Embed Size (px)
Citation preview
at SciVerse ScienceDirect
Food Microbiology 32 (2012) 135e146
Contents lists available
Food Microbiology
journal homepage: www.elsevier .com/locate/ fm
The first dairy product exclusively fermented by Propionibacterium freudenreichii:A new vector to study probiotic potentialities in vivo
Fabien J. Cousin a,b,c, Séverine Louesdon a,b, Marie-Bernadette Maillard a,b, Sandrine Parayre a,b,Hélène Falentin a,b, Stéphanie-Marie Deutsch a,b, Gaëlle Boudry d, Gwénaël Jan a,b,*
a INRA, UMR1253 Science et Technologie du Lait et de l’Œuf, F-35042 Rennes, FrancebAGROCAMPUS OUEST, UMR1253 Science et Technologie du Lait et de l’Œuf, F-35042 Rennes, FrancecCNIEL/Syndifrais, 42 rue de Châteaudun, F-75314 Paris 09, Franced INRA, UR1341 ADNC, F-35590 Saint Gilles, France
a r t i c l e i n f o
Article history:Received 9 September 2011Received in revised form14 March 2012Accepted 14 May 2012Available online 19 May 2012
Keywords:Dairy propionibacteriaPropionibacteriumProbioticFermented milkSurvival
Abbreviations: CIRMeBIA, Centre InternMicrobienneseBactéries d’Intérêt Alimentaire; CLADHNA, 1,4-dihydroxy-2-naphtoic acid; EPS, exopolysabowel diseases; OD, optical density; P.a, PropionibPropionibacterium freudenreichii subsp. freudenreichiidenreichii subsp. shermanii; P.j, Propionibacterium jensaline; SCFA, short chain fatty acid; UF, milk ultrafiltra* Corresponding author. INRA, UMR1253 STLO, 65
Rennes cedex, France. Tel.: þ33 (0)2 23 48 57 41; faxE-mail addresses: [email protected] (F.J.
grignon.inra.fr (S. Louesdon), [email protected]@rennes.inra.fr (S. Parayre), [email protected] (S.-M. Deutsch(G. Boudry), [email protected] (G. Jan).
0740-0020/$ e see front matter � 2012 Elsevier Ltd.doi:10.1016/j.fm.2012.05.003
a b s t r a c t
Dairy propionibacteria display probiotic properties which require high populations of live and meta-bolically active propionibacteria in the colon. In this context, the probiotic vector determines probioticefficiency. Fermented dairy products protect propionibacteria against digestive stresses and generallycontain a complex mixture of lactic and propionic acid bacteria. This does not allow the identification ofdairy propionibacteria specific beneficial effects. The aim of this study was to develop a dairy productexclusively fermented by dairy propionibacteria. As they grow poorly in milk, we determined theirnutritional requirements concerning carbon and nitrogen by supplementing milk ultrafiltrate (UF) withdifferent concentrations of lactate and casein hydrolysate. Milk or UF supplemented with 50 mM lactateand 5 g L�1 casein hydrolysate allowed growth of all dairy propionibacteria studied. In these new fer-mented dairy products, dairy propionibacteria remained viable and stress-tolerant in vitro duringminimum 15 days at 4 �C. The efficiency of milk fermented by the most tolerant Propionibacteriumfreudenreichii strain was evaluated in piglets. Viability and SCFA content in the colon evidenced survivaland metabolic activity of P. freudenreichii. This work results in the design of a new food grade vector,which will allow preclinical and clinical trials.
� 2012 Elsevier Ltd. All rights reserved.
1. Introduction
A growing number of studies concern probiotic potentialities ofdairy propionibacteria (Cousin et al., 2011). These bacteria, well-known for their role as ripening starters in the cheese industry,can for example trigger apoptosis of colorectal cancer cells via the
ational de Ressources, conjugated linoleic acid;ccharides; IBD, inflammatoryacterium acidipropionici; P.ff,; P.fs, Propionibacterium freu-senii; PBS, phosphate bufferte; YEL, yeast extract lactate.rue de Saint Brieuc, 35042
: þ33 (0)2 23 48 53 50.Cousin), [email protected] (M.-B. Maillard),[email protected] (H. Falentin),), [email protected]
All rights reserved.
production of short chain fatty acids (SCFAs), propionic and aceticacids (Jan et al., 2002a). They also synthesize 1,4-dihydroxy-2-naphtoic acid (DHNA) which can modulate the microbiota byenhancing intestinal bifidobacteria (Hojo et al., 2002). These pro-pionibacterial metabolites may also exert anti-inflammatory effectsin situ (Suzuki et al., 2006; Tedelind et al., 2007). Nevertheless, sucheffects require high populations of live and metabolically activepropionibacteria in the colon. Use of a well-adapted delivery vectoris decisive to ensure bacteria survival towards digestive stressesmet before the gut (Sanders and Marco, 2010).
Fermented dairy products constitute 90% of the products in theprobiotic market (Meyer, 2007) and a growing part in functionalfood market (Ozer and Kirmaci, 2010). They have revealed goodabilities to protect bacteria against digestive stresses (Ranadheeraet al., 2010). Fermented milks were shown to protect dairy pro-pionibacteria towards digestive stresses in vitro (Leverrier et al.,2005). Moreover, consumption of milk or cheese containing dairypropionibacteria led to survival of these bacteria and to anincreased production of propionic acid within the caecum of mice(Perez-Chaia and Zarate, 2005). In addition, Emmental cheese (Jan
F.J. Cousin et al. / Food Microbiology 32 (2012) 135e146136
et al., 2002b) and yoghurt supplemented with dairy propionibac-teria (Hervé et al., 2007) are efficient vectors to deliver live andmetabolically active Propionibacterium strains to the human gut.However, Emmental also contains lactic acid bacteria such asS. thermophilus and L. helveticus. Probiotic yogurts also contain highamounts of S. thermophilus and L. delbrueckii. Other probioticproducts were developed with dairy propionibacteria but containa mixture of bacteria, some of them being probiotic strains (El-Nezami et al., 2006; LeBlanc et al., 2006; Maity et al., 2008; Xuet al., 2005). Consequently, the identification of specific beneficialhealth effects due to propionibacteria using these products is notpossible. Dairy propionibacteria grow poorly in milk (Piveteauet al., 2000) and Propionibacterium freudenreichii cannot beconsidered as a milk-adapted species (Falentin et al., 2010a). Thishas been linked to their greater peptidase than protease activity,preventing P. freudenreichii utilizing the nitrogen frommilk (Dupuiset al., 1995; El-Ssoda et al., 1992; Gagnaire et al., 1999; Langsrudet al., 1995; Lemée et al., 1998; Sahlstrom et al., 1989) and, forsome strains, to their inability to ferment lactose (Piveteau, 1999).During Emmental cheese manufacture, dairy propionibacteria needthe presence of starter lactic acid bacteria which provide lactateand peptides, from lactose and milk proteins, respectively.
The aim of this study was to develop a dairy vector to efficientlydeliver propionibacteria to the gut without any other bacteria. First,a dairy product exclusively fermented by a dairy propionibacteriumwas designed. Its efficiency to provide propionibacteria digestivestress adaptation was then checked in vitro and in vivo by feedingpiglets.
2. Material and methods
2.1. Bacterial strains and growth conditions
Dairy propionibacteria strains (Table 1) were provided by theCIRMeBIA (Centre International de Ressources MicrobienneseBactéries d’Intérêt Alimentaire, INRA, Rennes, France). This panelcontained strains either able to ferment lactose (lactoseþ) or not(lactose�). They were routinely cultivated, at 30 �C without shaking,in Yeast Extract Lactate (YEL) broth containing 100 mM lactate and10 g L�1 yeast extract (Malik et al., 1968). This laboratory mediumwas also used as reference medium for dairy propionibacteriagrowth, survival at 4 �C and tolerance towards digestive stresses. Inthe first part of this study, 4 strains were used, representing the 4species/subspecies the most studied in literature. The reference
Table 1Bacterial strains and their origin.
