International Journal of Food Ali('n)hiolog)" 17 (It)t)3) 311-320 311 , It193 Elsc~ricr Science Publb, hcrs B.V. All right~, rc.,,cncd OIl)N-Iol)5./t,)3/S116.|10

FOOD 00547

Inhibition of the growth of yeasts in fermented salads

M.H. Bones t roo , J.C. de Wit, B.J.M. Kusters and F.M. R o m b o u t s I)cpartmc~t of l:ood S('i(w('('..-Igri('uhural Unit cr.',ity. II'(tg('ni~gcn Th(" ;~)'ctherhmd.~

(Received 1) i)cccn)bcr Itltil ' rc~b, ion rccci,.cd 17 June 1~1~,)2, accepted 3 Augu.~! It}g2)

5:|l;.ids (,:OlllpOsL.'d of ~cgchlhlc~, :.lntl/or mual ill an oil-ill-v.;ller cnzttlr..ioll x~,crt" prepared by l'~rmunt:_|liOl) for 7 h at 42'(" or 45"(" ~ith MrailP, of /.,l('l,)t,,t(illtt~ H~p. Their ,,tabilit.x Io~ards spoilage yea~,ts t~'as ~,tudicd I.l.~illg .~a('l'hllrottly('('.~ ('('r('tl.shl('. ,~d('('h0)r,)ttw,++'s ,'+ttrlol), ~md T, ruhtvnJra &'lhn~t'('kii, isohttcd from salads, as ~ell ;is Pi¢hh~ Ilh'll)lhrallat'til('iclP, alltl Z}'~oslll'(hlll'Otllt'(t'~ I)ailii. Salads fermented x~'ith good lactic shlrlcrs usually Ilad pll valuc~, of _< 4.2 and lactic acid conccntration,~ of ().2N Io 0.43~i (v./v,). Iligh nl.lnll')crs t)f spoilage yca~,l~, (and prod).lClion of htrgc ~.Ohlll'lc~ o[ (.'(.),) ~crc not attained ill these salad.~, provided the initial t'oll¢Clllldtion of ,~poil:lt~c ~C;.I~I>, ~.~,a~ ,qlllicicntl¥ h)v,- { _< IIX} ( 'FU/g ) . Inhibition of ~poilagc ycast~ ill I:tclic fcrnlenlcd .~alittb, it, probabl.~ title to h.tctic acid, the Io~. t, storage h:mpcraturc and Ihc Iov, rc~,idual oxygen con¢~:nlr:lliOl).

Key word,< Saladr,: FerlllCnlalion: l.~|clil: acid baclcria: I.at'lol)d(tl/It~: ~l'ci.|sl~,; .~a(tharonlyt'e.~; Tortdav)ora: ( ; ro~lh inhibition

Introduction

Salads, composed of vegetables and/or me~t in an oil-in-water cmui,,ion with a h)w pH ( < 4.8) due to the addition of acids, are popular in Western countries. "lhe shelf life of these salads is influenced mainly by low pH, storage temperature and the use of chemical preservatives, i.e. sorbic and benzoic acids. Higher demands on quality of raw materials, better preparation methods and especially the use of preservatives, have led to a shelf life of these salads of 6-8 weeks, if stored below 7°C. However, consumer trends arc towards salads without chemical preservatives and with a less sour taste. The omission of preservatives limits the shelf life of salads to 2-3 weeks, if stored below 7°C (Gromzik, 19t) l ). Micro-organisms causing spoilage arc lactic acid bacteria, yeasts, moulds and bacilli (Smittle, 1977: Kirsop and Brocklchurst, 1982: Brocklehurst et al., It)83: Lund et al., 1984). According to literature data. the most important spoilage yeasts arc Saccharomyces eriguus and Saccharomyces dairensix. Pichia membramt~]'(wienx. Debat3"omyces kloeckeri.

