HAL Id: hal-00929400https://hal.archives-ouvertes.fr/hal-00929400

Submitted on 1 Jan 1994

HAL is a multi-disciplinary open accessarchive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come fromteaching and research institutions in France orabroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, estdestinée au dépôt et à la diffusion de documentsscientifiques de niveau recherche, publiés ou non,émanant des établissements d’enseignement et derecherche français ou étrangers, des laboratoirespublics ou privés.

Growth promotion of Bifidobacterium animalis bybovine milk proteose-peptone

L Etienne, Jm Girardet, G Linden

To cite this version:L Etienne, Jm Girardet, G Linden. Growth promotion of Bifidobacterium animalis by bovine milkproteose-peptone. Le Lait, INRA Editions, 1994, 74 (5), pp.313-323. �hal-00929400�

Lait (1994) 74, 313-323© Elsevier/INRA

313

Original article

Growth promotion of Bifidobacterium animalisby bovine milk protecse-peptone 1

L Etienne, JM Girardet, G Linden

Laboratoire de Biochimie Appliquée associé à l'INRA, Faculté des Sciences,Université de Nancy t, BP 239, 54506 Vandœuvre-lès-Nancy, France

(Received 2 May 1994; accepted 13 July 1994)

Summary - The industrial strain Bifidobacterium animalis was used as assay organism to evaluatebifidobacterial growth-promoting activity of bovine milk proteose-peptone. This proved to be a bettergrowth-promoting factor than bovine casein. The bifidogenic activity was found mainly in the pro-teose-peptone hydrophobie fraction containing component 3, although the glycan moiety was a weakgrowth-promoter. Proteose-peptone digests by various proteolytic enzymes caused great enhancementof B animalis growth, particularly the Pronase digest. Size-exclusion chromatography of digests showedthat the more active peptides had a molecular mass distribution of 1000 to 5000 Da.

proteose-peptone 1component 31 Bifidobacterium animalis 1growth stimulation 1 proteolysis 1milk

Résumé - Activité stimulante des protéose-peptones de lait bovin sur la croissance de Biti-dobacterium animalis. La souche industrielle 8ifidobacterium animalis a été utilisée pour évaluerl'activité stimulante des protéose-peptones du lait bovin. Elles se sont montrées de meilleurs facteursbifidogéniques que la caséine. L'activité bifidogénique a été principalement trouvée dans la fraction hydro-phobe des protéose-peptones contenant le composant-3, bien que la partie glycannique du composant-3 n'ait qu'une faible activité stimulante. L 'hydrolyse des protéose-peptones par des enzymes protéo-lytiques provoque une forte augmentation de la croissance de 8 animalis, tout particulièrement avecl'hydrolyse pronasique. L'étude des hydrolysats par chromatographie d'exclusion stérique montre queles peptides les plus actifs ont une taille comprise entre 1000 et 5000 Da.

protéose-peptone / composant-3 /8ifidobacterium animalis / stimulation de la croissance /protéolyse / lait

for this strain (Bezkorovainy et al, 1976),but there is very little such growth-promotingactivity in the bovine milk proteose-peptone.However, this strain may not accuratelyreflect the biochemical response of thegenus Bifidobacterium to growth factors.Moreover, this model strain is uncommonin the intestinal tract and is considered as amutant, since it requires exogenousN-acetyl-glucosamine for cell wall synthe-sis (Poch and Bezkorovainy, 1988). Con-sequently, much of the growth-promotingactivity has been attributed to the N-acetyl-glucosamine-containing oligosaccharidesand glycoproteins. Human milkproteose-peptone contains more carbohy-drates (45% according to Bezkorovainy et al,1976) than bovine milk proteose-peptone.No study has been published on bovine milkproteose-peptone promotion of bifidobac-terial growth, although growth-promotingactivity of Bifidobacterium species by bovineor human whey proteins and caseins wastested (Petschow and Talbott, 1990, 1991).

It is now well-established that bifidobac-teria strains cannot grow in synthetic media,but require unknown factors for optimalgrowth (Poch and Bezkorovainy, 1988). Thepurpose of this work was to investigate thegrowth-promoting activity of bovine milk pro-teose-peptone for B animalis, one of thecommonly used strains in dairy industry(Roy and Ward, 1993).

314 L Etienne et al

INTRODUCTION

Bovine milk proteose-peptone representsabout 10% of the whey protein and is aheat-stable and acid-soluble fraction whichcan be separated into two groups (Pâquet,1989). The first includes polypeptides result-ing from proteolysis of casein by endoge-nous proteinase in milk among whichp-CN-5P (f1-1 05/1 07; 12 441 Da), p-CN-4P(f1-28; 3478 Da), and p-CN-1 P(f29-1 05/1 07; 8981 Da) are N-terminal frag-ments of p-casein (Eigel et al, 1984). Thesecond group can be separated from thefirst by hydrophobic interaction liquid chro-matography (Pâquet et al, 1988; Girardetet al, 1991) and contains three principal gly-coproteins with apparent Mr of 11 000,19 000, and 29 000 Da. The primary struc-ture of component 3 corresponding to the29 OOO-Daglycoprotein has been recentlydetermined (Serensen and Petersen, 1993)and shown to contain 135 amine acidresidues. Two O-linked (Tyr16 and Tyr86)and one N-linked (Asn77) glycans are boundon the polypeptide chain. This carbohydratemoiety represents 17-18% of the compo-nent 3 fraction (Kanno, 1989; Girardet et al,1993). Moreover, the 19 OOO-Daglycopro-tein was identified as the 54-135 C-terminalfragment of component 3 (Sarensen andPetersen, 1993).

