ARTICLE IN PRESS

1075-9964/$ - se

doi:10.1016/j.an

�CorrespondPharmaceutical

University of S

fax: +5511 381

E-mail addr

Anaerobe 13 (2007) 200–207

www.elsevier.com/locate/anaerobe

Food microbiology

Functional petit-suisse cheese: Measure of the prebiotic effect

Haıssa R. Cardarellia,�, Susana M.I. Saadb, Glenn R. Gibsona, Jelena Vulevica

aFood Microbial Sciences Unit, Department of Food Biosciences, The University of Reading, Whiteknights, Reading RG6 6AP, UKbDepartment of Biochemical Pharmaceutical Technology, Faculty of Pharmaceutical Sciences, University of Sao Paulo,

Av. Prof. Lineu Prestes, 580, B16, 05508-000, SP, Brazil

Received 13 February 2007; received in revised form 12 April 2007; accepted 4 May 2007

Available online 21 May 2007

Abstract

Prebiotics and probiotics are increasingly being used to produce potentially synbiotic foods, particularly through dairy products as

vehicles. It is well known that both ingredients may offer benefits to improve the host health. This research aimed to evaluate the

prebiotic potential of novel petit-suisse cheeses using an in vitro fermentation model. Five petit-suisse cheese formulations combining

candidate prebiotics (inulin, oligofructose, honey) and probiotics (Lactobacillus acidophilus, Bifidobacterium lactis) were tested in vitro

using sterile, stirred, batch culture fermentations with human faecal slurry. Measurement of prebiotic effect (MPE) values were generated

comparing bacterial changes through determination of maximum growth rates of groups, rate of substrate assimilation and production

of lactate and short chain fatty acids. Fastest fermentation and high lactic acid production, promoting increased growth rates of

bifidobacteria and lactobacilli, were achieved with addition of prebiotics to a probiotic cheese (made using starter+probiotics). Addition

of probiotic strains to control cheese (made using just a starter culture) also resulted in high lactic acid production. Highest MPE values

were obtained with addition of prebiotics to a probiotic cheese, followed by addition of prebiotics and/or probiotics to a control cheese.

Under the in vitro conditions used, cheese made with the combination of different prebiotics and probiotics resulted in the most

promising functional petit-suisse cheese. The study allowed comparison of potentially functional petit-suisse cheeses and screening of

preferred synbiotic potential for future market use.

r 2007 Elsevier Ltd. All rights reserved.

Keywords: In vitro fermentation; Prebiotic effect; Petit-suisse cheese; Synbiotic; Functional food

1. Introduction

The development of foods that promote health and wellbeing is one of the key research priorities of food industryand has favoured consumption of foods enriched withphysiologically active components such as probiotics andprebiotics [1,2]. They have an important role in gutfermentation by influencing the microbiota compositionand fermentation metabolites, and consequently by con-tributing to local and systemic effects in humans [3].Probiotics are ‘live organisms that, when administered in

e front matter r 2007 Elsevier Ltd. All rights reserved.

aerobe.2007.05.003

ing author. Present address: Department of Biochemical

Technology, Faculty of Pharmaceutical Sciences,

ao Paulo, SP, Brazil. Tel.: +55 11 30912378;

56386.

ess: hrcarda@terra.com.br (H.R. Cardarelli).

adequate amounts, confer a health benefit on the host’ [4].The beneficial influences of probiotics on human gutmicrobiota include factors such as antagonistic effectsagainst pathogens, competitive exclusion and immuneeffects [3]. Bifidobacteria and lactobacilli are the mainnatural inhabitants of the human gut and claimedprobiotics. They have been traditionally incorporated intoa variety of food products, including cheese [5–7]. Onefurther approach to increase the number of beneficialbacteria in the intestinal microbiota is through the use ofprebiotics. Gibson et al. [8] reviewed the prebiotic conceptas ‘a selectively fermented ingredient that allows specificchanges, both in the composition and/or activity in thegastrointestinal microbiota that confers benefits upon hostwellbeing and health’. Amongst the prebiotics, fructo-oligosaccharides (FOS) or oligofructose and inulin typefructans have been the most investigated thus far [9]. The

ARTICLE IN PRESSH.R. Cardarelli et al. / Anaerobe 13 (2007) 200–207 201

effects of both inulin and oligofructose upon the humanmicrobiota have been widely studied both in vivo and in

vitro, and selective fermentation by bacterial microbiotahas been frequently reported, mainly by bifidobacteria and,to a lesser extent, lactobacilli [10]. Honey also containsvarious oligosaccharides (4.2%) [11] and may possiblyserve as a prebiotic agent by suppressing potentiallydeleterious bacteria among the gastrointestinal microbiota[12,13]. Pre- and probiotics may be combined in a foodproduct, called synbiotics [14].

