Mangostan y Diabetes

7/31/2019 Mangostan y Diabetes

1/14

Journal ofAGRICULTURAL and

FOOD CHEMISTRYALVIN ENG KIAT LOO AND DEJIAN HUANG*

Department of Chemistry,National University of Singapore,

3 Science Drive 3,Singapore 117543, Republic of Singapore

Assay-guided

Fractionation Study ofr-AmylaseInhibitors from Garcinia mangostanaPericarp

Ensayo guiado

Estudio del Fraccionamiento de los Inhibidores de la r-Amilasa en la Garcinia del Pericarpio del Mangostn

R-Amylase inhibitor (R-AI) activity ofGarcinia mangostana, commonly known asmangosteen, pericarp

extracts was studied by assay guidedfractionations from lipophilic to hydrophilicusing combined solvent extraction and

Amberlite XAD2 adsorptionchromatography. Neither the lipophilic,

xanthone containing fraction, nor the highlypolar fraction, which has no affinity on

Amberlite XAD2, showedany R-AI. The fraction that shows very highinhibitory activity contains primarilypolyphenols and canbe adsorbed on Amberlite XAD2. The IC50of 5.4 g/mL of this fraction is comparable tothat of acarbose, a prescribed R-AI used inthe control of type II diabetes, at 5.2 g/mL.Total phenoliccontent (TPC) of each fraction was

measured and the TPC has no correlationwith the R-AI activity.The lipophilic fraction contains mainlyxanthones as revealed by HPLC-MSanalysis. Colorimetricanalysis coupled with UVvis and IRspectroscopic analysis demonstrated thatthe fractions withhigh R-AI activity are primarily oligomericproanthocyanidins (OPCs) with little gallate

R-amilasa inhibidor (R-AI) de la actividad de la Garcinia,comnmente conocido como el pericarpio del mangostn

Los extractos fueron estudiados por ensayos guiados defraccionamiento desde lipfilo al hidrfilo y uso combinadode extraccin con disolventes y Amberlite XAD2cromatografa de adsorcin. Ni el xantona lipfilo, quecontiene la fraccin, ni la fraccin altamente polar, que notiene afinidad sobre Amberlite XAD2, mostr

cualquier. R-AI La fraccin que muestra la actividadinhibidora de muy alta contiene principalmente polifenolesy puedeser adsorbidos sobre Amberlite XAD2. El IC50 de 5,4 ug /ml de esta fraccin es comparable a la de acarbosa, unaprescripcin R-IA utilizado en el control de la diabetes tipoII, en 5,2 mg / mL. fenoles totalescontenido (TPC) de cada fraccin se midi y el TPC notiene correlacin con la actividad de I-AI.La fraccin lipoflica contiene principalmente xantonassegn lo revelado por HPLC-MS anlisis. colorimtricoanlisis junto con el anlisis espectroscpico UV-Vis e IR

demostr que las fracciones conR-IA alta actividad son principalmente proantocianidinasoligomricas) con fraccin de galato de poco. hayTambin existen pruebas para demostrar que el R-IA porestos OPC no es puramente por formacin de complejosprotena inespecfica.Tanto el cido tnico y OPC G. mangostana precipitarBSA igual de bien pero los OPC G. mangostanason 56 veces ms eficaz en el Estudio de la inhibicin deFraccionamiento R-amylase.Assay guiada de la r-amilasa


2/14

moiety. Thereis also evidence to show that the R-AI bythese OPCs is not purely by nonspecificprotein complexation.Both tannic acid and G. mangostana OPCsprecipitate BSA equally well but G.mangostana OPCs

are 56 times more effective in inhibiting R-amylase.Assay-guided Fractionation Studyof r-AmylaseInhibitors from Garcinia mangostanaPericarp