Compone nt 3 has interesting functionalproperties, particularly its important emulsi-fying capablhty (Shimizu et al, 1989) andits biochemical role in the inhibition of milk'sspontaneous lipolysis (Cartier et al, 1990;Girardet et al, 1993). Recently, the mitogenicactivity of a hydrophobic fraction containingcomponent 3 on DNA synthesis in MARK 3hybridoma was shown (Mati et al, 1993).

The growth-promotion of Bifidobacteriumbifidum var pennsylvanicus has been exten-sively studied (Gyôrgy and Rose, 1955;Kehagias et al, 1977; Bezkorovainy andTopouzian, 1981). Human milk proteose-peptone contains growth-promoting factors

MATERIALS AND METHODS

Materials and reagents

The fast protein liquid chromatography (FPLC)system and TSK-Phenyl-5PW (21.5 x 150 mm)column were obtained from Pharmacia FineChemicals (Uppsala, Sweden). The HPLC L6200apparatus was from Merck (Darmstadt, Germany)and the TSK-2000 SW column from Pharmacia.Bio-gel P2 was from Bio-Rad Laboratories (Rich-mond, VA, USA). An ultrafiltration cell52-model

B animalis activation by proteose-peptone

and PM-10 membrane (eut-off 10 000 Da) werepurchased from Amicon (Danvers, MA, USA).Dialysis tubing was cellulose ester (eut-off 500Da; Medicellintemational, London, UK). Enzymesused were: immobilized neuraminidase(EC.3.2.1.18) of Clostridium perfringens, type VI-A (Sigma Chemicals Co, Saint-Louis, MO, USA),Pronase E (EC.3.4.24.4) of Streptomyces griseus(Merck, Darmstadt, Germany), N-tosyl-L-phenyl-alanine-chloromethylketone (TPCK)-treatedtrypsin (EC.3.4.21.4) and œ-chyrnotrypsln(EC.3.4.21.1) both from bovine pancreas andattached to beaded agarose (Sigma). Amine acids(Lys, His, Arg, Asp, Thr, Ser, Glu, Pro, Gly, Ala,Cys, Val, Met, Ile, Leu, Tyr, Phe, Trp), N-acetyl-glucosamine, lactose, lactulose were purchasedfrom Sigma, and the brain heart infusion (BHI)from Difco Laboratories (Detroit, MI, USA).

Fraetionation of milk proteins

Whole casein was prepared byacid precipitationof Holstein cows' raw skim milk at pH 4.6 by 1moill-HCI and soluble whey proteins were dia-Iyzed and freeze-dried. Proteose-peptone extractswere prepared after bulk skim milk treatment byheating (95°C, 30 min) and acidification with 1moi/l-HCI at pH 4.6 (Pâquet et al, 1988).

Fraetionation of proteose-peptone

Proteose-peptone fractionation was pertermedusing a TSK-Phenyl-5PW column connected tothe FPLC system (Girardet et al, 1991). Sample(150-200 mg) was loaded on the column equili-brated in 0.8 mol/l-NaH2P04 buffer (pH 6.8) at25°C. Flow rate was 6 ml/min and detection wasmeasured at 280 nm. An isocratic elution of 20min in 0.8 molll-NaH2P04 was pertormed to col-lect the non-hydrophobic fraction of proteose-peptone (NHF). After a second isocratic step of 18min at 0.15 molll-NaH2P04, the low hydrophobicfraction of proteose-peptone (LHF) was eluted.Last, the highly hydrophobic fraction of proteose-peptone (HHF) containing component 3 wasobtained in pure water (isocratic elution of 20 minin 0 rnolll-NaH2P04). After dialysis, each fractionwas freeze-dried. Between each chromatogra-phy step, the column was washed by injectionof 2.5 ml of 200 mmolll-NaOH.

315

Proteose-peptone was aise separated by ultra-filtration using a membrane with a nominal mole-cular mass cut-off of 10 000 Da. Filtrate and reten-tate fractions were referred as low and high Mrfractions respectively.

Neuraminidase treatmentof high hydrophobie fractionof proteose-peptone

Desialylation was carried out according to themethod of Coddeville et al (1989). HHF (10 mg)in 0.1 moili-citrate buffer (pH 5.5), was treated at37°C for 24 h with 5 units (one unit willliberate 1urnol of N-acetylneuraminic acid per min at pH 5at 37°C, using NAN-lactose as substrate) ofimmobilized neuraminidase.