Straina Taxonomy
CIRM Other name Genus Species
BIA1 CIP103027 Propionibacterium freudenrBIA64 CNRZ80 Propionibacterium acidipropBIA116 CNRZ81 Propionibacterium freudenrBIA125 ITG P14 Propionibacterium freudenreBIA127 ITG P18 Propionibacterium freudenreBIA129 ITG P20 Propionibacterium freudenreBIA136 LSP 121 Propionibacterium freudenreBIA138c ITG P9 Propionibacterium freudenreBIA455 CNRZ 87 Propionibacterium jenseniiBIA458 LSP 110 Propionibacterium freudenreBIA527 e Propionibacterium freudenreBIA703 LSP 103 Propionibacterium freudenre
The 4 species/subspecies reference strains used for the supplemented milk ultrafiltrate ta Culture collections: CIRMeBIA, Centre International de Ressources MicrobienneseBac
France; CNRZ: Centre National de Recherche en Zootechnie, INRA, Jouy-en-Josas, France; ICaen, France.
b þ: able to ferment lactose; �: unable to ferment lactose.c Strain used for the in vivo trial.
strain P. freudenreichii subsp. shermanii BIA1T has recently beensequenced and annotated (Falentin et al., 2010a).
2.2. Supplementations in milk ultrafiltrate (UF) or UHT milk
Milk ultrafiltrate (UF) was obtained by an ultrafiltrationprocess as described previously (Michalski et al., 2006). Briefly,raw milk was skimmed using a cream separator (Westfalia,Chateau-Thierry, France). Ultrafiltrationwas then performed usinga UF pilot equipment (T.I.A., Bollene, France) equipped withorganic spiral membrane with a molecular weight cut-off of 5 kDa(Koch International, Lyon, France). The temperature during theultrafiltration process was maintained around 50e55 �C. The UFcollected was then sterilized by 0.2 mm filtration (Nalgene, Ros-kilde, Denmark) and stored at 4 �C. In contrast with milk, this clearmedium allows bacteria growth kinetic monitoring, using classicalmethods, such as spectrophometry. In addition, its composition isvery close to milk, UF is the aqueous phase of milk free fromproteins, containing notably lactose (50 g L�1), minerals andvitamins.
This UF was supplemented with different amounts of food gradesodium lactate (>97% L-lactate, galaflow SL 60, Société Arnaud,Paris, France) and food grade casein hydrolysate (Casein PeptonePlus, Organotechnie, La Courneuve, France), neutralized to pH 7 byNaOH 5 M and sterilized by 0.2 mm filtration (Table 2).
UHT milk (half-skimmed milk, UHT, Agrilait, Cesson-Sévigné,France) was supplemented with 50mMof sterile lactate and 5 g L�1
of sterile casein hydrolysate.
2.3. Growth monitoring in milk ultrafiltrate and UHT milk,supplemented or not
Dairy propionibacteria were pre-cultivated in YEL at 30 �Cwithout shaking during 3 days, before inoculation (1%) in thedifferent pre-warmed supplemented UF tested (Table 2). Theinoculation in supplemented UHT milk was performed in the sameway except that the pre-cultures were performed in UF L100P10(Table 2). The growth of dairy propionibacteria was monitored byOD at 650 nm (only in UF), pH measurement and cfu counting. Thedairy propionibacteria growth was compared to the YEL referenceculture medium (Malik et al., 1968). The enumeration was per-formed using the micromethod previously described (Baron et al.,2006) adapted to dairy propionibacteria using YEL agar andanaerobic incubation (5 days, 30 �C) before counting.
Lactoseb Origin
Subspecies
eichii shermanii D Cheeseionici D Dairy producteichii freudenreichii � Dairy productichii shermanii þ Cheeseichii freudenreichii � Cheeseichii shermanii þ Cheeseichii shermanii þ Human fecesichii shermanii þ Cheese
± Buttermilkichii shermanii þ Cheeseichii freudenreichii � Cheeseichii shermanii þ Dairy product
ests appear in bold characters.téries d’Intérêt Alimentaire, INRA, Rennes; CIP: Collection de l’Institut Pasteur, Paris,TG: Institut Technique du Gruyère, Actilait, Rennes, France; LSP: Laboratoires Standa,
Table 2Media tested for dairy propionibacteria growth.
YELa Medium UF Casein hydrolysate supplementation Lactate supplementation
UF L100P0 UF L100P2 UF L100P5 UF L100P10 UF L100P15 UF L0P5 UF L50P5 UF L150P5
10b Casein peptone (g L�1) 0 0 2 5 10 15 5 5 5100 Sodium lactate (mM) 0 100 100 100 100 100 0 50 150
Media were neutralized (pH 7) with NaOH 5 M and sterilized by 0.2 mm filtration.The chosen concentrations for supplementation appeared in bold characters.
a YEL: yeast extract lactate; UF: milk ultrafiltrate; L: lactate; P: peptone.b Yeast extract.
F.J. Cousin et al. / Food Microbiology 32 (2012) 135e146 137
2.4. Survival during storage at 4 �C
Three-day cultures of dairy propionibacteria, in YEL, in supple-mented UF or in supplemented UHT milk, were analyzed duringstorage at 4 �C. Survival during long-term cold storagewas followedby enumeration and live/dead staining (Invitrogen, Carlsbad, USA).For live/dead staining, bacteria cultivated in supplemented UF werewashed in phosphate buffer saline (PBS) and stained according tothe instructions of the manufacturer, prior to observation under anOptiphot fluorescence microscope (Nikon, Champigny sur Marne,France).
2.5. Digestive stresses in vitro
Acid challenge was performed during 1 h at 37 �C by a ten-folddilution of the culture in pre-warmed acidified lactate broth (YELdevoid of yeast extract (Jan et al., 2001), adjusted to pH 2.5 usingHCl 37%). After this challenge, dairy propionibacteria wereenumerated on YEL agar as described below.
The bile salts challenge was performed in the same way in pre-warmed lactate broth (pH 7) supplemented with bile salts (cholicacid and deoxycholic acid sodium salt mixture, Fluka, Saint Louis,USA) at a final concentration of 1 g L�1, as previously described(Leverrier et al., 2003).
For these two in vitro digestive stresses, an enumeration of thesample was performed before and after challenge and allowedsurvival percentage calculations.
2.6. Animal procedure
The experimental protocol was designed in compliance withrecommendations of the French law (2001-464 29/05/01) and EEC(86/609/CEE) for the care and use of laboratory animals under thecertificate of authorization to experiment on living animals n�3569.The pigs provide a suitable model because the gastrointestinal tractis physiologically and anatomically similar to that of humans(Moughan et al., 1992).
Twenty-four ((Pietrain � Landrace) � Large White) pigs fromthe experimental herd of INRA St-Gilles (France) were used. Eighttriplets of sex- and weight-matched 6week-old littermates werehoused individually in stainless cages in a temperature-controlled(23 �C) and 12 h/12 h dark/light cycle room. Initial body weightof pigs was similar between the three groups (8.76 � 0.14 kg forsterile milk group, 8.98 � 0.21 kg lyophilizate group 8.89 � 0.23 kgfor fermented milk respectively, P> 0.05). Pigs were weighed twiceaweek. They were fed ad-libitumwith a pig dedicated diet (CooperlHundaye, Lamballe, France). Food intake was measured daily. Pigshad free access to water.
Pigs were gavaged every morning (between 9.00 and 10.00 am)during 14 days with 10mL of either 1) sterile milk, or 2) 2� 1010 cfuof lyophilized P.fs BIA138 (Laboratoires Standa, Caen, France),resuspended in sterile physiologic water, or 3) P.fs fermented milkcontaining 2 � 1010 cfu of P.fs BIA138 strain.
2.7. Counting of propionibacteria in fecal and caecal samples
Feces were analyzed at 3 specific times: at day 0 (before thetreatment), at day 7 (during the treatment) and at day 14 (end of thetreatment). At the end of the treatment period, pigs were sacrificed3 h after their last meal by electronarcosis and exsanguination. Theproximal colon was immediately dissected and colon contents werecollected. 3e4 g of feces (or colon contents) freshly collected wereimmediately analyzed. Their propionibacteria concentrations weremeasured by enumeration on Pal-propiobac� selective agar (Labo-ratoires Standa, Caen, France), supplemented with metronidazoleat 4 mg L�1 (Sigma, St Quentin Fallavier, France), as previouslydescribed (Hervé et al., 2007). At days 7 and 14, feces collectedduring 24 hwereweight to allow survival rate calculation of ingesteddairy propionibacteria. Results are expressed as survival rates infeces and log10 cfu g�1 of dried matter for colon contents.