Inhibidores de la Garcinia pericarpio

INTRODUCTIONIt has been suggested that insulin resistance is amanifestationof the bodys inability to effectively metabolize a highglycaemicload (GL) diet (1). A high GL diet causes excessive

stress onthe_ cells to secrete insulin at levels higher than theywereoriginally designed for, leading to hyperinsulinemiaand,subsequently, insulin resistance. Introducing R-AIinto the dietcan reduce the rate of postprandial glycaemiaincrease and thusrelieve stress on_ cells. Therefore, it is not surprisingthatacarbose, an R-amylase and R-glucosidase inhibitor,

had beenfound to be effective in treating and possiblypreventingnoninsulin-dependent diabetes (2). Howeveracarbose is aprescribed drug and cannot be added to foods orused as asupplement. Plants are a rich source of variousantinutrients (3),including R-AI, which are not prescribed drugs.Therefore theyare ideal candidates for novel R-AI to be used in

functionalfoods.Garcinia mangostana, commonly known asmangosteen, isa tropical fruit belonging to the Guttiferae family (4). Itis widelycultivated in Thailand, Malaysia, and Indonesia.Worldwideproduction ofG. mangostana is about 150000 tonsper annum

Se ha sugerido que la resistencia a la insulina esuna manifestacinde la incapacidad del cuerpo para metabolizarcon eficacia un alto ndice glucmicocarga (GL) en la dieta (1). Una dieta alta en GL

causa tensin excesiva enel? las clulas para secretar insulina en losniveles ms altos de lo que erandiseado originalmente para, dando lugar a lahiperinsulinemia y,Posteriormente, resistencia a la insulina.Presentacin de R-IA en la dietapuede reducir la tasa de aumento de la glucemiapostprandial y por lo tantoaliviar la tensin sobre? clulas. Por lo tanto, noes sorprendente queacarbosa, un inhibidor de R-amilasa y R-

glucosidasa, haba sidoencontrado para ser eficaz en el tratamiento yprevencin posiblementela diabetes no insulino-dependiente (2). Sinembargo, la acarbosa es unprescrito de drogas y no se puede aadir a losalimentos o se utiliza como uncomplementarlos. Las plantas son una fuenterica de antinutrientes diversos (3),incluyendo R-IA, lo cual no se les prescribenfrmacos. Por lo tantoson candidatos ideales para la nueva R-AI para

ser utilizados en funcionalalimentos.Garcinia, comnmente conocido como elmangostn, es una fruta tropical que pertenece ala familia Guttiferae (4). Es ampliamentecultivado en Tailandia, Malasia e Indonesia.Mundial la produccin de G. mangostana es dealrededor de 150.000 toneladas anuales (5). Elpericarpio de la fruta es de aproximadamentedos tercios de la totalidad la fruta por peso, de


3/14

(5). The pericarp of the fruit is about two thirds of thewhole fruit by weight, bright red in colour, and usuallynot eaten. Ithad been used for treating diarrhoea, wounds, andskin infectionin traditional medicine (6). The pericarp is rich inpolyphenols

and contains 715% tannins (3). The polyphenols inthe pericarp ofG. mangostana and various otherfruits could have evolved as a defensive mechanismagainst herbivores. The high tannincontent not only makes the pericarp unpalatable butalso inhibitsvarious digestive enzymes (7). Thus it is common forinedibleplant parts, such as the seeds, to contain high levelsofpolyphenols in the form of tannins to prevent it frombeing eaten

(8). Therefore, the G. mangostana pericarp may alsocontainantinutrients such as R-AI as a defense mechanism.There had been no known studies on the R-AIactivity of phenolic compounds in G. mangostanapericarp. Utilization ofthe pericarp polyphenols is economically feasible asthe pericarpis a waste material. In addition there is also anincreased interest in low GL foods (9). Therefore, thepurpose of this study is to determine if the R-AIactivity of polyphenolic compounds from

G. mangostana pericarp is comparable to otherpolyphenols and known R-AI using in vitro assays.The chemical properties ofthe polyphenolic compounds would also bedetermined so as to understand the structure andcharacteristics of these compounds.