Preparation of glyeopeptides

Proteolysis of proteose-peptone and purificationof glycopeptides were pertormed according toMontreuil et al (1987). The proteose-peptone (1 g)in 10 mmoill-calcium acetate buffer (pH 8) wastreated with Pronase E (enzyme:substrate, 1:50)(4 000 000 PUIg, one unit (PU) will hydrolysecasein to produce color equivalent to 1 urnol oftyrosine per 10 min at pH 7.4 at 40°C, color byCiocalteu reagent), at 37°C for 6 h. Enzymic activ-ity was stopped by adding ace tic acid to pH 4.5.Cold ethanol (9 vol, -10°C) was then added toprecipitate glycopeptides. This fraction was thendissolved in calcium acetate buffer and hydroly-sis with Pronase was pertormed twice more.Then, trichloroacetic acid to a final concentrationof 12% (wlv) was added to precipitate excessenzyme and, after adjustment to neutral pH byNaOH, glycopeptides were desalted by passingthrough a Bio-gel P2 (1.5 x 53 cm) column. Thosefractions, which stained at 80°C for 5 min with0.2% (vlv) 3,5-dihydroxytoluol in 20% (vlv) H2S04,were pooled and freeze-dried.

Proteolytie treatmentsof proteose-peptone

Freeze-dried proteose-peptone (50 mg) in 25 mlof a mixture of 43 mmoi/l-HCOOH and

316 L Etienne et al

57 mmolll-NH3 was treated with 10 N-cx-ben-zoyl-L-arginine-ethylester (SAEE) units of trypsinor 10 N-acetyl-L-tyrosine-ethylester (ATEE) unitsof cx-chymotrypsin both attached to beadedagarose. Hydrolysis occurred during 2.5 h at37°C under gentle mixing. The reaction wasstopped by centrifugation (1800 g, 5 min). Fordouble hydrolysis the second step is then real-ized under the same conditions. Afterwards, thesupernatant was evaporated under partial vacu-um and freeze-dried in order to remove ammo-nium formate.

Digestion by Pronase E (4 000 000 PUig) wasperformed during 2.5 h at 37°C in the sameammonium formate buffer (pH 8.5) (enzyme:sub-strate, 1:50). The reaction was stopped by addingacetic acid to pH 4.5. The product was ultrafil-tered (eut-off 10 000 Da) to remove Pronase andthe permeate was evaporated under partial vacu-um and freeze-dried.

Ana/ytica/ methods

Protein concentration was determined by the pro-cedure of Lowry et al (1951), with bovine serumalbumin as standard.

The degree of hydrolysis of each fraction wasmeasured in triplicate from the amount of freeamino groups with the trinitrobenzene sulphonicacid spectrometric method (Church et al, 1985),using glycine as the standard.

The molecular mass distribution profiles ofenzymic hydrolysates were determined by highperformance size-exclusion chromatography(HPSEC; Vijayalakshmi et al, 1986). Proteinstandards (Serva Fine Siochemicals, Heidel-berg, Germany) were œ-lactalburnin (14 200 Da),trypsin inhibitor (6500 Da), melittin (2847 Da),bradykinin (1060 Da), and Pro-Leu-Gly-Pro-D-Arg (795 Da).

Bacteria/ strain and growth conditions

The Bifidobacterium strain, kindly supplied byChr Hansen (Copenhagen, Denmark), has beenindustrially employed to prepare fermented milks.The fructose-6-phosphate-phosphoketolase (spe-cific enzyme of bifidobacteria) test was done asdescribed by Scardovi (1986) and the strain was

genetically identified by its ribosomal RNA generestriction patterns (Grimont and Grimont, 1986)as Bifidobacterium animalis ATCC 27536 (resultnot shown). The microorganism was routinelypropagated and subcultured three times beforeexperiments in SHI broth (Romond et al, 1980).The standard bacterial inoculum was preparedfrom cultures of 24 h at 37°C. Sifidobacteria werecentrifuged 15 min at 5000 g, suspended in 0.9%(w/v) NaCI and 0.1% (w/v) peptone (Collins andHall, 1984), then standardized to a turbidity of0.65 measured at 550 nm (approximately109 cells/ml; Roy et al, 1990).

Bifidobacteria/ growth promotion assay

Growth of microorganisms in the presence ofpotential growth promoters was based in part onthe method of Poch and Sezkorovainy (1991),using synthetic Garches medium (Romond et al,1980).

After compounding with or without the growthpromoters to be tested (1 mg/ml), the test mediumwas sterilized by autoclaving (110°C, 20 min).The medium (10 ml) was inoculated with 0.1 ml ofthe standard bifidobacterial inoculum and incu-bated aerobically at 37°C for 24 h without agita-tion. Growth was determined by measurementof turbidity at 635 nm (Proulx et al, 1992) (tar-bidity of 0.15 ± 0.05 for synthetic Garches mediumalone). In order to standardize the results, ailgrowth experiments in the synthetic Garchesmedium were accompanied by growing themicroorganism in SHI broth (reference standard;turbidity of 1.4 ± 0.15), because growth variedfrom day to day (Poch and Sezkorovainy, 1988).The formula: growth promotion (%) =[A test medium - A synthetic medium 1 A refer-ence medium - A synthetic medium] x 100, wasused to express growth, where A syntheticmedium corresponded to the turbidity obtainedwith the Garches medium alone. Ali assays weredone three times in duplicate.