2.8. Short chain fatty acid (SCFA) analysis in caecal samples
Immediately after colon content sampling, SCFAs were extractedin a cold Tris buffer (50 mM, pH ¼ 7.5) and stored at �20 �C untilanalysis. Proteins were precipitated by incubation for 1 h at 4 �C inthe presence of oxalic acid (0.03 M final concentration). SCFAs wereseparated on a BP21 column and quantified by flame ionizationdetector as previously described (Thierry et al., 2002). Isocaproicacid was used as an internal standard because it was absent frompiglet caecal samples (data not shown). Samples were analyzed induplicates.
2.9. Statistical analysis
Statistical analyses were performed using the General LinearModel procedure of Statistical Analysis Systems software (SASInstitute, Cary, NC, USA), testing the piglet pair and treatment effect,with t-test as a subsequent multiple comparisons when appro-priate. All results are presented as means � SEM. Differencesbetween groups were declared significant at P < 0.05.
3. Results
3.1. Dairy propionibacteria need casein hydrolysate to grow in milkultrafiltrate
As dairy propionibacteria grow poorly in milk (Piveteau et al.,2000), their nutritional requirements concerning nitrogenwere determined by supplementing UF with different concentra-tions of casein hydrolysate, with a constant concentration of100 mM lactate (same concentration as in YEL). The suitabledairy propionibacteria population had to reach 1E þ 09 cfu mL�1 inthe stationary phase (population comparable to Emmental cheese).
Dairy propionibacteria grew very slowly in UF or UF only sup-plemented with lactate (Fig. 1). Indeed, the OD650 nm did notincrease (OD650 nm ¼ 0.106 after 4 days of culture) for the lactose�
strain in the non supplemented UF (Fig. 1C). In YEL, OD650 nm
1
10
1
10 BA P.fs BIA1 P.a BIA64
0.1
OD
65
0n
m
0.1
OD
65
0n
m
10
0.010 1 2 3 4
Time (days)
10
0.010 1 2 3 4
Time (days)
DCP.ff BIA116 P.j BIA455
1
OD
65
0n
m
1
D6
50
nm
0.01
0.1
0 1 2 3 40.01
0.1
0 1 2 3 4
O
Time (days)
25
30
h)
1.E+10
.m
L-1
)
Time (days)
FE
10
15
20
Gen
eratio
n t
im
e (
1.E+08
1.E+09
op
io
nib
acteria (
cfu
0
5
BIA1 BIA64 BIA116 BIA455
1.E+07BIA1 BIA64 BIA116 BIA455
Pr
P.fs P.a P.ff P.j P.fs P.a P.ff P.j
Fig. 1. Impact of casein hydrolysate supplementation on dairy propionibacteria growth in milk ultrafiltrate. Growth of P. freudenreichii subsp. shermanii (A), P. acidipropionici (B),P. freudenreichii subsp. freudenreichii (C) and P. jensenii (D) were assessed by OD650 nm monitoring. (E) Generation time was calculated during the exponential phase of growth.(F) Propionibacteria final concentration (cfu mL�1) was determined after 3 days of incubation (stationary phase). Different casein hydrolysate supplementations in milk ultrafiltrate(UF) were tested for propionibacteria growth. A constant lactate concentration of 100 mM was used with different concentrations of casein hydrolysate: 0 g L�1 ( ), 2 g L�1 ( ),5 g L�1 ( ), 10 g L�1 ( ) and 15 g L�1 ( ). Reference media used were YEL (C) and non supplemented UF (D). The results are means of at least two independent experiments. Thechosen concentration of 5 g L�1 casein hydrolysate is indicated by an arrow.
F.J. Cousin et al. / Food Microbiology 32 (2012) 135e146138
decreased after the third day of culture, especially for the BIA1strain (Fig. 1A), suggesting autolysis. In all the supplemented UFtested, the OD650 nm remained stable or even increased duringstationary phase, whatever the strain (Fig. 1). With casein hydro-lysate addition, the highest concentration is, the shortest genera-tion time is (Fig. 1E) and the highest final propionibacteriaconcentration is (Fig. 1F). The generation times for the 4 strainsbecame even shorter than those in YEL medium classically used fordairy propionibacteria culture. However, the final propionibacteriaconcentrations obtained were always higher in YEL (Fig. 1F).
Maximal dairy propionibacteria population levels reached after3 days of culture were not improved with casein hydrolysatesupplementation superior to 5 g L�1 (Fig. 1F), except for the BIA455
strain. Consequently, this 5 g L�1 casein hydrolysate concentrationwas chosen for the next step of the study.
3.2. Lactate modulates dairy propionibacteria growth in milkultrafiltrate
After casein hydrolysate supplementation, the nutritionalrequirements in carbon were determined by supplementing UFwith different concentrations of lactate, with a constant concen-tration of 5 g L�1 casein hydrolysate.
The addition of lactate facilitates the growth, especially for thelactose� strains (Fig. 2CeD). For the lactoseþ strains (Fig. 2AeB), thelactate supplementations did not affect the OD curves but
1
10
1
10P.fs BIA1 P.a BIA64BA
0.1
OD
65
0n
m
0.1
OD
65
0n
m
0.010 1 2 3 4
Time (days)
0.010 1 2 3 4
Time (days)
10 10P.ff BIA116 P.j BIA455
DC
1
D6
50
nm
1
65
0n
m
0.01
0.1
0 1 2 3 4
O
0.01
0.1
0 1 2 3 4
OD
25
30
)
Time (days) Time (days)
1.E+10
.m
L-1)
FE
10
15
20
Gen
eratio
n t
im
e (h
1.E+08
1.E+09
op
io
nib
ac
te
ria
(c
fu
0
5
BIA1 BIA64 BIA116 BIA455
1.E+07BIA1 BIA64 BIA116 BIA455
Pr
j.Pff.Pa.Psf.Pj.Pff.Pa.Psf.P
Fig. 2. Impact of lactate supplementation on dairy propionibacteria growth in milk ultrafiltrate. Growth of P. freudenreichii subsp. shermanii (A), P. acidipropionici (B), P. freudenreichiisubsp. freudenreichii (C) and P. jensenii (D) were assessed by OD650 nm measurement. (E) Generation time was calculated during the exponential phase of growth. (F) Propionibacteriafinal concentration (cfu mL�1) was determined after 3 days of incubation (stationary phase). Different lactate supplementations in milk ultrafiltrate (UF) were tested for propio-nibacteria growth. A constant casein hydrolysate concentration of 5 g L�1 was used with different concentrations of lactate: 0 mM ( ), 50 mM ( ), 100 mM ( ) and 150 mM ( ).Reference media used were YEL (C) and non supplemented UF (D). The results are means of at least two independent experiments. The chosen concentration of 50 mM lactate isindicated by an arrow.
F.J. Cousin et al. / Food Microbiology 32 (2012) 135e146 139
decreased the population reached after 3 days of culture withthe highest lactate concentrations. Contrary to casein hydrolysateaddition, generation times (Fig. 2E) and final propionibacteriaconcentrations (Fig. 2F) remained unchanged with high concen-trations of lactate. Addition of sodium lactate can even lead togrowth inhibition in Propionibacterium acidipropionici and Propio-nibacterium jensenii (Fig. 2F).
The maximal propionibacteria concentrations were reached witha 50mM lactate supplementation. This 50mM lactate concentrationwas chosen for final supplementation, with 5 g L�1 of casein hydro-lysate (Table 2). Series of 5 successive passages were performed onthismediumwithout change of the growth kinetics (data not shown).
3.3. Dairy propionibacteria reached high concentrations insupplemented UHT milk
The aim of our study was to develop a fermented dairy productthat will efficiently deliver propionibacteria without any otherbacterial species. The nutritional requirements in nitrogen andcarbon having been determined, 50 mM lactate and 5 g L�1 caseinhydrolysate were added to UHT milk and improved the growth ofall the dairy propionibacteria species tested (Fig. 3). The growthcurves in supplemented UHTmilk were close to those in YEL and allthe tested strains reached 1E þ 09 cfu mL�1 after 3 days of culture.Dairy propionibacteria reached higher levels in supplemented UHT
BA
1 E+09
1.E+10
u.m
L-1)
1 E+09
1.E+10
u.m
L-1)
P.fs BIA1 P.a BIA64
1.E+08
1.E+09
pio
nib
ac
te
ria
(c
f
1.E+08
1.E+09
op
io
nib
ac
te
ria
(c
f
1.E+070 1 2 3 4
Pro
Time da s
1.E+070 1 2 3 4
Pr
Time da s
1.E+10DC
( y ) ( y )
1.E+10 P.ff BIA116 P.j BIA455
klimdetnemelppuSklimTHULEY Supplemented UFUF
1.E+09
ac
te
ria
(c
fu
.m
L-1)
1.E+09
eria (
cfu
.m
L-1)
1.E+08P
ro
pio
nib
1.E+08
Pro
pio
nib
act
1.E+070 1 2 3 4
Time (days)
1.E+070 1 2 3 4
Time (days)
Fig. 3. Growth kinetic of dairy propionibacteria in UF or UHT milk, supplemented or not with 5 g/L hydrolysate casein and 50 mM lactate. Growth kinetic of P. freudenreichii subsp.shermanii (A), P. acidipropionici (B), P. freudenreichii subsp. freudenreichii (C) and P. jensenii (D) was studied in YEL (C), milk ultrafiltrate (UF) supplemented ( ) or not (D) and UHTmilk supplemented ( ) or not (,). The supplementation was 50 mM of lactate and 5 g L�1 of casein hydrolysate. The results are means of at least two independent experiments.
