HAL Id: hal-00929631https://hal.archives-ouvertes.fr/hal-00929631

Submitted on 1 Jan 1999

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.

Propionic acid bacteria as protective cultures infermented milks and breads

Tarja H. Suomalainen, Annika M. Mäyrä-Makinen

To cite this version:Tarja H. Suomalainen, Annika M. Mäyrä-Makinen. Propionic acid bacteria as protective cultures infermented milks and breads. Le Lait, INRA Editions, 1999, 79 (1), pp.165-174. <hal-00929631>

Lait (1999)79, 165-174© InralElsevier, Paris

Original article

Propionic acid bacteria as protective culturesin fermented milks and breads

Tarja H. Suomalainen*, Annika M. Mâyrâ-Mâkinen

Valio Ltd, R & D, P.O. Box 30, 00039 Valio, Finland

Abstract - Strains of lactic acid bacteria (LAB) and propionic acid bacteria (PAB) were screenedas biopreservatives against food spoilage yeasts, molds and Bacillus spp. singly and in combina-tion. A combination of Lactobacillus rhamnosus strain LC705 and Propionibacterium freudenreichiissp. shermanii strain JS was found to be the most active against yeasts, molds and Bacillus spp. Thecombination was tested for its activity against yeasts and molds in different food applications and thebest results were obtained in fermented milks and in bakery products. An initiallevel of 107 cells.g "fermented milk product and a level of 1as cells.g " sour dough were found to be effective' against yeastsand Bacillus spp. Technology for the use of the combination in different applications was devel-oped. © Inra/Elsevier, Paris.

propionic acid bacteria / lactic acid bacteria / protective culture / antimicrobial activity

\

Résumé - Des bactéries propioniques comme agent de conservation dans les laits fermentés etles pains. Des souches de bactéries lactiques (LAB) et de bactéries propioniques (PAB) ont été tes-tées séparément ou en culture mixte, comme agent de conservation pour leur activité antimicro-bien ne envers des contaminants comme les levures, moisissures et Bacillus. La combinaison dessouches Lactobacillus rhamnosus LC705 et Propionibacterium freudenreichii ssp. Shermanii JSs'est révélée la plus active à l'encontre des levures, moisissures et Bacillus. L'activité de cette com-binaison a été testée envers les levures et les moisissures dans différents produits alimentaires et lesmeilleurs résultats ont été obtenus sur des laits fermentés et des produits de boulangerie, Un niveauinitial de 107 cellules.gtde lait fermenté et de 108 cellules.g " de pâte à pain acidifiée étaient efficacecontre les levures et Bacillus. La technologie d'emploi de cette combinaison bactérienne a été déve-loppée pour différentes applications. © Inra/Elsevier, Paris.

bactérie propionique / bactérie lactique / culture protectrice / activité antimicrobienne /conservation

Oral communication at the 2nd Symposium on Propionibacteria, Cork, Ireland, June 25-27, 1998.* Correspondence and reprints. Tarja.Suomalainen@Valio.Fi

165

166 T.H. Suomalainen, A.M. Mâyrâ-Mâkinen

1. INTRODUCTION

TraditionaIly, propionic acid bacteria(PAB) are known for their ability to con-vert lactate to propionate, acetate and CO2,

which is responsible for the formation ofeyes in Swiss-type chee se [Iû]. A potentialnew role for PAB in food preservation hasbeen recently introduced. 'Microgard TM' isa weIl known PAB based biopreservativein which inhibitory activity has been asso-ciated with diacetyl, propionic, acetic andlactic acid and to a heat-stable 700-Da pep-tide [5]. 'Microgard TM' inhibits most Gram-negative bacteria and sorne fungi [3]. Bac-teriocins of PAB have been studied andreviewed recently by Barefood and Grin-stead [4]. Other uses of PAB in the controlof undesirable micro-organisms have beenreported by Odame-Darkwah and Marshall[13] who showed that the bread spoilagebacterium, Bacillus pumilus, was inhibitedby P. freudenreichii ssp. shermanii. Propi-onic acid and its salts are accepted as preser-vatives for industrial use in bread manufac-ture because of their inhibitory activityagainst mold and Bacillus ssp. [8]. In addi-tion to food applications, PAB and propi-onic acid have also been used as preserva-tives in silage [9].