Statistica/ methods

Multiple comparisons were made using a multiplerange test based on confidence intervals and dif-ferences between sample means.

B animalis activation by proteose-peptone

RESULTS

Growth promotion activityof milk protein fractions

Casein, whey proteins and proteose-pep-tone, prepared from bovine milk were testedfor B animalis growth promotion activity(table 1). Ali significantly enhanced growth incomparison with the synthetic medium alone(0.15 ± 0.05%). Both whey proteins and pro-teose-peptone were better growth-promot-ing factors of B animalis than casein (P <0.01).

Distribution of activity among proteose-peptone fractions

Proteose-peptone was divided in terms ofits hydrophobie behavior (fig 1). NHF, LHFand HHF were identical to fractions l,1I+1I1+IV and V respectively, isolated in ourearlier work (Girardet et al, 1991). So, NHFwas composed by hydrophilic peptides orpolypeptides. LHF contained principallyp-CN-5P (f1-1 05/1 07), and the high

Table 1.Bifidobacterium animalis growth-promo-ting activity of bovine milk protein fractions.Activité stimulante des fractions protéiques dulait bovin sur la croissance de Bifidobacteriumanimalis.

Milk protein fraction Growth-promoting activity°(1 mg / ml) (%)

CaseinWhey proteinsProteose-peptone (PP)

6.0 ± 1.2 at9.5 ± 1.8 b

10.9 ± 2.1 b

o Defined in text, t Different lelters show significant dif-ference (P< 0.01).oDéfini dans le texte. t Des lettres différentes montrentune différence significative (P < 0,01).

317

hydrophobie fraction (HHF) contained mainlycomponent 3. The B animalis growth pro-moting activities of NHF and LHF were notsignificantly different (P> 0.05), but lower (P< 0.01) than those of proteose-peptone andHHF (table Il).

Moreover the low and high Mr fractions ofproteose-peptone displayed significant activ-ity, especially the latter (P < 0.01).

Therefore, the activation capability of theproteose-peptone could be mainly due tofractions containing hydrophobie compo-

NaHzP04: 1 molli 0.1 molli o molliNaOa0.2 molli

HHF,j, WF1l.K 3.7%

r---~r----r----111----

NHF17.6%

LHF67.2%

0.5

.Oo':-------=---~-----i~

Fig 1. Fractionation of proteose-peptone by semi-preparative hydrophobic interation FPLC in aT8K-Phenyl 5PW column (method adapted fromGirardet et al, 1991). Nitrogen content of eaeheollected fraction was determined by the Kjeldahlmethod. NHF, non-hydrophobie fraction; LHF,low hydrophobie fraction; HHF, high hydropho-bic fraction; WF, wash fraction.Fractionnement des protéase-peptones par FPLCd'interactions hydrophobes semi-préparative surcolonne T5K-Phenyl 5PW (méthode adaptéed'après Girardetet al, 1991). Le contenu en azotede chaque fraction collectée est déterminé parla méthode de Kjeldahl. NHF: fraction non bydn»phobe ; LHF : fraction faiblement hydrophobe:HHF: fraction fortement hydrophobe; WF: frac-tion de lavage.

318 L Etienne et al

Table Il. Bifidobacterium animalis growth-pro-moting activity of proteose-peptone fractions.Activité stimulante des fractions de protéose-pep-tones sur la croissance de Bifidobacterium ani-malis.

Proteose-peptone fraction(1 mg/ml)

Growth-promotingactivity' (%)

Proteose-peptone (PP)NHFLHFHHFPP high Mr fractionPP low Mr fractionDesialylated HHFGlycopeptide fraction from HHFN-Acetyl-glucosamineLactuloseLactose

10.9 ± 2.1 at7.3±1.8b

7.4±1.6b

11.8 ± 2.3 a16.2 ± 2.3 c3.7± 0.5 d

11.9 ± 2.5 a6.6± 2.1 b0.8 ± 0.2 e0.2 ± 0.1 f0.2 ± 0.1 f

Legend: NHF, non-hydrophobie fraction; LHF, low hydro-phobie fraction; HHF: high hydrophobie fraction .• Defi-ned in text. t Different letters show signifieant differenee(P<O.01).Légende: NHF: fraction non hydrophobe; LHF: fractionfaiblement hydrophobe; HHF: fraction fortement hydro-phobe .• Défini dans le texte. t Des lettres différentesmontrent une différence significative (P < 0,01 J.

nent 3 (present in HHF and in the high Mrfraction).

Growth response to carbohydratepromoters

To determine the role of component 3 sialicacids on the activation of the B animalisgrowth, neuraminidase treatment of HHFwas performed, but no change wasobserved after desialylation (P> 0.05). Theactivity of the glycopeptide fraction was sig-nificantly lower than the HHF and proteose-peptone activities (P < 0.01; table Il). N-Acetyl-glucosamine and lactulose, knownas growth promoters for bifidobacteria,

showed no growth promotion of B animalis(table Il).