('orrc,v~mdeswe addrc.~,~: M.]I. l]onc~troo. Dcpartmcn! of Food Science, Wagcningcn Agricultural University. Bomcn~'eg 2. ()703 l i d Wageni,lgcn. The Netherlands. Fax: + 31 $370 N4S93.

312

Geotrichum cam~Mum, some CamlMa spp. and Zygosaccharomyces bailii are a l~ found in salads (Smittle, 1977: Baumgart, 1977: Smittle and Flowers, 1982; Brock- lehurst et al., 1983; Brocklehurst and Lund, 1984).

It is generally accepted that the antimicrobial activity of weak acids such as acetic acid and lactic acid is due to the undissociated acid molecules. An increase in concentration of these molecules, independent from a decrease in pH levels, will improve the microbial stability of salads. This can be achieved by using buffer systems of weak organic acids and their salts (Debevere, 1987). it has been suggested that an additional advantage would be that taste problems are dimin- ished. The shelf life of salads could also be extended by technological solutions, such as the production of salads under strict hygienic conditions, in closed systems with rigorous exclusion of oxygen and the use of modified atmospheres. These methods require, however, large capital investments. Moreover. the growth of yeasts having a fermentative ability and especially lactic acid bacteria will not be prevented (Buick and Damoglou, 1989: Farber, 1991).

Lactic acid fermentation of foods generally is a low-cost method of food preservation in which spoilage and pathogenic organisms are inhibited, mostly by acid production and pH reduction but also by lowering the oxidation-reduction potential, competition for essential nutrients and possibly by the production of inhibitory compounds: antibiotics and hydrogen peroxide. In a previous paper (Boncstroo et al., 1992) lactic acid fermentation was proposed as a novel way of preparation of salads, if sufficiently attractive, a certain interest may be expected from the market fi~r these fermented salads, because of their mildly sour taste, their possible health effects and especially because the use of preservatives can be abandoned.

The objective of this study, as part of an investigation on lactic acid fermented salads, was to investigate the possible inhibitory effect of fermentation on the growth of yeasts. Therefi)re, yeasts, isolated amongst others from spoiled, industri- ally produced salads, were deliberately added to salads, which were then fer- mented. Changes in microbiological condition and chemical composition were monitored during storage at 7°C.

Materials and methods

Organisnls and culture comlitions Lactobacillus platHariln~ and Lactobacillus acidophih¢s wcrc obtained from our

laboratory collection (Boncstroo ct al., 1992). Prior to use the strains were cultivated for 24 h at 30°C or 42°C in a broth with the same composition as MRS broth (Merck, Darmstadt, FRG), but containing sucrose (Merck) as carbohydrate, in,~tcaa~ ef glucose, followed by subsequent cultivation in the same broth for 16 h at 30°C or 42°C. Pichia tnembranaefaeciens CBS 107 and Zygosaccharomyces bailii QST2877 were kindly provided by Dr W.J. Middelhoven (Department of Microbi- ology, Wageningen Agricultural University, The Netherlands) and Mr P. Breeuwer (Department of Food Science, Wageningen Agricultural University, The Nether-

313

lands), respectively. Other yeast strains were isolated, using Oxytetracycline Glu- cose Yeast Agar (OGYA, CM545, Oxoid Ltd., Basingstoke, UK) incubated at 25°C, from fresh salads, sell-by date salads and spoiled salads with bulging packings, all produced without preservatives. Strains were purified and after morphological screening, preliminary characterized with the API ATB 32 C identification kit (AP! System S.A., France). Predominant yeast species were identified by Dr M. Th. Smith (CBS Yeast Division, Identification Service. Delft, The Nethcrla~|ds). Yeast cultures were maintained at 4°C on Malt Extract Agar (CM59, Oxoid). Prior to usc the yeast cultures were cultivated twice, at 25°C for 48 h. consecutively, in Malt Extract Broth (CM57, Ox,,~id).