In Brazil, petit-suisse cheese is a food dessert designed totarget children as consumers. It is pleasantly accepted andhas been increasingly consumed [15]. The production of apetit-suisse cheese capable of generating a potentiallysynbiotic effect, due to the incorporation of L. acidophilus,

B. lactis and the prebiotics inulin, oligofructose and honeyis promising.

The prebiotic functionality of different substrates, forexample, dairy products, may be evaluated throughfermentation studies. Fermentation studies describe theprebiotic capabilities of different substrates throughswitches in the composition of purportedly beneficial anddetrimental microbiota [16]. The most commonly used in

vitro models to study anaerobic fermentation of carbohy-drates by mixed bacterial populations, particularly faecalbacteria, are batch and continuous culture fermentationsystems [8]. Such models are valuable for studying the gutmicrobiota as fermentation substrates can be added undercontrolled condition, which may not be feasible in a humantrial [17]. Moreover, these studies provide measurementsthat can be used to assess the prebiotic effect byquantifying the microbial population changes. The pre-biotic index (PI), based upon changes in bifidobacteria,lactobacilli, clostridia and bacteroides [18], was the firstand simplest example of quantitative analysis. PI values arecalculated through an equation that involves the relativemicrobial counts over the total bacterial numbers of fourgenera of bacteria found in the human colon which aregiven equal weight in the equation. It is assumed that anincrease in populations of bifidobacteria and/or lactobacilliis a positive effect while an increase in clostridia(hystoliticum subgroup) and bacteroides is a negativeeffect [18]. However, the beneficial effect of prebiotics isnot only seen through their ability to increase or influencenumbers of bacteria but also through activities of thesebacteria. Namely, the production of short-chain fatty acids(SCFA) and particularly lactic acid in the colon are seen asimportant factors when determining a positive prebioticeffect and have generated much interest [19]. The measureof the prebiotic effect (MPE) is an extension of the PIequation to include quantitative changes in more bacterialgroups, fermentation end products, such as SCFA andsubstrate assimilation [16]. MPE, therefore, provides abetter in vitro evaluation of the prebiotic potential ofvarious carbohydrate-based substrates. The present studyaimed to apply the MPE theory to evaluate the in vitro

prebiotic potential of novel petit-suisse formulations.

2. Materials and methods

2.1. MPE theory

The MPE concept and theory involved are described byVulevic et al. [16]. The equations necessary for theapplication of the MPE concept are listed below:

Substrate assimilation:

St ¼ S0 � Art, (1)

where St is the substrate concentration after the timeinterval t (hours), S0 is the initial substrate assimilation andAr is the rate of substrate assimilation (h�1). The maximumAr is calculated during the exponential phase of bacterialgrowth.

Changes in bacterial populations:

ln Nt ¼ ln N0 þ mt, (2)

where N is the number of bacteria after the time interval t

(h), N0 is the initial number of bacteria and m is the specificgrowth rate (h�1).During the exponential phase nutrients are in excess and

bacteria are growing at their maximum specific growthrate, mmax, under the prevailing conditions. Therefore, mmax

describes the ability of bacteria to grow under specificconditions and would vary for different bacteria andsubstrates. Changes in bacterial populations may becalculated through incorporating the mmax into the PIequation [18] yielding a modified prebiotic index (PIm),according to the following:

PIm ¼ mmaxBif þ mmaxLacþ mmaxErec� mmaxBac

� mmaxClos� mmaxEcþ mmaxSRB, ð3Þ

where Bif is bifidobacteria, Lac is lactobacilli, Erec iseubacteria, Bac is bacteroides, Clos is clostridia (histolyti-

cum subgroup), Ec is Escherichia coli and SRB is sulphate-reducing bacteria. An increase in the maximum growth rateof bifidobacteria, lactobacilli and eubacteria is a positiveeffect whilst an increase in the remaining groups gives anegative effect.

SCFA production: Maximum SCFA production isexpected at the end of the exponential growth phase.Therefore, by calculating the mass of acetate, propionate,butyrate and lactate produced by the microbiota as awhole, at this time point, it is possible to compare the effectof different substrates upon total SCFA:

TSCFA ¼ Aþ Bþ Pþ L, (4)

where A is acetate, B is butyrate, P is propionate and L islactate.The ratio of lactic acid production over the total SCFA

production may provide a qualitative as well as quantita-tive assessment of each substrate tested. Lactic acid isusually associated with the prebiotic effect as principalmetabolite of beneficial lactic acid bacteria [20,21]. Thisratio is calculated as follows:

Ratio ¼ dL=dTSCFA, (5)

ARTICLE IN PRESSH.R. Cardarelli et al. / Anaerobe 13 (2007) 200–207202

where d is the difference between the initial mass and themass at the sampling time point.