Enzymic digests of proteose-peptone

The Mr distribution of the proteose-peptonedigests were related approximately to thedegree of hydrolysis (table III) and also tothe growth-promoting activity of B animalis(table IV). ppT and ppc consisted mainlyof large peptides (5000-10 000 Da) whereasppTC and ppCT, obtained by double trypsin-a-chymotrypsin (PPTC) or a-chymotrypsin-trypsin (ppCT) hydrolysis, contained equalamounts of large (5000-10 000 Da) andsmall (2000-5000 Da) peptides. ppTC andppCT, with more amino groups th an ppTand ppc, were better growth-promoting fac-tors than the simple hydrolysates (table IV).Pro nase digestion of proteose-peptone(PPP) gave the highest degree of hydrolysis,and contained only sm ail peptides « 5000Da). ppP was a good stimulant of B ani-malis growth (table IV).

By dialysis of ppT (eut-off 500 Da), a sig-nificant increase of the activity was observed(P> 0.05; table IV).

ln order to test the effect of free ami noacids on the growth promotion of B animalis,a mixture of 18 amino acids was added tothe Garches medium. Amino acids wereadded at 50 mgll except for cysteine (6.25mg/I). The growth-promoting activity wasweak.

DISCUSSION

Proteose-peptone represents a milk frac-tion that still remains to be valorized in dairyindustry. In this study, we have tested itsgrowth-promoting activity on B animalis, astrain commonly used in fermented milk pro-duction, contrary to label information (Biavatiet al, 1992; Roy and Ward, 1993).

B animalis activation by proteose-peptone 319

Table III. Hydrolysis of proteose-peptone by various proteases and molecular mass distribution ofpeptides produced.Hydrolyses protéasiques des protéose-peptones et distribution de la masse moléculaire des peptidesproduits.

Proteose-peptonedigest

Hydrolysis degree Mf distribution of peptides' (%)mmolll

equivalent Gly >10 000 Da 5000-10 000 Da 2000-5000 Da 1000-2000 Da < 1000 Da

pp 0.40 95.0 5.0ppT 1.30 12.0 53.0 35.0Dialyzed ppT 0.95 13.0 53.0 34.0ppc 1.20 21.5 59.0 19.5ppTC 1.99 5.5 48.0 45.0 1.5ppCT 2.20 6.4 45.8 43.8 4.0ppP 4.30 67.5 24.5 8.0

PP, proteose-peptone; ppT or ppc, trypsin or a-chymotrypsin hydrolysate; ppTC or PPCT, trypsin-a-chymotrypsinor a-chymotrypsin-trypsin double hydrolysate; ppP, pronase hydrolysate .• Molecular mass distribution is catcu-lated Irom the Integration 01 the surface 01 the chromatogram. Results were expressed in percentage 01 the totalsurface.PP: proteose-peptones ; ppT OUPPc : hydrolysat trypsique ou a-ehymotrypsique ; ppTC OU ppcT: double hydroly-sat trypsique-a-chymotrypsique ou a-ehymotrypsique-trypsique ; PPP : hydrolysat pronasique .• La distribution dela masse moléculaire est estimée à partir de l'intégration de la surface totale du chromatogramme. Les résultatssont exprimés en pourcentage de la surface totale.

Bovine whey proteins and proteose-pep-tone proved to be better growth-promotingfactors of B animalis than casein (table 1),and these results are in good agreementwlth the work of Petschow and Talbott(1990). They showed that the growth-pro-moting activity of several strains (B bifidum,B infantis, B breve, B longum) is containedin the acid whey rather than in the acidcasein fraction, concluding that cow milkcontains beat-stable growth promoting fac-tors in whey. We believe that these factorscou rd be present in the heat-stable and acld-soluble fraction called proteose-peptone.

Although the presence of the growth-pro-moting activity in both the non-protein(< 10000 Da; low Mr) and the protein (highMr) fractions of whey has been reported(Petschow and Talbott, 1991), we did notfind much activity in the low Mr fraction ofproteose-peptone. This fraction could con-

tain small hydrophilic peptides such as~-CN-4P (f1-28). The two principal com-pounds ~-CN-5P (f1-105/107) and cornpo-nent 3 (in an aggregate state inmilk > 100 000 Da; Pâquet, 1989) remainingin the protein proteose-peptone fraction afterultrafiltration showed a good growth-pro-moting activity.

The glycan moiety of compone nt 3 con-tains fucose, galactose, mannose, N-acetyl-galactosamine, N-acetylglucosamine andN-acetyl-neuraminic acid (Kanno, 1989;Girardet et al, 1994). Moreover, the corn-ponent 3 N-linked glycan (bounded toAsn77; Sarensen and Petersen, 1993)seems to be a biantennary N-acetyllac-tosamine-type structure (Girardet et al,1994). So, the glycan moiety has an het-erogeneous structure composed of N- andO-linked chains. N-acetyl-neuraminic acidsbound at the free extremity of the glycan

320 L Etienne et al

Table IV. Bifidobacterium animalis growth-promoting activity 01 proteose-peptone digestsActivité stimulante d'hydrolysats de protéose-peptones sur la croissance de Bilidobacteriumanimalis.