color rojo brillante, y no suele comer. Lohaba sido utilizado para tratar la diarrea, lasheridas, infeccin de la piel y elen la medicina tradicional (6). El pericarpio esrico en polifenoles y contiene taninos 7-15% (3).Los polifenoles en el pericarpio de G.mangostana y varias otras frutas podran haber

evolucionado como mecanismo de defensacontra los herbvoros. El alto en taninoscontenido no slo hace que el sabordesagradable del pericarpio, pero tambin inhibela diversas enzimas digestivas (7). As, es comnpara no comestible partes de las plantas, talescomo las semillas, que contiene altos niveles depolifenoles en forma de taninos que le impidenser comido(8). Por lo tanto, la G. mangostana pericarpiotambin puede contenerantinutrientes como el R-IA como un mecanismo

de defensa.No ha habido estudios conocidos sobre laactividad de I-AI los compuestos fenlicos en G.mangostana pericarpio. La utilizacin de lospolifenoles del pericarpio es econmicamenteviable, como el pericarpioes un material de desecho. Adems, existetambin un inters crecienteen los alimentos de baja GL (9). Por lo tanto, elpropsito de este estudio esdeterminar si la actividad de I-AI de compuestospolifenlicos de

G. mangostana pericarpio es comparable a otrospolifenoles y conocida R-IA utilizando ensayos invitro. Las propiedades qumicas de los loscompuestos polifenlicos tambin sedeterminarn de tal modopara entender la estructura y caractersticas deestos compuestos.

* To Whom correspondence should be addressed: e-mail: [email protected], fax: 65-6775-7895

MATERIALS AND METHODS MATERIALS Y MTODOS

Instruments. FT-IR spectrum was acquired with Perkin-ElmerSpectrum One (Boston, MA). UVvis spectroscopy andabsorbancereadings were carried out on a BioTek Synergy HTmultidetectionmicroplate reader (Winooski, VT) or on Mini 1240 Shimadzu(Kyoto,Japan) UVvis spectrophotometer. LC/MS chromatograms, andspectrumswere acquired using Finnigan/MAT LCQ ion trap massspectrometer(San Jose, CA) equipped with TSP 4000 HPLC system, which


4/14

includes UV6000LP PDA detector, P4000 quaternary pump,andAS3000 autosampler. The heated capillary and spray voltage

were maintained at 250 C and 4.5 kV, respectively. Nitrogenis operated at80 psi for sheath gas flow rate and 20 psi for auxiliary gasflow rate.The full scan mass spectra from m/z502000 were acquiredboth in

positive and negative ion mode with a scan speed of one scanpersecond. Chromatographic separations were accomplished ona Shim-Pack VP-ODS column (250 mm 4.6 mm i.d.) (Shimadzu,Kyoto,Japan) with the following elution condition: 1.0 mL/min; roomtemperature (25 C); solvent A, 0.1% formic acid in water;solvent B,0.1% formic acid in methanol, starting from 30 to 60% B in 10min,from 60 to 100% B in 40 min, from 100 to 60% B in 5 min,and from60 to 30% B in 5 min, and lastly, reconditioning of the columnfor 20

min. Absorbance was recorded from 320nm.Materials. Ripe G. mangostana from Thailand werepurchased fromlocal fruit stores and stored frozen in a -20 C freezer forfurther use.All solvents used were of analytical reagent grade unlessotherwisespecified. Folin Ciocalteu reagent and KIO3 were purchasedfrom Merck(Darmstadt, Germany). Amberlite XAD-2, 4-dimethylamino-cinnamaldehyde,methyl gallate, starch from potatoes, amylase from hogpancreaswith 46.3 U/mg, (+)- Catechin, bovine serum albumin, andtannic acidwere purchased from Sigma chemical company (St Louis,

MO). Wepurchased 4-hydroxybenzhydrazide 98% from Alfa Aesar(Ward Hill,MA). Gallic acid was purchased from Acros Organics (Geel,Belgium).Acarbose was obtained by dissolving a Glucobay tablet fromBayerscontaining 50 mg acarbose in phosphate buffer (pH 6.9, 50mM,containing 6.7 mM saline), at the desired concentration. Afterstandingfor 15 min, the mixture was then centrifuged to remove theinsolublefiller.


