


Guanine-GCE Linear range (mg L-1) 0.50–4.00 0.20–0.60 0.40–1.20 0.50–0.90 0.10–0.50

Slope (A mg L-1L) 1.14x10-6 8.84x10-6 6.41x10-6 9.17x10-6 9.87x10-6

Intercept (A) 4.69x10-6 2.44x10-6 1.99x10-6 2.39x10-7 5.80x10-6


RSD (%) (mg L-1) 4.71 (3.00) 6.58 (0.50) 2.51 (1.00) 2.77 (0.60) 5.05 (0.10) LOD (mg L-1) 0.47 0.02 0.15 0.04 0.02

Adenine-GCE

Linear range (mg L-1) 0.50–4.00 0.10–0.50 0.30–0.80 0.50–1.00 0.10–0.60

Slope (A mg L-1) 5.58x10-7 1.00x10-5 5.02x10-6 8.96x10-6 9.21x10-6

Intercept (A) 6.28x10-6 5.07x10-6 6.52x10-6 4.05x10-7 3.74x10-6 Correlation coefficient (n=5) 0.990 0.975 0.990 0.998 0.990

RSD (%) (mg L-1) 0.54 (1.00) 4.45 (0.20) 3.86 (0.50) 7.26 (0.60) 5.44 (0.20)

LOD 0.50 0.10 0.10 0.50 0.10 The purine-based biosensor was applied for the determination of TAC on flavour and

flavoured waters. Table 2 shows the TAC values expressed in mg L-1 of ascorbic acid,

gallic acid, caffeic acid, coumaric acid and resveratrol.

It was verified that all flavours and flavoured waters presented antioxidant capacity; the

natural waters, like it was expected, not presented antioxidant capacity. Flavours

presented the highest values of TAC, indeed, flavours are fruit extract and have in its

composition several concentred antioxidant compounds, so these results was expected.

Using the adenine and guanine GCE the higher TAC values were found with the

standard ascorbic acid. The flavours that presented the highest TAC when it was used the

guanine-biosensor was the apple flavour fallowed by tangerine, strawberry, lemon, lime

and gooseberry. With the adenine-biosensor apple had also the highest TAC level

fallowed by lemon, gooseberry, lime, tangerine and strawberry.

When it was used, the guanine-biosensor applied to the analysis of TAC in flavoured

waters, TAC values ranged from 2 to 16 mg L-1, 0.2 to 3 mg L-1, 0.7 to 5 mg L-1, 1.2 to 4.5

mg L-1 and 0.7 to 2 mg L-1 with the antioxidant ascorbic acid, gallic acid, caffeic acid,

coumaric acid and resveratrol, respectively. Using the ascorbic acid as standard

antioxidant it was observed higher TAC values in sparkling flavoured waters (brand E to J)

than in still waters (brand A to D). The higher TAC content were from brand F (sample 21,

22, 24 and 25), brand G (sample 27, 29 and 30) and brand H (sample 32). Samples from

brand F have several added ingredients, such as tea, ginseng and ginkgo beloba, brand

G and H have also vitamin C.

.

18

1

Tab

le 2

. T

AC

val

ues

obta

ined

for

the

flavo

urs

and

flavo

ured

wat

ers

usin

g a

guan

ine-

GC

E a

nd a

deni

ne-G

CE

(m

g L-1

).

Gua

nine

-GC

E

Ade

nine

-G

CE

Bra

nd

Sam

ple

Asc

orbi

c ac

id

Gal

lic a

cid

Caf

feic

aci

d C

oum

aric

aci

d R

esve

ratr

ol

Asc

orbi

c ac

id

Gal

lic a

cid

Caf

feic

aci

d C

oum

aric

aci

d R

esve

ratr

ol

Lem

on

154.

39 ±

5.3

2 40

.72

± 3.

78

69.5

8 ±

9.84

67

.73

± 6

.88

19.7

6 ±

2.40

23

5.35

± 4

.89

50.1

3 ±6

.86

235.

35 ±

10.

54

77.8

0 ±

9.30

36

.81

± 8.

64

Tan

gerin

e 20

9.12

± 1

0.12

47

.09

± 5.

31

79.3

1 ±

6.30

74

.54

± 11

.39

26.0

8 ±

3.62

91

.40

± 8.

04

27.8

8 ±

3.7

6 91

.40

± 5.

87

53.9

2 ±

8.46

14

.63

± 2.

86

App

le

281.

89 ±

7.3

2 55

.52

± 7.

41

92.2

5 ±

3.76

83

.58

± 2

.63

34.4

8 ±

2.44

23

9.12

± 2

1.36

50

.57

± 5.

95

239.

12 ±

15.

23

78.2

7 ±

6.38

37

.24

± 5.

93

Str

awbe

rry

184.

21 ±

15.

12

44.2

1 ±

1.60

74

.88

± 4.

67

71.4

4 ±

4.96

23

.20

± 4.

75

24.6

1 ±

2.44

25

.72

± 3.

76

24.6

1 ±

2.56

51

.60

± 4.

03

12.4

7 ±

3.75

Goo

sebe

rry

101.

75 ±

2.4

5 34

.65

± 2.

65

60.2

2 ±

5.78

61

.19

± 6.

23

13.6

8 ±

2.32

14

0.23

± 9

.15

7.16

± 1

.43

14

0.00

± 1

2.43

66

.05

± 6.

94

25.8

0 ±

2.97

Fla

vour

Lim

e 12

6.93

± 4

.32

37.5

7 ±

4.87

64

.70

± 8.

76

64.3

2 ±

9.42

16

.59

± 4.

32

110.

13 ±

4.4

0 35

.63

± 3.

50

110.

13 ±

4.4

0 62

.23

± 5.

47

22.3

5 ±

5.09

A

1 Le

mon

2.

57 ±

0.5

4 1.

44 ±

0.1

3 2.

56 ±

0.1

9 2.

75 ±

0.1

3 0.

39 ±

0.0

2 3.

14 ±

0.2

3 1.

24 ±

0.0

4 2.

26 ±

0.5

8 2.

53 ±

0.0

6 0.

84 ±

0.0

2

2 M

ango

6.

63 ±

0.7

2 1.

91 ±

0.4

0 3.

28 ±

0.1

9 3.

25 ±

0.4

3 1.

14 ±

0.0

8 2.

26 ±

0.0

7 1.

16 ±

0.0

3 2.

12 ±

0.0

4 2.

44 ±

0.0

2 0.

74 ±

0.0

7

3 S

traw

berr

y 5.

79 ±

0.6

2 1.

81 ±

0.7

1 3.

14 ±

0.0

9 3.

15 ±

0.0

6 1.

27 ±

0.0

2 2.

81±

0.04

1.

34 ±

0.0

4 2.

46 0

.28

2.64

± 0

.05

0.68

± 0

.03

4

Nat

ural

- -

- -

-

B

5 P

inea

pple

/ora

nge

5.87

± 0

.26

1.82

± 0

.06

3.15

± 0

.09

3.16

± 0

.07

0.77

± 0

.06

2.56

± 0

.04

1.24

± 0

.03

2.

25 ±

0.0

4 2.

53 ±

0.0

8 0.

78 ±

0.0

3

6

Lem

on

9.66

± 0

.41

2.26

± 0

.09

3.82

± 0

.15

3.63

± 0

.10

1.21

± 0

.09

2.11

± 0.

02

1.06

± 0

.04

1.97

± 0

.23

2.33

± 0

.01

0.72

± 0

.05

7

Nat

ural

-

- -

- -

C

8 Le

mon

/Mag

nesi

um

9.19

± 0

.43

2.21

± 0

.56

3.74

± 0

.87

3.57

± 0

.17

1.16

± 0

.07

4.22

± 0

.56

1.63

± 0

.07

2.86

± 0

.03

2.95

± 0

.02

0.97

± 0

.05

9

App

le/w

hite

tea

9.18

± 0

.45

2.21

± 0

.34

3.74

± 0

.52

3.57

± 0

.36

1.16

± 0

.03

5.23

± 0

.26

0.82

± 0

.02

1.61

± 0

.02

2.08

± 0

.05

0.30

± 0

.03

10

Pin

eapp

le/fi

bre

11.2

5 ±

1.58

2.

45 ±

0.3

7 4.

11 ±

0.0

6 3.

82 ±

0.3

9 1.

40 ±

0.0

7 3.

12 ±

0.7

2 1.

16 ±

0.0

5 2.

13 ±

0.0

6 2.

44 ±

0.0

6 0.

84 ±

0.0

7

11

Nat

ural

-

- -

- -

D

12 A

pple

3.

77 ±

0.4

7 0.

15 ±

0.1

1 2.

78 ±

0.1

6 2.

90 ±

0.1

2 0.

53 ±

0.0

2 -

0.54

± 0

.02

1.51

± 0

.03

1.89

± 0

.23

0.96

± 0

.07

13

Ora

nge/

peac

h 2.

43 ±

0.6

7 0.

23 ±

0.0

8 0.

71 ±

0.0

9

1.45

± 0

.23

0.66

± 0

.08

- 0.

95 ±

0.0

6 1.

34 ±

0.0

6 2.

06 ±

0.0

3 0.

28 ±

0.0

4

14

Lem

on

2.88

± 0

.30

0.28

± 0

.05

0.79

± 0

.07

1.51

± 0

.31

0.76

± 0

.04

1.82

± 0

.05

2.20

± 0.

10

2.73

± 0

.05

3.56

± 0

.05

1.53

± 0

.15

15

Nat

ural

- -

- -

E

16 L

emon

10

.82

± 0.

03

1.60

± 0

.13

2.80

± 0

.07

2.91

± 0

.21

1.35

± 0

.12

5.00

± 0

.34

2.12

± 0

.06

3.61

± 0.

04

3.48

± 0

.34

1.36

± 0

.05

17

Ora

nge/

rasp

berr

y 6.

36 ±

0.3

7 1.

88 ±

0.0

8 3.

24 ±

0.1

3 3.

22 ±

0.0

9 0.

83 ±

0.0

9 0.

83 ±

0.0

9 1.

59 ±

0.0

7 2.

79 ±

0.0

5 1.

39 ±

0.0

7 0.

54 ±

0.0

4

18

Pea

ch/p

inea

pple

4.

73 ±

0.3

0 1.

69 ±

0.0

3 2.

95 ±

0.0

2 3.

02 ±

0.0

5 1.

02 ±

0.0

3 7.

56 ±

0.8

4 1.

29 ±

0.0

1

2.32

± 0

.05

2.58

± 0

.05

0.62

± 0

.05

19

Gua

va/li

me

2.33

± 0

.19

1.41

± 0

.07

0.48

± 0

.08

1.2

9 ±

0.02

0.

90 ±

0.0

7 4.

23 ±

0.0

5 0.

72 ±

0.0

5 1.

46 ±

0.0

3 1.

98 ±

0.0

4 0.

14 ±

0.0

1

20

Nat

ural

-

- -

- -

F

21 L

emon

/gre

en te

a 11

.89

± 0.

09

2.52

± 0

.02

4.22

± 0

.04

3.91

± 0

.02

1.47

± 0

.02

1.31

± 0

.07

1.24

± 0

.01

2.25

± 0

.18

2.53

± 0

.03

0.58

± 0

.05

22

Ras

pber

ry/g

inse

ng

16.8

0 ±

0.58

3.

09 ±

0.0

6 5.

09 ±

0.3

2 4.

52 ±

0.1

4 2.

04 ±

0.0

6 2.

86 ±

0.0

4 0.

86 ±

0.0

3 1.

68 ±

0.0

9 2.

13 ±

0.3

5 0.

20 ±

0.0

3

18

2

Gua

nine

-GC

E

Ade

nine

-G

CE

Bra

nd

Sam

ple

Asc

orbi

c ac

id

Gal

lic a

cid

Caf

feic

aci

d C

oum

aric

aci

d R

esve

ratr

ol

Asc

orbi

c ac

id

Gal

lic a

cid

Caf

feic

aci

d C

oum

aric

aci

d R

esve

ratr

ol

23

Pea

ch/w

hite

tea

9.81

± 0

.10

2.27

± 0

.23

3.85

± 0

.36

3.65

± 0

.25

1.23

± 0

.03

0.98

± 0

.03

1.16

± 0

.06

2.

13 ±

0.0

5 2.

44 ±

0.0

4 0.

50 ±

0.0

4

24

Man

go/g

inkg

o be

loba

11

.78

± 0.

65

2.51

± 0

.15

4.20

± 0

.23

3.89

± 0

.16

1.45

± 0

.15

5.56

± 0

.07

1.13

± 0

.01

1.52

± 0

.04

2.02

± 0

.43

0.45

± 0

.01

25

Mel

on/m

int

14.0

5 ±

1.36

2.

77 ±

0.3

2 4.

61 ±

0.0

8 4.

17 ±

0.0

4 1.

72 ±

0.0

1 4.

00 ±

0.4

5 0.

98 ±

0.0

4

1.02

± 0

.07

1.67

± 0

.05

0.37

5 ±

0.0

26

Nat

ural

-

- -

- -

- -

- -

-

G

27 L

emon

10

.85

+ 2

.90

2.40

± 0

.07

4.04

± 0

.03

3.7

8 ±

0.0

8 1.

35 ±

0.0

7 4.

00 ±

0.0

5 1.

24 ±

0.0

3 2.

25 ±

0.0

9 2.

53 ±

0.0

4 0.

57 ±

0.0

4

28

Lim

e 12

.14

± 0.

80

2.55

± 0

.19

4.27

± 0

.28

3.94

± 0

.19

1.50

± 0

.09

1.00

± 0

.03

0.98

± 0

.01

1.65

± 0

.05

2.11

± 0

.04

0.15

± 0

.03

29

App

le

13.7

8 ±

2.72

2.