Proteose-peptone fraction Growth-promoting(1 mg/ml) activity' (%)

ppppTDialyzed ppTHHFTppcppTCppCTppPFree amine acids i

10.9 ± 2.1 at26.3± 2.7 b33.7 ± 3.6 c.h28.7 ± 3.1 b,h21.3 ± 2.4 d46.8 ± 4.2 9

38.3 ± 3.9 C

72.3 ± 5.8 f7.2±1.19

PP, proteose-peptone; HHF, high hydrophobie fraction;ppT or PPc, trypsin or a-ehymotrypsin hydrolysate; ppTeor pPCT, trypsin-a-ehymotrypsin or a-chymotrypsin-tryp-sin double hydrolysate; ppP, Pronase hydrolysate ." Defi-ned in text. t Different lellers show significant difference(P < 0.01). i Free amino aeids are a mixture of Lys, His,Arg, Asp, Thr, Ser, Glu, Pro, Gly, Ala, Cys, Val, Met, Ile,Leu, Tyr, Phe, Trp.PP: protéase-peptones; HHF: fraction fortement hydro-phobe ; ppT ou PPc : hydrolysats trypsique ou a-chy-motrypsique ; ppTC ou ppcT : double hydrolysat tryp-sique-a-ehymotrypsique ou a-ehymotrypsique trypsique ;ppP: hydrolysat pronasique .• Défini dans le texte. tDes lettres différentes montrent une différence signifi-cative (P < 0,0/). t Acides aminés libres constitués d'unmélange de Lys, His, Arg, Asp, Thr, Ser, Glu, Pro, Gly,Ala, Cys, Val, Met, Ile, Leu, Tyr, Phe, Trp.

chains may protect the glycan moietyagainst hydrolysis by the recurrent glycolyticsystem of the bifidobacteria strains (Nicolaiand Zilliken, 1972). Indeed, most of thespecies of Bifidobaeterium do not have neu-raminidase (Desjardins et al, 1990). How-ever, in spite of the removal of the glycanresidues after enzymic desialylation of HHF,the growth of B animalis was not increased.ln the case of B bifidum var pennsylvani-eus, the oligosaccharides tested are goodgrowth-promoting factors only after neu-

raminidase treatment (Nicolai and Zilliken,1972; Gyërgy et al, 1974). N-Acetyl-glu-cosamine contained in these oligosaccha-rides is available after hydrolysis of the ter-minai sialic acids. However, it ls known thatN-acetyl-glucosamine is only needed for thegrowth of B bifidum var pennsylvanieus; theother species do not take up this glycanresidue. The growth-promoting factors con-tained in the proteose-peptone should bedistinguished from the oligosaccharide andglycoprotein activations of the mutant Bbifidum var pennsylvanieus.

Glycopeptides purified after Pronasetreatment of proteose-peptone are gener-ated by proteolysis of compone nt 3, the onlyknown glycosylated compound of proteose-peptone (Pâquet, 1989). Because of theweak growth-promoting activity of these gly-copeptides, the growth factors were mostIikely peptides.

The enzymic specificity of proteinasesaffects the hydrolysis kinetics of proteose-peptone and the Mrdistribution of peptidesproduced. The trypsin digestion was greaterth an the a-chymotrypsin one. Basic aminoacids (Lys and Arg) located mainly at thesurface of the molecule are more accessibleth an aromatic amino acids and are moreplentiful in the primary structure of compo-nent 3 (serensen and Petersen, 1993).

After proteose-peptone digestion byendopeptidases, the growth of B animaliswas more active. It is reported that bifido-genic activity is increased by enzymic hydrol-ysis especially of bovine casein (Kehagiaset al, 1977; Pech and Bezkorovainy, 1988;Proulx et al, 1992). Moreover, the growth ofB animalis was enhanced when the degreeof hydrolysis was increased (by trypsin-e-chymotrypsin double digestion). Pronase,a non-specifie proteolytic system composedof various types of endopeptidases andexopeptidases, can generally proceed tothe level of single ami no acids. The largespecificity of Pronase explains the highdegree of proteose-peptone hydrolysis and

B animalis activation by proteose-peptone

the large amount of small peptides gener-ated (ail peptides < 5000 Da). In general,the growth-promoting activity was correlatedwith the ratio of medium (2000-5000 Da)and small (1000-2000 Da) peptides. Proulxet al (1992) have shown that bovine caseinpeptides < 2000 Da are the best activatorsof the growth of B infantis, B breve, Blongum, and B adolescentis. Peptides iso-lated from cellular filtrate of Lactobacilluscasei (948-231 9 Da) enhanced the growthof B infantis and B breve (Cheng and Naga-sawa, 1984).

The deficiency of free amino acids in thegrowth-promoting activity could be explainedby possible competition in free amino acidtransport systems of B anima lis.