Most of the research and developmentwork in biopreservation has concentratedon lactic acid bacteria (LAB). These pro-duce various antimicrobial compounds suchas organic acids, H202, diacetyl, bacteri-ocins and bacteriocin-like substances. Stud-ies on bacteriocin-producing LAB werereviewed by De Vuyst and Vandamme [7].The activity of LAB cultures and their bac-teriocin production has been studied in var-ious food systems and sorne bacteriocins(nisin and pediocin) are already used in foodpreservation. Unfortunately, Gram-negativebacteria, yeasts and molds are not inhibitedby LAB bacteriocins, although yeasts andmolds are frequently the cause of hygienicproblems in fermented milk, products andbread.

The objective ofthis work was to developa protective culture which would improvethe shelf life of fermented milks and bread,and which would replace chemical additivessuch as sorbic and acetic acids in differentapplications. Further aims were to optimizethe amount of antagonistic culture neededfor the inhibition of undesirable microbesand to develop a technology for using theculture in fermented milks and in wheatbread.

2. MATERIALS AND METHODS

2.1. Bacterial strains and preparationof the protective culture

Lactobacillus rhamnosus LC705 (DSM 7061)and Propionibacterium freudenreichii ssp. sher-manii JS (DSM 7067) were from the Valio cul-ture collection. The protective culture, calledBioprofit, was produced by growing both strainstogether in a medium consisting of 5 % (w/v)whey permeate (Valio Ltd, Helsinki, Finland),2 % (w/v) casein hydrolysate (Valio Ltd), and1 % (w/v) yeast extract (LAB M, Bury, UK) in aBIOSTAT (B. Braun, Melsungen, Germany) fer-mentor connected to a microfiltration system(Millipore, Bedford, USA) for 48 h at 30 "C atpH 5.5. The protective culture was used in appli-cations as a fresh concentrated culture.

Commercial starters for quark (Probat 505)and for yogurt (V2) were obtained from WiesbyGmbH, Niebull, Germany. The DL-culture forquark was composed of undefined multiplespecies of Le. lactis ssp. cremoris, Le. lactis ssp.lactis, Le. lactis ssp. lactis biovar. diacetylactisand Leuconostoc mesenteroides ssp. cremoris(Probat 505). The yogurt starter consisted of dif-ferent strains of Streptococcus thermophilus andLaetobacillus bulgaricus.

Rhodotorula rubra RHO and Pichia qui/er-mondii PQ were both isolated from spoiled fer-mented milk products and were from the Valioculture collection. They were grown in BactoYM-broth (Difco Laboratories, Detroit, USA) at25 "C for 2 d before use. Baeil/us subtilis P.2.94and Bacillus licheniformis P.l.94 were isolatedfrom a ropy wheat bread. Bacillus strains weresub-cultured twice in BHI-broth (LAB M) for24 h at 37 "C before use. Spores were obtained bygrowing these Baeil/us strains on nutrient agar

Propionibacteria as biopreservatives

plates fortified with 0.003 % (w/v) MnS04·4Hpand 0.02S % (w/v) KH2P04, aerobically at 37 "Cfor 48 h. The spores were collected by centrifu-gation at 4 SOOg for 10 min at 4 "C, washed twicewith sterile water and then resuspended in sterilewater (SO mL) to form a stock solution. The stocksolution was heated at 80 "C for 10 min beforestorage at 4 oc. The stock solution contained9 X 109 spores.ml..:':

2.2. Manufacture of quark

Quark was manufactured according to thethermo-quarg process [IS] using Probat SOS asstarter. The protective culture concentrate wasmixed with the fresh fermented quark mass, aftermanufacture at levels of either 0.01 % (v/v) or0.1 % (v/v). Con trois were prepared without andwith calcium sorbate which was suspended inwater and added to the fresh fermented quarkmass at a final concentration of 0.06 % (w/v).The quark masses were then artificially contam-inated with R. rubra RHO before packing.Yeasts, molds, LAB and PAB were analyzedonce per week during storage at 6 "C for S weeksand organic acids were analyzed at the begin-ning and after 3 weeks of storage at 6 oc.