74 ±

0.0

2 4.

56 ±

0.0

7 4.

14

± 0

.07

1.69

± 0

.03

1.52

± 0

.027

1.

28 ±

0.0

4 2.

31

± 0.

24

2.57

± 0

.03

0.62

± 0

.04

30

Pea

ch

11.6

1 ±

2.11

2.

49 ±

0.0

9 4.

17 ±

0.0

5 3.

87

± 0

.03

1.44

± 0

.05

2.56

± 0

.15

1.18

± 0

.05

2.16

± 0

.06

2.46

± 0

.05

0.78

± 0

.05

31

Nat

ural

-

- -

- -

- -

- -

-

H

32 L

emon

10

.06

± 0.

72

2.31

± 0

.04

3.90

± 0

.07

3.6

8 ±

0.0

7 1.

26 ±

0.0

2 4.

29 ±

0.0

5 1.

23 ±

0.0

3 2.

24 ±

0.0

5 2.

52 ±

0.0

1 0.

98 ±

0.0

3

33

Nat

ural

-

- -

- -

- -

- -

-

I 34

Lem

on

12.8

3 ±

0.52

2.

63 ±

0.1

2 4.

39 ±

0.1

8 4.

02

± 0

.13

1.58

± 0

.11

8.01

± 1

.23

2.07

± 0

.17

2.54

± 0

.04

2.73

± 0

.05

1.4

± 0.

07

35

Gre

en A

pple

13

.72

± 0.

53

2.73

±0.

59

4.55

± 0

.06

4.13

± 0

.04

1.68

± 0

.09

0.75

± 0

.02

1.08

± 0

.02

2.

01 ±

0.0

3 2.

36 ±

0.0

6 0.

57 ±

0.0

4

36

Str

awbe

rry

13.6

0 ±

1.46

2.

72 ±

0.3

4 4.

52 ±

0.5

2 4.

12 ±

0.3

6 1.

66 ±

0.0

4 0.

98 ±

0.0

5 0.

79 ±

0.0

1

1.56

± 0

.06

2.05

± 0

.03

0.12

± 0

.06

37

Nat

ural

-

- -

- -

- -

- -

-

J 38

Lem

on

11.4

5 ±

1.86

2.

47 ±

0.4

3 4.

14 ±

0.6

7 3.

85

± 0

.47

1.42

± 0

.03

1.89

± 0

.04

0.69

± 0

.03

1.40

± 0

.05

1.94

± 0

.02

0.87

± 0

.09

39

Nat

ural

-

- -

-

-

-

-


183

These ingredients are antioxidants, so it was expected an increase of the TAC values is

these samples due to the dual antioxidant effect on the protection of DNA. The lowest

TAC content was from Brand D. Using the gallic acid as a standard antioxidant the higher

TAC were from brand F (sample 22) fallowed by brand G (sample 27-29), and I (sample

34-36), the lowest TAC values was from brand D. With the others antioxidants (caffeic

acid, coumaric acid and resveratrol) it was obtained a similar TAC behaviour. TAC values

obtained with the four antioxidant, gallic acid, caffeic acid, coumaric acid and resveratrol

are narrower than the values obtained with the standard ascorbic acid antioxidant. Theses

differences obtained between the ascorbic acid and the others antioxidants can be

elucidated by the fact that ascorbic acid is a powerful antioxidant and in this study

presented a larger linear range.

Using the adenine-GCE all TAC values were very similar. Using the ascorbic acid as

standard, TAC contents ranged from 0.83 to 8 mg L-1. The highest TAC value was from

sample 34 (brand I) followed by sample 18 (brand E), sample 25 (brand F), and sample 9

(brand C). Sample 12 and 13 from brand D did not presented TAC activity. With the

standard gallic acid TAC values ranged from 0.5 to 2.2 mg L-1, 1.0 to 3.6 mg L-1 with the

caffeic acid, 1.6 to 3.5 mg L-1 with the coumaric acid and 0.20 to 1.41 mg L-1 with the

standard resveratrol. With the resveratrol standard it was observed a narrow range of TAC

values. Using the other antioxidants (gallic acid, caffeic acic and coumaric acid) the TAC

behaviour was similar.

Analysing results from Table 2 it is possibly to confirm that the purine bases immobilized

on GCE can be used for the quantification of TAC in beverages, however using the

guanine-GCE and ascorbic acid as antioxidant standard it was obtained the highest TAC

values.

The purine-based biosensors described in this study have some advantages over the

conventional methodologies previously reported in the literature, such as determination of

TAC based on the electrochemical properties do not require the use of reactive

compounds, since it is based on electrochemical behaviour and consequently on their

chemical-physical properties. Furthermore, the electrochemical method has some

advantages over the commonly used optical method, such as a shorter detection time, a

smaller sample volume, higher accuracy and a high simplicity. In addition, colored

samples can be directly used for the measurement without pretreatment. The use of the

purine-based biosensor are closer to biological systems, with a nucleotide being damaged

by free radical, in this case, the radical sulphate. These radical may develop oxidative

attack against DNA in biological systems which may generate replication errors and

subsequent misleading protein synthesis. These advantages indicate that a purine-based


184

biosensors can be used as a useful tool for a rapid screening in the determination of TAC

in food matrices.

4. Conclusion

A purine-based biosensor for the evaluation of TAC in beverages was developed. The

adenine and guanine bases immobilized on the GCE surface were damage by the sulfate

radical generated by the persulfate /iron (II) system. The protective effect on the DNA

bases performed by the presence of five antioxidants was confirmed and allows to the

development of a methodology for the quantification of TAC in food samples. The purine-

based biosensor developed is disposable, and requires a very easy, rapid, reproducible

preparation and also the advantage to combine with portable equipment.

Acknowledgements


grant (SFRH/BD/29440/2006). The authors thank Frize for providing flavours samples.


185

References

[1] C.J. Burrows, J.G. Muller, Chem. Rev. 98 (1998) 1109.

[2] C.J. Burrows, R.J. Perez, J.G. Muller, S.E. Rokita, Pure Appl. Chem. 70 (1998) 275.

[3] J.G. Muller, P. Zheng, S.E. Rokita, C.J. Burrows, J. Am. Chem. Soc. 118 (1996) 2320.

[4] M.S. Swaraga, M. Adinarayana, Indian J. Biochem. Biophys. (40) 2003 27.

[5] M.S. Swaraga, L. Charitha, M. Adinarayana, J. Chem. Sci. (117) 2005 345.

[6] T.L. Luke, H. Mohan, V.M. Manoj, P. Manoj, J.P. Mittal, C.T. Aravindakumar, Res.

Chem. Intermed. 29 (2003) 379.

[7] G. Chu, S. Zhang, S. Yao, Z. Han, Z. Du, Z. Zhang, Sci. China, Ser. B 2002, 45 (2002)

398.

[8] X.-Y. Yu, Z.-C. Bao, J.R. Barker, J. Phys. Chem. A 108 (2004) 295.

[9] G.P. Anipsitakis, D.D. Dionysiou, Appl. Catal. B 54 (2004) 155.

[10] G.P. Anipsitakis, D.D. Dionysiou, M.A. Gonzalez, Environ. Sci. Technol. 40 (2006)

1000.

[11] E.N. Frankel, Antioxidants in food and biology facts and fiction. Publisher Oily Press.

Bridgwater, England, vol 20 2007.

[12] C. Sánchez-Moreno, Food Sci. Technol. Int. 8 (2002) 121.

[13] F.R. Ferreira, E.G. Silva, L.P.M. De Leo, E.J. Calvo, E.S. Bento, M.O.F. Goulart, F.C.

Abreu, Electrochim. Acta 56 (2010) 797.

[14] V.C. Sanz, M.L. Mena, A.G. Cortés, P.Y. Sedeño, J.M. Pingarrón, Anal. Chim. Acta

528 (2005) 1.

[15] L.D. Mello, L.T. Kubota, Talanta 72 (2007) 335.

[16] M.F. Barroso, N. de-los-Santos-Álvarez, M.J. Lobo-Castañón, A.J. Miranda-Ordieres,

C. Delerue-Matos. M.B.P.P. Oliveira, P. Tuñón-Blanco, Biosens. Bioelectron. 26

(2011) 2396.

[17] J.F. Liu, C. Roussel, G. Lagger, P. Tacchini, H.H. Girault, Anal. Chem. 77 (2005)

7687.

[18] L.D. Mello, S. Hernandez, G. Marrazza, M. Mascini, L.T. Kubota, Biosens.

Bioelectron. 21 (2006) 1374.

[19] P. Qian, S.Y. Ai, H.S. Yin, J.H. Li, Microchim. Acta 168 (2010) 347.

[20] M. Fojta, T. Kubicarova, E. Palecek, Biosens. Bioelectron. 15 (200) 107.

[21] V.C. Diculescu, A.M.C. Paquim, A.M.O. Brett, Sensors 5 (2010) 377.

[22] A. Ferancová, L. Heilerová, E. Korgová, S. Šilhár, I. Štepánek, J. Labuda, Eur. Food

Res. Technol. 219 (2004) 416.

[23] C. Liang, Z.-S., C.J. Bruell, Chemosphere 66 (2007) 106.

[24] C. Liang, I-L. Lee, J. Contam. Hydrol. 100 (2008) 91.


186

[25] C. Liang, C.J. Bruell, M.C. Marley, K.L. Sperry, Chemosphere 55 (2004) 1213.

[26] M.E. Rice, Z. Galus, R.N. Adams, J. Electroanal. Chem. 143 (1983) 89.

[27] G. Marrazza, I. Chianella, M. Mascini, Anal. Chim. Acta 387 (1999) 297.

[28] G. Chiti, G. Marrazza, M. Mascini, 2001. Anal. Chim. Acta 427 (2001) 155.

[29] M.F. Barroso, A. Silva, S. Ramos, M.T. Oliva-Teles, C. Delerue-Matos, M.G.F. Sales,

M.B.P.P. Oliveira, Food Chem. 116 (2009) 580.

[30] M.F. Barroso, S. Ramos, M.T. Oliva-Teles, C. Delerue-Matos, M.G.F. Sales, N.B.P.P.

Oliveira, Food Addit. Contam. Part B-Surveill. 2 (2009) 121.

[31] N. de-los-Santos-Álvarez, P. de-los-Santos-Álvarez, M.J. Lobo-Castañón, R. López,

A.J. Miranda-Ordieres, P. Tuñon-Blanco, Electrochem. Commun. 9 (2007) 1862.

[32] ANIRSF, 2010. Associação Nacional dos industriais de refrigerantes e sumos de

frutos, 2010. Informar nº29. from the website www.anirsf.pt (accessed 03.09.2010).

[33] M.L. Rodríguez-Méndez, C. Apetrei, J.A. de Saja, Electrochim. Acta 53 (2008) 5867.

[34] M. Fátima Barroso, J.P. Noronha, Cristina Delerue-Matos, M. B.P.P. Oliveira.

Flavoured waters: influence of ingredients in antioxidant capacity and terpenoid

profile by HS-SPME/GC-MS. J. Agric. Food Chem. 2011 (59) 5062.

[35] V. Thavasi, L.P. Leong, R.P.A. Bettens, J. phys.Chem. A 110 (2006) 4918.

[36] S. Fulda, Drug Discov. Today 15 (2010) 757.

187

Capítulo 5

Construção de biossensores de ADN

5.1.

DNA-based biosensor for the electrocatalytic determination of antioxidant capacity in beverages

M.F. Barroso, N. de-los-Santos-Álvarez, M.J. Lobo-Castañón, A.J. Miranda-Ordieres,

C. Delerue-Matos, M.B.P.P. Oliveira, P. Tuñón-Blanco

Biosensors and Bioelectronics, 2011, 26 (5), 2396 - 2401

5.2.

Electrocatalytic evaluation of DNA damage by superoxide radical for antioxidant capacity

M.F. Barroso, N. de-los-Santos-Álvarez, M.J. Lobo-Castañón, A.J. Miranda-Ordieres,

C. Delerue-Matos, M.B.P.P. Oliveira, P. Tuñón-Blanco

Journal of Electroanalytical Chemistry, 2011, em publicação

doi:10.1016/j.jelechem.2011.04.022

5.1. Biossensores de ADN – radical hidroxilo

189

DNA-based biosensor for the electrocatalytic determ ination of

antioxidant capacity in beverages

M.F. Barrosoa,b,c, N. de-los-Santos-Álvareza, M.J. Lobo-Castañóna, A.J. Miranda-

Ordieresa, C. Delerue-Matosc, M.B.P.P. Oliveirab, P. Tuñón-Blancoa aDepartamento de Química Física y Analítica, Universidad de Oviedo, Julián Clavería 8,

33006 Oviedo, Spain bRequimte, Serviço de Bromatologia, Faculdade de Farmácia, Universidade do Porto, R.

Aníbal Cunha n. 164, 4050-047 Porto, Portugal cREQUIMTE/Instituto Superior de Engenharia do Porto, Dr. Bernardino de Almeida 431,

4200-072 Porto, Portugal

Abstract

Reactive oxygen species (ROS) are produced as a consequence of normal aerobic

metabolism and are able to induce DNA oxidative damage. At the cellular level, the

evaluation of the protective effect of antioxidants can be achieved by examining the

integrity of the DNA nucleobases using electrochemical techniques. Herein, the use of an

adenine-rich oligonucleotide (dA21) adsorbed on carbon paste electrodes for the

assessment of the antioxidant capacity is proposed. The method was based on the partial

damage of a DNA layer adsorbed on the electrode surface by OH• − radicals generated by

Fenton reaction and the subsequent electrochemical oxidation of the intact adenine bases

to generate an oxidation product that was able to catalyze the oxidation of NADH. The

presence of antioxidant compounds scavenged hydroxyl radicals leaving more adenines

unoxidized, and thus, increasing the electrocatalytic current of NADH measured by

differential pulse voltammetry (DPV). Using ascorbic acid (AA) as a model antioxidant

species, the detection of as low as 50 nM of AA in aqueous solution was possible. The

protection efficiency was evaluated for several antioxidant compounds. The biosensor

was applied to the determination of the total antioxidant capacity (TAC) in beverages.