Dialysis of ppT resulted in importantdecrease of degree of hydrolysis while themodification of Mr distribution of peptidesvaried by only 1% (table III). This differencein degree of hydrolysis of dialysed ppT couldbe explained by the elimination of ammo-nium (ammonium formate buffer) during dia-Iysis. Moreover, elimination of compounds< 500 Da by dialysis of the proteose-pep-tone trypsin hydrolysate (PPT) resulted inan enhancement of the B animalis growth.Poch and Bezkorovainy (1991) showed thatdialysis of x-casein digest (eut-ott 1000 Da)increased the activity of the dialyzed frac-tion. Moreover, no correlation was observedbetween the growth-promoting activity andthe proportion of very small peptides« 1000 Da), as suggested by the resultsobtained with ppTC and PPCT. Therefore,peptides produced by enzymic hydrolysisof proteose-peptone (or of bovine casein;Poch and Bezkorovainy, 1988) seem not tobe simply ami no acid suppliers for thegrowth of B anima lis. Peptides with a Mrrange of 1000 to 5000 Da seem to be abetter nitrogen source of growth factorsthan very small peptides « 1000 Da) orfree amino acids. This result agrees withthose reported previously (Poch andBezkorovainy, 1988; Proulx et al, 1992) and

321

with those of Ibrahim et al (1993) who foundno growth promotion of B longum with x-casein trypsin hydrolysates « 500 or> 6000Da). Growth activators were located in pep-tides having Mrbetween 1000 and 6000 Da.Meanwhile, Cheng and Nagasawa (1984)showed that the upper size limit for peptidetransport through lactic streptococcalmembranes is five or six residues. Accord-ingly, extracellular and intracellular pepti-dases of bifidobacteria could play an impor-tant role in the peptide uptake for theirgrowth (Desjardins et al, 1990).

To conclude, bovine milk proteose-pep-tone appears to contain factors able to pro-mote the growth of this industrial strain ofbifidobacteria. These factors seem to beassociated with fractions containing com-ponent 3. The carbohydrate moiety of com-ponent 3 has a weak efficiency on thegrowth activation that seems to be due tothe peptide. Moreover, proteolytic digestionof proteose-peptone or HHF considerablyincreases this stimulation. The most effi-cient peptides seemed to be of hydrophobicnature and may have a Mr distribution of1000 to 5000 Da; both size and origin ofpeptides could be important for its growthpromotion activity. However, the eventualpresence of unknown substances bound tothe hydrophobic peptides must be consid-ered. Separation and charactetization ofgrowth-promoting peptides are now inprogress.

ln dairy industry, the increase of the ratioof proteose-peptone (previously treated byproteolytic enzymes) in cow milk-based for-mulae could result in an enhancement ofthe B animalis growth.

ACKNOWLEDGMENTS

We acknowledge L Miclo, E Perrin, G Humbertand F Saulnier for their goOOadvice. We expressour gratitude to 1Mangin (Laboratoire de Géné-tique Microbienne, Vandœuvre-lès-Nancy,

322 L Etienne et al

France) for her help in identification of B animalisATCC 27536 by RNA gene restriction patterns.We also thank P Bracquart and H GonzalezMarquez for their constructive comments on thepreparation of this manuscript.

REFERENCESBezkorovainy A, Topouzian N (1981) Bifidobacterium

bifidumvar pennsylvanicusgrowth promoting activ-ity 01 human milk casein and its derivatives. Int JBiochem 13,585-590

Bezkorovainy A, Nichols JH, Sly DA (1976) Pro-teose-peptone Iraction 01 human and bovine milk.Int J Biochem 7, 639-645

Biavali B, Mallareli P, Crociani F (1992) Identification01 bilidobacteria Irom lermented milk products. Micro-biologica 15, 7-14

Cartier P, Chilliard Y, Paquet 0 (1990) Inhibiting andactivating effects 01 skim milks and proteoss-pep-tone Iractions on spontaneous lipolysis and puriliedlipoprotein lipase activity in bovine milk. J Dairy Sci73,1173-1177

Cheng CC, Nagasawa T (1984) Effect 01 peptides andamino acids produced by Lactobacillus casei in milkon the acid production 01 bilidobacteria. Jpn JZootech Sci 55, 339-349

Church FC, Porter OH, Catignani GL, Swaisgood HE(1985) An O-phtaldehyde spectrophotometric assaylor proteinases. Anal Biochem 146,343-348

Coddeville B, Stratil A, Wieruszeski JM, Strecker G, Mon-treuil J, Spik G (1989) Primary structure 01 horseserotranslerrin glycans. Eur J Biochem 186, 583-590

Collins EB, Hall BJ (1984) Growth 01 bilidobacteria inmilk and preparation 01 Bifidobacterium infantis lordietary adjunct. J Dairy Sci 67, 1376-1380

Desjardins ML, Roy D, Goulet J (1990) Growth 01 bili-dobacteria and their enzyme profiles. J Dairy Sci73,299-307

Eigel WN, Butler JE, Ernstrom CA, Farrell HM, Har-walkar VR, Jenness R, Mc Whitney R (1984) Nomen-clature 01 proteins 01 cow's milk: lifth revision. J DairySci67,1599-1631

Girardet JM, Mati A, Sanogo T, Etienne L, Linden G(1991) Fast protein liquid chromatography purificationof hydrophobie Iraction 01 bovine milk proteose-pep-tone and characterization by bidimensional élee-trophoresis. J Dairy Res 58, 85-98

Girardet JM, Linden G, Loye S, Courthaudon JL, Loriento (1993) Study of mechanism 01 lipolysis inhibition bybovine milk proteose-peptone component-3. J DairySci76,2156-2163