2.3. Manufacture of yogurt

Control yogurt was fermented with a com-mercial yogurt culture (V2). Test yogurt was pre-pared as for control yogurt with the addition ofthe protective culture to give an initial levelof> 107 cells of protective culture-g " yogurt.The protective culture was added at the samelime as the starter. Yogurt was fermented at 42 "Cfor about 3 h until the pH reached pH 4.S. Straw-berry jam was contaminated either with R. rubraRHO or P. quilermondii PQ at concentrations of1OO-SOO cells-g:". Contaminated jam was addedto yogurt at 14 % (v/v). Basic starters, LAB, PABand yeasts were enumerated once per week dur-ing storage at 6 "C for 4 weeks and organic acidswere analyzed at the beginning and after 3 weeksof storage at 6 oc.

2.4. Manufacture of wheat breadwith sour dough

Bread was made by a natural sour dough pro-cess without intenlional addition of starter. Test

167

bread was manufactured by mixing equal amountsof wheat flour and water with the protective cul-ture to a homogenous mas s, which was then fer-mented for either 4, 10 or 20 h at 30 oc. This sourdough was added at different levels to the finaldough. Control bread was manufactured for eachfermentation lime in the same way but withoutthe protective culture. The final dough was madeof 1 700-2 100 g wheat flour, 170 g Iiquid salt(26 % w/v), 170 g liquid yeast (67 % w/v), 96 gbutter, S00-8oo g water, 172 g Iiquid sugar (77 %w/v) and 400-1 200 g sour dough. An inoculumof 3 X 102 B. subtilis and B. licheniformis sporess'dough was added to each dough. The finaldoughs were baked at 220 "C for 30 min. Breadswere stored at 20-22 "C at an RH of < 20 % at28-30 "C and at an RH of 70 % for 6 d. Thebread was analyzed for Baeil/us spp., organicacids, pH and total titratable acidity (TT A) dur-ing storage for 6 d.

2.5. Microbiological analyses

Sampi es were analysed for lactobacilli onMRS-agar (LAB M, Bury, UK) and for L. rham-nosus LC70S on MRS supplemented withO.OOS % (w/v) of vancomycin (MRSV) (SigmaChemical Co., St Louis, USA), for propionicacid bacteria on sodium lactate agar (YEL) withthe addition of 1 % (w/v) of I)-glycerophosphate(Merck, Darmstadt, Germany), for yeasts andmolds on YCG (LAB M) and for Bacillus spp. onPhenol Red Egg Yolk Polymyxin agar (LAB M)by spread-plating. MRS and MRSV plates wereincubated anaerobic aily at 37 "C for 3 d, YEL-plates were incubated anaerobically at 30 "C for7 d, YCG plates were incubated aerobically at2S "C for 3 d and Bacillus cereus plates aerobi-cally at 37 "C for 20-24 h.

2.6. Physico-chemical analyses

pH and total titratable acidity (TT A) wereanalysed using a modified standard procedure[1] by suspending lOg bread in 2 mL of acetointo break the dough structure and 98 mL of waterand titrating the sample with 0.1 N NaOH to pH8.S with a titrator (Mettler DL 20, GWB, Zurich,Switzerland). Lactic acid concentrations weredetermined enzymatically (Boehringe Mannheim,Mannheim, Germany). Propionic and acetic acidswere determined by gas chromatography (Perkin-Elmer Sigma 3B GC coupled to a AS 300 auto-

168 T.H. Suomalainen, A.M. Mâyrâ-Makinen

sampler) using Chromosorb WAW 80/100 as thestationary phase (3 m x 2 mm, id.) [16]. Ethanoland diacetyl were determined by capillary gaschromatography (Carlo Erba GC 6000 Vegacoupled to a HS 250 autosampler) using a 30 mx 0.32 mm (id.) SPB-I silica column (4.0 mmfilm, Supelco) as the stationary phase [17]. Ben-zoic acid was determined by isocratic reversedphase Iiquid chromatography using a C-18 col-umn (3.2 x 250 mm, packed with 5 mmSpherisorb ODS) equipped with an UV -detec-tor (Waters, Milford, USA) according to [2].