F.J. Cousin et al. / Food Microbiology 32 (2012) 135e146140
milk than in supplemented UF (Fig. 3). This was also observed fordairy propionibacteria in non supplemented milk compared to nonsupplemented UF (Fig. 3).
3.4. Dairy propionibacteria remained viable in the fermented milkduring storage at 4 �C
The probiotic definition stipulates that probiotics must beconsumed “live” (FAO/WHO, 2002). Survival of dairy propionibac-teria at cold temperature in this new fermented milk is so animportant parameter, as fermented dairy products are usuallystored during a shelf life of 2 weeks at 4 �C. All dairy propioni-bacteria strains tested remained viable during at least 15 days at4 �C, whatever the considered medium (Fig. 4). Indeed, no decreasein propionibacteria concentrations was observed in YEL, in sup-plemented UF or in supplemented milk (Fig. 4AeD). The propio-nibacteria concentrations reached at least 1E þ 09 cfu mL�1 after 3days of culture (except for BIA455 in supplemented UF) andremained stable during the storage at 4 �C for a minimum of 15days. After a 6-month storage at 4 �C, the propionibacteriaconcentrations were still over 1E þ 09 cfu mL�1 in fermented milk,except for BIA64 (data not shown). Besides, live/dead stainingconfirmed the very little proportion of dead propionibacteria after15 days of storage at 4 �C (Fig. 4E).
3.5. Dairy propionibacteria provided in the fermented milk toleratedigestive stresses in vitro
Another critical parameter of probiotic efficiency is theircapacity to withstand the digestive tract stresses and to remainmetabolically active in the gut. The tolerance towards acid and bile
salts stresses was compared in vitro for the 4 dairy propionibacteriain YEL, in supplemented UF or in supplemented milk. The P.fs BIA1strain presented a good survival rate to acidic environment in thethree tested media and a very high tolerance towards bile salts inYEL and fermented milk (Fig. 5A). The P.a BIA64 strain did notsurvive to pH 2.5 and have a low survival rate in bile salts presence,except in fermented milk with a survival rate between 2 and 8%(Fig. 5B). The P.ff BIA116 strainwas themost resistant strain towardsthe digestive stresses studied in vitro milk (Fig. 5C). Whatever thetime, medium or stress, the survival was always high and very closeto 100%. The P.j BIA455 strain displayed a similar tolerance profile tothe BIA64 strain towards acidic challenge (Fig. 5D). For the bile saltsstress, survival of BIA455 strain was low in YEL. It was improved insupplemented UF. The best survival rate was reached in fermentedmilk (58e89%). Altogether, these data showed a similar orimproved tolerance towards bile salts stress and a comparablesurvival to acidic challenge in the fermented milk, compared to theYEL reference laboratory medium, whatever the strain.
3.6. Fermented milk allows screening of dairy propionibacteriastrains
In order to select the best candidate strain for the in vivo trial,the tests (maximal population reached, survival at 4 �C and towardsdigestives stresses) applied to the 4 reference strains wereextended to 8 other strains in fermented milk (Table 1). All thesestrains reached a 1Eþ 09 cfumL�1 concentration in fermentedmilkafter 3 days of culture at 30 �C and during storage at 4 �C for at least15 days (Fig. 6A). By contrast, the P. freudenreichii strains displayedvarious tolerances towards acidic challenge while P. acidipropioniciand P. jensenii did not survive this challenge (Fig. 6B). The strains
1.E+10
(cfu
.m
L-1)
P.fs BIA1 P.a BIA641.E+10
(cfu
.m
L-1)
1.E+08
1.E+09
Pro
pio
nib
acteria
1.E+08
1.E+09
Pro
pio
nib
acteria
3-day cultures D+1 D+8 D+15
YEL Supplemented UF
Supplementedmilk
P.ff BIA1161.E+10 1.E+10
YEL Supplemented UF
Supplementedmilk
BA
DC P.j BIA455
1.E+09
acteria (
cfu
.m
L-1)
1.E+09
cteria (
cfu
.m
L-1)
1.E+08YEL Supplemented
UFSupplemented
milk
Pro
pio
nib
1.E+08YEL Supplemented
UFSupplemented
milk
Pro
pio
nib
a
E
P.fs BIA1
N ti t l
P.a BIA64
P.ff BIA116
Negative control
P.j BIA455
Fig. 4. Viability of dairy propionibacteria during storage at 4 �C. Concentration of P. freudenreichii subsp. shermanii (A), P. acidipropionici (B), P. freudenreichii subsp. freudenreichii (C)and P. jensenii (D) were determined in YEL, supplemented milk ultrafiltrate (UF) and supplemented UHT milk. These concentrations were determined after 3 days at 30 �C (,) andafter storage at 4 �C during 1 (D þ 1, ), 8 (D þ 8, ) and 15 (D þ 15, ) days. The results are means of at least two independent experiments. (E) Viability of the dairy pro-pionibacteria strains in supplemented UF after 15 days at 4 �C, was assessed using live/dead kit. Live bacteria emit green fluorescence. Negative control was obtained by heatingculture at 70 �C. Magnification: �100. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
F.J. Cousin et al. / Food Microbiology 32 (2012) 135e146 141
BIA116, 138 and 527 were the most acid-tolerant ones witha viability reduction of 0.5 log. The strain BIA703was themost acid-sensitive with a viability reduction of more than 2 logs. All thetested strains resisted to bile salts in fermented milk with valuesvery close to 100% (Fig. 6C). The most digestive stress-tolerantstrain chosen for the in vivo trial was thus the P.fs BIA138 strain.
3.7. Fermented milk delivers live and metabolically activeP. freudenreichii in the gut
Pigs were daily gavaged with sterile milk, or with P.fs BIA138(2E þ 10 cfu day�1), either lyophilized or in fermented milk, during
14 days. Propionibacteria were undetected (below the threshold of1E þ 03 cfu g�1) in pig feces the day before gavage beginning (datanot shown). P.fs delivery form, lyophilizate or in fermented milk,did not significantly affect the propionibacteria survival rates in thefeces, which were 12% and 16% for the lyophilizate group and 18%and 17% for the fermented milk group (Fig. 7A). The propionibac-teria concentrations between these two groups did not differ incolon content either (Fig. 7A), with populations superior to 5log10 cfu g�1 of dried colon contents (superior to 6 log10 cfu g�1). Asa control, propionibacteria remained undetectable in the feces andcolon content in the sterile milk group, throughout the experiment(data not shown).