Keywords : NADH; DNA biosensor; Electrocatalytic oxidation;Total antioxidant capacity

(TAC); Ascorbic acid; Reactive oxygen species (ROS).


Biosensor and Bioelectronics 2011, 26 (5), 2396-2401


191

1. Introduction

Oxidative lesions in DNA are the primary risk factor for gene mutations, which plays a

key role in carcinogenesis and aging (Freidman and Heller, 2004). Reactive oxygen

species (ROS) are continuously generated in living cells, such as, in the inner

mitochondrial membrane, outer membrane, and in several metabolic pathways in

mammalian cells, for instance, in the microsomal electron transport. Hydroxyl radical, OH•

is one of the most powerful oxidant known in a biological setting, and upon formation, it

oxidizes indiscriminately and site-specifically any biomolecule (Laranjinha, 2009). In living

systems, most of the hydroxyl radicals are generated from the metal (M) ion-dependent

breakdown of hydrogen peroxide. In the presence of reduced transition metal ions, e.g.

ferrous or cupric ions, hydrogen peroxide is turned into OH• and OH− through a one-

electron redox reaction commonly called Fenton reaction (Eq. (1)) (Mello et al., 2006). The

Fenton chemistry is important because it is involved in oxidative damage in vivo leading to

changes in DNA that induce mutagenesis, and eventually, carcinogenesis:

H2O2 +Fe2+→ OH• + OH− +Fe3+ (1)

Most living organisms have developed complex endogenous and exogenous antioxidant

systems to counteract and prevent the deleterious effects of ROS. Antioxidants act as

reductants agents (free radical terminators), metal chelating and singlet oxygen

quenchers (Vertuani et al., 2004). Endogenous antioxidant systems include enzymes,

such as, superoxide dismutase, glutathione peroxidase, glutathione reductase,

glutathione-S-transferase and catalase (Huang et al., 2005). An additional protection can

be provided by exogenous antioxidant compounds, such as, vitamins (A, E, C, β-

carotene), phenolic compounds, minerals (Se, Zn) or proteins (transferrin, ceruloplasmin,

albumin). Foodstuffs constitute an excellent exogenous source of natural antioxidants. It is

well-known that vegetables, fruits, whole-grain and some beverages (tea, juice, wine) are

rich in many antioxidant and bioactive compounds. Recently, to answer to consumer’s

preferences, flavoured waters produced from mineral and spring waters were developed

and commercialized. In the first semester of 2009, 6.23 million litters of this kind of water

were consumed by Portuguese population (ANIRSF, 2009). This kind of water consists of

the addition of flavours, juices and sugar or sweeteners that provide water with singular

tastes and aromas appreciated by consumers. Considering that flavours/aromas are fruit

extracts, they contain natural antioxidants, transferring them to the bottled water. So,


192

drinking this type of water can increase the daily intake of natural exogenous antioxidants

or may contribute to the protective system against ROS.

Several methods have been reported to evaluate the total antioxidant capacity (TAC) in

biological and food samples, defined as the moles of a given free radical scavenged by a

sample solution disregarding the antioxidant present (Mello and Kubota, 2007). These

methods rely on the inhibition of the oxidation of a suitable substrate by the antioxidant

agent. After reaction, the extent of the oxidation is measured at a fixed time by UV–vis

spectrometry, chemiluminescence, fluorescence and after chromatographic separation

(Sanchez-Moreno, 2002).

The protective effect of antioxidants at a cellular level could only be achieved by

monitoring the DNA integrity. In recent years, several electrochemical DNA-based sensors

have been developed in order to assess the antioxidant capacity (Labuda et al., 2002,

2003; Mello and Kubota, 2007; Qian et al., 2010). UV irradiation (Liu et al., 2005) or most

commonly Fenton reaction were used for OH• generation. Recently, a ruthenium complex

was also used as an electrogenerated oxidant to cause the oxidation of DNA in the

presence of TiO2 nanoparticles (Liu et al., 2006).

Taking advantage of the two main lesions caused by ROS in DNA, two detection

estrategies have been proposed. On one hand, the use of redox active dsDNA

intercalators allowed evaluating the oxidative damage on dsDNA layers because of a

significant decrease in the current intensity of the intercalator upon strand scission

(Labuda et al., 2002, 2003; Liu et al., 2005). On the other hand, the intrinsic electroactivity

of DNA can be exploited. The oxidation of nucleobases on solid electrodes, mainly

guanine, and also adenine in a lesser extent, allowed the use of their decreasing oxidation

current after damage on carbon electrodes (Mello et al., 2006; Qian et al., 2010).

In this work, an electrocatalytic voltammetric method to assess TAC using DNA-

modified carbon paste electrodes (CPE) was developed. It has been reported that the

electrochemical oxidation of both adenine and guanine homopolynucleotides in neutral or

alkaline conditions led to the formation of a common oxidized product that catalyzed the

oxidation of NADH (de-los-Santos-Alvarez et al., 2007). Therefore, the oxidative lesions

generated after immersion of the DNA-CPE in the Fenton mixture were indirectly

quantified after the electrochemical oxidation of the adenines that remained unoxidized on

the electrode surface. The increase of this electrocatalytic current in the presence of

several antioxidant species was studied. The biosensor developed was applied to the

determination of TAC in several beverages and the results were compared with those

attained using other methodologies to obtain an overall picture of the antioxidant profile.


193


2.1. Chemicals

Deoxyadenylic acid oligonucleotide (dA21) was purchased as a desalted product from

Sigma-Genosys (London, UK). Concentrated saline sodium phosphate EDTA (20× SSPE;

0.2 mol l−1 sodium phosphate, 2 mol l−1 NaCl, 0.02 mol l−1 EDTA), tris–HCl pH 9.0,

phosphate buffer pH 9.0, iron (II) sulphate heptahydrate, hydrogen peroxide (30%, w/v),

gallic acid, resveratrol, nicotinamide adenine dinucleotide disodium salt, reduced form

(NADH), were acquired from Sigma–Aldrich (Spain). L(+)-Ascorbic acid was from Riedel-

de-Haën. Caffeic acid, and trolox (6-hydroxy-2,5,7,8- tetramethylchroman-2-carboxylic

acid, a water-soluble derivative of vitamin E) were from Fluka (Spain). Other chemicals

employed were of analytical grade.

Stock solutions of 1 g l−1 dA21 were stored at 4 ºC and diluted with 2× SSPE buffer

solution (prepared by dilution of 20× SSPE solution) prior to use. Fenton solution

(generation of hydroxyl radical) was prepared by mixing Fe2+:EDTA:H2O2 (µmol l−1:2 µmol

l−1:40 µmol l−1) in the molar ratio of 1:2:40. Mello et al., 2006 reported that when an excess

of hydrogen peroxide was added in the reaction a high DNA damage was obtained. EDTA

was added for solubility reasons. All solutions were prepared with water purified with a

Direct-Q (Millipore) system.


Cyclic voltammetry (CV) and differencial pulse voltammetry (DPV) were performed with

a µAutolab II controlled by GPES software, version 4.8 (EcoChemie, The Netherlands). A

conventional three-electrode cell was used, which included a home-made CPE (3 mm in

diameter) as a working electrode, a platinum wire counter electrode and a Ag|AgCl|KClsat

reference electrode to which all potentials were referred. The CPE was prepared by

mixing 1.8 g of paraffin oil as a pasting liquid with 5 g of spectroscopic grade graphite

powder (Ultracarbon, Dicoex, Spain). The unmodified carbon paste was introduced into

the well of a teflon electrode body provided by a stainless steel piston. The surface was

smoothed against a plain white paper while a slight manual pressure was applied to the

piston. Unless otherwise stated, after each experiment, the CP was discarded and a new

electrode surface was freshly prepared.


194


Unless otherwise mentioned, most experiments consisted of four steps: DNA

immobilization, damage of oligonucleotide by the immersion of DNA-CPE on the Fenton

mixture and study of the effect of the presence of antioxidants in the system;

electrooxidation of the remaining unoxidized adenines on the CPE, and detection in a

Ca2+ containing-NADH solution.

DNA immobilization was performed by dry adsorption placing a 5-µl droplet of dA21 (180

mg l−1) in 2× SSPE solution on the electrode surface and evaporating it to dryness under a

stream of warm air.

DNA damage was carried out by immersing the modified electrode in a freshly prepared

Fenton mixture in the absence or the presence of antioxidant in 2× SSPE buffer.

After a fixed period of reaction time, the DNA-CPE was washed with water and

immediately immersed in a 0.1 mol l−1 phosphate buffer (pH 9.0) to carry out the electro-

oxidation of the remaining unoxidized adenine bases. A hundred potential scans were

performed between −0.2 and +1.4 V at 500 mV s−1 to ensure a complete oxidation.

For detection, the DNA-CPE was placed in a NADH solution (0.5 mmol l−1 in 0.1 mol l−1

tris–HCl pH 9.0 containing 0.01 mol l−1 CaCl2). The electrocatalytic current of NADH was

obtained by sweeping the potential between −0.2V and +0.5 V at 50 mV s−1 when CV was

used as a detection technique. For DPV experiments, the potential was swept from −0.2

to +0.3 V. The step potential was 0.005 V and the modulation amplitude 0.05 V.

2.4. Samples and alternative methods description

Two lemon sparkling flavoured water samples corresponding to two different brands

were collected in a supermarket and stored in the dark at +4 ºC. Sonication was used to

eliminate gas from the sample. A lemon flavour used in the formulation of some water

brands, was also analysed. This flavour had no description about their chemical or aroma

composition, only knowing that they are present in some brands of flavoured water.

Label information from brand A indicated the presence of vitamin C, some

preservatives, such as, sodium benzoate, potassium sorbate and the acidifying regulator

citric acid. Label from brand B sample indicated the presence of green tea and citric acid.

In order to compare the results obtained with the DNA based sensor developed, the

total phenolic contents (TPC) of these samples was determined by a colorimetric assay

based on procedures previously described (Singleton and Rossi, 1965). Folin-Ciocalteu

reagent was used, and the reduced phenols produced a stable blue product at the end of

reaction and the results were given as milligram of gallic acid l−1.


195

Other two methods were used to obtain a complete profile of the antioxidant capacity.

The radical scavenging ability of these samples was tested by DPPH (1,1-diphenyl-2-

picrylhydrazyl) stable radical assay (Hatano et al., 1988) and reducing power method

(Oyaizu, 1986). DPPH values were expressed in mg trolox l−1 and reducing power in mg

of gallic acid l−1. A comprehensive study on the chemistry behind these methods has been

reviewed (Huang et al., 2005).


Previous studies reported by our group indicated that the electro-oxidation of different

adenine nucleosides and nucleotides, such as adenosine (de-los-Santos-Álvarez et al.,

2001), (dA)20 (delos- Santos-Álvarez et al., 2002), SAMe (de-los-Santos-Álvarez et al.,

2004), or FAD (de-los-Santos-Álvarez et al., 2005) on pyrolytic graphite electrodes (PGE)

or CPE (Alvarez-González et al., 2000) in phosphate buffer ranging from pH 7 to 12, led to

the formation of a diimine species, strongly adsorbed on the electrode surface. The

common oxidized compound exhibits a quasi-reversible redox process at low potentials

and is able to efficiently catalyze the oxidation of NADH reducing the overpotential by

more than 300 mV.

The oxidizability of purine bases in DNA depends, predominantly, on the secondary

structure of the polynucleotide. Accordingly, ssDNA leads to higher oxidation currents

than dsDNA because of their flexibility and better accessibility of nucleobases to the

electrode surface (de-los-Santos-Álvarez et al., 2002). Since polynucleotides strongly

adsorb on carbon electrodes, the oxidation of a layer of dA21 was carried out after physical

adsorption of the ssDNA on the electrode surface. The oxidation of adenine bases in DNA

takes place at about 1.2 V at CPE. In Fig. 1 (inset), the redox process associated to the

oxidation products arisen from dA21 after 100 potential scans at 500 mV s−1 from −0.2 to

+1.4 V on CPE is depicted. As anticipated, the redox process has a formal potential (Eo`)

of 0.035 V, which corresponds to the diimine species. After addition of NADH in the

presence of calcium ions, a great enhancement in the anodic current was observed at low

potentials associated to the mediated oxidation of NADH (Fig. 1, curve a). The onset of

the electrocatalytic wave appeared at potentials as low as 0.010 V with a plateau at 0.184

V. A further increase in the magnitude of the anodic current at more positive potentials

was adscribed to the direct uncatalyzed oxidation of NADH at CPE, as it is shown in curve

b, where the oxidation of NADH at a bare unmodified CPE (Ep = 0.70 V) is depicted. The

use of calcium ions has been reported to greatly improve the electrocatalytic current of

NADH (de-los-Santos-Alvarez et al., 2006; Mano and Kuhn, 1999; Toh et al., 2003). This

effect is not completely understood at the molecular level, although it is speculated that


196

divalent cations can effectively counterbalance the negative charge of the catalyst

favouring the approach of the negatively charged NADH to form a complex between

NADH, Ca2+ and the catalyst. In addition to this, calcium ions can interact directly with

DNA contributing to DNA stabilization and conformation (de-los-Santos-Alvarez et al.,

2006).