3. RESULTS

3.1. Inhibition of yeasts in fermentedmilk products

Cell numbers of L. rhamnosus LC70Sand P. freudenreichii ssp. shermanii JS inthe protective culture concentrate were2 x l O'? cfu-ml.r! and 2 x 1010 cfu-rnl.:",respectively. A level of 0.1 % (v/v) of the pro-tective culture equivalent to 2 x 107 cells-g"of both strains, inhibited the growth ofR. rubra RHO whereas 0.01 % (v/v), equiv-aient to levels of 2 x 106 cells of both strainss:'. had no effect on the growth of the con-taminant lfigure 1). The initiallevel of yeastwas 2 to 3 X 102 cellsg! quark whichreached 8 x 106 cells-g' in the control andin test quark containing 0.01 % (v/v) of theprotective culture after storage at 6 oc forS weeks. In test quark containing 0.1 % (v/v)of the protective culture, the yeast countsremained at a level of -1 x 102 cells-g"' dur-ing storage at 6 oc for S weeks. The organ-isms in the protective culture did not growduring the storage (data not shown). Thecell numbers of L. rhamnosus LC70S andP. freudenreichii ssp. shermanii JS in theprotective culture concentrate used foryogurt studies were 4 x 1010 cfu-ml.' ' and3 x 10'0 cfu-rnl.:', respectively. The growthof R. rubra RHO was totally inhibited andthe growth of P. quilermondii PQ wasretarded (figure 2) by 0.1 % (v/v) of the pro-tective culture during storage for 4 weeksat 6 oc. This level resulted in initial numbersof 4 x Î 07 cells-g! and 3 x 107 cells-g "

yogurt of L. rhamnosus LC70S and P. freud-enreichii ssp. shermanii JS, respectively.The strains of the protective culture did notgrow in the yogurt during the manufactureand the cell numbers remained at _107

cfu g! yogurt during storage for 4 weeksat 6 oc (data not shown).

3.2. Physico-chemical and sensoryproperties of fermented milks

No difference in the concentrations oflactic acid was obtained in either the quarkor the yogurt, manufactured with or with-out the protective culture after storage at6 oc for 3 weeks. In yogurt, the concentra-tion of propionic acid manufactured withthe protective culture was 2 mg-Iûùg "whereas the concentration of propionic acidin the control yogurt was less than the detec-tion limit « O.S mg-I OOg-I). In quark man-ufactured with the protective culture, nopropionic acid was formed by P. freuden-reichii ssp. shennanii JS as the concentrationof propionic acid was less than the detec-tion limit. The concentration of acetic acidwas higher in quark and in yogurt to whichthe protective culture had been added than inthe controls. The concentration of diacetylwas higher in both quark and yogurt con-taining the protective culture than in thecontrols. A high amount of ethanol wasfound in control yogurt, which was probablyproduced by P. quilermondii PQ (table 1and figure 2).

3.3. Effective process for wheatbread

The initiallevels of L. rhamnosus LC70Sand P. freudenreichii ssp. shennanii JS wereof 1 x 108 and 3 x 108 cells-g :' of sour-dough, respectively. Their numbers did notincrease during the fermentation of the sourdough for 4 or 10 h but after fermentation for20 h, the numbers of L. rhamnosus LC70Sincreased by one log-unit while those of

....1~5.2(.)4Cl

..Q

uf3ltl~ 2 ~

Propionibacteria as biopreservatives 169

7

6 ~-

oo 2 4 53

Time (weeks)