Prt'l~ttrtltion and storage of scd, ds Four types of salads wcrc prcparcd, onc containing 65c; (w/w) pcclcd, sliced (3

mm) and ctR~ked (5-10 rain, t~)-I00°C) potato tubers (cultivar Bintje} and 35c~ {w/w) dressing; the second containing 23~/" (w/w) shredded (10 x 6 × 3 ram) leek, 23~ {w/w) shredded ( 15 × ill × 3 ram) cabbage, 10~; (w/w} cooked ham and 44~ dressing: the third containing 5{}c; (w/w) gratcd {30 × 3 × 3 ram) and blanched ( I min, I(X}°C} carrots and 5()¢; (w/w) dressing: the fi~urth containing 50c/,; (w/w) shredded ( 15 × 10 × 3 ram) cabbage and 5(}c,~ dressing. Thc dressing was com- posed of wlter (35.0-52.2~ (w/w)), soybean oil (35-5t}r~; {w/w)), sucrose (6¢; (w/w)), salt (1-1.8~ {w/w)), cgg yolk {3-3.5r,~ (w/w)) and thickening agents (2.8r~, (w/w)). Occasionally small amounts of seasonings were added to the salads. The salads were inoculated with 10"-!0 v cells of lactic acid bacteria per g and filled in glass jars (liHI g salad, hcadspacc approx. 50 ml) air-tightly sealed with metal Vapor-Vacuum"" Twist-Off" caps (Whitc Cap international, USA). Thc jars were incubated in a watcrbath for 7 h at 42 or 45°C, and then chilled in ice watt! to a temperature of 7°C and stored at this temperature. Uninoculatcd, acidified (0.4~ (w/w) lactic acid) salads were used as control.

Growth at dif]'t'rent tonperclturt',~ Yeast strains, grown twice at 25°C fi~r 48 h, consecutively, were inoculated into

Malt Extract Broth (CM57, Oxoid) and cultivated at different temperatures. Growth was monitored either by pipetting aliquots in micro-titer plates and reading optical densities at 620 nm or by enumerating the yeasts on Oxytetracy- cline Glucose Yeast Agar (CM545, Oxoid).

Et'ah~ation of the quafiO" oj'li'rmer:ted ,scd, d,~ Me, surement oJ'pH. The pH of the salads was detcrmiJlcd with a pH-eiectrode

(NOl, Schott Ger[itc GmbH, Hofheim a. Ts., FRG) coupled to a combined pH/mV-metcr (pH522, Wisscnschaftlich-Technische Werkst~itten GmbH, Weil- heim, FRG).

Microbiological qualiO'. The vegetable salads were judged on the basis of microbiological analyses directly after fermentation and up to three weeks of storage at 7°C. Mesophilic aerobic spoilage bacteria were enumerated using Gelysate Agar (GEL): 5 g Bacto-peptone (Difco, Detroit, Michigan, USA), 5 g

314

sodium chloride (Merck) and 14 g agar (LI3. Oxoid) per !, pH 7.6 _+ 0.1, incubated at 30°C for 3 days. Enterobacteriaceae were counted on Violet Red Bile Glucose Agar (VRBG: CM485, Oxoid) with a top layer of the same medium and incubated fi~r 24 h at 30°C. Lactic acid bacteria were selectively enumerated on MRS medium (Merck) with 12 g of agar (L13, Oxoid) and 2 g of Delvocid (Gist Brocades. Delft, The Netherlands) per I. incubated under anaerobic conditions (Anaerocult system, Merck) at 30°C for 3 days. Yeasts and moulds were counted on Oxytetracyclinc Glucose Yeast Agar (CM545. Oxoid), incubated at 25°C fi)r 5 days, as recommended by Banks and Board (1987).

Determbtation of sugars, etha,ol a ,d orgallic acMs. Sugars. ethanol and organic acids in fermented salads were determined by means of HPLC as described earlier (Bonestr,ao ct al.. 1992).