Fig. 1. CVs obtained at 50 mV s−1 in tris–HCl pH 9.0 in the presence of 0.5 mmol l−1 NADH + 0.01 mol l−1

CaCl2 with (solid line) a DNA-CPE and (dotted line) a bare CPE after electrochemical oxidation up to 1.4 V at

500 mV s−1. Inset: CV obtained with a DNA-CPE at 50 mV s−1 in tris–HCl pH 9.0 in the absence of NADH+

CaCl2 solution after identical electrochemical oxidation.

It is well-known that hydroxyl radicals generated in the close proximity to DNA, can

attack both the deoxyribose sugar moiety and the nucleobases resulting intermediate

radicals, which are precursors of DNA base damage such as base oxidation, sugar

fragmentation and DNA strand structural changes (Jaruga and Dizdaroglu, 1996). In order

to verify that OH• generated by a Fenton-type reaction are able to oxidize dA21 on the

electrode surface, the DNA-CPE was placed in the Fenton mixture for 120 s. After

transferring to a phosphate buffer solution (pH 9), no redox process at low potentials was

observed. An extended voltammetric scan up to 1.4 V did not show any oxidation peak at

1.2 V indicating that, at least, part of the adenine bases were effectively oxidized by the

generated radicals. This result suggested that Fenton-generated hydroxyl radicals

induced oxidative damage on adenine bases following a pathway somehow different from

the electrochemical oxidation because the adsorbed compound was not formed or the

yield was so low that cannot be detected by CV. At this point it is important to remark that

the NADH catalyst generated on the electrode surface after oxidation of adenine residues

E / VE / V

- 0.20 0.20 0.60 1.00-- 0.50

0.75

2.00

3.25

4.50

- 0.20 0.20 0.60 1.00-- 0.50

0.75

2.00

3.25

4.50

I/ µ

A

-0.20 0.15 0.50-0.04

0

0.40

E / V

I / µ

A

a) b)

E / VE / V

- 0.20 0.20 0.60 1.00-- 0.50

0.75

2.00

3.25

4.50

- 0.20 0.20 0.60 1.00-- 0.50

0.75

2.00

3.25

4.50

I/ µ

A

-0.20 0.15 0.50-0.04

0

0.40

E / V

I / µ

A

- 0.20 0.20 0.60 1.00-- 0.50

0.75

2.00

3.25

4.50

- 0.20 0.20 0.60 1.00-- 0.50

0.75

2.00

3.25

4.50

I/ µ

A

-0.20 0.15 0.50-0.04

0

0.40

E / V

I / µ

A

a) b)


197

is not the main oxidation product but only one of the several electrogenerated products

reported so far (Dryhurst and Elving, 1968; Goyal et al., 1991; Goyal and Sangal, 2002).

The damaged DNA-CPE was subjected to the above mentioned procedure to oxidize

the remaining undamaged adenine bases and immersed in a Ca2+-containing NADH

solution. An apparent electrocatalytic wave was observed with a plateau at about 0.18 V

(Fig. 2, curve a). This current was clearly smaller than that obtained when no Fenton

reaction was carried out (Fig. 2, curve b) indicating that only few adenine bases remained

unoxidized after exposure to Fenton mixture, confirming the ability of the generated OH• to

partially oxidized the DNA layer. When ascorbic acid was added to the Fenton mixture, an

electrocatalytic current higher than that obtained in its absence but smaller than when no

Fenton reaction was performed was observed (Fig. 2, curve c). This behaviour was in

good agreement with a scavenging effect of the antioxidant that prevented the DNA

damage to occur. As a consequence, the number of lesions diminished yielding a larger

number of adenine available for electrochemical oxidation. A positive correlation between

the partial oxidation of DNA and the concentration of antioxidant species in the tested

solution would allow the use of the electrocatalytic current of NADH to evaluate the TAC.

Fig. 2. CV obtained with a DNA-CPE at 50 mV s−1 in tris–HCl buffer pH 9.0 in 0.5 mmol l−1 NADH with 0.01

mol l−1 CaCl2 after: (a) immersion in Fenton solution and electrochemical oxidation (b) only electrochemical

oxidation; (c) immersion in Fenton solution with 10 µmol l−1 AA and electrochemical oxidation.

It is worth noting that a common oxidation product for the electrochemical oxidation of

both adenine and guanine bases was proposed (de-los-Santos-Alvarez et al., 2007).

Therefore, a layer of DNA containing guanine could be also used, in principle, for the

preparation of the biosensor layer. However, guanines are the primary nucleobase target

for hydroxyl radicals suggesting that the number of intact guanine bases for the

subsequent electrochemical oxidation would be smaller than in the case of adenine.


198

Besides, the yield of the catalyst generated from guanine derivatives is much lower than in

the case of adenine derivatives (de-los-Santos- Alvarez et al., 2007). Thus, the selection

of adenine-rich oligos for the preparation of biosensors seems to be advantageous.


Firstly, the influence of the number of potential scans to obtain a complete oxidation of

the dA21 was studied. It was observed that when increasing the number of potential scans

between −0.2 V and +1.4 V at 500 mV s−1, the amount of catalyst produced also

increased, and consequently, an enhancement in the electrocatalytic current was verified.

This increase in the peak current of the catalyst was observed until the 100th scan after

that, the amount of catalyst generated reached the highest value (surface coverage (Γ) of

1.1×10−11 mol cm−2) and remained constant, indicating the complete oxidation of dA21

adsorbed on CPE. Therefore, this number of potential scans was chosen for the next

optimization steps.

In order to evaluate the TAC on beverages, some parameters concerning the damaging

reaction (iron concentration, reaction time between hydroxyl radical and the target

molecule) at a fixed concentration of antioxidant compound has to be optimized in order to

achieve the maximum DNA effect, but without a complete damage.

The ratio between the electrocatalytic current obtained after exposing the DNA-CPE

electrode to the Fenton mixture in the presence of a fixed amount of antioxidant (ascorbic

acid as a model molecule) (Ia) and the electrocatalytic current obtained in the absence of

the antioxidant (Id) was selected as a criteria for optimization. Id is the minimum value of

the electrocatalytic current under each experimental condition because the absence of

antioxidants in the damaging solution precluded the protection of the nucleobases from

radical oxidation, leaving the lowest amount of adenine bases for further electrochemical

oxidation. Therefore, the maximum value for this ratio will be selected in each

optimization.

The level of DNA damage was evaluated as a function of the variation of the

concentration of Fe2+ keeping constant the molar ratio Fe2+:EDTA:H2O2 used (1:2:40)

(Mello et al., 2006). Fig. 3 shows the effect of Fe2+ concentration (from 1 to 100 µmol l−1)

on the amount of adenine lesions generated, indirectly evaluated using the electrocatalytic

current of the NADH. When increasing the Fe2+ concentration in the absence of

antioxidant compound, the electrocatalytic current decreased indicating an increased

damaging power attributable to the generation of a higher concentration of ROS.

However, when experiments were carried out in the presence of 10 µmol l−1 ascorbic acid,

the scavenging effect was apparent (much higher currents) especially at lower Fe2+


199

concentrations. From this results it is clear that the protective role of ascorbic acid strongly

depended on the concentration of radicals generated, and thus, at high OH•

concentrations, ascorbic acid could not longer prevent DNA damage (Ia very similar to Id).

The maximum value for the Ia/Id ratio was obtained at a Fe2+ concentration of µmol l−1,

which was chosen for further experiments.

Fig. 3. Influence of Fe2+ concentration (ratio Fe:EDTA:H2O2; 1:2:40) on the electrocatalytic current intensity of

0.5 mmol l−1 NADH with 0.01 mol l−1 CaCl2 in tris–HCl buffer at pH 9.0.

Considering that the reaction time between hydroxyl radicals and DNA depends on the

half-time of the generated free radicals, this parameter was optimized. The reaction time

was studied from 10 to 120 s. In the absence of antioxidant, a decrease in the

electrocatalytic current was observed when increasing the incubation time from 10 to 30 s.

At longer reaction times the electrocatalytic signal remained almost constant. On the

contrary, a continuous increase in the analytical signal when increasing the reaction time

in the presence of ascorbic acid was observed. The maximum value for the Ia/Id ratio was

obtained after 120 s, so this time was used in subsequent experiments.

3.2. Determination of antioxidant capacity

Foodstuff constitutes an excellent exogenous source of natural antioxidants to

counteract and prevent the deleterious effects of ROS. This protective effect is analytically

defined as TAC. In this work, ascorbic acid was used as a model antioxidant compound.

Ascorbic acid is a potent reductant agent that can reduce metal transition ions, thus, a

potential pro-oxidant role in vivo was suggested (Podmore et al., 1998). However, most

evidences point out to a predominant antioxidant role of ascorbic acid (Carr and Frei,

1999; Evans and Halliwell, 2001). Owing to the fact that DPV is a technique more

sensitive than CV, it was used for calibration purposes. The concentration of ascorbic acid

0.0

0.1

0.2

0.3

0 25 50 75 100

[Fe2+] / µmol l-1

I / µ

A

0.0

0.1

0.2

0.3

0 25 50 75 100

[Fe2+] / µmol l-1

I / µ

A


200

was varied from 0.05 to 1.00 µmol l−1. Fig. 4 shows the catalytic current obtained after

immersion of the DNA-CPE on Fenton mixtures containing increasing concentrations of

ascorbic acid.

Fig. 4. DPVs obtained after immersion of DNA-CPE in Fenton solution containing a standard solution of AA:

(a) 0.05, (b) 0.10, (c) 0.50, (d) 0.80 and (e) 1.00 µmol l−1. Inset: relationship between Ia and AA concentration.

As expected, the electrocatalytic current of NADH increased when the concentration of

ascorbic acid increased up to µmol l−1 because of the availability of a larger number of

undamaged adenines for electrochemical oxidation. In the inset, the I measured by DPV

was plotted against the concentration of AA. A linear range from 0.05 to 1.00 µmol l−1 of

AA was found (I (nA) = 9.34 [AA (µmol l−1)] + 5.72; r = 0.995, n = 5). A limit of detection of

50 nmol l−1 was estimated from the regression parameters (concentration at which the

analytical signal is equal to the y-intercept plus three times the standard deviation of the

regression). The RSD was 3.2% at 1 µmol l−1.

Zhang et al. (2008) reported the study of DNA damage induced by Fenton system on a

glassy carbon electrode (GCE) and its protection with the antioxidant ascorbic acid. These

authors verified that ascorbic acid promoted protective effect on the DNA in a narrow

concentration range (from 1.5 to 2.5 mmol l−1) while at lower concentrations, a pro-oxidant

role was observed. This behavior was not observed in our experiments carried out on the

electrode surface and not in solution as in that study. Other recent study also reported the

protective effects on the DNA by applying ascorbic acid as a scavenging antioxidant

(Nobushi and Uchikura, 2010). Enzyme-modified electrodes using ascorbate oxidase and

I / n

A

E / V

-0.05 0.10 0.25- -0.20

20

30

40

50

a

e

[AA] / µmol l-1

I / n

A

0

9

18

0.0 1.0 2.0 3.0

I / n

A

E / V

-0.05 0.10 0.25- -0.20

20

30

40

50

a

e

[AA] / µmol l-1

I / n

A

0

9

18

0.0 1.0 2.0 3.0[AA] / µmol l-1

I / n

A

0

9

18

0.0 1.0 2.0 3.0


201

peroxidase enzymes for the detection of AA showed linear ranges in the submM level

(Mello and Kubota, 2007), several orders of magnitude higher than the exhibited by this

DNA-based sensor, which emphasizes the good analytical performance.

Fig. 5 shows a comparison of the efficiency of different antioxidant compounds on

hydroxyl radical scavenging.

Fig. 5. Efficiency of hydroxyl radical scavenging of several antioxidant compounds: AA – ascorbic acid; GA –

gallic acid; CA – caffeic acid; RES - resveratrol.

The efficiency was expressed as the percentage of the electrocatalytic current according

to the following expression: % efficiency = Ia/Ib ×100, where Ia is the current intensity

measured after DNA damage in the presence of the antioxidant compound and Ib is the

electrocatalytic current measured when no damage was done (maximum expected value).

The compounds used were ascorbic acid, gallic acid, trolox, caffeic acid, and resveratrol.

Using an antioxidant concentration of 0.5 µmol l−1, it was observed that AA presented the

highest protective role among all compounds tested (58.6%). Hydroxyl radicals had the

ability to produce 87.6% of damage on the dA21 layer. The protective effect of antioxidants

ranged from 19.3 to 58.6%. The lowest values were found for gallic acid and trolox, 19.3

and 20%, respectively. Caffeic acid and resveratrol presented a similar protective role of

34.1 and 37.9%, respectively. Using chemiluminescence detection, gallic acid and trolox

showed higher protection than AA (Nobushi and Uchikura, 2010).