Figure 1. Growth of Rhodotorula rubra RHO at 6 "C in quark to which 0.01 % (v/v) protective cul-ture (0), 0.1 % (v/v) protective culture (e), or 0.06 % (w/v) calcium sorbate (0) was added after man-ufacture, and control without any additive (.).Figure 1. Croissance de Rhodotorula rubra RHO à 6 "C dans du fromage blanc additionné après fabri-cation de (0) 0,01 % (v/v) de culture protectrice; (e) 0,1 % (v/v) de culture protectrice; (0) 0,06 %(w/v) sorbate de calcium; (.) contrôle sans additif.

8.-------------------::::a7

.... 6 ~1

Cl

.2 5 -o

~4-uf3ltl ~al>- 2 -

oo 1 3 42

Time (weeks)

Figure 2. Growth of Pichia quiliermondii PQ at 6 "C in yogurt fermented without (e) or with (0)0.1 % (v/v) of the protective culture and of Rhodotorula rubra RHO in yogurt fermented without (.)or with (0) 0.1 % (v/v) of the protective culture.Figure 2. Croissance de Pichia quiliermondii PQ à 6 "C dans un yaourt fermenté sans (e) ou avecaddition (0) de 0,1 % (v/v) de culture protectrice et de Rhodotorula rubra RHO dans un yaourtfermenté sans (.) ou avec addition (0) de 0,1 (v/v) de culture protectrice.

170 T.H. Suomalainen, A.M. Mâyrâ-Mâkinen

Table I. Levels of organic acids, diacetyl and ethanol in quark and yogurt manufactured with basicstarter and with the addition of 0.1 % protective culture after storage for 3 weeks at 6 oc. (Con-trol = product without protective culture; protective culture = product with 0.1 % protective culture[v/v D.Tableau I. Acides organiques, diacetyle et éthanol dans du fromage blanc et du yaourt produitsavec le levain additionné de 0,1 % de culture protectrice, après stockage trois semaines à 6 oc.(Contrôle = produit sans culture protectrice; protective culture = produit avec 0,1 % de culture pro-tectrice [v/v D.

Quark Yogurt

Control Protective culture Control Protective culture

Lactic acid, % 0.72 0.73 0.94 0.94Propionic acid, mg-l OOg-1 <0.5 <0.5 <0.5 2.0Acetic acid, mg-I OOg-1 38.0 64.0 5.0 10.0Diacetyl, mg-kg" 0.6 49.0 <0.5 24.0Ethanol, mg-kg! 10.0 15.0 220.0 7.08enzoic acid, rng-kg! nt nt <10 <10

nt: nottested / nt : non déterminé.

P. freudenreichii ssp. shermanii JS did not(data not shown). The duration of the fer-mentation of sour dough with the protee-tive culture had an effect on the inhibition ofBacillus spp. in the bread. No growth ofBacillus spp. was observed in bread manu-factured with 10 % of sour dough fermented

....1Cl 5

~Cl4.Q

cri 3~'u~ 2

7 r------

6

for 10 or 20 h with the protective culture,whereas Bacillus spp. started to grow after2 d of storage at 20-22 "C in bread manu-factured with 10 % of sour dough fermentedfor 4 h with the protective culture and in thecontrol bread (figure 3). Similar results wereobtained with 20 % sour dough. The vol-

1.dayo2.day.6.day

oPC4h c4h PC10h c10h PC20h c20h

Figure 3. Growth of Bacillus spp. at 20-22 "C for 6 d in wheat bread made with 10% of sour doughwhich was fermented with the protective culture for 4, 10 or 20 h. Control bread was made withoutthe protective culture (PC= protective culture; c= control).Figure 3. Dénombrement de Bacillus à 20-':'22 "C pendant 6 j dans du pain réalisé avec 10 % depâte acide fermentée par la culture protectrice pendant 4, 10 ou 20 h. Le pain de contrôle était réalisésans culture protectrice (PC = culture protectrice; c = contrôle).