Results and Discussions

In most fermentation processes a rapid pH decrease, in addition to a low final pH, is necessary to achieve a microbiologically stable product. As can bc seen in Table I a low pH ( < 3.9) was reached in the salads after 7 h of fermentation at 42°C. Although the fermented salads contained raw ingredients, i.e. cabbage, no growth of yeasts occurred. During storage at 7°C the pH remained low and the fermented sahtds were microbiologically stable fl~r at least 3 weeks. The fermented salads contained 11.35 ___ ().113C/¢ (w/w) lactic acid alter 7 h of fermentation at 42°C, increasing to 0.46 ± ll.I)9¢/~ (w/w) ;.liter subsequent storage at 7°C for 3 weeks. Little or no acetic acid could be found in the fermented salads, whilst ethanol

TABI,[" I

Microbioh)gical condition and pl I of fermented salads composed of cabbage and dressing

Star ter p i t ('tiLl d after 7 h I, (log N / g ) ( ' F U al ter s torage c (log N / g )

no. " Inili~.l Al'lcr After E N T " Y[-A I,AB MAB E N T Y E A LAB MAll 7 h i, s torage "

9 5.117 3.~11 3.ll4 < I.I)l) < 1.1111 9.12 < 3.711 < I.|1t1 < 1.1)1) S.~)9 ,,~ I.|H) 21) 5.112 3.82 3.57 < I.Ill) < I.()l) 8.71 < 3.711 < I.II(| < I.qll) 8.42 2.114 21 .1.1.)3 3.8¢7 3.62 < I.IIt) < I.(lll 8.83 < 3.711 < I.(ll) < I.|HI 8.5f~ < 1.1111 23 5.(17 3.811 3.(14 < I.(li) < I.I)0 8.8~ < 3.71) < I.(lll < I.(ltl 8.69 < I.(ll| 24 5.13 3.84 3.78 < 1.1)1) < 1.11t) 8.91 < 3.711 < I.II11 < I.Illl 8.27 < I.II{)

Resttlb~ ;are the nle;.ins of deterfflin;.itions of dttplicilte s;.tmples. ~ Ntmlber,', refer to l.at'lo~at'ilhl.~ I~ho,ar, t , strain numbel:., tlcscribcd ill Bol|¢Mroo cl it(. (Iqq2L h After 7 h of fermentat ion at 42°C. ¢ Al ter 7 II of fermental ion at 42°C iliad subsequent s torage ill 7°C for 21 days. d CFU, Colony filt'lning units. c ENT. ]',',terobacteritlcette" YEA, yeasts" LAFL lactic acid bacteria: MAB. mes~,philic aerobic spoilage

bacteria.

315

production was absent and minor amounts of sugars, generally less than ().Sr~ (w/w). were used. by the lactic acid bacteria (results not shown).

,~'a('('haronlyce.~ t:~'igtltt.s, Sat't'hto'omyt't'.s cercrisiac, "l~icho.woron beigelii and Tortt- laspora delbnwckii were isolated from fresh salads, sell-by date salads and spoiled salads with bulging packings, all industrially produced without prese~'alivcs. The strains failed to grow in Malt Extract Broth at 3-3.5°C, but they showed g ~ d growth at 4-4.5°C (data not shown). They showed good growth at higher tempera- lures, except for Saccharomyces (:riguu.~" and Toruh~xlu}ra dclhrueckii, of which no growth was observed at 3(1°C. None of these strains showed growth at 42°C. At a temperature of 7°",_ ", all strains reached maximum cell numbers within 7-14 days (results not shown). Most strains possess a fermentative mcta~flism and arc thus able to grow in the absence of oxygen, except Trit'ho, vporotl I~cigelii, which possesses an oxidative metabolism.