1
10
100
urviv
al (%
)
1
10
100
e su
rvival (%
)
P.fs BIA1A P.fs BIA1
egnellahcstlaseliBegnellahcdicA
0.0001
0.001
0.01
0.1
YEL Supplemented Supplemented
Bile
s
alts
s
0.0001
0.001
0.01
0.1
YEL Supplemented Supplemented
Acid
c
ha
lle
ng
UF milk
0.1
1
10
100
lts su
rvival (%
)
0.1
1
10
100
ng
e su
rvival (%
)
UF milkP.a BIA64B P.a BIA64
0.0001
0.001
0.01
YEL Supplemented UF
Supplementedmilk
Bile sa
0.0001
0.001
0.01
YEL Supplemented UF
Supplementedmilk
Acid
c
ha
lle
¥ ¥ ¥ ¥¥ ¥ ¥ ¥ ¥ ¥ ¥ ¥
P ff P ff
¥ ¥ ¥ ¥
C
0.1
1
10
100salts su
rvival (%
)
0.1
1
10
100
allen
ge su
rvivla (%
)
P.ff BIA116 P.ff BIA116
1001)D
0.0001
0.001
0.01
YEL Supplemented UF
Supplementedmilk
Bile
0.0001
0.001
0.01
YEL Supplemented UF
Supplementedmilk
Acid
c
h
P.j BIA455P.j BIA455
0.01
0.1
1
10
ile salts su
rvival (%
)
0.01
0.1
1
10
00
ch
allen
ge su
rvival (%
0.0001
0.001
YEL Supplemented UF
Supplementedmilk
B
0.0001
0.001
YEL Supplemented UF
Supplementedmilk
Acid
3-day cultures D+1 D+8 D+15
¥ ¥ ¥ ¥ ¥ ¥ ¥ ¥ ¥ ¥ ¥ ¥
Fig. 5. Survival of dairy propionibacteria in digestive stress conditions in vitro. P. freudenreichii subsp. shermanii (A), P. acidipropionici (B), P. freudenreichii subsp. freudenreichii (C) andP. jensenii (D) were cultivated in YEL, supplemented milk ultrafiltrate (UF) and supplemented UHT milk during 3 days at 30 �C. These cultures were acidified to pH 2.5 using HCl oradded with bile salts (1 g L�1). After 1 h incubation (37 �C), survival was determined by cfu counting. These survival rates were determined after 3 days at 30 �C (,) and afterstorage at 4 �C during 1 (D þ 1, ), 8 (D þ 8, ) and 15 (D þ 15, ) days. U: undetectable propionibacteria survival (<0.0001%). The results are means of at least two independentexperiments.
F.J. Cousin et al. / Food Microbiology 32 (2012) 135e146142
The SCFA content measured in the colon contents at slaughteringdiffered between groups (Fig. 7B). Significant differences wereobserved concerning the propionibacterial metabolites: acetic andpropionic acids. These two SCFAs were higher in the fermented milkgroup compared to the two other groups (P < 0.05). By contrast, nosignificant difference was observed between the sterile milk and thelyophilizate group, and no difference between groups in terms ofbutyric, valeric acid and caproic acid content. Concerning thebranched-chained SCFAs, both branched butyric and valeric acidswere reduced in the fermented milk group. Moreover, the totalbranched SCFAs were lower (P ¼ 0.02) in the fermented milk groupwith 4.91 � 0.47 mmol g�1 of dried colon content compared to thelyophilizate groupwith 8.28� 1.28mmol g�1 of dried colon content.
4. Discussion
4.1. Dairy propionibacteria growth in milk environment
Dairy propionibacteria are known for their probiotic aptitudes(Cousin et al., 2011). They are commonly used in the Swiss cheeseindustry and many strains have been isolated from dairy products.In cheese, dairy propionibacteria grow thanks to the presence oflactate and peptides released by lactic acid starters (Langsrud andReinbold, 1973). Propionibacteria grow poorly in milk in theabsence of other microorganisms (Piveteau et al., 2000). Accord-ingly, clinical studies using dairy propionibacteria involvedmixtures of probiotic propionibacteria and other bacteria. The
A1.E+10
ia (
cfu
.m
L-1)
1.E+09
Pro
pio
nib
acter
10
100
al (%
)
B
1.E+081 64 116 125 127 129 136 138* 455 458 527 703
0.01
0.1
1
ch
alle
ng
e s
urviv
100
0.0001
0.001
1 64 116 125 127 129 136 138* 455 458 527 703
Ac
id
¥¥¥¥¥¥¥¥
0.1
1
10
s s
urviva
l (%
)
C
0.0001
0.001
0.01
Bile s
alt
1 64 116 125 127 129 136 138* 455 458 527 703
3-day cultures D+1 D+8 D+15
P.fs P.a P.ff P.fs P.ff P.fs P.fs P.fs P.j P.fs P.ff P.fs
Dairy propionibacteria strains
Fig. 6. Stability of bacterial population and digestive stress tolerance of 12 dairy propionibacteria strains in fermented milk. Twelve dairy propionibacteria strains were cultivated inmilk during 3 days at 30 �C (,) and stored at 4 �C during 1 (D þ 1, ), 8 (D þ 8, ) and 15 (D þ 15, ) days. (A) Concentration of dairy propionibacteria strains in fermented milkwas determined. (B) Survival following acid challenge was determined as in Fig. 5. U: undetectable propionibacteria survival (<0.0001%). (C) Survival following bile salts challengewas determined as in Fig. 5. The results are means of at least two independent experiments. * Strain used for the in vivo trial.
F.J. Cousin et al. / Food Microbiology 32 (2012) 135e146 143
probiotic effect of a particular bacterial species is then difficult toelucidate. The aim of our study was to develop a dairy productfermented exclusively by a dairy propionibacterium.
Milk and dairy products are widely considered as a good pro-biotic vectors as they protect bacteria from stresses undergone inthe digestive tract (Ranadheera et al., 2010). This has been evi-denced for dairy propionibacteria in human trials (Hervé et al.,2007; Jan et al., 2002b). In addition to digestive stress protection,several studies reported that dairy products favour dairy propio-nibacteria functional properties in yogurt (Ekinci and Gurel, 2008),conjugated linoleic acid (CLA) synthesis (Xu et al., 2004), trehalosesynthesis (Cardoso et al., 2004), vitamin synthesis (Holasova et al.,2004; Hugenschmidt et al., 2011; Kalfirtova and Sovjak, 2005;LeBlanc et al., 2006; VanWyk et al., 2011) and pathogenic organisminhibition (Suomalainen and Mayra-Makinen, 1999; Lyon et al.,1993).
The first step of this work was to determine the nutritionalrequirements in nitrogen and carbon for dairy propionibacteria togrow inmilk, as the sole microorganisms. UF was used to determinethe appropriate concentrations in casein hydrolysate and lactate. Insupplemented UF, generation time was shorter and maximalOD650 nm was higher than in the reference medium YEL. However,the propionibacteria concentrations reached, revealed by counting,was higher in YEL. A higher production of exopolysaccharides (EPS)in UF than in YEL, perturbing OD650 nmmeasurements, could explainthis result. Indeed, dairy propionibacteria are able to synthesize EPS(Deutsch et al., 2008, 2010), which production occurs in UF and ishighly variable as a function of environmental factors (Gorret et al.,2001b, 2001a). Lactate supplementation facilitates the growth oflactose� strains. Surprisingly, the lactoseþ strain concentrationsreached decreased with high amounts of lactate. Indeed, the ener-getic conversion ratio is better with lactose thanwith lactate, which
A
20
25
30
iva
l in
fe
ce
s (
%)
4
5
6
eria
lo
n c
on
te
nt)
5
10
15
ro
pio
nib
ac
te
ria
su
rv
1
2
3
Pro
pio
nib
act
lo
g1
0c
fu
/g
o
f d
rie
d c
o
Lyophilizate group Fermented milk group
0Day 7 Day 14
P
0Colon content
(
Sterile milk group
4
5
6
te
nt)
400
500
600
ra
tio
n
n c
on
te
nt)
B
*
*
*
*
2
3
CF
As
co
nc
en
tra
tio
n
l/g
o
f d
rie
d c
olo
n c
on
200
300
SC
FA
s c
on
ce
nt
(m
mo
l/g
o
f d
rie
d c
olo
*
#
0
1
Branchedbutyric acid
Branchedvaleric acid
Caproic acid
S
(m
mo
0
100
Acetic acid
Propionicacid
Butyric acid
Valeric acid
Fig. 7. Fermented milk allows P. freudenreichii survival and activity in vivo. Piglets were daily gavaged with sterile milk (,, n ¼ 8) or P.fs BIA138 (2E þ 10 cfu day�1) either asa lyophilizate ( , n ¼ 8) or within fermented milk ( , n ¼ 8). (A) In vivo survival of P. freudenreichii. Propionibacteria populations were determined on selective medium in feces(days 7 & 14 of gavages) and colon content (at slaughter). The results are means of survival rate for feces and means of log10 propionibacteria concentrations in colon contents.(B) Concentrations of SCFAs in colon contents. The results are means � SEM of two analyses per piglets. * P < 0.05, # P < 0.1. Isocaproic acid was added as internal standard to checkthe repeatability of analysis. The repeatability coefficient of isocaproic acid amount was 2.27%.