MPE equation: Eqs. (1), (3) and (5) are combined into asingle equation which results in a single number describingthe activity of the microbiota. This number is then ascribeda positive or negative value, based upon the microbiota,and referred to as the MPE value:

MPE ¼1

2

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffix2y2 þ x2z2 þ y2z2

p, (6)

where x is the rate of substrate assimilation (Ar) (Eq. (1)), y

is the modified PI (PIm) (Eq. (3)) and z is the ratio of lactateover the total SCFA (Eq. (5)).

2.2. Petit-suisse cheese manufacture

Two pilot-scale petit-suisse cheese-making trials wereperformed (C and P), both using Streptococcus thermo-

philus (TA 040, Danisco, France) as a starter culture, oneof them as a single culture (control cheese ¼ C). Only P(probiotic cheese) was supplemented with the probioticculture of L. acidophilus (La-5, Christian Hansen,Denmark), and the potentially probiotic culture ofB. lactis (BL04, FloraFit, Danisco). Cheese-base (quark

cheese) for subsequent mixture with the further ingredientswas manufactured from commercial pasteurised skimmedmilk (Salute, Brazil) heated to 37 1C, after which additionof cultures proceeded. All cultures employed were freeze-dried commercial DVS cultures, added at 30mg l�1

(S. thermophilus and L. acidophilus) and at 20mg l�1

(B. lactis). As soon as pH reached 6.3, commercial rennetHa-la (88–92% bovine pepsin+8–12.5% bovine chymosin;Christian Hansen, Brazil, 50mg l�1) was added to thecheese-milk. A curd was formed when the pH reached 5.6and cut into cubes, placed in sterilised cotton cheesecloth,and allowed to drain at 15 1C for 15 h. After draining,63.9 g 100 g�1 cheese-base was homogenised with thefurther ingredients (g 100 g�1): pasteurised strawberrywhole pulp (Maisa, Brazil) – 11, commercial sucrose(Uniao, Brazil) – 10, commercial sterilised milk cream(25% lipids, Nestle, Brazil) – 14, powdered unflavouredjelly (Oetker, Brazil) – 0.5, xanthan gum (Rhodigel 80,Rhodia, France) – 0.5, natural colouring agent (PluryQuımica, Brazil) – 0.06, strawberry flavour (Mylner,Brazil) – 0.04. After 1 day refrigerated storage, the cheeseswere freeze-dried, grated and vacuum-packed.

2.3. Petit-suisse cheese digestion

The simulated human digestion of the petit-suisse cheeseswas made using five different mixtures, combining or noteither one of the two petit-suisse cheese trials obtainedpreviously (control cheese C and probiotic cheese P) withfurther ingredients. These ingredients included the sameprobiotic bacteria added during production of cheese P and10% (w/w) of a prebiotic blend containing the prebioticingredients FOS (BeneoTM P95, Orafti, Belgium), inulin

(BeneoTM ST, Orafti), and honey (eucalyptus honey,Biosciences Institute-University of Sao Paulo, Brazil) inequal proportions. Thus, C and P were digested solely andalso mixed with further ingredients as follows: C plus theprobiotic bacteria (C+P); C plus the probiotic bacteriaand the prebiotic blend (C+P+PP) and, finally P plus theprebiotic blend (P+PP). Levels of probiotic bacteria wereverified and adjusted, assuming that all trials whichreceived them presented the same viability (1.0� 109 and3.0� 109 cfu g�1, respectively for L. acidophilus and B.

lactis). Plate counts were made by pour-plating 1ml of eachdilution in appropriate medium. DeMan-Rogosa-Sharpe(MRS) agar, modified by the substitution of glucose formaltose as the main carbohydrate source, as described bythe International Dairy Federation, after three days ofanaerobic incubation (Anaerobic System Anaerogen,Oxoid) at 37 1C was used for L. acidophilus enumeration[22]. MRS agar (Oxoid Ltd. Basingstoke, UK) to whichsodium propionate (0.3%, w/v) and lithium chloride(0.2%, w/v) solutions were added, after three days ofanaerobic incubation (Anaerobic System Anaerogen,Oxoid) at 37 1C, was used to enumerate B. lactis [23].For the petit-suisse cheese digestion, a method described

by Minihane et al. [24] was used with following modifica-tions. Samples (60 g) were diluted in 325ml of distilledwater, 6.25ml of a-amylase solution (0.0313 g of a-amylasein 10ml of filter sterilised 0.001mol l�1 CaCl2, pH 7) wasthen added and the mixture was incubated while stirring at37 1C for 30min. pH was then adjusted to 2, 25ml ofpepsin solution (2.7 g pepsin in 25ml 0.1mol l�1 HCl) wasadded and the mix was further incubated at 37 1C for 2 h.The pH of the digestion mix was then adjusted to 6.0,125ml of the pancreatin-bile solution (0.675 g pancreatin;7.21 g bile extract in 150ml filter-sterilised 0.5mol l�1