Trichosporon hdgelii, isolated from fresh salads, failed to grow in Malt Extract Broth at pt t 4.5 and ptt 4.0 at 7°C in the presence of I).5C; acetic acid. The other strains mentioned ab~wc showed moderate to good growth under these conditkms. With the exception of .~ilt't'htlronlyt't's ccreri.shw, which could tolerate IC~ acetic acid at pH 4.5, but not at pH 4.(I, they tolerated I.O(:~; acetic acid at both pH levels. It wits clear that acetic acid is more inhibitory to the yeasts than lactic acid. Addition of lactic acid in concentrations of 0.5~ and l.l) to t}.5c~ , acetic acid did not change the growth patterns of th~ 4 yeaq s!rains (data not included).

Various salads, deliberately inoculated with a mixture of tile above mentioned yeasts (approx. 10[} cel ls /g salad), fermented with l, at'tobat'ilhls spp. tbr 7 h at 42°C and stored at 7°C, remained microbiologieally stable ibr at least 3 weeks, as filr its yeast growth is concerned. Growth of yeasts was obviously suppressed. However, with higher inoculation levels ( > 10t]O ccl ls /g salad), growth of yeasts within 2 weeks generally could not bc prevented (data not included).

Other experiments, in which carrot salads were inoculated with single strains of the above yeasts ( >_ 250 cel ls /g salad) and the yeasts Pichhl memhranaefaciens and Zygosacchatwnlyces hailii, fermented fi,r 7 h at 42°C with different Lactohacillus plantanon strains and subsequently stored at 7°C, gave comparable results (Table !!). Strains of Tonlht.~pora delhnwckii, Pit'hhl membnmaclbcit'ns and especially Sat'charomyce.s t'elrri,shle showed abundant growth in the fermented salads after ll} days of storage at 7°C, while growth of other strains was more or less inhibited. After 20 days storage at 7°C, most salads were spoiled by yeasts, except fi}r the salads inoculated with Stlt'charoltl)'t't'x t:~gulls 2 and Zygostlt'charonlyces btlilii, and the salad inoculated with "l)'it'ho,~poron be(gelii fermented with Lactobacilhts pltm- tarum no. 20. The fermented salads contained 0.28 ± (I.(}2q; (w/w) lactic acid after l0 days storage at 7°C, increasing to 0.35 +_ O.07f~ (w/w) after 20 days. Little or no acetic acid could be found in the fermented salads with little or no yeast growth (_~ 0.09_+ 0.02c~ (w/w)). whilst ethanol production in these salads was nearly absent ( _< 0.02 + I}.Ol~i (w/w)). Minor amounts of sugars, generally less than 0.5% (w/w), were used. However, in the salads deliberately con'aminated with Saccha- romyces cererisiae and Torula,~pora delbrueckii larger amounts of acetic acid and

311)

TABLI- I!

pl I alld number of yeasts in lermellletl cam)! salads inoculated ~ith _> 250 spoilage yeasts per g

Species Slarler pl l at difl'crenl limes i, ( 'FU • at differenl l imes

m). " A B (" l ) A (" D

Saccharomyc('.~ (aigm1.~ 1 d (1 0.51 4.52 4.02 3.81 2.4~ < 2.(X) 4.64 20 6.51 -1.3t~ 4.117 3.85 2.48 < 2.IX) 2.78

Sacchar(uny('('~ (:~iguu.~ 2 9 h.4N 4.3N 3.92 3.73 2.40 < 2.IX) < 2.IX) 20 0.4~ 4.47 4.(X) 3.73 2.40 2.52 < 2.(X)

Sacchur(unyc('.~ ('('rcci.~hw ~J 6.5! 4.47 4.(X) 3.69 2.77 > O.IX) > O.IX) 20 I).51 4.33 3.04 3.1) (} 2.77 > I).(X) :> (,.00

Tricho.vn)ron h('ig('lii () 0.45 4.23 3.84 3.72 3.31 < 2.(X) 4.78 20 t).51 4.32 4.04 3.Sl 2.05 < 2.00 < 2.(X)