Girardet JM, Saulnier F, Driou A, Linden G, CoddevilleB, Spik G (1994) High performance electrophoresis

chromatography 01 bovine milk component-3 glyco-proteins. J DairySci77, 1205-1215

Grimont F, Grimont PAO (1986) Ribosomal ribonucleicacid gene restriction pattems as potential taxonomietools. Ann Inst Pasteur-Microbio/137B, 165-175

GyOrgy P, Rose CS (1955) Further observations onmetabolic requirements 01 Lactobacillus bifidus varpennsylvanicus. J Bacterio/69, 483-490

GyOrgy P, Jeanloz RW, Von Nicolai H, Zilliken F (1974)Undialysable growth lactors lrom Lactobadllus bifidusvar pennsylvanicus protective effect 01 stalle acidbound to glycoproteins and oligosaccharides againstbacterial degradation. Eur J Biochem 43, 29-33

Ibrahim SA, Bezkorovainy A, Hicks CL (1993) Effect 01kappa casein trypsin hydrolysates on growth 01 bifi-dobacteria. J Dairy Sci76 (suppl 1), 123

Kanno C (1989) Characterization 01 multiple lorms 01lactophorin isolated Irom bovine milk whey. J DairySci72,1732-1739

Kehagias C, Jao YC, Mikolajcik EM, Hanse PMT (1977)Growth response 01 Bifidobacterium bifidum to ahydrolytic product isolated Irom bovine casein. JFood Sci42, 146-150

Lowry OH, Rosebrough NJ, Farr AL, Randall RJ (1951)Protein measurement with the Folin phenol reagent.J Biol Chem 193, 265-275

Mati A, Moulti-Mati F, Girardet JM, Fokou E, Belleville-Nabet F, Nabet P, Linden G (1993) Mitogenic acnv-ity 01 hydrophobie lraction 01 proteose-peptone Iromcows', ewes' and goats' milk measured with MARK3 hybridoma culture. J Dairy Res 60, 443-448

Montreuil J, Bouquelet S, Debray H, Foumet B, Spik G,Strecker G (1987) Glycoproteins. In: CarbohydrateAnalysis. A Practical Approach (Chaplin MF,Kennedy JF, eds),IRL Press, Washington, 143-204

Nicolai Von H, Zilliken F (1972) Neuraminidase lromBifidobacterium bifidum var pennsytvanicus. Hoppe-Seylers Z Physiol Chem353, 1015-1016

Pâquet 0 (1989) Revue bibliographique: la Iraction pro-téose-peptone du lait. Lait 69, 1-21

Pâquet D, Nejjar Y, Linden G (1988) Study 01 ahydrophobie protein Iraction isolated lrom milk pro-teose-peptone. J DairySci71 , 1484-1471

Petschow BW, Talbott RD (1990) Growth promotion 01Bifidobacterium species by whey and casein Irac-tlons lrom human and bovine milk. J Clin Microbiol28,287-292

Petschow BW, Talboll RD (1991) Response 01 Bifi-dobacterium species to growth promoters in humanand cow milk. Pediatr Res 29, 208-213

Poch M, Bezkorovainy A (1988) Growth-enhancing sup-plements lor various species 01 the genus Bifi-dobacterium. J Dairy Sci 71, 3214-3221

Poch M, Bezkorovainy A (1991) Bovine milk x-caseintrypsin digest is a growth enhancer lor the genusBifidobacterium. J Agric Food Chem 39, 73-77

B animalis activation by proteose-peptone

Proulx M, Gauthier SF, Roy D (1992) Utilisationd'hydrolysats enzymatiques de caséine pour la crois-sance des bilidobactéries. Lait 72, 393-404

Romond C, Beerens H, Neut C, Montreuil J (1980) Con-tribution à l'étude de la matemisation des laits: inllu-ence in vitro du lait maternel, du lait de vache et delaits matemisés sur la croissance de bilidobactéries.Ann Microbiollnst Pasteur 131A, 309-314

Roy D, Ward P (1993) Genomic lingerprinting 01 dairy-related bifidobacteria by pulsed·field gel élee-trophoresis. FEMS Microbio/12, 151

Roy D, Dussault F, Ward P (1990) Growth requirements01 Bifidobacterium strains in milk. Milchwissenschaft45,500-502

323

Scardovi V (1986) Genus Bifidobacterium Or/a Jensen1924. In: Bergey's manual of systematic bacteriology.vol 2 (Sneath PHA, Mair NS, Sharpe ME, Holt JG,eds) The Williams and Wilkins Company, Baltimore,MD 1418-1434

Shimizu M, Yamauchi K, Saito M (1989) Emulsilyingproperties 01 the proteose-peptone lraction obtainedlrom bovine milk. Milchwissenschaft 44,497-500

sarensen ES, Petersen TE (1993) Phosphorytation, gly-cosylation and amino acid sequence 01 componentPP3 Irom the proteose-peptone Iraction 01 bovinemilk. J Dairy Res 60, 535-542

Vijayalakshmi M, Lemieux L, Amiot J (1986) HPLC ana-Iytical separation 01 peptides a=rding to !heir molec-uiar weight. J Liq Chromatogr9, 3559-3576