F.J. Cousin et al. / Food Microbiology 32 (2012) 135e146144
is the preferential substrate of dairy propionibacteria (Thierry et al.,2011). So, with no or a few lactate addition, the lactoseþ strainsfermented lactose and reached the highest population levels. Finally,a supplementation of 50 mM lactate and 5 g L�1 casein hydrolysatewas chosen and applied tomilk, as its allows growth of all the testedstrains and species.
In this study, we developed a dairy product exclusively fer-mented by a dairy propionibacterium strain. In this new fermentedmilk, dairy propionibacteria reached 1E þ 9 cfu mL�1. This is closeto the 1E þ 8 cfu g�1e1E þ 9 cfu g�1 concentration generally foundin Swiss-type cheeses ready for consumption (Parayre et al., 2007;Bachmann et al., 2002; Falentin et al., 2010b). To our knowledge,this new fermented milk is the first food grade dairy vector con-taining dairy propionibacteria, as the sole microorganism. Thedesign of appropriate vectors for probiotics delivery is a permanentfield of research (Sanders and Marco, 2010). The European Unionreports that the minimum level of probiotic bacteria should be setto 1E þ 07 cfu g�1 (EU/AGRI/38743/2003rev3), for a commonlyprobiotic daily dose between 0.1 and 10 billion bacteria (Sandersand Marco, 2010). Cummings et al. showed that the needed doseof probiotics to exert a health benefit is around 1010 to 1011 bacteria(Cummings, 2009). This allows to reach a population of 1 millionbacteria per gram in the colon, population level required for anexpected probiotic effect (Marteau et al., 1993).
4.2. Survival during storage at 4 �C
Cold adaptation is an important parameter for strain selectionbecause of probiotic storage at 4 �C. Indeed, the probiotic definitionstipulates that they must be consumed “live” (FAO/WHO, 2002). Inthe present study, the new fermented milk preserved dairy pro-pionibacteria viability during storage at 4 �C, as shown byenumeration and live/dead labelling. The stability of dairy propio-nibacteria population during cold storage has previously beenevidenced in dairy products. Maity et al. showed that a fermentedwhey contained 2.9Eþ 08 cfu mL�1 of dairy propionibacteria at theend of the fermentation and 9.3Eþ 07 cfumL�1 after 15 days at 5 �C(Maity et al., 2008). In addition, Kalfirtova et al. showed thatP. freudenreichii population stayed at the same level in differentdairy products during 14 days of storage at 4 �C (Kalfirtova andSovjak, 2005). During Emmental cheese ripening in a cold room,dairy propionibacteria are still alive even after a long-term storageand remain metabolically active, contributing to the aromacompound production in cheese (Dalmasso et al., 2012).
4.3. Tolerance towards digestive stresses in vitro and in vivo
Tolerance to gut environment (pH, bile salts, enzymes.)constitute the bottleneck of probiotic efficiency. Indeed, dairy
F.J. Cousin et al. / Food Microbiology 32 (2012) 135e146 145
propionibacteria probiotic potentialities need live and metaboli-cally active bacteria. Dairy propionibacteria survival under diges-tive stress conditions can be grealty enhanced by a brief exposure tothe same stress at a non-lethal level (Jan et al., 2000, 2001;Leverrier et al., 2004; Leverrier et al., 2003) or by inclusion in anappropriate vector, such as dairy product (Leverrier et al., 2005).Consequently, an essential part of our study was to determine theefficiency of our new fermented milk to protect propionibacteriafrom digestive stresses. Using this fermented milk, the populationlevels of live dairy propionibacteria in the piglet colon were inaccordance with the probiotic population level required for anexpected probiotic effect (Marteau et al., 1993). In human trials,dairy propionibacteria presented similar survival rate during guttransit (Hervé et al., 2007; Jan et al., 2002b;Myllyluoma et al., 2005;Suomalainen et al., 2008), and metabolic activity has been evi-denced (Hervé et al., 2007). Bifidobacterium animalis DN-173 010was also provided in a probiotic yogurt versus lyophilizate andshowed no difference in the survival rates in a randomized study inhealthy adults (Rochet et al., 2008). In a human trial, Saxelin et al.studied the survival of different probiotics administered as cellu-lose capsules, yoghurt or cheese. They showed that highest fecalquantity of P. freudenreichii subsp. shermanii JS was obtained afteryogurt consumption (Saxelin et al., 2010). In the same study, thesurvival of Lactobacillus strains was similar with the 3 vectors whileBifidobacterium survival was highest with yoghurt. The presentwork confirmed indeed the efficiency to deliver a high amount oflive dairy propionibacteria in the gut using a dairy product. It couldbe interesting to test the in vivo survival rates of P. freudenreichiiwith a cheese, such as Emmental, containing exclusively one strainof this probiotic bacterium. In addition, the UF fermented byP. freudenreichii was also used in another in vivo trial involving pigsand allowed similar survival rates as the fermented milk (Cousinet al., unpublished results).
The end-products of propionibacterial obligatory fermentativemetabolism are the SCFAs acetate and propionate. These metabo-lites were significantly enhanced in colon content as a result offermented milk consumption in this work. By contrast, this was notthe case for lyophilized dairy propionibacteria consumption. Thiscan translate a metabolic activity of dairy propionibacteria in thecolon, when provided under the form a fermented dairy product.Indeed, it has been shownpreviously that themost robust strains ofP. freudenreichii, provided under the form of live cultures or infermented milk, keep an active metabolism within the intestine ofhumanized rats (Lan et al., 2007) and of humans (Hervé et al.,2007). However, the results obtained here suggest that the samestrain, provided under the form of a lyophilizate, does not exert thesame effect. This confirms the impact of the probiotic deliveryvehicle on bacterial fitness andmetabolic activity when the colon isreached. Another explanationwould be the amounts of acetate andpropionate contained in the fermented milk, but this seems lessprobable, as ingested SCFAs are rapidly absorbed by the smallintestine mucosa of pigs andwould not reach the colon (Claus et al.,2007). Recent reviews focused on the health promoting SCFAs,including propionic acid (Al-Lahham et al., 2010; Arora et al., 2011;Hosseini et al., 2011). The production of propionic acid in the gutcan be considered of probiotic potentialities in different areas suchas the maintenance of a normal colonic epithelium, the preventionof colorectal cancer (Jan et al., 2002a; Lan et al., 2008), the healingof inflammatory disorders such as inflammatory bowel diseases(IBD) (Tedelind et al., 2007; Uchida and Mogami, 2005), and colo-nization by pathogenic bacteria and cholesterol. Regarding aceticacid, the improvement of its concentration in gut could also displayhealth benefits for diabetes (Sakakibara et al., 2006), hypercholes-terolemia (Fushimi et al., 2006) or hypertension (Kondo et al.,2001).
5. Conclusions
This work led to the development of a new fermented milkcontaining exclusively a dairy propionibacterium strain andallowing preclinical and clinical trials in human. Moreover, all thetested strains and species grow in both products (UF and milk) andit will permit strain selection in vitro. Most of the human trialsinvolving dairy propionibacteria have been conducted withmixtures of probiotic bacteria from different genera. The develop-ment of a pure culture of dairy propionibacteria in a food gradevector will allow identification of specific dairy propionibacteriaprobiotic potential. Depending of the sought properties, the fer-mented product can be either close to neutral pH (lactose� strainand lactate supplementation) or acidified (lactoseþ strain withoutlactate). The fermented UF can also be used to study the impact ofmilk proteins on probiotic physiology (such as digestive stresstolerance) by comparing to the fermented whole milk. In addition,this fermented UF can also be used in studies in which the milkproteins could interfere with the expected health benefit, such asthe immunomodulation.
Acknowledgements
We thank Jacques Fauquant for milk ultrafiltrate preparationandMarie-Noëlle Madec for microscopy. This workwas financed byConseil Régional de Bretagne through the CBB développement callfor projects and by the Centre National Interprofessionnel del’Économie Laitière (CNIEL) through by the Scientific Committee ofSyndifrais call for projects. F.J.C. received a grant from CNIEL. Theauthors also thank the Knock for kindly lending meeting room.
References
Al-Lahham, S.H., Peppelenbosch, M.P., Roelofsen, H., Vonk, R.J., Venema, K., 2010.Biological effects of propionic acid in humans; metabolism, potential applica-tions and underlying mechanisms. BBA-Mol. Cell. Biol. L. 1801, 1175e1183.
Arora, T., Sharma, R., Frost, G., 2011. Propionate. Anti-obesity and satiety enhancingfactor? Appetite 56, 511e515.