NaHCO3 solution) was added and the pH adjusted to 7.The digestion mix was then distributed into several dialyses(Spectra/Por 1 kDa MWCO) tubes (100ml each or more),placed in a bucket containing 2 l of sterile NaCl solutionand left at 37 1C. Conductivity was determined after fewhours and the digestion mix was left overnight. When theconductivity became stable, dialysis fluid was replaced withfresh NaCl solution and the mixture was left under thesame conditions until no changes in conductivity (ca.27–30 h) were detected. The removed solution was nanofil-trated (according to Goulas et al. [25]) and reincorporatedto the dialysis content later. Dialysis bag contents weretransferred into plastic containers and frozen at �20 1C.Following freeze-drying, the digested material was va-cuum-packed and stored at �20 1C until used for thefermentation studies.

2.4. Preparation and collection of faecal samples for

fermentation studies

Fresh faecal samples were obtained from two healthyhuman volunteers who had not been prescribed antibioticsfor at least 6 months prior to the study and had no history

ARTICLE IN PRESSH.R. Cardarelli et al. / Anaerobe 13 (2007) 200–207 203

of any gastrointestinal diseases. Samples were collected onsite and used immediately following collection. A one tenthdilution in anaerobic phosphate buffer (0.1mol l�1, pH 7.4)was prepared and the samples homogenised in a stomacherfor 2min.

Table 1

Rate of substrate breakdown (Ar), ratio of lactate over total SCFA

production and MPE values from stirred pH-controlled batch culture

fermentations when 1% (w/v) cheeses in the presence of human faecal

bacteria were used

Cheese triala Arb Lactate ratioc MPEd

C 0.40 0.36 0.25

C+P 0.34 0.64 0.63

C+P+PP 0.34 0.53 0.60

P 0.40 0.42 0.47

P+PP 0.43 0.62 0.84

2.5. Batch fermentations, bacterial enumeration, SCFA and

total carbohydrate analyses

Sterile, stirred, batch cultures were run in triplicate usingfaecal samples from two volunteers as described previously[16]. Briefly, six (C, P, C+P, C+P+PP, P+PP and anegative control) fermentation vessels (300ml volume),were filled with a basal nutrient medium and gassedovernight with oxygen-free nitrogen. Prior to the additionof faecal slurry, 1% (w/v) digested petit-suisse cheeseformulations plus prebiotic ingredients according to pre-vious determined formulations were added to differentvessels, culture temperature was set at 37 1C and pH wasmaintained at 6.8. The vessels were inoculated with 15mlof fresh faecal slurry (one tenth dilution w/v) andcontinuously sparged with oxygen-free nitrogen. Batchcultures were ran for a period of 24 h and samples collectedat 0, 4, 6, 8, 10, 12 and 24 h. At each sampling period,0.3ml of sample was collected from each vessel forbacterial enumeration via fluorescence in situ hybridisation(FISH), analysis of SCFA by high performance liquidchromatography (HPLC) and total carbohydrate measure-ments [16,26].

aC, control petit-suisse cheese; C+P, control cheese plus probiotics;

C+P+PP, control cheese plus pro-and prebiotics; P, probiotic cheese;

P+PP, probiotic cheese plus prebiotics.bAr calculated using Eq. (1).cLactate ratio calculated using Eq. (5).dMPE calculated using Eq. (6).

3. Results and discussion

The present study investigated the functional capacitiesof novel petit suisse cheese formulations supplemented with

0.000

1.000

2.000

3.000

4.000

5.000

0 4 8

Tim

Glu

co

se

eq

uiv

ale

nt (g

l-1)

Fig. 1. Concentrations of residual substrate (mean values, n ¼ 6) in stirred

Control petit-suisse cheese (~); Control plus probiotics petit-suisse cheese (}suisse cheese ( ); Probiotic plus prebiotics petit-suisse cheese (’); Negative co

probiotics and/or prebiotics, previously made using pro-biotic cultures (Lactobacillus acidophilus and Bifidobacter-

ium lactis) plus a starter culture (Streptococcus

thermophilus) or just the starter culture, through switchesin the composition and activity of purportedly beneficialand detrimental microbiota. The gut microbiota inoculumwas faeces, which probably gives an accurate representa-tion of events in the distal colon, but not in more proximalareas [8]. Nevertheless, it would be laborious and invasiveto collect representative samples of the whole intestinalareas for in vitro tests.