5/14

Extraction ofr-AI.

The extraction and fractionation steps aresummarized in scheme 1. The G. mangostana werethawed,and the pericarps were removed manually.The extraction method used was adapted from


6/14

methods used in the extraction of tannins in cocoa(10).The first extraction was carried out by blending fruithulls (250 g) with water (295 mL) in a Warringblender. After which, 5 mL acetic acid and 700 mLacetone was added to the slurry. Extraction wascontinued by stirring on an orbital shaker for 3 h. The

slurry was then filtered using a suction pump andthe residue was collected. Residues collectedfrom two separate extractions were pooledtogether and further extracted with 1 L ofacetone: water: acetic acid (70:29.5:0.5). Theslurry was again stirred using an orbital shakerfor 3 h. The filtration and extractionsteps were repeated twice with less solvent(500 mL). All the fltrate was pooled andacetone was removed by evaporation at40 under reduced pressure using a rotaryevaporator. The extract was then centrifuged at

3000g to remove the insoluble solids that hadprecipitatedupon removal of acetone. The sediments in thecentrifuge tubes werecarefully washed withwater to remove the trace amount of thewatersoluble extract and then dissolved indiethyl ether. The disolved sediments were

pooled together and evaporated at 30 C underreducedpressure to dryness. The powder collected waslabelled as fraction I.

The water-soluble extract was pooled together

and a small simple of it was evaporated todryness at 60 C in vacuo. The powder collectedwas named as fraction IV. The remaining water-soluble extract was fractionated using XAD2resin.

Column Fractionation of Water Soluble Extract.

Fractionation by adsorption to XAD2 resin wascarried out with slight modifications to instructionsprovided by the manufacturer. Resins were packedand equilibrated with water acidified with HCl to pH2.0. The water-soluble extract was adjusted to pH 2.0with HCl before loading into the column.Thirty milligrams of gallic acid equivalent (GAE),which is the binding capacity of the resin forG.mangostana pericarp extract, was loadedper mL of swollen XAD2 resin. Resin (500 g) wasused during each separation. The column was thenwashed with 2 bed volume (B.V.) of acidified (pH 2.0)water. All flow rates were kept constant at 2 B.V. perhour. A sample of the extract not bounded to the


7/14

column was collected and evaporated to dryness at60 C under reduced pressure.The powder collected was named fraction V.Desorption was carried out by washing the columnwith 2 B.V. of methanol followed by 2 B.V. ofacetone. The flow was kept at a constant 2 B.V. perhour. The solvent was removed by evaporating to

dryness at 40 C under reduced pressure. Thepowder collected was named fraction

VI.Solvent Fractionation. Fraction I and VI were furthersolventfractionated. The sample (1.0 g) was accuratelyweighed,and hexane(5.0 mL) was added. The mixture was allowed tostand for 15 minwith occasional shaking before centrifuging at 3000g.

The hexane andresidue was separated and the above processrepeated twice more. Thehexane-extracted solid was obtained by evaporatinghexane to drynessat 40 C under reduced pressure. This extractionprocess was repeatedon the hexane-insoluble residue using diethyl etherfollowed by ethylacetate, methanol and DMSO in running order. Forfraction I, only thediethyl ether and methanol fraction yield significant

quantities for furthercharacterisation and were termed fraction II and IIIrespectively. Forfraction VI, only the methanol fraction yield sufficientquantities andit was termed fraction

VII.Determination of Phenolic Contents. Totalphenolic content wasmeasured using a modified Folin-Ciocalteu assay asreported previously