Bachmann, H.P., Bütikofer, U., Isolini, D., 2002. Swiss-type cheeses. In: Roginski, H.,Fuquay, J.W., Fox, P.F. (Eds.), Encyclopedia of Dairy Sciences. Academic Press,London, pp. 363e371.
Baron, F., Cochet, M.F., Ablain, W., Grosset, N., Madec, M.N., Gonnet, F., et al., 2006.Rapid and cost-effective method for microorganism enumeration based onminiaturization of the conventional plate-counting technique. Lait 86, 251e257.
Cardoso, F.S., Gaspar, P., Hugenholtz, J., Ramos, A., Santos, H., 2004. Enhancement oftrehalose production in dairy propionibacteria through manipulation of envi-ronmental conditions. Int. J. Food Microbiol. 91, 195e204.
Claus, R., Gþþnthner, D., Letzguþf , H., 2007. Effects of feeding fat-coated butyrateon mucosal morphology and function in the small intestine of the pig. J. Anim.Physiol. Anim. Nutr. 91, 312e318.
Cousin, F.J., Mater, D.D.G., Foligné, B., Jan, G., 2011. Dairy propionibacteria as humanprobiotics: a review of recent evidence. Dairy Sci. Technol. 91, 1e26.
Cummings, J.H., 2009. Probiotics: better health from ‘good’ bacteria? Nutr. Bull. 34,198e202.
Dalmasso, M., Aubert, J., Briard-Bion, V., Chuat, V., Deutsch, S.M., Even, S., et al.,2012. A temporal -omic Study of Propionibacterium freudenreichii CIRM-BIA1adaptation strategies in conditions mimicking cheese ripening in the cold.PLoS ONE 7, e29083.
Deutsch, S.M., Falentin, H., Dols-Lafargue, M., LaPointe, G., Roy, D., 2008. Capsularexopolysaccharide biosynthesis gene of Propionibacterium freudenreichii subspshermanii. Int. J. Food Microbiol. 125, 252e258.
Deutsch, S.M., Le Bivic, P., Hervé, C., Madec, M.N., LaPointe, G., Jan, G., et al., 2010.Correlation of the capsular phenotype in Propionibacterium freudenreichii withthe level of expression of gtf, a unique polysaccharide synthase-encoding gene.Appl. Environ. Microbiol. 76, 2740e2746.
Dupuis, C., Corre, C., Boyaval, P., 1995. Proteinase activity of dairy Propionibacterium.Appl. Microbiol. Biotechnol. 42, 750e755.
Ekinci, F.Y., Gurel, M., 2008. Effect of using propionic acid bacteria as an adjunctculture in yogurt production. J. Dairy Sci. 91, 892e899.
El-Nezami, H.S., Polychronaki, N.N., Ma, J., Zhu, H.L., Ling, W.H., Salminen, E.K., et al.,2006. Probiotic supplementation reduces a biomarker for increased risk of livercancer in young men from Southern China. Am. J. Clin. Nutr. 83, 1199e1203.
El-Ssoda, M., Ziada, N., Ezzat, N., 1992. The intracellular peptide-hydrolase system ofpropionibacterium. Microbios 72, 65e74.
F.J. Cousin et al. / Food Microbiology 32 (2012) 135e146146
Falentin, H., Deutsch, S.M., Jan, G., Loux, V., Thierry, A., Parayre, S., et al., 2010a.The complete genome of Propionibacterium freudenreichii CIRM-BIA1, a hardyactinobacterium with food and probiotic applications. PLoS ONE 5, e11748.
Falentin, H., Postollec, F., Parayre, S., Henaff, N., Le, B.P., Richoux, R., et al., 2010b.Specific metabolic activity of ripening bacteria quantified by real-time reversetranscription PCR throughout Emmental cheese manufacture. Int. J. FoodMicrobiol. 144, 10e19.
FAO/WHO, 2002. Guidelines for the Evaluation of Probiotics in Food. Food andAgriculture Organization of the United Nations and World Health OrganizationWorking Group Report, Geneva, Switz.
Fushimi, T., Suruga, K., Oshima, Y., Fukiharu, M., Tsukamoto, Y., Goda, T., 2006.Dietary acetic acid reduces serum cholesterol and triacylglycerols in rats fedacholesterol-rich diet. Br. J. Nutr. 95, 916e924.
Gagnaire, V., Molle, D., Sorhaug, T., Leonil, J., 1999. Peptidases of dairy propionic acidbacteria. Lait 79, 43e57.
Gorret, N., Maubois, J.L., Engasser, J.M., Ghoul, M., 2001a. Study of the effects oftemperature, pH and yeast extract on growth and exopolysaccharides produc-tion by Propionibacterium acidi-propionici on milk microfiltrate using a responsesurface methodology. J. Appl. Microbiol. 90, 788e796.
Gorret, N., Maubois, J.L., Ghoul, M., Engasser, J.M., 2001b. Exopolysaccharideproduction by Propionibacterium acidi-propionici on milk microfiltrate. J. Appl.Microbiol. 90, 779e787.
Hervé, C., Fondrevez, M., Cheron, A., Barloy-Hubler, F., Jan, G., 2007. Trans-carboxylase mRNA: a marker which evidences P. freudenreichii survival andmetabolic activity during its transit in the human gut. Int. J. Food Microbiol. 113,303e314.
Hojo, K., Yoda, N., Tsuchita, H., Ohtsu, T., Seki, K., Taketomo, N., et al., 2002. Effect ofingested culture of Propionibacterium freudenreichii ET-3 on fecal microflora andstool frequency in healthy females. Biosci. Microflora 21, 115e120.
Holasova, M., Fiedlerova, V., Roubal, P., Pechacova, M., 2004. Biosynthesis of folatesby lactic acid bacteria and propionibacteria in fermented milk. Czech J. Food Sci.22, 175e181.
Hosseini, E., Grootaert, C., Verstraete,W., Van deWiele, T., 2011. Propionate as a health-promoting microbial metabolite in the human gut. Nutr. Rev. 69, 245e258.
Hugenschmidt, S., Schwenninger, S.M., Lacroix, C., 2011. Concurrent high productionof natural folate and vitamin B12 using a co-culture process with Lactobacillusplantarum SM39 and Propionibacterium freudenreichii DF13. Process Biochem.46, 1063e1070.
Jan, G., Belzacq, A.S., Haouzi, D., Rouault, A., Metivier, D., Kroemer, G., et al., 2002a.Propionibacteria induce apoptosis of colorectal carcinoma cells via short-chainfatty acids acting on mitochondria. Cell Death Differ. 9, 179e188.
Jan, G., Leverrier, P., Pichereau, V., Boyaval, P., 2001. Changes in protein synthesis andmorphology during acid adaptation of Propionibacterium freudenreichii. Appl.Environ. Microbiol. 67, 2029e2036.
Jan, G., Leverrier, P., Proudy, I., Roland, N., 2002b. Survival and beneficial effects ofpropionibacteria in the human gut: in vivo and in vitro investigations. Lait 82,131e144.
Jan, G., Rouault, A., Maubois, J.L., 2000. Acid stress susceptibility and acid adaptationof Propionibacterium freudenreichii subsp shermanii. Lait 80, 325e336.
Kalfirtova, P., Sovjak, R., 2005. Improve the nutritional value of fermented dairyproducts by nutraceutical-producing food-grade microorganisms. Agric. Trop.Subtrop. 38, 22e27.
Kondo, S., Tayama, K., Tsukamoto, Y., Ikeda, K., Yamori, Y., 2001. Antihypertensiveeffects of acetic acid and vinegar on spontaneously hypertensive rats. Biosci.Biotechnol. Biochem. 65, 2690e2694.
Lan, A., Bruneau, A., Bensaada, M., Philippe, C., Bellaud, P., Rabot, S., et al., 2008.Increased induction of apoptosis by Propionibacterium freudenreichii TL133 incolonic mucosal crypts of human microbiota-associated rats treated with1,2-dimethylhydrazine. Br. J. Nutr. 100, 1251e1259.
Lan, A., Bruneau, A., Philippe, C., Rochet, V., Rouault, A., Hervé, C., et al., 2007.Survival and metabolic activity of selected strains of Propionibacterium freu-denreichii in the gastrointestinal tract of human microbiota-associated rats. Br. J.Nutr. 97, 714e724.