Tor.lavmra (h'll)rtw('kii 9 1).4~ 4.22 4.1)t) 3.77 2.42 4.16 > 6.(X) 20 6.4S 4.22 4.12 3.73 2.42 3.81 ~ O.(X)

l)ichia mcml,rana(~l~wh'n.~ q 6.45 4.30 3.911 3.75 2J)l) 4.61 > O.IX) 21) I).45 4.30 3.82 3.77 2.1)0 4.81) >/).IX)

Zyg+aa('t+haromyc('~ built| 9 h.45 4.44 4.111 3.7S 3.37 3.h5 3.73 21) 6.45 4.33 4.(11 3.~3 3.37 3.31 3.34

:' Nunlbcrs refer to I,actohucilh~x phmlarmta Mrain n),mlbers described ill Boncsl roo el al. (1902). h A. boil)re fermentat ion: B, after 7 h fermentat ion at 42~'(': ( ' , afler lermentation and subsequent

storage at 7:( ' fi)r 10 d;l.Vs: o . after fcrmcnlalion Lllld subsequent storage ~ll 7~( ' f i)r 2{) d~lVs. *" ( 'o lony forming tlnils of yeasts ( log N / g ) . average o f tltlpli¢;t~¢ dulcrmil | : l lhms. d .~gl('t'htlroltll'('()?i Ct~llll,S I was isolaled 1ton1 salad. Sm'charomyces ¢.l'(¢uus 2 was isolaled from

equipmcnl.

ethanol were found, with corresponding hlrger sugar consumption (results not included).

(h'o~,~th of ,~LIct'hLIroIHY('CS cererisiae was further studied in potato salads and salads containing leek, cabbage and hanl el'able Ill). When inoculated with 70-90 cclls/g salad, 11o growth of yeasts occurred within 4 weeks at 7°C in the fermented salads, compared to non-fermented, acidified control salads. Potato and leek-cab- bage-ham salads contained 0.41 +_ 0.()iqf (w/w) and 0.43_+ (l.t)2~ (w/w) lactic acid after fermentation fi)r 7 h at 45°(7, respectively. After 42 days storage at 7°C the lactic acid concentrations were 0.72 _+ 0.l)8e/~ and 0.46 _+ 0.()7g; (w/w), respec- tively. Little or no acetic acid could be found in the fermented salads (_< 0.(13q~ (w/w)), whilst ethanol production was nearly absent (_< (I.(I5C~ (w/w)). Minor amounts of sugars, generally less than ().5r~ (w/w), were utilized. The fermented salads had a mildly sour taste and good texture. A striking difference of the fermented salads with commercially available salads, acidified with acetic acid, was that the ingredients kept more of their own characteristic taste. The addition of small amounts of seasonings further contributed to this taste.

it is frequently reported that raw vegetables can contain high numbers of ~easts (Hartog and Jansen, 1986: King Jr. et al., 1991; T6r6k and King, 1991), mostly being asporogcnous, non-fermentative, and representing the genera Ct3"plot'oct'zls, Rhodolorlda, SI)oroholomyces and, in some cases, Toruh~psis. Cllndidtl species are

317

TABLE III

pl ! and numbers of .~easts in salads inoculated with 711-t111 cells of 5a~ chanmtyt'ex cereri~ia¢ per g salad

Salad Starter pl[ at different times i, ('FU ' at diffi:rcnt times type no. " A B (" l) E A (" D E

I a 4 5.11t) 3.t/t) 3.~}1 3.tH 3.92 1.86 < 1.711 < 1.711 > 3.1111 control ~ 4.1111 4.1111 3.t~4 3.80 3.88 I.SI) 4.15 > 5.1111 > 5.1111 control * 4.1)5 3.8, v, 4.112 3.t,b3 3.t,~S I.S(~ 5.25 4.65 > 3,1111