Antioxidant compound

0

Blank fenton AA Trolox GA CA RESBlank fenton AA Trolox GA CA RES

20

40

60

80

100

20

40

80

120

An

tio

xid

an

t e

ffic

ien

cy

(%

)


0


20

40

60

80

100

20

40

80

120

0


20

40

60

80

100

20

40

80

120

20

40

60

80

100

20

40

80

120

An

tio

xid

an

t e

ffic

ien

cy

(%

)


202

3.3. Application to TAC assessment in beverages

The methodology developed was used to quantify TAC on a lemon flavour and two

different brands of lemon flavoured waters samples. A lemon flavour was chosen because

this fruit is an important source of antioxidants such as vitamin C, phenolic compounds

and it is the most commercialised flavour in the world (Orak, 2009; Xu et al., 2010). For

the quantification of TAC in beverages, 5 ml of the flavor water or 20 µl of flavour were

diluted in 2× SSPE to a final volume of 10 ml. Then, the DNA-CPE was immersed in the

solution and a freshly prepared Fenton solution was added for 120 s. After this period of

time the DNA-CPE was washed and immersed in phosphate buffer pH 9 to carry out the

electro-oxidation of the remaining unoxidized adenine bases. The detection was carried

out in a Ca2+-containing-NADH solution. In Table 1, TAC values are expressed in ascorbic

acid content, both µmol l−1 and mg l−1. It was observed that all samples exhibited

antioxidant activity. As expected, the highest level was found in the flavour because is a

concentrated product. Brand B lemon flavoured water presented an order of magnitud

higher TAC value than brand A. In fact, water sample from brand B has lemon flavor and

also green tea. Green tea is a bioactive compound and contains numerous components

with antioxidant activity, such as polyphenols (catechins, epicatechin, epigallocatechin)

and vitamins (Cabrera et al., 2006; Neyestani et al., 2009) so, this may explain the high

TAC value.

Table 1. TAC of flavoured waters

Samples TAC (expressed in ascorbic acid)

(µmol l-1) (mg l-1)

Lemon flavour 480 ± 20 85 ± 4

Lemaon flavoured water

Brand A

Brand B

0.19 ± 0.08

2.8 ± 0.3

0.03 ± 0.01

0.50 ± 0.05

The reducing capacity of samples was measured by the total phenolic content assay

(TPC). This parameter is commonly assumed to be equivalent to the antioxidant capacity,

so a good correlation is expected (Huang et al., 2005). The TPC level of the flavour and

water from brand A and B was 380, 7.2 and 39.7 g of gallic acid l−1, respectively. As it was

expected, the highest TPC values were from lemon flavour. The water sample from brand


203

B exhibited a higher TPC value than water from brand A, in good agreement with TAC

results. Finally, the DPPH radical scavenging activity and the reducing power activity were

both evaluated. All samples presented DPPH scavenging activity and reducing power

activity. DPPH values were 38.9, 44.88 and 41.5 g trolox l−1 for lemon flavour and water

from brand A and B, respectively. Reducing power activity values for the flavour and water

samples from brand A and B were 10.7, 3.8 and 9.6 mg of gallic acid l−1, respectively.

These four different methodologies applied constituted a powerful tool to elucidate a full

profile of TAC in food samples.

4. Conclusions

The natural electrochemistry of adenine bases was exploited to develop a DNA-based

biosensor for the assessment of total antioxidant capacity in beverages. A layer of dA21

adsorbed on CPE was damaged by hydroxyl radicals generated in a Fenton-type reaction.

The remaining unoxidized adenine bases were electrochemically oxidized to give rise to

an adsorbed oxidation product that was able to catalyze the oxidation of NADH in the

presence of calcium ions. Several antioxidant compounds were tested as hydroxyl

radicals scavengers exhibiting efficiencies ranging from 19 to 59%. Ascorbic acid showed

the highest protective role, so the DNA-CPE biosensor was used for the detection of this

molecule. The biosensor developed was disposable and required a very easy, rapid and

reproducible preparation. In addition to this, the low detectability (50 nM) allowed its

advantageous used for TAC evaluation in foodstuffs as it was sucessfully shown in

several beverages.

Acknowledgements

N.S.A. thanks to MICINN for a Ramón y Cajal contract. M.F.B. is grateful to the

Fundação para a Ciência e a Tecnologia for a PhD grant (Grant Number

SFRH/BD/29440/2006). This work was cofinanced by Projects CTQ2008-02429 and

FICYT IB08-087 and the European Regional Development Fund.


204

References

Alvarez-González, M.I., Saidman, S.A., Lobo-Castañón, M.J., Miranda-Ordieres, A.J.,

Tuñón-Blanco, P., 2000. Anal. Chem. 72 (3), 520–527.

ANIRSF, 2009. Associação Nacional dos industriais de refrigerantes e sumos de frutos,

2009. Informar n 27. from the website www.anirs.pt (in 01/06/2010).

Cabrera, C., Artacho, R., Gimenez, R., 2006. J. Am. Coll. Nutr. 25 (2), 79–99.

Carr, A., Frei, B., 1999. Faseb J. 13 (9), 1007–1024.

de-los-Santos-Alvarez, N., de-los-Santos-Alvarez, P., Lobo-Castañon, M.J., Lopez,

R.,Miranda-Ordieres, A.J., Tuñon-Blanco, P., 2007. Electrochem. Commun. 9 (8),1862–

1866.

de-los-Santos-Álvarez, N., de-los-Santos-Álvarez, P., Lobo-Castañón, M.J., Miranda-

Ordieres, A.J., Tuñón-Blanco, P., 2005. Anal. Chem. 77 (13), 4286–4289.

de-los-Santos-Álvarez, N., Lobo-Castañón, M.J., Miranda-Ordieres, A.J., Tuñón-Blanco,

P., 2004. Electroanalysis 16 (11), 881–887.

de-los-Santos-Álvarez, N., Muñiz-Ortea, P., Montes-Pañeda, A., Lobo-Castañón,

M.J.,Miranda-Ordieres, A.J., Tuñón-Blanco, P., 2001. J. Electroanal. Chem. 502 (1–

2),109–117.

de-los-Santos-Alvarez, P., de-los-Santos-Alvarez, N., Lobo-Castanon, M.J., Miranda-

Ordieres, A.J., Tunon-Blanco, P., 2006. Biosens. Bioelectron. 21 (8), 1507–1512.

de-los-Santos-Álvarez, P., Lobo-Castañón, M.J., Miranda-Ordieres, A.J., Tuñón- Blanco,

P., 2002. Anal. Chem. 74 (14), 3342–3347.

Dryhurst, G., Elving, P.J., 1968. J. Electrochem. Soc. 115 (10), 1014-&.

Evans, P., Halliwell, B., 2001. Br. J. Nutr. 85, S67–S74.

Freidman, K.A., Heller, A., 2004. J. Am. Chem. Soc. 126 (8), 2368–2371.

Goyal, R.N., Kumar, A., Mittal, A., 1991. J. Chem. Soc. Perkin Trans. 2 (9), 1369–1375.

Goyal, R.N., Sangal, A., 2002. J. Electroanal. Chem. 521 (1–2), 72–80.

Hatano, T., Kagawa, H., Yashura, T., Okuda, T., 1988. Chem. Pharm. Bull. 36, 2090–

2097.

Huang, D.J., Ou, B.X., Prior, R.L., 2005. J. Agric. Food Chem. 53 (6), 1841–1856.

Jaruga, P., Dizdaroglu, M., 1996. Nucleic Acids Res. 24 (8), 1389–1394.

Labuda, J., Buckova, M., Heilerova, L., Caniova-Ziakova, A., Brandsteterova, E.,

Mattusch, J., Wennrich, R., 2002. Sensors 2 (1), 1–10.

Labuda, J., Buckova, M., Heilerova, L., Silhar, S., Stepanek, I., 2003. Anal. Bioanal.

Chem. 376 (2), 168–173.

Laranjinha, J., 2009. Oxidative stress: From the 1980s to Resent Update in Oxidative

Stress, Inflammation and Angiogenesis in the Metabolic Syndrome, 1st ed. Springer.


205

Liu, J.F., Roussel, C., Lagger, G., Tacchini, P., Girault, H.H., 2005. Anal. Chem. 77 (23),

7687–7694.

Liu, J.F., Su, B., Lagger, G., Tacchini, P., Girault, H.H., 2006. Anal. Chem. 78 (19), 6879–

6884.

Mano, N., Kuhn, A., 1999. Electrochem. Commun. 1 (10), 497–501.

Mello, L.D., Hernandez, S., Marrazza, G., Mascini, M., Kubota, L.T., 2006. Biosens.

Bioelectron. 21 (7), 1374–1382.

Mello, L.D., Kubota, L.T., 2007. Talanta 72 (2), 335–348.

Neyestani, T.R., Gharavi, A., Kalayi, A., 2009. Int. J. Food Sci. Nutr. 60, 79–88.

Nobushi, Y., Uchikura, K., 2010. Chem. Pharm. Bull. 58 (1), 117–120.

Orak, H.H., 2009. Int. J. Food Sci. Nutr. 60 (3), 266–266.

Oyaizu, M., 1986. Jpn. J. Nutr. 44, 307–315.

Podmore, I.D., Griffiths, H.R., Herbert, K.E., Mistry, N., Mistry, P., Lunec, J., 1998. Nature

392 (6676), 559–559.

Qian, P., Ai, S.Y., Yin, H.S., Li, J.H., 2010. Microchim. Acta 168 (3–4), 347–354.

Sanchez-Moreno, C., 2002. Food Sci. Technol. Int. 8 (3), 121–137.

Singleton, V.L., Rossi, J.A., 1965. Am. J. Enol. Viticult. 16, 144–158.

Toh, C.S., Bartlett, P.N., Mano, N., Aussenac, F., Kuhn, A., Dufourc, E.J., 2003. Phys.

Chem. Chem. Phys. 5 (3), 588–593.

Vertuani, S., Angusti, A., Manfredini, S., 2004. Curr. Pharm. Des. 10 (14), 1677–1694.

Xu, G., Liu, D., Chen, J., Ye, X., Ma, Y., Shi, J., 2010. Food Chem. 106, 545–551.

Zhang, J.J., Wang, B., Li, Y.F., Jia, W.L., Cui, H., Wang, H.S., 2008. Electroanalysis 20

(15), 1684–1689.

5.2. Biossensores de ADN – radical superóxido

207

Electrocatalytic evaluation of DNA damage by supero xide radical

for antioxidant capacity assessment

M.F. Barrosoa,b,c, N. de-los-Santos-Álvareza, M.J. Lobo-Castañóna, A.J. Miranda-

Ordieresa, C. Delerue-Matosc, M.B.P.P. Oliveirab, P. Tuñón-Blancoa aDepartamento de Química Física y Analítica, Universidad de Oviedo, Julián Clavería 8,

33006 Oviedo, Spain bRequimte, Serviço de Bromatologia, Faculdade de Farmácia, Universidade do Porto, R.

Aníbal Cunha n. 164, 4050-047 Porto, Portugal cREQUIMTE/Instituto Superior de Engenharia do Porto, Dr. Bernardino de Almeida 431,

4200-072 Porto, Portugal

Abstract

The integrity of DNA purine bases was herein used to evaluate the antioxidant capacity.

Unlike other DNA-based antioxidant sensors reported so far, the damaging agent chosen

was the O2• - radical enzymatically generated by the xanthine/xanthine oxidase system. An

adenine-rich oligonucleotide was adsorbed on carbon paste electrodes and subjected to

radical damage in the presence/absence of several antioxidant compounds. As a result,

partial damage on DNA was observed. A minor product of the radical oxidation was

identified by cyclic voltammetry as a diimine adenine derivative also formed during the

electrochemical oxidation of adenine/guanine bases. The protective efficiency of several

antioxidant compounds was evaluated after electrochemical oxidation of the remaining

unoxidized adenine bases, by measuring the electrocatalytic current of NADH mediated

by the adsorbed catalyst species generated. A comparison between O2• −

and OH• radicals

as a source of DNA lesions and the scavenging efficiency of various antioxidant

compounds against both of them is discussed. Finally, the antioxidant capacity of

beverages was evaluated and compared with the results obtained with an optical method.

Keywords : NADH electrocatalysis; DNA damage; Antioxidant capacity; Ascorbic acid;

Reactive oxygen species; Superoxide radical

Available online at pubs.acs.org

Journal of Electroanalytical Chemistry In press doi:10.1016/j.jelechem.2011.04.022


209

1. Introduction

Deleterious oxidative processes mediated by free radicals, such as ROS, are involved in

aging and in a vast array of diseases, including cancer, inflammation, cardiovascular and

neurodegenerative diseases [1]. Therefore, overproduction of ROS can be dangerous for

cells [2]. The superoxide anion radical (O2•-) is the primary component of ROS and the

most abundant radical in biological systems, resulting from the single electron reduction of

oxygen [3]. This cytotoxic species is enzymatically produced by xanthine oxidase (XOD),

a metalloenzyme that catalyzes the oxidation of hypoxanthine and xanthine to uric acid

generating O2•- during the respiratory burst of phagocytic cells (Eq. (1)) [1]. Under normal

physiological conditions, the highly reactive superoxide radical undergoes dismutation by

non-catalytic and enzymatic reactions, thus the physiological concentration is rather low

[4].

(1) O 2H aciduric OOHXanthine -2

XOD22

•+ ++ →++

The biological effects of highly reactive ROS are controlled in vivo by a variety of

nonenzymatic and enzymatic antioxidant mechanisms. Superoxide radical is easily

attacked by other active biomolecules and scavenged by enzymes and antioxidants [5].

The major scavenger of this radical in vivo is the superoxide dismutase enzyme (SOD)

that catalyzes its disproportionation to H2O2. Subsequently, catalase detoxifies H2O2, and

glutathione peroxidase detoxifies H2O2 and converts lipid hydroperoxides into non-toxic

alcohols [1]. An additional protection can be provided by exogenous antioxidant

compounds, such as low molecular weight molecules, vitamin (A, E, C, β-carotene), and

minerals (Se, Zn). This exogenous protective effect can be achieved by the intake of

foodstuff and beverages, like vegetables, fruit, whole-grain, tea, juice and wine.