Langsrud, T., Reinbold, G.W., 1973. Flavor development and microbiology of Swisscheese. A review. Part 3 ripening and flavor production. J. Milk Food Technol.36, 593e609.
Langsrud, T., Sorhaug, T., Vegarud, G.E., 1995. Protein degradation and amino acidmetabolism by propionibacteria. Lait 75, 325e330.
LeBlanc, J.G., Rutten, G., Bruinenberg, P., Sesma, F., de Giori, G.S., Smid, E.J., 2006.A novel dairy product fermented with Propionibacterium freudenreichiiimproves the riboflavin status of deficient rats. Nutrition 22, 645e651.
Lemée, R., Gagnaire, V., Maubois, J.L., 1998. Strain variability of the cell-freeproteolytic activity of dairy propionibacteria towards beta-casein peptides.Lait 78, 227e240.
Leverrier, P., Dimova, D., Pichereau, V., Auffray, Y., Boyaval, P., Jan, G.L., 2003.Susceptibility and adaptive response to bile salts in Propionibacterium freu-denreichii: physiological and proteomic analysis. Appl. Environ. Microbiol. 69,3809e3818.
Leverrier, P., Fremont, Y., Rouault, A., Boyaval, P., Jan, G., 2005. In vitro tolerance todigestive stresses of propionibacteria: influence of food matrices. Food Micro-biol. 22, 11e18.
Leverrier, P., Vissers, J.P.C., Rouault, A., Boyaval, P., Jan, G., 2004. Mass spectrometryproteomic analysis of stress adaptation reveals both common and distinctresponse pathways in Propionibacterium freudenreichii. Arch. Microbiol. 181,215e230.
Lyon, W.J., Sethi, J.K., Glatz, B.A., 1993. Inhibition of psychrotrophic organisms bypropionicin Plg-1, a bacteriocin produced by Propionibacterium thoenii. J. DairySci. 76, 1506e1513.
Maity, T., Rakesh, K., Misra, A., 2008. Development of healthy whey drink withLactobacillus rhamnosus, Bifidobacterium bifidum and Propionibacterium freu-denreichii subsp. shermanii. Mljekarstvo 58, 315e325.
Malik, A.C., Reinbold, G.W., Vedamuthu, E.R., 1968. An evaluation of the taxonomy ofPropionibacterium. Can. J. Microbiol. 14, 1185e1191.
Marteau, P., Pochart, P., Bouhnik, Y., Rambaud, J.C., 1993. The fate and effects oftransiting, nonpathogenic microorganisms in the human intestine. World Rev.Nutr. Diet. 74, 1e21.
Meyer, H.L., 2007. Les probiotiques dans le monde. Revue Laitière Française 676,20e23.
Michalski, M.C., Leconte, N., Briard-Bion, V., Fauquant, J., Maubois, J.L.,Goudedranche, H., 2006. Microfiltration of raw whole milk to select fractionswith different fat globule size distributions: process optimization and analysis.J. Dairy Sci. 89, 3778e3790.
Moughan, P.J., Birtles, M.J., Cranwell, P.D., Smith, W.C., Pedraza, M., 1992. The pigletas a model animal for studying aspects of digestion and absorption in milk-fedhuman infants. World Rev. Nutr. Diet. 67, 40e113.
Myllyluoma, E., Veijola, L., Ahlroos, T., Tynkkynen, S., Kankuri, E., Vapaatalo, H.,et al., 2005. Probiotic supplementation improves tolerance to Helicobacter pylorieradication therapy e a placebo-controlled, double-blind randomized pilotstudy. Aliment. Pharmacol. Ther. 21, 1263e1272.
Ozer, B.H., Kirmaci, H.A., 2010. Functional milks and dairy beverages. Int. J. DairyTechnol. 63, 1e15.
Parayre, S., Falentin, H.þ., Madec, M.N. l., Sivieri, K., Le Dizes, A.S., Sohier, D. l, et al.,2007. Easy DNA extraction method and optimisation of PCR-TemporalTemperature Gel Electrophoresis to identify the predominant high and lowGC-content bacteria from dairy products. J. Microbiol. Methods 69, 431e441.
Perez-Chaia, A., Zarate, G., 2005. Dairy propionibacteria from milk or cheese dietsremain viable and enhance propionic acid production in the mouse cecum. Lait85, 85e98.
Piveteau, P., 1999. Metabolism of lactate and sugars by dairy propionibacteria:a review. Lait 79, 23e41.
Piveteau, P., Condon, S., Cogan, T.M., 2000. Inability of dairy propionibacteria togrow in milk from low inocula. J. Dairy Res. 67, 65e71.
Ranadheera, R.D.C.S., Baines, S.K., Adams, M.C., 2010. Importance of food in pro-biotic efficacy. Food Res. Int. 43, 1e7.
Rochet, V., Rigottier-Gois, L., Ledaire, A., Andrieux, C., Sutren, M., Rabot, S., et al.,2008. Survival of Bifidobacterium animalis DN-173 010 in the faecal microbiotaafter administration in lyophilised form or in fermented product e A rando-mised study in healthy adults. J. Mol. Microbiol. Biotechnol. 14, 128e136.
Sahlstrom, S., Espinosa, C., Langsrud, T., Sorhaug, T., 1989. Cell-wall, membrane, andintracellular peptidase activities of Propionibacterium shermanii. J. Dairy Sci. 72,342e350.
Sakakibara, S., Yamauchi, T., Oshima, Y., Tsukamoto, Y., Kadowaki, T., 2006. Aceticacid activates hepatic AMPK and reduces hyperglycemia in diabetic KK-A(y)mice. Biochem. Biophys. Res. Commun. 344, 597e604.
Sanders, M.E., Marco, M.L., 2010. Food formats for effective delivery of probiotics.Annu. Rev. Food Sci. Technol. 1, 65e85.
Saxelin, M., Lassig, A., Karjalainen, H., Tynkkynen, S., Surakka, A., Vapaatalo, H.,et al., 2010. Persistence of probiotic strains in the gastrointestinal tract whenadministered as capsules, yoghurt, or cheese. Int. J. Food Microbiol. 144,293e300.
Suomalainen, T., Sigvart-Mattila, P., Matto, J., Tynkkynen, S., 2008. In vitro andin vivo gastrointestinal survival, antibiotic susceptibility and genetic identifi-cation of Propionibacterium freudenreichii ssp shermanii JS. Int. Dairy J. 18,271e278.
Suomalainen, T.H., Mayra-Makinen, A.M., 1999. Propionic acid bacteria as protectivecultures in fermented milks and breads. Lait 79, 165e174.
Suzuki, A., Mitsuyama, K., Koga, H., Tomiyasu, N., Masuda, J., Takaki, K., et al., 2006.Bifidogenic growth stimulator for the treatment of active ulcerative colitis:a pilot study. Nutrition 22, 76e81.
Tedelind, S., Westberg, F., Kjerrulf, M., Vidal, A., 2007. Anti-inflammatory propertiesof the short-chain fatty acids acetate and propionate: a study with relevance toinflammatory bowel disease. World J. Gastroenterol. 13, 2826e2832.
Thierry, A., Deutsch, S.M., Falentin, H., Dalmasso, M., Cousin, F.J., Jan, G., 2011. Newinsights into physiology and metabolism of Propionibacterium freudenreichii. Int.J. Food Microbiol. 149, 19e27.
Thierry, A., Maillard, M.B., Yvon, M., 2002. Conversion of L-leucine to isovaleric acidby Propionibacterium freudenreichii TL 34 and ITGP23. Appl. Environ. Microbiol.68, 608e615.
Uchida, M., Mogami, O., 2005. Milk whey culture with Propionibacterium freu-denreichii ET-3 is effective on the colitis induced by 2,4,6-trinitrobenzenesulfonic acid in rats. J. Pharmacol. Sci. 99, 329e334.
Van Wyk, J., Witthuhn, R.C., Britz, T.J., 2011. Optimisation of vitamin B-12 and folateproduction by Propionibacterium freudenreichii strains in kefir. Int. Dairy J. 21,69e74.
Xu, S., Boylston, T.D., Glatz, B.A., 2004. Effect of lipid source on probiotic bacteriaand conjugated linoleic acid formation in milk model systems. J. Am. Oil Chem.Soc. 81, 589e595.
Xu, S., Boylston, T.D., Glatz, B.A., 2005. Conjugated linoleic acid content andorganoleptic attributes of fermented milk products produced with probioticbacteria. J. Agric. Food Chem. 53, 9064e9072.