In vitro fermentation of petit-suisse cheese formulationsresulted in some differences. Fig. 1 shows the meanconcentrations of residual substrate in stirred pH-con-trolled batch cultures as measured by total carbohydrateassay for the petit-suisse cheeses tested. These data wereused to calculate the rate of assimilation (Table 1), for each

12 16 20 24

e (hours)

pH-controlled batch cultures as measured by total carbohydrates assay.

); Control plus pro-and prebiotics petit-suisse cheese (m); Probiotic petit-

ntrol (K).

ARTICLE IN PRESS

5.56.06.57.07.58.08.59.09.5

10.0

Log10 c

ells

ml-1

5.56.06.57.07.58.08.59.09.5

10.0

0 4 8 12 16 20 24

0 4 8 12 16 20 24

0 4 8 12 16 20 24

Log10 c

ells

ml-1

5.56.06.57.07.58.08.59.09.5

10.0

Log10 c

ells

ml-1

5.56.06.57.07.58.08.59.09.5

10.0

0 4 8 12 16 20 24

Log10 c

ells

ml-1

5.56.06.57.07.58.08.59.09.5

10.0

0 4 8 12 16 20 24

Time (hours)

Total Bif Bac Lab Clos EC Erec SRB

Log10 c

ells

ml-1

(a)

(b)

(c)

(d)

(e)

Fig. 2. Growth performance of bacterial groups (mean values, n ¼ 6) assessed in the presence of 1% (w/v): (a) control petit-suisse cheese; (b) control

cheese plus probiotics; (c) control cheese plus pro-and prebiotics; (d) probiotic cheese; (e) probiotic cheese plus prebiotics. Total (~); Bif is bifidobacteria

(’); Bac is bacteroides (m); Lab is lactobacilli (n); Clos is clostridia (histolyticum subgroup) ( ); EC is Escherichia coli (K); Erec is eubacteria (}); SRB is

sulphate-reducing bacteria (&).

H.R. Cardarelli et al. / Anaerobe 13 (2007) 200–207204

petit-suisse cheese, using Eq. (1). Ar values thus generatedshowed small variation in fermentation times betweendifferent samples. P+PP cheese resulted in the fastestfermentation. Similar to other batch fermentations studiesinvolving FOS and inulin [19], there were almost nosubstrates available after 6 h of fermentations. Never-theless, the rates of substrate breakdown (Ar results) were

lower than those observed by Vulevic et al. [16]. Onepossible reason for this difference is that pure oligosac-charides at 1% (w/v) concentration were used previously,whilst in the present study digested petit-suisse cheesesamples contained no more than 0.25% (w/v) concentra-tion of the oligosaccharides mixture (inulin, FOS andhoney).

ARTICLE IN PRESSH.R. Cardarelli et al. / Anaerobe 13 (2007) 200–207 205

The growth performances of bacterial groups assessedvia FISH are shown in Fig. 2. Maximum growth rates, foreach bacterial group assessed, were calculated through Eq.(2), taking different time points for each microorganismand formulation, and used to obtain PIm values by Eq. (3).Different PIm values (Table 2) were obtained for thedifferent formulations tested. All had positive values,although the absence of probiotics and prebiotics in thecontrol cheese (C) resulted in the lowest PIm and, on thecontrary, P+PP presented the highest PIm. Positive ornegative PIm value was determined by the proportion ofdesired bacterial groups (bifidobacteria, lactobacilli andeubacteria) versus less desirable bacterial groups and by theselectivity of each substrate [16]. For example, despite thepositive PIm value, control cheese supported growth ofSRB, which did not occur for the other cheeses. On theother hand, P+PP cheese supported both bifidobacteria,lactobacilli and eubacteria, but almost no growth for thedetrimental bacteria, showing it to be selective for thesebeneficial bacteria. Considering the selectivity aspect, whencomparing PIm values obtained in this study (Table 2) withthe best PIm values reported by Vulevic et al. [16] whichvaried from 1.3 to 0.8, all cheese formulations tested seemsto be satisfactory to some extent, except the control cheese.The presence of probiotics and prebiotics in P+PP cheeseeffectively promoted increased maximum growth rates ofbifidobacteria and lactobacilli compared to control cheese(Table 2). This behaviour is in agreement with other studiesinvolving the simultaneous addition of pro- and prebioticingredients during simulated in vitro fermentations byhuman gut microbiota [27,28].