Photometric, chemiluminescent, fluorimetric, chromatographic and electrochemical

methods have been proposed for in vitro quantification of the antioxidant capacity (AOC)

in biological and food samples [6]. Electrochemical biosensors use two main sources of

ROS: OH• and O2

•-. The former can be generated photocatalytically [7] or by Fenton

reaction in DNA-based antioxidant sensors [8, 9], and the latter is mostly enzymatically [2,

10, 11] but also chemically [3, 12, 13] or electrochemically [14] formed for the

determination of both superoxide radical and AOC. Sensors based on O2•- commonly rely

on the immobilization of cytochrome c, which is reduced by superoxide radical, on gold [2-

4], carbon [15] or screen printed-Au-electrode [16] surfaces, where it is reoxidized. To

enhance the electrical contact between cytochrome c and the electrode and to increase

II. Investigação e Desenvolvimento

210

the surface coverage of this compound, several immobilization strategies have been

proposed mostly based on SAMs of thiols of different length [2-4, 15] and hemin modified

electrodes [17]. However, these sensors present the interference of H2O2, uric acid and

also some electrical communication problems between the protein and the electrode.

Another strategy is the immobilization of SOD by physical adsorption or through SAM

[18-20] on the electrode surface in order to follow the disproportionation of superoxide

radical by measuring the O2 and H2O2 formed. These biosensors presented interferences

derived from the high potential at which the generated H2O2 is detected, limiting the

practical application of the sensor.

Nonetheless, the protective effect of antioxidants at a cellular level could only be

achieved by monitoring the DNA integrity. To the best of our knowledge, all

electrochemical DNA-based antioxidant sensors developed so far used the hydroxyl

radical as a damaging agent, which caused strand scission or oxidative lesions in

nucleobases (guanine or adenine). Superoxide radical has not been used for this purpose

probably because the mechanism of O2•- damage on DNA is not completely understood. It

is believed that its participation is limited to promote the production of OH• radicals [21-

23]. However, it is important to develop assays to study other radical sources active in

cells and tissues and the way antioxidants eliminate it preventing its deleterious effect.

Antioxidants can react by different mechanisms depending on the free radical/oxidant

source or by multiple pathways against a single oxidant [24]. This observation implies that

there is no a universal assay for the detection of all antioxidants. To obtain a full profile of

antioxidant capacity against various ROS, the development of methods specific for each

ROS is needed.

In this work, the effectiveness of superoxide radical generated by the enzymatic reaction

between XOD and xanthine to induce damage on a DNA-based sensor is studied. Based

on previous work on electrochemical oxidation of adenine and guanine derivatives [25-28],

a minor product of the radical oxidation was identified. The oxidative lesions were

indirectly quantified after electrochemical oxidation of the remaining intact adenine bases

to generate a well-known catalyst species that mediates the oxidation of NADH. CV was

used to measure the electrocatalytic current after the subsequent immersion of the

damaged DNA-modified CPE in a NADH-Ca2+ containing solution. A dependence of the

electrocatalytic current on the concentration of antioxidant in the damaging solution was

found, which allowed the development of a voltammetric method for the determination of

AOC in flavored waters.


211

2. Material and methods

2.1. Chemicals

Deoxyadenylic acid oligonucleotide (dA21) purchased as a desalted product, xanthine

oxidase (XOD) and xanthine were from Sigma-Aldrich (Madrid, Spain). Concentrated

saline sodium phosphate EDTA (20 × SSPE; 0.2 M sodium phosphate, 2 M NaCl, 0.02 M

EDTA), tris-HCl pH 9.0, phosphate buffer pH 9.0, gallic acid (GA), resveratrol (RES),

nicotinamide adenine dinucleotide disodium salt, reduced form (NADH), were also

acquired from Sigma-Aldrich. L(+)-ascorbic acid (AA) was from Riedel-de-Haën

(Germany). Caffeic acid (CA), and trolox (6-hydroxy-2,5,7,8-tetramethylchroman-2-

carboxylic acid, a water-soluble derivative of vitamin E) were from Fluka (Madrid, Spain).

Other chemicals employed were of analytical grade. Stock solutions of 1 g L-1 dA21 were

stored at 4 ºC and diluted with 2 × SSPE buffer solution (prepared by dilution of 20 ×

SSPE solution) prior to use. All solutions were prepared with water purified with a Direct-Q

(Millipore) system.


Cyclic voltammetry was performed with a µAutolab II controlled by GPES software,

version 4.8 (EcoChemie, The Netherlands). A conventional three electrode cell was used,

which includes a home-made CPE (3 mm in diameter) as a working electrode, a platinum

wire counter electrode and a Ag|AgCl|KClsat reference electrode to which all potentials are

referred. The CPE was prepared by mixing 1.8 g of paraffin oil as pasting liquid with 5 g of

spectroscopic grade graphite powder (Ultracarbon, Dicoex, Spain). The unmodified

carbon paste was introduced into the well of a Teflon electrode body provided by a

stainless steel piston. The surface was smoothed against a plain white paper while a

slight manual pressure was applied to the piston. Unless otherwise stated, after each

experiment, the CP was discarded and a new electrode surface was freshly prepared.

For temperature-controlled experiments a circulating thermostat HAAKE DC1 (Thermo

Electron GmbH, Germany) was used.


Unless otherwise mentioned, experiments were structured in four steps: DNA layer

preparation, damage of oligonucleotide by immersion of the DNA-CPE on a XOD/xanthine

solution in the absence/presence of several antioxidants; electro-oxidation of the


212

remaining unoxidized adenines on the CPE, and detection in a Ca2+-containing NADH

solution.

DNA immobilization was performed by dry adsorption placing a 5-µL droplet of dA21 (180

mg L-1) in 2 × SSPE solution on the electrode surface and evaporating it to dryness under

a stream of warm air.

DNA damage was carried out by immersing the dA21-CPE in a freshly prepared

XOD/xanthine mixture (superoxide radical generating solution) in the absence or the

presence of antioxidant under controlled temperature (27.0 ± 0.1 ºC). The superoxide

radical was generated by the addition of XOD (0.1 U mL-1) to oxygen-saturated 2 × SSPE

solutions at pH 7.4 containing xanthine (4.4 × 10-5 M).

After a fixed reaction time, the DNA-CPE was washed with water and immediately

immersed in a 0.1 M phosphate buffer (pH 9.0) to carry out the electro-oxidation of the

remaining unoxidized adenine bases. 100 potential scans were performed between -0.2

and +1.4 V at 500 mV s-1 to ensure a complete oxidation [29].

For detection, the DNA-CPE was placed in a NADH solution (5.0 × 10-4 M in 0.1 M tris-

HCl pH 9.0) containing 0.01 M CaCl2. The electrocatalytic current of NADH was obtained

by CV sweeping the potential between -0.2 V and 0.5 V at 50 mV s-1.

2.4. Samples and description of alternative methods

Two lemon sparkling flavored water samples corresponding to two different brands were

purchased in a supermarket and stored in the dark at +4 ºC. Sonication was used to

eliminate gas from the sample. Label information from brand A indicates the presence of

vitamin C, some preservatives, such as sodium benzoate, potassium sorbate and the

acidifying regulator citric acid. Label from brand B sample indicates the presence of green

tea and citric acid. A lemon flavor used in the formulation of some water brands was also

analyzed. This flavor had no description about its chemical or aroma composition.

For the measurement of AOC in beverages, 200 µl of the flavored water or 10 µl of

flavor were diluted in 2 x SSPE to a final volume of 500 µl. Then, the DNA-CPE was

immersed in the solution and a freshly prepared superoxide radical was added for 10 min.

After this period of time the DNA-CPE was washed and immersed in a phosphate buffer to

carry out the electro-oxidation of the remaining unoxidized adenine bases. The detection

was carried out in a Ca2+-containing NADH solution.

A colorimetric assay, based on a procedure previously reported [30], was used to

elucidate the antioxidant profile of the simples, expressed as the total phenolic content

(TPC). Folin-Ciocalteu reagent was used, and the reduced phenols produced a stable

blue product at 760 nm. The results were expressed as mg of GA L-1.


213


Oxygen and its reactive species are very important in oxidative metabolism. ROS induce

oxidative damage producing a variety of modifications at DNA level including base and

sugar lesions, strand breaks, DNA-protein cross-linking and base-free sites [31]. In order

to verify that O2•- generated by a xanthine/XOD reaction is able to oxidize dA21 on the

electrode surface, the DNA-CPE was placed in a freshly prepared xanthine/XOD solution

in 2 × SSPE buffer (pH 7.4) for 15 min. After transferring to a phosphate buffer solution

(pH 9), a small quasi-reversible redox process was observed at low potentials, Eº’ = 0.041

V (Fig. 1a). The amount of compound generated (surface coverage, Γ) was estimated to

be 1.2 × 10-11 mol cm-2 from the integrated change under the anodic wave. An extended

voltammetric scan up to 1.4 V did not show any oxidation peak at 1.2 V (oxidation

potential of adenine bases in DNA) but a gradual increase in the magnitude of the redox

process at low potentials was observed after several potential scans (Fig. 1b).

Fig. 1 . CVs obtained at 50 mV s-1 in tris–HCl pH 9.0 after: (a) immersion of dA21-CPE in a superoxide radical

generating solution ([XOD] = 0.3 U mL-1, [xanthine] = 4.4 x 10-5 M) for 15 min and (b) subsequent

electrochemical oxidation of the undamaged adenine bases adsorbed on the dA21-CPE.

Since the only oxidizable species was the oligonucleotide adsorbed on CPE, this

behavior is in good agreement with a partial oxidation of adenine bases by the superoxide

radical. Therefore, the intact adenines can be further electrochemically oxidized at the

electrode surface at 1.2 V leading to the product responsible for the redox couple at low

potentials. It is worth mentioning that the oxidation current of the remaining adenines was

not observed because it was overlapped by the rising background current at the high

potential at which it takes place. Additionally, the Eº’ of redox processes originated from

radical attack (Fig. 1a) and electrochemically (Fig. 1b) were virtually identical suggesting


214

that the same compound is formed in both types of oxidation. According to previous

studies on the electro-oxidation of adenine derivatives on carbon electrodes in phosphate

buffer [28, 32, 33], the compound responsible for redox process at +0.041 V is a diimine

species strongly adsorbed on the electrode surface. This compound was also identified

after oxidation of guanine derivatives [25-27]. Therefore, it can be concluded the existence

of a common lesion on DNA generated by O2•- generated by the xanthine/XOD system

and by electrochemical oxidation. However, from Fig. 1 it is apparent that the adsorbed

diimine species was a minor product of the radical oxidation and the yield was much lower

than in the electrochemical oxidation. This result indicated that the product profile and

compound distribution differed, thus, both oxidations are mechanistically different.The fact

that the O2•- attack on DNA led to the generation of this adsorbed compound is remarkable

because the oxidation of adenine bases through OH• radicals generated by Fenton-type

reaction was recently demonstrated not to occur via the formation of the diimine species,

at least, at levels detectable by CV [29]. Given that both radical attacks led to different

products, the reported primary OH• radical promoter role of O2•- remains uncertain.

The adsorbed species was shown to efficiently catalyze the oxidation of NADH reducing

the overpotential by more than 300 mV at pyrolytic graphite electrodes [27, 28, 32]. This

ability can be exploited, in principle, to detect the DNA damage. However, the low yield

achieved by radical oxidation did not allow observing an electrocatalytic current sufficiently

high to be used as an analytical signal. In fact, no significant current was observed at

potentials close to the redox process when NADH was added to the solution after DNA

damage by superoxide radicals (data not shown). To solve this problem an indirect

method was tested. The unoxidized adenine bases were electrochemically oxidized to

generate a larger amount of diimine (catalyst) species. Therefore, the higher the damage,

the lower the intact adenine available for further electrocatalytic measurement in the

presence of NADH. To electro-oxidize the remaining adenine adsorbed on the CPE,

several cyclic scans were carried out up to 1.4 V. After this step, the damaged DNA-CPE

was immersed in a NADH-Ca2+ solution. The use of calcium ions was reported to greatly

improve the electrocatalytic current of NADH [34, 35]. An apparent electrocatalytic wave

was observed at a potential as low as 0.011 V with a plateau at about 0.14 V (Fig. 2,

curve a). Given that the oxidation peak of the uncatalyzed oxidation of NADH at a bare

unmodified CPE is 0.70 V [29], a decreased of more than 550 mV is achieved. A low

potential is advantageous for analytical purposes because of the diminution of potential

oxidizable interferent compounds present in real food samples. Under these conditions,

this was the lowest electrocatalytic current possible because it arose from the maximum

damage. In the presence of antioxidant compounds a diminution in the damage was

expected along with an increase in the electrocatalytic current. When an antioxidant, AA


215

(10 µM), was added to the superoxide radical generating solution, a high augment of the

electrocatalytic current was observed (Fig. 2, curve b).

Fig. 2 . CVs obtained with a dA21-CPE at 50 mV s-1 in tris–HCl pH 9.0 containing 0.5 mM NADH + 0.01 M

CaCl2 after; immersion in O2•- generating solution ([XOD] = 0.1 U mL-1, [xanthine] = 4.4 x 10-5 M) for 10 min (a)

in the absence of antioxidant (b) in the presence of 10 µM of AA; and further complete electrochemical

oxidation in both cases.

This anticipated behavior was related to the ability of antioxidant compounds to

scavenge or inactivate the ROS and prevent the damage on DNA. As a consequence, the

number of lesions diminished, yielding a larger number of adenine available for

electrochemical oxidation. A positive correlation between the partial oxidation of DNA by

O2•- and the concentration of antioxidant species in the tested solution would allow the use

of the electrocatalytic current of NADH to evaluate the AOC on flavored waters.