(11). Briefly, a 10-fold diluted Folin Ciocalteu solution(100L), sample(20 L) or standard solution of gallic acid was addedtogether, followedby sodium carbonate solution (7%, w/v, 80 L). Thesolution was incubated for 30 min at 37 andabsorbance read at 765 nm. Phenoliccontent was expressed as gram gallic acidequivalents (GAE).Oligomeric proanthocyanidins (OPCs) was measured


8/14

using 4-dimethylamino-cinnamaldhyde (DMCA) assay as reported byShahidi and Naczk (12), using catechin as standard.Hydrolysable tannins wasdetermined by reacting with KIO3 using a modifiedmethod (13). Sample or standard (2.0 mL), methylgallate, dissolved in water was added to methanol

(2.0 mL) and 5% KIO3 (100L). Absorbance wasread at 525 nm after incubating in a 30 C water bathfor 30 min, the time was found to be optimal for colordevelopment in the standards.

Anthocyanins were determined using a pH-differential spectrophotometry method as describedby a method from the Institute of Nutraceutical

Advancement (INA) method 116.00 (14).

Bovine Serum Albumin Precipitation Assay. BSAprecipitation for the determination of tannins wascarried as described by Hagerman (15, 16). BSA was

dissolved in acetic acid buffer (pH 4.9, 0.20M,containing 0.17 M NaCl). Sample or tannic acidstandard (100L) dissolved in methanol was addedto BSA (200L), were mixed thoroughly for 15 minbefore centrifuging for 5 min. The supernatant wasdiscarded and the pellet washed with deionizedwater. The pellet was dissolved in SDS (400L, 1%w/v containing 5% (v/v) triethanolamine).

Ferric chloride (100L of 0.01 M in 0.01 Mhydrochloric acid solution) was added to thedissolved pellets and the mixture was allowed to

stand at room temperature for 15 min for theformation of the ferricphenolate complex. Theabsorbance readings were acquired using amicroplate reader at 510 nm with 200L ferric-phenolate complex in each well of a 96-well plate.The results were expressed as tannic acid equivalent(TAE). The TAE refers to the amount of tannic acid(g) required to produce an equivalent amount of BSAprecipitation as caused by 1 g of sample.


9/14

Amylase Inhibition Assay. A modified version of the assay introduced by Lever

for the determination of reducing sugar was used(17). A 1% starch solution was prepared bysuspending starch from potatoes in phosphate buffer(pH 6.9, 50 mM, containing 6.7 mM NaCl), andgelatinized at 100 C for 10 min on a hotplate. Anyevaporation losses were corrected by replacing thelost with water. R-Amylase solution was prepared bydissolving R-amylase from hog pancreas (35 mg) incalcium chloride (10 mL, 20 mM) topped up to 100mL with phosphate buffer. A stock solution (5% (w/v)in 0.5 M HCl) ofp-hydroxybenzoic acid hydrazide(PAHBAH) was prepared and stored at 4 C. The

stock solution was diluted five times with NaOH (0.5M) to give the working PAHBAH reagent which wasprepared daily.

Control assays contained phosphate buffer (900L)and R-amylase solution (100L) and the reactionwas started by the addition of gelatinised starchsolution (500L). To monitor the production ofreducing termini, sample (50L) was transferred outfrom each assay at 30 s intervals into PAHBAHreagent (1.0 mL). The reaction was monitored for 3.5min. After heating for 10 min in a boiling wter bath,

the absorbance was acquired at 410 nm with amicroplate reader.The initial rate of reaction was determined by usinglinear regression on Microsoft Excel softwarepackage. Inhibition assays contained phosphatebuffer (400L), R-amylase solution (100L) andinhibitor (500L) dissolved in buffer at the desiredconcentration and with pH adjusted to 6.9. Thereaction was started by the addition of gelatinizedstarch solution (500L) after the inhibitor was


10/14

incubated together with R-amylase for 15 min.Production of reducing termini was monitored usingthe same method as that of the control assay. Blanksfor control and inhibition assays was prepared byreplacing R-amylase with phosphate buffer, so as todetermine the amount of reducing termini present at0 min. The percent inhibition was determined by

comparing the initial rate of the inhibited assay to thatof a control assay conducted on the same day.