II 25 5.69 4.114 4.35 4.13 4.07 l.t~t~ -, 1.711 < 1.711 > 2.80 control " 3.52 4.lhr) 4.18 4.18 4.22 1316 *, 1.711 --> 4,1111 > 4.1111 control i 3.57 4.11 4.1t,~ 4.24 4.111 1316 <- 1.711 > 4.(lil "-. 6.1111

" Lactobacilht~ acidophiht~ strains. ~' A. heft)re fermentation: B, after 7 h t'ermcntation al 42'('- (', after fermentation and subsequent

storage at 7°C for 14 days: D, after fermentation :rod snhseqncnt storage at 7~(." fl~r 28 days: l-, after fermentation and suhscqncnt storage at 7~(" ft)r 42 days.

r Colony ft)rming units of yeasts (log N/g), average of duplicate determinations, 'l Sah|d type I. potato salad; salad type II, s.'dad containing leek, cabbage and hi~m, " Not inoculated wilh L. achh~phih¢s, acidified with 11.4~; lactic acid. i Not inoculated with L. m'ichqduhes, acidified with 11.4~; lactic acid, snhj,:cted to a temperature of

45°C during 7 h.

usual ly in the minor i ty ( M i d d c l h o v e n and van Baalen. 1988: Phaff and S ta rmcr . 1990). In f e rmen ta t i on expe r imen t s with sa lads con ta in ing raw ingredients , i.e. cabbage , no growth of yeas ts occur red , indica t ing that most o f the yeasts , na tura l ly occur r ing on fresh vegetab les , do not cause spoi lage o f f e r m e n t e d salads.

Salads , inocu la ted with acid to le ran t yeasts , and f e r m e n t e d with lact ic acid bac te r ia , r ema in microbiologica l ly s table for at least 3 weeks, as far as yeas t growth

is concerned . A l t h o u g h the re la t ively low number s of yeas ts c o m p a r e d to the lactic ac id bac te r i a may account in par t fi, r this p h e n o m e n o n , o t h e r factors, a m o n g which the amoun t of acid, the low s to rage t e m p e r a t u r e and the oxygen concen t r a t ion in the salads , have to be cons ide red .

R e g a r d i n g the inhibi tory effect o f ac ids on the growth of micro-organisms , not only the a m o u n t o f acid is impor tan t , but also the deg ree o f d issocia t ion of the

ac ids which is in f luenced by pH. i t is genera l ly accep t ed that the an t imicrob ia l activity o f weak acids is due largely to und i ssoc ia ted acid molecules , in the i r s tud ies conce rn ing the spoi lage of m a y o n n a i s e - b a s e d salads. Kirsop and Brockle- hurs t (1982), showed that yeas ts were to be found in the aqueous phase o f the mayonna i se emuls ion and were not to any subs tan t ia l deg ree assoc ia ted with the

sur face o r the in te rna l par t s of the vege tab le t issue p resen t in the mayonnaise . The re fo r e , it s eems r ea sonab le to ca lcula te the concen t r a t ion o f und issoc ia tcd ac ids in the aqueous phase o f the salad. Ca lcu la t ed values for the concen t r a t ion o f und i ssoc ia ted lactic acid in the aqueous phase for ca r ro t (Table Ii) , l eek-cabbage- ham and p o t a t o sa lads (Tab le 111) a re 0.27 + 0.116, 0.34 + 0.01 and 0.24 + 0.03, respect ively. The concen t r a t i ons of und issoc ia ted acet ic acid in the aqueous phase o f the above m e n t i o n e d sa lads genera l ly were < 0.1)8c~ ( w / w ) . T h e s e values a re

31,~

significantly lower than the concentrations of undissociated acids in Malt Extract Broth with 0.5c.,; acetic and 0.5% t~r l.(lq~ - lactic acid. Regarding the level of acid in the non-fermented, acidified control salads and the extreme levels of acid toler- ance. when grown in Malt Extract Broth with acetic and lactic acids, inhibition of yeasts in fermented salads is not solely due to the amount of acid.