3.1. Selection of the experimental conditions for the damaging reaction

In order to determine AOC on beverages, some parameters concerning the damaging

reaction (xanthine and XOD concentration, reaction time between superoxide radical and

the target molecule) at a fixed concentration of antioxidant compound were varied in order

to achieve the maximum effect on DNA without a complete damage. For this reason, for

each experiment the ratio between the electrocatalytic current obtained after exposing the

DNA-CPE to the superoxide radical in the presence of a fixed amount of AA as antioxidant

(Ia) and the electrocatalytic current obtained in the absence of AA (Id, minimum value

expected) was estimated. The highest value for this ratio was always selected for further

experiments. The level of DNA damage was evaluated as a function of the amount of

radical formed through the variation of the concentration of XOD and xanthine. XOD

concentration was studied between 0.05 and 0.20 U mL-1. Fig. 3A shows the influence of

XOD concentration on the electrocatalytic current of NADH.


216

Fig. 3. Influence on the electrocatalytic current of 0.5 mM NADH + 0.01 M CaCl2 in tris–HCl at pH 9.0 of (A): XOD concentration: (B) xanthine concentration: and (C) time reaction: (o) without AA and (·) with 10 µM of AA to the O2

·−.

When increasing the XOD concentration in the absence of antioxidant compound (open

circles), the electrocatalytic current decreased until a XOD concentration of 0.10 U mL-1.

At higher concentrations the current remained constant. This behavior suggested that an

increase in the enzyme concentration implied a larger number of lesions attributed to the


217

superoxide radical attack. The damage on the dA21 layer exhibited a maximum (minimum

electrocatalytic current) at a XOD concentration of 0.10 U mL-1. When the same

experiments were carried out in the presence of ascorbic acid (10 µM), the protective

effect on the DNA was apparent because the electrocatalytic currents were virtually

constant up to 0.10 U mL-1, within the experimental error (Fig. 3A, filled circles). Only at

higher concentrations of enzyme the analytical signal diminished suggesting that the

ascorbic acid concentration is not sufficient to compensate the increase in the amount of

superoxide radicals generated. In addition to this, it is worth noting that, with the addition

of this powerful antioxidant, consistently higher currents were measured at all XOD

concentrations (Fig. 3A). The highest value for the Ia/Id ratio was obtained at a XOD

concentration of 0.10 U mL-1, which was chosen for the next optimization steps.

Xanthine concentration was varied from 4.40 × 10-6 to 4.40 × 10-4 M. The influence of

this parameter within the range assayed was very limited. A slight decrease in the

electrocatalytic current was observed when increasing the xanthine concentration in the

absence of AA, which is not significant within the experimental error (Fig. 3B open circles).

In the presence of antioxidant species, all currents were clearly higher and a small but

relevant increase was apparent at 4.40 × 10-5 M (filled circles). At higher concentrations a

further diminution was observed. This behavior is in good agreement with a scavenging

activity of AA. The highest Ia/Id ratio was observed at a xanthine concentration of 4.40 ×

10-5 M, and this value was used for the next experiments.

The reaction time between the superoxide radical and dA21 layer depends on the

halftime on the generated ROS, so, this parameter is an important feature to select. The

reaction time between the free radical, the superoxide, and the DNA adsorbed on the CPE

was studied between 5 and 30 min. Increasing the incubation time, the electrocatalytic

current of NADH decreased during the first 10-15 min. (Fig. 3C, open circles). At longer

reaction time the current remained constant. With the introduction of ascorbic acid (10 µM)

on the reaction system, the electrocatalytic current measured was higher than in its

absence at all reaction times assayed in good agreement with the radical scavenging role

(Fig. 3C, filled circles). Nevertheless, a decrease is observed up to 15 min although the

remaining electrocatalytic current is significantly higher than in the absence of AA (Fig.

3C). This behavior indicated that, even at very long times, AA was able to partially protect

the integrity of DNA from O2•-

radical attack. The highest value of Ia/Id ratio was found

when an incubation time of 10 min. was used, so, this value was selected for further

studies.


218

3.2. Determination of AOC

In this work, the antioxidant ascorbic acid was used as a model for the study of the

behavior of antioxidants on the protection of DNA against O2•- radicals generated by

XOD/xanthine reaction. The feasibility of measuring the antioxidant concentration was

investigated varying the concentration of AA from 10 to 100 µM. A linear range wasfound

for the entire range (I (nA) = (0.85 ± 0.07) [AA (µM)] + (16 ± 5); r = 0.990 n = 5). The limit

of detection was estimated using the regression parameters obtaining a value of 10 µM.

The reproducibility expressed as RSD was 4.2% at 50 µM. Fig. 4 shows CVs obtained in a

Ca2+-containing NADH solution after immersing the DNA-CPE in a superoxide radical

solution with increasing concentrations of AA. The catalytic current of NADH increased up

to 100 µM due to the availability of a larger number of undamaged adenines for

electrochemical oxidation. At concentrations above this value the electrocatalytic current

remained constant indicating the saturation of the ability of AA to counterbalance the

radical attack.

Fig. 4. CVs at 50 mV s-1 obtained in tris–HCl at pH 9 containing 0.5 mM NADH + 0.01 M CaCl2 after

immersion of DNA-CPE in O2·−)generating solution ([XOD] = 0.1 U mL-1, [xanthine] = 4.4 x10-5 M) containing a

standard solution of AA: (a) 10, (b) 30, (c) 50, (d) 80 and (e) 100 µM for 10 min. Inset panel: relationship

between Ia and AA concentration.

Other authors have also used AA in order to study its protective effect on the DNA

(adsorbed at an electrode surface) against free radicals. However, all these reports only

described the scavenging role of AA towards hydroxyl radicals [29, 36, 37].


219

As mentioned before, no DNA sensors for antioxidant assessment using other ROS,

such as superoxide radical, have been reported so far. Two reports described the use of

AA as a standard antioxidant against superoxide radical, but the biolayer on the electrode

was formed by cytochrome c or SOD [2]; or the electrochemically generated radical was

detected directly on a glassy carbon disk electrode [14]. From our previous work on

antioxidant activity against OH• on DNA-CPE, it can be concluded that AA seemed to be

less efficient as a scavenger of superoxide radical than hydroxyl radicals. In fact, the

minimum AA concentration able to show a protective action is more than two order of

magnitude lower in the case of OH• [29].

In order to compare the efficiency of radical scavenging, several antioxidants (AA, GA,

trolox, CA, and RES) were tested at a concentration of 10 µM under the same

experimental conditions and the results are shown in Fig. 5.

Fig. 5. O2

•- scavenging efficiency of several antioxidant compounds: AA – ascorbic acid, GA – gallic acid, CA –

caffeic acid, RES – resveratrol. Values are expressed as percentage of the electrocatalytic current obtained

with an intact (not damaged) dA21 layer that remained after exposure to a damaging solution containing 10 µM

of antioxidant species.

The efficiency was expressed as the percentage of the electrocatalytic current according

to the following expression: % efficiency = Ia/Ib x 100, where Ia is the current intensity

measured after DNA damage in the presence of the antioxidant compound, and Ib is the

electrocatalytic current measured when no damage was done (maximum expected value).

It was found that the superoxide radical generated 85% of damage on the dA21 layer, that

is, in the absence of a scavenging molecule. The protective effect of antioxidants ranged

from 33% to 63%. The lowest values were found for trolox and CA, 33% respectively.

0

20

40

60

80

100

120

Blank superoxideradical

AA Trolox GA CA Res


An

tio

xid

an

t e

ffic

ien

cy

(%

)

0

20

40

60

80

100

120

Blank superoxideradical

AA Trolox GA CA Res


An

tio

xid

an

t e

ffic

ien

cy

(%

)


220

RES presented the highest protective effect (62.5%). AA and GA presented a protective

role of 53.8% and 53.2 % respectively.

At this point, it is interesting to note that superoxide radicals caused a similar degree of

damage on DNA adsorbed on CPE to hydroxyl radical [29]. Although efficiency values

were similar or much higher than those obtained with OH•, the antioxidant concentration

employed is much higher, which is in good agreement with the lower scavenging activity

above found. This result was not unexpected because it is commonly accepted that not all

antioxidants behaves equally against different radicals [24]. It was clear that the efficiency

order differed from that obtained against. Whereas AA and CA exhibited similar protecting

roles against both radicals (about 55% and 33% respectively), the effectiveness of RES

dramatically increased from 38% for OH• to 62.5% for O2•-. In any case both compounds

were the most effective antioxidants assayed. Similarly, GA was much more active for O2•-

than for OH• shifting from 19.3% (the worst one) to 53.2% virtually identical to AA within

the experimental error.

Once the analytical features of the electrocatalytic voltammetric method were

characterized in aqueous solution, it was applied to the determination of AOC in real

samples. A lemon flavor and two different brands of lemon flavored water samples were

chosen because this citrus fruit is used and commercialized all over the world and is rich

in antioxidants such as vitamin C and phenolic compounds. As it is shown in Table 1, all

samples presented antioxidant capacity. Lemon flavor exhibited the highest level of AOC

expressed in mg L-1 of AA. This finding was expected because this flavor is extracted from

the fruit along with essential oils, and has several substances at high concentration in its

composition. Lemon water from brands A and B had a similar AOC value. However, the

composition of both samples was different because brand B had green tea in addition to

vitamin C.

Table 1. AOC values of flavored waters obtained using the electrochemical and optical methods

Samples DNA-CPE (mg l-1 AA) TPC (mg l-1 GA)

Lemon flavor 124 ± 13 380

Lemon flavored water

Brand A

Brand B

30.2 ± 0.7

31.0 ± 7.5

7.2

39.7


221

Among the methods used for antioxidant capacity assessment, the Folin-Ciocalteu

method for the quantification of the phenolic content is widely used because its

robustness, simplicity and cost-effectiveness [24]. In general, phenolic compounds

content correlates with antioxidant activity and seems to have an important role in

stabilizing lipid oxidation. Therefore, the TPC of these samples was evaluated and

expressed in mg L-1 of GA. As expected, the highest TPC value was found in the lemon

flavor (Table 1). The water sample from brand B exhibited a significantly higher TPC value

than water from brand A. This difference on the TPC values was attributed to the

presence of green tea in brand B. Green tea contains polyphenols (catechins, epicatechin,

epigallocatechin) in addition to vitamins [38, 39]. Some phenols react with TPC reagents

although they may not necessarily be efficient radical scavengers [40]. The presence of

this type of phenolic compounds might explain the discrepancy between values obtained

for both voltammetric and TPC methods.

4. Conclusions

A DNA-CPE antioxidant biosensor for the assessment of AOC in beverages was

developed. For the first time in this type of devices, the effectiveness against damage of

superoxide radical on DNA was evaluated testing different antioxidant compounds.

Although the damage in terms of adenine oxidative lesions was similar to that found using

hydroxyl radicals, the scavenging activity of the antioxidant tested was lower because a

much higher concentration was needed to obtained similar efficiencies. The order of

protective efficacy was also different and as follows, RES> AA> GA> trolox ~ CA. A minor

product of the radical oxidation was identified by CV as a diimine compound that did not

appear when the oxidant source was the OH• radical. This result suggested that the

mechanism of O2•- attack on DNA is more complex that the reported promotion / source of

OH• radicals. In spite of the lower efficiency of AA as O2•- scavenger, the indirect

electrocatalytic method described allowed the quantification of ascorbic acid from 10 µM

and AOC determination in flavored waters and extracts.

Acknowledgements

N.S.A thanks to MICINN for a Ramón y Cajal contract. M. Fátima Barroso is grateful to

the Fundação para a Ciência e a Tecnologia for a PhD grant (Grant Number

SFRH/BD/29440/2006). This work was co-financed by Projects CTQ2008-02429 and

FICYT IB08-087 and the European Regional Development Fund.


222

References

[1] J. Laranjinha, Oxidative Stress: From the 1980s to Resent Update in OxidativeStress,

Inflammation and Angiogenesis in the Metabolic Syndrome, Springer, 2009.

[2] S. Ignatov, D. Shishniashvili, B. Ge, F.W. Scheller, F. Lisdat, Biosens. Bioelectron. 17

(2002) 191–199.

[3] B. Ge, F. Lisdat, Anal. Chim. Acta 454 (2002) 53–64.

[4] K.V. Gobi, F. Mizutani, J. Electroanal. Chem. 484 (2000) 172–181.

[5] L.D. Mello, L.T. Kubota, Talanta 72 (2007) 335–348.

[6] C. Sanchez-Moreno, Food Sci. Technol. Int. 8 (2002) 121–137.

[7] J.F. Liu, C. Roussel, G. Lagger, P. Tacchini, H.H. Girault, Anal. Chem. 77 (2005)

7687–7694.

[8] L.D. Mello, S. Hernandez, G. Marrazza, M. Mascini, L.T. Kubota, Biosens.

Bioelectron. 21 (2006) 1374–1382.

[9] P. Qian, S.Y. Ai, H.S. Yin, J.H. Li, Microchim. Acta 168 (2010) 347–354.

[10] M. Beissenhirtz, F. Scheller, F. Lisdat, Electroanalysis 15 (2003) 1425–1435.

[11] A.V. Krylov, H. Adamzig, A.D. Walter, B. Lochel, E. Kurth, O. Pulz, J. Szeponik, F.

515 Wegerich, F. Lisdat, Sensor Actuat. B-Chem. 119 (2006) 118–126.

[12] K. Hyland, C. Auclair, Biochem. Biophys. Res. Commun. 102 (1981) 531–537.

[13] V. Lvovich, A. Scheeline, Anal. Chem. 69 (1997) 454–462.

[14] C. Le Bourvellec, D. Hauchard, A. Darchen, J.L. Burgot, M.L. Abasq, Talanta 75

(2008) 1098–1103.

[15] M.K. Beissenhirtz, F.W. Scheller, F. Lisdat, Anal. Chem. 76 (2004) 4665– 4671.