RESULTSFractionation of Pericarp Extract and Screeningofr-Amylase Inhibition Activity. Extracting G.mangostana pericarp with acetonewater-acetic acid(70:29.5:0.5) gave clear dark red solution.Subsequent removal of the acetone resulted inprecipitation of a dark brown solid as fraction I(Scheme 1).Defatting treatment of fraction I removed yellowish

green substances, possibly carotenoids orchlorophyll, which yielded too little for R-amylaseinhibition studies. The remaining fraction I wasseparated into diethyl ether soluble (fraction II) andmethanol soluble (Fraction III) fractions. Fraction IIwas bright yellow with no measurable R-AI activity. Insharp contrast, fraction III has very high inhibitionactivity with IC50 of 5.4g/mL. This value iscomparable with that of acarbose, a benchmarkcompound for R-AI. The water soluble portionFraction IV, shows considerable inhibitory activity(IC50 44.8g/mL). Fraction IV was treated with

Amberlite XAD2 resin and separated into those thathave no affinity to amberlite (fraction V) and thefraction that binds Amberlite XAD2 (Fraction VI).Fraction V was shown to contain mainlyanthocyanins (data not shown) and has rather lowinhibition activity (IC50 of 391.6g/mL), whereas theR-AI of fraction VI has been enriched (IC50, 27.5

g/mL). The IC50 was further improved to 10.1g/mL when fraction VI was extracted with methanoland yielding a methanol soluble fraction VII (Table1).No inhibition was detected for (() catechin and gallic

acid for levels up to 1000g/mL. Tannic acid, acommercially available hydrolysable tannin, wasfound to be a weak R-AI with an IC50 of 305.7

g/mL. The red pigment, anthocyanins does notexhibit significant R-AI activity. Typical inhibitionassay plots are shown in Figure 1. The doseinhibition relationship is linear and the IC50 wasobtained by interpolation of points closest to 50%inhibition.


11/14

Total Phenolics Content and r-Amylase InhibitionActivity.The total phenolic contents (TPC) of the fractions andthe results are tabulated in Table 1. Apparently, thereis no relationship between TPC and R-AI activity. Forexample, fraction II has rather high TPC but there isno measurable inhibition activity. Fraction VI has

about 5 times higher TPC than fraction V but has 14times more R-AI. Tannic acid has similar TPC tofraction VII but its R-AI IC50 is thirty fold higher thanthat of fraction VII. Therefore, the specific chemicalstructure profiles of the fractions are more importantfactors tan the TPC for determining R-AI activity,indicating there is specific interactions between theenzymes and the polyphenolic compounds, insteadof none-specific binding as commonly believed.

Protein Precipitation Capacity. The ability of thefractions containing high R-AI activity (III and VII) to

precipitate protein was measured and expressed asTAE. Both fraction III and VII precipitated BSA andtheir TAE is tabulated in Table 1. Fraction III hasTAE of 1.03 ( 0.06 TAE while fraction VII has TAE of1.22 ( 0.06. TAE values of above 1.0 imply that thesamples were either able to precipitate BSA moreefficiently than tannic acid or able to chelate more Fe(III) ions as compared to tannic acid. Catechin andgallic acid were not found to cause any precipitationof BSA, which had also been reported in literature(16). Therefore, fraction III and VII are polyphenolsthat can

precipitate proteins.


12/14

OPC Contents. While the BSA precipitation assayconfirmed that both fraction III and VII are tannins,the DMACA assay was carried out to determine ifthey are OPCs. The DMACA reagent reacts with theterminal unit of OPC or catechin at position C-8 of the

A ring to produce a green chromophore.