Fig. 3 shows the production of lactate (g l�1), acetate(g l�1), propionate (g l�1) and butyrate (g l�1) measured byHPLC for 1% (w/v) digested cheeses. The ratio of lactateover the total SCFA production (Table 1) was then

Table 2

Maximum specific growth rates (mmax) and modified prebiotic index (PIm) from

in the presence of human faecal bacteria were used

Cheese triala mmax (h�1)b

Bif 164c Lab 158d Erec 482e Ba

C 0.19 0.22 0.22 0.1

C+P 0.32 0.26 0.15 0.0

C+P+ PP 0.35 0.19 0.12 0.0

P 0.32 0.41 0.10 0.0

P+PP 0.32 0.35 0.25 0.0

aC, control petit-suisse cheese; C+P, control cheese plus probiotics; C+P+

probiotic cheese plus prebiotics.bmmax calculated using Eq. (2).cBif 164, Bifidobacterium spp.dLab 158, Lactobacillus spp.eErec 482, Eubacterium rectale/Clostridium coccoides group.fBac 303, Bacteroides spp.gChis 150, Clostridium histolyticum group.hEc 1531, Escherichia coli.iSrb 687, Desulfovibrio spp.jPIm calculated using Eq. (3).

calculated, at the point of maximum production (4 h),using Eq. (5). It is important to note that the initial lactateconcentrations (more than 0.8 g l�1) were higher in thedigested cheeses than in the negative control (medium andfaecal inoculum) (Fig. 3) due to lactic acid productionduring cheese-making. Moreover, the Student’s t-test wasused to evaluate differences in means between lactateproduction at 4 h compared to initial lactate levels and thiswas significant (Po0.05) (Fig. 3). Interestingly, the lactateproduced was rapidly consumed after 4 h fermentation,showing that this metabolite, besides being highly pro-duced, was fundamental for maintenance of the gutmicrobiota during cheese fermentation in the simulatedconditions. It is known that human faecal microbiota iscapable of converting lactate to end-products such aspropionate, acetate and butyrate [29]. Moreover, Bour-riaud et al. [29] suggested that in human colonic lumenwhen highly fermentable non-digestible carbohydrates likeprebiotic oligosaccharides (e.g. FOS) are present, there isan induced accumulation of lactate which could greatlyaffect butyrate production in some individuals, dependingon their microbiota. These statements are favourable forthe use of petit-suisse cheese as pro- and prebiotics vehicle,as most probiotic bacterial species are lactate producersand some prebiotics may interfere in both lactate andSCFA production. Moreover, cheeses are consideredtraditional sources of lactic acid bacteria. Proportionallymore lactate over the total SCFA was produced duringfermentation of C+P and P+PP cheeses, according to thehighest lactate ratio achieved (Table 1). The lactateproduction did not vary considerably between the othercheeses (C, C+P+PP and P), and the values generated inthe present study were similar to those obtained by Vulevicet al. [16] for FOS, trans-galactooligosaccharides andsoyaoligosaccharides. These data suggest that both cheese

stirred pH-controlled batch culture fermentations when 1% (w/v) cheeses

c 303f Chis 150g Ec 1531h Srb 687i PImj

4 �0.11 �0.12 0.12 0.59

5 �0.26 �0.37 0.01 1.42

5 �0.19 �0.36 �0.02 1.46

4 �0.41 �0.34 �0.01 1.18

5 �0.35 �0.52 �0.03 1.82

PP, control cheese plus pro-and prebiotics; P, probiotic cheese; P+PP,

ARTICLE IN PRESS

0.0

0.5

1.0

1.5

2.0

0 4 8 12 16 20 24

P < 0.05Lactic a

cid

(gl-1

)

0.0

0.5

1.0

1.5

2.0

0 4 8 12 16 20 24

Acetic a

cid

(gl-1

)

0.0

0.5

1.0

1.5

2.0

0 4 8 12 16 20 24

0 4 8 12 16 20 24

Pro

pio

nic

acid

(gl-1

)

0.0

0.5

1.0

1.5

2.0

Time (hours)

Buty

ric a

cid

(gl-1

)

(a)

(b)

(c)

(d)

Fig. 3. Production of SCFA as determined by HPLC in stirred pH-controlled batch cultures (mean values, n ¼ 6). (a) lactic acid; (b) acetic acid; (c)

propionic acid; (d) butyric acid. Control petit-suisse cheese (~); Control plus probiotics petit-suisse cheese (’); Control plus pro-and prebiotics petit-suisse

cheese (m); Probiotic petit-suisse cheese (}); Probiotic plus prebiotics petit-suisse cheese ( ); Negative control (K). There were significant differences in

lactic acid production for all cheeses between 0 and 4 h (Po0.05).