[16] M. Cortina-Puig, X. Munoz-Berbel, R. Rouillon, C. Calas-Blanchard, J.L. Marty,

Bioelectrochemistry 76 (2009) 76–80.

[17] J. Chen, U. Wollenberger, F. Lisdat, B.X. Ge, F.W. Scheller, Sensor Actuat. B-Chem.

70 (2000) 115–120.

[18] K. Endo, T. Miyasaka, S. Mochizuki, S. Aoyagi, N. Himi, H. Asahara, K. Tsujioka, K.

Sakai, Sensor Actuat. B-Chem. 83 (2002) 30–34.

[19] E. Emregul, Anal. Bioanal. Chem. 383 (2005) 947–954.

[20] L. Campanella, A. Bonanni, E. Finotti, M. Tomassetti, Biosens. Bioelectron. 19 (2004)

641–651.

[21] K. Keyer, J.A. Imlay, Proc. Natl. Acad. Sci. USA 93 (1996) 13635–13640.

[22] K. Keyer, A.S. Gort, J.A. Imlay, J. Bacteriol. 177 (1995) 6782–6790.

[23] J. Cadet, T. Douki, J.L. Ravanat, Acc. Chem. Res. 41 (2008) 1075–1083.

[24] R.L. Prior, X.L. Wu, K. Schaich, J. Agric. Food Chem. 53 (2005) 4290–4302.


223

[25] N. de-los-Santos-Alvarez, P. de-los-Santos-Alvarez, M.J. Lobo-Castanon, R. Lopez,

A.J. Miranda-Ordieres, P. Tunon-Blanco, Electrochem. Commun. 9 (2007) 1862–

1866.

[26] E. González-Fernández, N. de-los-Santos-Alvarez, M.J. Lobo-Castañón, A.J.

Miranda-Ordieres, P. Tuñón-Blanco, Electroanalysis 20 (2008) 833–839.

[27] N. de-los-Santos-Alvarez, M.J. Lobo-Castañón, A.J. Miranda-Ordieres, P. Tuñón-

Blanco, Electrochim. Acta 53 (2007) 829–836.

[28] N. de-los-Santos-Álvarez, P. Muñiz-Ortea, A. Montes-Pañeda, M.J. Lobo- Castañón,

A.J. Miranda-Ordieres, P. Tuñón-Blanco, J. Electroanal. Chem. 502 543 (2001) 109–

117.

[29] M.F. Barroso, N. de-los-Santos-Alvarez, M.J. Lobo-Castañón, A.J. Miranda- Ordieres,

C. Delerue-Matos, M.B.P.P. Oliveira, P. Tuñón-Blanco, Biosens. Bioelectron. 26

(2011) 2396–2401.

[30] V.L. Singleton, J.A. Rossi, Am. J. Enol. Viticult. 16 (1965) 144–158.

[31] M. Dizdaroglu, P. Jaruga, M. Birincioglu, H. Rodriguez, Free Radical Bio. Medicine 32

(2002) 1102–1115.

[32] P. de-los-Santos-Álvarez, M.J. Lobo-Castañón, A.J. Miranda-Ordieres, P. Tuñón-

Blanco, Anal. Chem. 74 (2002) 3342–3347.

[33] N. de-los-Santos-Álvarez, M.J. Lobo-Castañón, A.J. Miranda-Ordieres, P. Tuñón-

Blanco, Anal. Chim. Acta 504 (2004) 271–277.

[34] P. de-los-Santos-Alvarez, N. de-los-Santos-Alvarez, M.J. Lobo-Castañón, A.J.

Miranda-Ordieres, P. Tuñón-Blanco, Biosens. Bioelectron. 21 (2006) 1507–1512.

[35] N. Mano, A. Kuhn, Electrochem. Commun. 1 (1999) 497–501.

[36] J.J. Zhang, B.Wang, Y.F. Li,W.L. Jia, H. Cui, H.S.Wang, Electroanalysis 20 (2008)

1684–1689.

[37] Y. Nobushi, K. Uchikura, Chem. Pharm. Bull. 58 (2010) 117–120.

[38] C. Cabrera, R. Artacho, R. Gimenez, J. Am. Coll. Nutr. 25 (2006) 79–99.

[39] T.R. Neyestani, A. Gharavi, A. Kalayi, Int. J. Food Sci. Nutr. 60 (2009) 79–88.

[40] D.J. Huang, B.X. Ou, R.L. Prior, J. Agric. Food Chem. 53 (2005) 1841–1856.

225

CONSIDERAÇÕES FINAIS

Considerações finais

227

Considerações gerais

Com muita frequência são lançados no mercado novos produtos alimentares e bebidas

com o objetivo de proporcionar ao consumidor uma alimentação diversificada e saudável.

Geralmente, no momento de distribuição, os consumidores não têm informações relativas

à completa composição destes novos produtos e aos seus benefícios para a saúde.

Ainda recentemente, assistimos ao aparecimento no mercado de uma grande diversidade

de águas com sabores. Quase todas as marcas de águas engarrafadas, prepararam as

respetivas águas de sabores os refrigerantes.

Este projeto de doutoramento centrou-se essencialmente no estudo de águas com

sabores designadamente (i) na avaliação do teor de macrominerais, microminerais, e

elementos vestigiais, (ii) na análise dos aromas usados nas formulações, identificando a

presença de alguns compostos recorrendo à cromatografia gasosa com deteção MS; (iii)

na determinação da capacidade antioxidante por métodos convencionais; e (iv) no

desenvolvimento de metodologias alternativas com recurso a biossensores

eletroquímicos para a avaliação da capacidade antioxidante total. Este último

procedimento foi o mais moroso mas ao mesmo tempo o mais inovador por envolver

pesquisa, desenvolvimento, otimização e validação das metodologias.

Inicialmente fez-se uma prospeção de mercado de modo a saber quantas marcas

comercializavam estas bebidas e quais os sabores disponíveis. Foram encontradas 10

marcas comerciais diferentes e 12 sabores diferentes. Ao longo do trabalho foram

surgindo novos sabores enquanto, outros aromas iam sendo retirados do mercado.

Considerando que os rótulos existentes nas garrafas destas águas com sabores não

contêm informações sobre o seu conteúdo mineralógico, fez-se uma análise química às

águas tendo-se analisado 18 minerais (4 macrominerais, 3 microminerais e 11 elementos

vestigiais). O critério de escolha destes minerais teve ainda em consideração o fato de os

minerais poderem ser interferentes (a nível do sinal elétrico) na análise eletroquímica da

capacidade antioxidante das amostras. Os resultados comprovam que os teores destes

minerais encontram-se dentro do limite estipulado por lei variando contudo de marca para

marca, o que está relacionado com a litologia e geoquímica do local de captação da

água. De um modo geral verificou-se que as águas minerais com gás natural (água

mineral natural gasosa) têm um teor mais elevado em minerais do que as águas de

nascente ou as águas minerais com gás adicionado (água mineral natural gaseificada).

Na avaliação da capacidade antioxidante das águas com sabores usaram-se

metodologias óticas convencionais (UV-vis). O perfil antioxidante foi obtido através da


228

determinação do teor fenólico e de flavonóides total, poder redutor e atividade anti-

radicalar. A tabela 1 apresenta os valores mínimos e máximos referentes ao perfil

antioxidante.

Tabela 1 . Quadro resumo do perfil antioxidante das águas com sabores e aromas por métodos

convencionais.

amostras Teor fenólico total

(mg GA L-1)

Poder redutor

(mg GA L-1)

DPPH

(mg trolox L-1)

Aromas 8,53-380,20 10,72-11,80 38,90-213,53

Agua com 1 aroma 0,29-17,62 0,28-6,10 8,05-54,21

Água com 2 aromas 1,51-39,70 0,14-13,78 13,49-48,66

Agua com vitamina C 147,0-284,0 3,77-154,04 44,78-268,89

GA – gallic acid

Como se pode verificar cada método conduz ao seu resultado pois embora todos eles

meçam a capacidade antioxidante, cada um reporta-se a uma família de compostos. Por

exemplo, o teor fenólico total e o poder redutor medem a capacidade antioxidante

equivalente ao ácido gálico, e a atividade antiradicalar (radical DPPH) usando o

antioxidante trolox. Genericamente, os aromas apresentam um elevado teor fenólico,

poder redutor e atividade antiradicalar. Relativamente às águas com sabores, de um

modo geral os valores mais elevados obtiveram-se com amostras de águas contendo 2

sabores mas em que o segundo aroma é um composto bioativo (chá verde, ginseng,

Ginkgo biloba, menta). Algumas amostras contendo vitamina C também apresentam

composição com altos valores de capacidade antioxidante. Em nenhuma das amostras

foram detetados flavonóides.

Considerando que o laboratório onde decorreram os trabalhos tem uma larga

experiência no domínio da eletroquímica, neste trabalho pretendeu-se também

desenvolver métodos alternativos para a avaliação da capacidade antioxidante das águas

com sabores, tendo-se construído um biossensor e usado técnicas voltamétricas. De

acordo com o que se encontra publicado, os radicais livres provocam danos oxidativos no

ADN, mas, como mecanismo de defesa, a célula usa os antioxidantes que conseguem

desativar os radicais livres e evitar as reações de propagação destes e assim proteger as

macromoléculas. De acordo com este princípio, a construção dos biossensores baseou-

se na imobilização de bases púricas ou de cadeia simples de ADN na superfície do

elétrodo, na danificação deste biossensor com radicais livres e na proteção do material

imobilizado no elétrodo com antioxidantes. O sinal eletroquímico obtido na voltametria de

onda quadrada foi usado para avaliar a eficácia dos biossensores.


229

Na tabela 2 apresenta-se um quadro resumo da capacidade antioxidante obtida

(valores mínimos e máximos) com os biossensores de adenina e guanina, e de cadeia

simples de ADN usando-se como danificador o radical hidroxilo, superóxido e sulfato e

como antioxidante padrão o ácido ascórbico (AA), o ácido gálico (AG), o ácido cafeico

(AC), o ácido cumárico (ACu) e o resveratrol (RES).

Tabela 2. Quadro resumo da utilização de biossensores para a determinação da capacidade

antioxidante de aromas e águas com sabores.

Amostras Biossensor Radical Antioxidante

Aromas Águas

AA (mg L-1) 100,8-185,6 2,9-40,3

GA (mg L-1) 9,2-30,1 0,1-1,4

CA (mg L-1) 39,2-66,7 1,5-4,3

Acu (mg L-1) 32,7-59,0 1,2-3,9

Hidroxilo

RES (mg L-1) 7,8-28,4 0,4-1,7

AA (mg L-1) 74,8-220 0,7-15,9

GA (mg L-1) 4,4-32,7 0,4-3,2

CA (mg L-1) 2,3-41,5 0,5-3,2

Acu (mg L-1) 32,2-68,4 0,6-4,7

Superóxido

RES (mg L-1) 15,3-29,0 0,3-2,1

AA (mg L-1) 101,8-281,9 3,8-16,8

GA (mg L-1) 34,7-55,5 0,2-2,8

CA (mg L-1) 60,2-92,3 0,5-5,1

Acu (mg L-1) 61,2-83,6 1,3-4,5

Guanina

Sulfato

RES (mg L-1) 13,7-34,5 0,4-1,7

AA (mg L-1) 211,6-571,9 3,3-37,5

GA (mg L-1) 14,6-32,5 0,1-2,0

CA (mg L-1) 19,9-31,9 0,9-2,3

Acu (mg L-1) 12,9-35,4 0,8-5,3

Hidroxilo

RES (mg L-1) 10.7-23,6 0,1-1,7

AA (mg L-1) 126,0-202,7 0,3-19,3

GA (mg L-1) 37,4-55,2 0,3-3,4

CA (mg L-1) 13,1-25,9 0,7-1,7

Acu (mg L-1) 6,4-35,3 0,6-2,5

Superóxido

RES (mg L-1) 12,1-26,9 0,7-1,9

ÁA (mg L-1) 24,6-239 0,8-5,6

GA (mg L-1) 7,2-50,6 0,5-2,1

CA (mg L-1) 24,6-235,4 1,3-3,6

Acu (mg L-1) 51,6-77,8 1,4-3,5

Adenina

Sulfato

RES (mg L-1) 12,5-37,2 0,2-1,5

Hidroxilo AA (mg L-1) 85 0,5 dA21

Superóxido AA (mg L-1) 124 31,0


230

Estes valores, não são comparáveis com os obtidos pelos métodos convencionais já

que a unidade de medida é diferente, mas a metodologia desenvolvida, à semelhança

dos métodos convencionais permite, comparar a capacidade antioxidante relativa nas

diferentes amostras. Como se pode verificar, as águas com sabores apresentam alguma

capacidade antioxidante sendo que os maiores teores foram encontrados nos aromas

utilizados nas formulações.

Assim sendo, traçou-se o perfil antioxidante das águas com sabores, e constatou-se

que em princípio terão algum efeito positivo na saúde.

A matriz chá encontra-se em estudo, estando-se a efetuar o mesmo procedimento que

foi realizado às águas.

Este trabalho permitiu um avanço significativo na área da eletroquímica,

concretamente na construção de biossensores, no grupo onde foi desenvolvido. Como

trabalho futuro seria interessante otimizar a utilização dos biossensores, nomeadamente

reduzir o tempo de construção recorrendo-se para isso a elétrodos descartáveis (screen-

printed) e pré-fabricados com o material biológico imobilizado e assim usar estes

biossensores para análises de rotina, da capacidade antioxidantes de alimentos e

bebidas.

Documents

Capacidade antioxidante de bebidas aromatizadas: … · contrebalancé par des mécanismes de défense antioxydante. ... Ensuite, une évaluation des paramètres de la capacité antioxydante