Catechin was used as the standard and fraction IIIand VII reacted positively with DMACA and thereforeconsist of flavonoids or OPCs. They contain 0.321( 0.009 and 0.424 (0.008 mmol terminal units/grespectively. Fraction II does not react with DMACAwhich confirm that it is not a flavonoid.

Tannic acid, a gallotannin, was used as a negativecontrol also does not react to DMACA.

Hydrolysable Tannin Contents. The oxidation ofgallate esters by KIO3 produces a red intermediate(max 525nm) before further oxidation to a yellowproduct (13). The red intermdiate is used as a testfor hydrolysable tannins which contain large numberof galloyl esters moiety. Methyl gallate was used asthe standard and fraction II, III, and VII does notcontain any gallate esters. Tannic acid, the positivecontrol, reacted strongly with KIO3.


13/14

Spectroscopic Properties. The UVvis spectra offraction II, III, and VII are shown in Figure 2. FractionII showed maximum absorption at 317 and 352 nmwhich is similar to that of mangostin (318 and 351nm) (18), the main xanthone in G. mangostana.Fraction III and VII hasmax at 275nm, whichis highly characteristic of flavonoids such as catechin

(19) which are structural components of OPCs.However the composition of fraction III and VIIappears to be slightly different. Fraction III showssome absorption at 370 nm, which could becontributed by small amounts of xanthones present inthem.Fraction VII on the other hand has a weak absorptionband at 450nm, possibly due to residualanthocyanins not removed during sorption to XAD2.From the IR spectra (Figure 3), it can be seen thatthe transmittance bands from 1700 to 900 cm-1 forfraction III and VII are almost identical. This further

suggests that they are either structurally similar orperhaps identical. Fraction II on the other handclearly has a different structure compared tothe other two fractions. FT-IR had been used as atechnique for differentiating OPCs and hydrolysabletannins (20). OPCs were found to have strongabsorptions at around 1280 cm-1 due to theirethereal C-O stretching vibrations, whereas nonecould be observed in hydrolysable tannins. Inagreement with the chemical assays, fraction III andVII are OPCs as they exhibit strong absorption in thatregion whereas tannic acid, a hydrolysable tannin,

does not. In addition, as these two fractions havepreviously tested positive for flavonoids but does notcontain the C)O stretching vibrations of flavones at16701625 cm-1 (21), it can be concluded that theyare either flavans, flavonoids lacking the carbonylfunctional group (22), or polymers of flavans, i.e.OPCs. The C-ring structure of xanthones is similarto the C-ring of flavonoids so they are expected tohave the IR absorption bands for the ethereal COand the flavone C)O stretching vibrations. Thestructural identification of the OPCs in mangosteenpericarps is further accomplished using MALDITOF

mass spectroscopy, 13C NMR, LC-MS, and thiolysisreactions. the results are reported in a separatepaper (23). Our results revealed that there are littlemonomers like epicatechin or catechin in thepericarp. As expected, fraction II showed strongabsorption bands in these two regions as well. LC-MS studies showed that fraction II contains mainly R-mangostin and other xanthones, which have alsobeen reported by other groups (24).


14/14

DISCUSSIONR-AI were detected in fraction III and VII but none infraction II. From LC-MS analysis, fraction II wasfound to constitute of xanthones of which mangostinsare predominant.Fraction III and VII are found to be OPCs as theyreacted positively to DMACA and had FT-IR

spectrum similar to that of OPCs. We have furthercharacterized the structural informationof the OPCs G. mangostana and the results shownthat they are mainly B-type oligomers (degree ofpolymerization ranging from 2 to 10) of epicatechinwith low content of gallates and monomers. Aplausible reason why OPCs could be found amongthe water insoluble extract (fraction I) as well as thewater soluble extract (fraction IV) is that they mighthave complexed with the xanthones and precipitatedtogether with them.While the KIO3 and DMACA assays are not strictly

quantitative, we used them as assay to differentiatebetween OPC and

Documents

Mangostan y Diabetes