Propionibacteria as biopreservatives

ume of sour dough added to the dough hadan effect on the Bacillus spp. counts espe-cially in warm and moist conditions. Breadwith lOto 20 % of sour dough fermentedwith the protective culture for 10 h inhib-ited the growth of Bacillus spp. totally,whereas in the control bread, the number ofBacillus reached a level of 104 g-l after 6 dof storage at 20-22 "C at an RH of < 20 %

A)

171

(figure 4a). In warm and moist conditions(28-30 oC; RH of70 %) Bacillus spp. wereinhibited for at least 6 d in bread made with15 to 20 % of sour dough fermented for 10 hwith the protective culture, whereas in thecontrol bread, Bacillus spp. reached a levelof 104 cells-g " after 2 d. Bread made with10 % of sour dough showed high counts ofBacillus spp. after 3 d storage (figure 4b).

7 .r--------------------,

6...~5'ti0>4.Q

lIÎ3~.lij 2al

o

Bc 10 %oBc 15 %IIBc 20 %

• Control

B)

2 3

Time (days)

6

7,.....----------------,,....------,6. mBc 10 %

OBc 15 %IIIiIIBc20 %

• Control

3 6O.

2

Time (days)

Figure 4. Growth of Bacillus spp. in wheat bread during (A) storage for 6 d at 20-22 "C at an RHof < 20 % and (B) at 28-30 "C at an RH of 70 % for 6 d. Breads were made by adding 10, 15 or 20 %of sour dough ferrnented with or without the protective culture for la h. Control bread was made with-out the protective culture (PC = protective culture).Figure 4. Dénombrement de Bacillus dans du pain conservé: (A) 6 j à 20-22 "C à une humidité rela-tive < 20 % ; (B) 6 j à 28-30 "C à une humidité relative de 70 %. Les pains étaient réalisés avec addi-tion à la pâte de la, 15 ou 20 % de pâte acide fermentée avec ou sans la culture protectrice pendantla h. Le pain de contrôle était réalisé sans culture protectrice (PC = culture protectrice).

172 T.H. Suomalainen, A.M. Mâyrâ-Mâkinen

3.4. Physico-chemical qualityofbread

The duration of the fermentation time ofthe sour dough had an effect on the pH andon the lactic and propionic acid but not theacetic acid concentrations of bread manu-factured with the protective culture (table Il).It was observed that the longer the fermen-tation time, the higher the concentration oflactic acid and the lower the pH of the breadmanufactured with the protective culture.The level of propionic acid did not increasebeyond JO h fermentation. No differencesin the concentrations of acetic acid wereobtained in either test bread or control bread,except for the lower concentration of aceticacid (lS mg-I Oüg') in the test bread fer-mented for 20 h than in other breads(18-19 mg.Iûôg ").

4. DISCUSSION

Sorbic, benzoic and acetic acids are usedwidely in the food industry as preservativesbec au se of their antimicrobial activityagainst various bacteria, yeasts and molds[6]. Our studies focused on replacing the seadditives with an antagonistic culture con-taining L rhamnosus LC70S and P.freuden-reichii ssp. shermanii JS. The combinedeffect of strains of the protective cultureagainst spoilage yeasts and molds [12] andBacillus ssp. was stronger than the effect ofeither culture alone (data not shown) pre-sumably due to a synergistic effect of thedifferent metabolites of this mixed culture.The mechanism and the mode of inhibitoryaction has not been characterized.

The shelf life of fermented milks wasprolonged by initiallevels of 2 x 107 cells of

Table II. The effect of duration of fermentation of the sour dough on pH, total titratable acidity(TTA) and concentrations of organic acids in wheat bread. Bread was made with sour dough fermentedwith an initiallevel of both L. rhamnosus LC705 and P. freudenreichii ssp. shermanii JS 1-3 x 108

cells·g-I dough for 4, 10 or 20 h by adding 20 % of the sour dough to the final dough. Control breadwas fermented without protective culture for 4, 10 or 20 h by adding 10% sour dough to the dough.Tableau II. Effet de la durée de fermentation de la pâte à pain (levain) sur le pH, l'acidité titrable (TTA)et la concentration en acides organiques dans le pain. Le pain était réalisé à l'aide d'une pâte avec unniveau initial de L. rhamnosus LC705, ainsi que de P.freudenreichii ssp. Shennanii JS de 1-3 x 108

cellules-g " de pâte, fermentée pendant 4, 10 ou 20 h avec addition de 20 % de cette pâte acidifiée àla pâte finale. Le pain de contrôle était fermenté sans culture protectrice pendant 4, JOou 20 h avecaddition de 10 % de pâte acidifiée.