H.R. Cardarelli et al. / Anaerobe 13 (2007) 200–207206

formulations were able to support the growth of lactic acidproducing bacteria.

The MPE values (Table 1), calculated using Eq. (6),summarise PIm, Ar and lactate ratio, previously obtained.The rational for the interpretation of the MPE values isthat whilst it is important to have a high SCFA productionand/or ratio of lactate over total SCFA, as well as fastfermentation times, it is imperative that all of these factorsare driven by the desired microbiota [16]. Moreover, Arand lactate ratio play an important role in situations wheresubstrates result in similar PIm values [9]. In the presentstudy, although, both Ar and lactate ratio contributedtowards the MPE values to corroborate that P+PP cheese

produced a good in vitro prebiotic effect, followed closelyby the C+P and the C+P+PP cheeses and the worst wasproduced by the control cheese. The magnitude of theMPE reported by Vulevic et al. [16] for the best in vitro

prebiotic effect of the substrates they studied varied from1.4 to 0.4 and all formulations tested in the present studyfalls into this range, except the control cheese. These resultsconfirm a high to moderate prebiotic effect of theformulations, in the order, P+PP, C+P, C+P+PP and P.It seems that MPE may be a useful tool for evaluation of

the in vitro prebiotic effect in a preliminary way, beforeundertaking human studies. In the present study, the MPEconcept was useful for this purpose, and led to the

ARTICLE IN PRESSH.R. Cardarelli et al. / Anaerobe 13 (2007) 200–207 207

conclusion that under the in vitro conditions used, thecombination of different prebiotics associated with pro-biotic bacteria previously added during cheese-making(P+PP cheese) resulted in the most promising petit-suisse

cheese in terms of its potentiality as a functional product. Itwould therefore be useful to design human trials using thisnovel synbiotic petit-suisse cheese developed to furtherdemonstrate its functionality.

Acknowledgements

We gratefully acknowledge the contributions of Dr.A.K. Goulas, Dr. P.G. Kapasakalidis and X. Tzounis. Wewish to thank Fundac- ao de Amparo a Pesquisa do Estadode Sao Paulo (FAPESP) (Projects 03/13748-1, 02/14185-8),Coordenac- ao de Aperfeic-oamento de Pessoal de NıvelSuperior (CAPES), Pro-Reitoria de Pesquisa de Pos-Graduac- ao da Universidade de Sao Paulo and OraftiActive Food Ingredients Company-Belgium, for financialsupport. We also wish to thank Salute, Danisco, Rhodia,Mylner, and Plury Quımica for providing part of thematerial resources employed in the present study.

References

[1] Klaenhammer TR, Kullen MJ. Selection and design of probiotics. Int

J Food Microbiol 1999;50:45–57.

[2] Betoret N, Puente L, Dıaz MJ, Pagan MJ, Garcıa MJ, Gras ML, et

al. Development of probiotic-enriched dried fruits by vacuum

impregnation. J Food Eng 2003;56:273–7.

[3] Puupponen-Pimia R, Aura AM, Oksman-Caldentey KM, Myllarinen

P, Saarela M, Mattila-Sanholm, et al. Development of functional

ingredients for gut health. Trends Food Sci Technol 2002;13:3–11.

[4] Food and Agriculture Organization of the United Nations, World

Health Organization. Evaluation of health and nutritional properties

of probiotics in food including powder milk with live lactic acid

bacteria. Cordoba, 2001. 34p. Available at: /ftp://ftp.fao.org/docrep/

fao/meeting/009/y6398e.pdfS. [Report of a Joint FAO/WHO Expert

Consultation].

[5] Buriti FCA, Rocha JS, Saad SMI. Incorporation of Lactobacillus

acidophilus in Minas fresh cheese and its implications for textural and

sensorial properties during storage. Int Dairy J 2005;15:1279–88.

[6] Ong L, Henriksson A, Shah NP. Development of probiotic Cheddar

cheese containing Lactobacillus acidophilus, Lb. casei, Lb. paracasei

and Bifidobacterium spp. and the influence of these bacteria on

proteolitic patterns and production of organic acid. Int Dairy J

2006;16:446–56.

[7] Buriti FCA, Cardarelli HR, Saad SMI. Biopreservation by Lactoba-

cillus paracasei in coculture with streptococcus thermophilus in

potentially probiotic and synbiotic fresh cream cheeses. J Food

Protect 2007;70:228–35.

[8] Gibson GR, Probert HM, Van Loo J, Rastall RA, Roberfroid MB.

Dietary modulation of the human colonic microbiota: updating the

concept of prebiotics. Nutr Res Rev 2004;17:259–75.

[9] Roberfroid MB. Introducing inulin-type fructans. Br J Nutr 2005;

93:S13–25.

[10] Kolida S, Tuohy K, Gibson GR. Prebiotic effects of inulin and

oligofructose. Br J Nutr 2002;87:S193–7.

[11] National Honey Board. Honey information kit of the food and

beverage industries. Longmont: National Honey Board; 1996.

[12] Kajiwara S, Gandhi H, Ustunol Z. Effect of honey on the growth of

and acid production by human intestinal Bifidobacterium spp.: an in

vitro comparison with commercial oligosaccharides and inulin. J

Food Protect 2002;65:214–8.

[13] Sanz ML, Polemis N, Morales V, Corzo N, Drakoularakou A,

Gibson GR, et al. An in vitro investigation into the potential prebiotic

activity of honey oligossacharides. J Agric Food Chem 2005;

53:2914–21.

[14] Holzapfel WH, Schillinger U. Introduction to pre- and probiotics.

Food Res Int 2002;35:109–16.

[15] Veiga PG, Cunha RL, Viotto WH, Petenate AJ. Caracterizac- ao

quımica, reologica e aceitac- ao sensorial de queijo petit suisse

brasileiro. Ciencia e Tecnologia de Alimentos 2000;20:349–57.

[16] Vulevic J, Rastall RA, Gibson GR. Developing a quantitative

approach for determining the in vitro prebiotic potential of

dietary oligosacchrarides. FEMS Microbiol Lett 2004;236:

153–9.

[17] Bruck WM, Graverholt G, Gibson GR. A two-stage continuous

culture system to study the effect of supplemental a-lactalbumin

and glycomacropeptide on mixed cultures of human gut

bacteria challenged with enteropathogenic Escherichia coli and

Salmonella serotype Typhimurium. J Appl Microbiol 2003;95:

44–53.

[18] Palframan R, Gibson GR, Rastall RA. Development of a quantita-

tive tool for the comparison of the prebiotic effect of dietary

oligosaccharides. Lett Appl Microbiol 2003;37:281–4.

[19] Rossi M, Corradini C, Amaretti A, Nicolini M, Pompei A, Zanoni S,

et al. Fermentation of fructooligosaccharides and inulin by bifido-

bacteria: a comparative study of pure and faecal cultures. Appl

Environ Microbiol 2005;71:6150–8.

[20] Shah NP. Functional foods from probiotics and prebiotics. Food

Technol 2001;55:46–53.

[21] Macfarlane S, Macfarlane GT. Food and large intestine. In: Fuller R,

Perdigon G, editors. Gut flora, nutrition, immunity and health.

Oxford: Blackwell; 2003. p. 24–51.

[22] International Dairy Federation. Fermented and non-fermented milk

products. Detection and enumeration of Lactobacillus acidophilus.

Culture Media. Bulletin of the IDF 306. Brussels: International Dairy

Federation; 1995.

[23] Vinderola CG, Reinheimer JA. Culture media for the enumeration of

Bifidobacterium bifidum and Lactobacillus acidophilus in the presence

of yoghurt bacteria. Int Dairy J 1999;9:497–505.

[24] Minihane AM, Fox TE, Fairweather-Tait SJ. A continuous flow in

vitro method to predict bioavailability of iron from foods. Bioavail-

ability 1993;93:175–9.

[25] Goulas AK, Kapasakalidis PG, Sinclair HR, Rastall RA, Grandison

AS. Purification of oligosaccharides by nanofiltration. J Membr Sci

2002;209:321–35.

[26] Al-Tamimi MAHM, Palframan RJ, Cooper JM, Gibson GR, Rastall

RA. In vitro fermentation of sugar beet arabinan and arabino-

oligosaccharides by human gut microflora. J Appl Microbiol

2006;100:407–14.

[27] Gmeiner M, Kneifel W, Kulbe D, Wouters R, De Boever P, Nollet L,

et al. Influence of a synbiotic mixture consisting of Lactobacillus

acidophilus 74-2 and a fructooligosaccharide preparation on the

microbial ecology sustained in a simulation of the human intestinal

microbial ecosystem (SHIME reactor). Appl Microbiol Biotechnol

2000;53:219–23.

[28] Fooks L, Gibson GR. Mixed culture fermentation studies on the

effects of synbiotics on the human intestinal pathogens Campylo-

bacter jejuni and Escherichia coli. Anaerobe 2003;9:231–42.

[29] Bourriaud C, Robins RJ, Martin L, Kozlowski S, Tenailleau E,

Cherbut C, et al. Lactate is mainly fermented to butyrate by human

intestinal microfloras but inter-individual variation is evident. J Appl

Microbiol 2005;99:201–12.