Fermentation time

4h JOh 20 h

Control PC Control PC Control PC

Lactic acid, % <0.04 0.04 <0.04 0.17 <0.04 0.23Propionic acid, mg-l OOg-1 \.0 4.0 \.0 5.0 \.0 5.0Acetic acid, rng-I 00g-1 19.0 19.0 18.0 19.0 19.0 15.0pH 5.5 5.5 5.5 4.8 5.5 4.6TTA,mL 3.0 3.2 3.0 4.5 2.8 4.7

PC = protective culture / PC = culture protectrice.

Propionibacteria as biopreservatives

both L. rhamnosus LC70S and P. freuden-reichii ssp. shermanii JS·g-l product. Thecell numbers of L. rhamnosus LC70S andP. freudenreichii ssp. shermanii did notincrease during the storage of fermentedmilks at 6 "C for 4 weeks, but the protee-tive strains continued to metabolize as theconcentrations of diacetyl and acetic acidin quark and the concentrations of diacetyl,propionic and acetic acids in yogurtincreased during storage. In a productionscale test of quark, the protective culture ata level of 2 x 107 cellsg :' inhibited thegrowth of molds (data not shown). The sen-sory quality of this product was superior tothe control product due to the production ofdiacetyl from citrate by the protective cul-ture. The protective culture did not inter-fere with the basic starters in yogurt as thecell counts of S. thermophilus and L. bul-garicus were similar in both the controlyogurt and in the yogurt manufactured withthe protective culture.

Inhibition of yeasts and molds by the pro-tective culture cannot be based on pH andacids alone. Propionic acid and its salts areprimarily inhibitory to molds and Bacillusspp. at concentrations of O.I-S % [8]. In ourstudies, using the protective culture, the con-centration of propionic acid reached 0.002 %(yogurt), < O.OOS % (quark) and O.OOS %(wheat bread). Acetic acid is effectiveagainst yeasts and bacteria at concentrationsof 0.4-0.8 % (w/v) [8], while, the concen-trations in the present study were 0.06 %(w/v) in quark and < 0.02 % (w/v) in bread.Concentrations of diacetyl in fermentedmilks made with the protective culture werebelow the inhibitory concentration of300 mg-kg! [S]. Concentrations of lacticacid were 0.78-0.94 % in fermented milksprepared with or without the protective cul-ture. Sorne reports [S, Il] suggest that theformation of benzoic acid by lactic acid bac-teria may be partially responsible for inhi-bition of yeasts but according to our results,the level of benzoic acid in yogurt did notincrease with the protective culture.

173

The optimal sour dough process wasdeveloped with the protective culture ofLC70S and JS. The initiallevel of 1-3 x 108

ofboth L. rhamnosus LC70S and P.freuden-reichii ssp. shermanii JS g-l sour doughwith la fermentation time of 10 h and withthe addition of over 10 % to the doughimproved the shelf life of the wheat bread.Inhibition of Bacillus spp. in wheat breadmay be partially explained by lower pH andhigher amounts of lactic acid in test breadcompared to control bread. In addition, thesensory quality of the test bread was optimal.The pH of the wheat bread made with sourdough is optimal around pH S while, belowpH 4.2, the product is considered too acidic[14]. The growth of Bacillus species causingropiness in wheat bread was found to bedependent on temperature and moi sture ofthe environment and was faster at an RH of70 % at 28-30 -c than at an RH of <. 20 %at 20-22 oc.

ACKNOWLEDGEMENTS

We thank Mrs H. Mâkelâ and Mrs P. Sirviëfor their technical help. The work was supportedby Valio Ltd, and Technology Development Cen-tre of Finland (TEKES).

REFERENCES

[1] Anon., Sâuregradbestimrnung von Brot, in: Stan-dard-Methoden für Getreide, Mehl und Broth,6th ed., Arbeitsgemeinshaft Getreideforsehung,Verlag Moritz Schâfer, Detmold, 1978,pp. 106--107.

[2] Anon., Benzoie aeid, sorbie aeid andp-hydrokxybenzoie aeid esters, Liquid chro-matographie determination in foods, NMKL(Nordie Commitee on Food Analysis) methodNo 124, 2nd ed., 1997.

[3] AI-Zoreky N., Ayres J.W., Sandine W.E.,Antimierobial aetivity of Microgardts, J. DairySei. 74 (1991) 758-763.

[4] Barefood S., Grinstead D.A., Baeteriocins fromdairy propionibaeteria and inducible baeteri-ocins of laetie acid baeteria, in: Hoover D.G.,Steenson L.R. (Eds.), Bacteriocins of LaetieAeid Baeteria, Academie Press ine., San Diego,1993, pp. 219-231.

174 T.H. Suomalainen, A.M. Mâyrâ-Mâkinen

[5] Daeschel M.A., Antimicrobial substances fromlactic acid bacteria for use as food preservatives,Food Technol. 43 (1989) 164-167.

[6] Davidson P.M., Branen A.L., Antimicrobials inFoods, Marcel Dekker Inc., New York, 1993.

[7] De Vuyst L., Vandamme E.1., Bacteriocins ofLactic Acid Bacteria. Microbiology, Geneticsand Applications, Blackie Academie & Profes-sional, Glasgow, 1994.

[8] Doores S., Organic Acids, in: Davidson P.M.,Branen A.L. (Eds.), Antimicrobials in Foods,2nd ed., Marcel Dekker Inc., New York, 1993,pp. 95-136.

[9] Florez-Galarza R.A., Glantz B.1., Bem C.1., VanFossen L.D., Prevention of high moisture cornby microbiological fermentation, J. Food Prot. 48(1985) 407-411.

[101 Hettinga D.H., Reinbold G.W., The propionicacid bacteria - a review II. Metabolism, J. MilkFood Technol. 35 (1972) 358-372.

[1 1] Holzapfel W.H., Geisen R., Schillinger U., Bio-logical preservation of foods with reference toprotective cultures, bacleriocins and food-gradeenzymes, Int. J. Food Microbiol. 24 (1995)343-362.

[12] Mâyra-Mâkinen A., Suomalainen T., Lacto-bacillus casei spp. rhamnosus, Bacterial Prepa-ration Comprising Said Strain, and Use of SaidStrain and Preparations for the Controlling ofYeast and Moulds, United States Patent US 5378458, 1995.

113] Odame-Darkwah J.K., Marshall D.L., Interactivebehavior of Saccharomyces cerevisiae, Baeil-lus pumilus, and Propionibacterium freudenre-ichii ssp. shermanii, Int. J. Food Microbiol. 19(1993) 259-269.

[14] Pagenstedt B., Mechanisierung von Weizen-mehl- Vorteig-Stufen, Getreide, Mehl u. Brot 37(1983) 147-153.

[15] Spreer E., Cheese Manufacture, in: Spreer E.(Ed.), Milk and Dairy Product Technology, Mar-ceI Dekker Inc., New York, 1998, pp. 243-334.

[16] Svensen A., Gassfraktometrisk metode til kval-itativ og kvantitativ bestemmelse av flyktigemonokarbonsyrer i ost, Melding nr. 97 fraMeieriinstituttet, Norges Landbrukhogskole,1961.

[17] Ulberth F., Headspace gas chromatographie esti-mation of sorne yogurt volatiles, J. Assoc. Off.Anal. Chem. 74 (1991) 630.