We are aware that acetic acid generally is more inhibitory to yeasts than lactic acid. According to several researchers a synergistic antimicrobial effect of combi- nations of these acids exists (Mtrm. 1983: Adams et al.. 1989). However. in our experiments with Malt Extract Broth synergistic effects of the two acids were not observed. These findings were confirmed by various experiments done in our laboratory in which dressings were inoculated with Saccharomyces cererisiae, which showed good growth at 7°C and pH 4.5 in the presence of 0.78% (w/w) acetic and 1.17r~ (w/w) lactic acid (values calculated on the aqueous phase).

In our opinion the effect of a storage temperature of 7°C on the growth of spoilage yeasts in the salads is also limited. Growth of spoilage yeasts is only inhib!tcd at storage temperatures below 3.5°C. in broth cultures all yeast strains tested reached maximum cell numbers within 7-14 days at 7°C. At fermentation temperatures ~ > 40°C), however, growth of these yeasts in broth was retarded or completely inhibited, possibly due to their psychrotrophic nature. These effects were intensified by the inclusion of acids in the broth cultures. However. in control salads, acidified with 0.4% (w/w) lactic acid, and subjected to a temperature of 45°C during 7 h, growth of Saccharomyces cererisiae was abundant (Table i11). Problems with growth of thermotrophic yeast species, such as Kluyreromyces spp., during fermentation or storage were never encountered.

In view of the considerable amount of sucrose added to the dressings used, i.e. generally 6c~ (w/w),thc amounts of monosaccharides commonly found in vegeta- bles, and the limited conversion of sugars to acids ( < I}.5% (w/w)), competition for fermentable sugars does not contribute to the observed inhibition of yeasts in fermented salads.

Oxygen is absent in fermented dairy products after culturing (Langeveld and Bollc, 1989). One might assume that the reducing effect of the fermentation processes could have an inhibitory effect on the growth of yeasts. This is certainly true for oxidative yeasts, i.e. Trichosl~oron beigelii. However, most yeasts isolated from spoiled salads possess a fermentative metabolism and are able to grow in the absence of oxygen, although at a lower rate. it is generally assumed that out of a low initial contamination massive growth of yeasts can occur under aerobic conditions, duc to a higher energy efficiency. Upon depletion of oxygen this large population of yeasts may bc abic to ferment the residual sugars, with attendant massive production of CO 2, resulting in bulging of packings {Scheffers, 1987). As stated earlier, during fermentation above a temperature of 40°C growth of yeasts is retarded or inhibited. Simultaneously, the concentration of oxygen in the salads decreases by approximately 85% (unpublished results). Therefore, high numbers of yeasts, neccessary for the production of large volumes of CO, will probably not be attained in fermented salads, provided the initial yeast contamination is sufficiently low ( < I(XI CFU/g of salad).

31 ~}

The consumption of oxygen by lactic acid bacteria could result in a simultaneous production of hydrogen peroxide (Condon, 1987), a compound which might be inhibitory, to yeasts. However, in broth cultures of various aerobically grown lactic acid bacteria, production of hydrogen peroxide could not bc detected (unpublished results).

In conclusion, inhibition of spoilage yeasts in fermented salads can be achieved by using starter cultures with high fermenting capacity, provided that measures are taken to attain absence or low initial contamination (< IIH) CFU/g salad) with these spoi lage yeasts. The inhibitory effect of lactic fermentat ion on the growth o f yeasts in salads is probably due to several hlctors, including the amount of lactic acid, the low storage temperature and the low residual oxygen concentrat ion.

In a subsequent paper it will be shown that spoi lage (l-nterotawleriaceae) and pathogenic bacteria (Staphylococc,s aureus, l.isleria mollot:vtogelit's ). del iberately added to salads can be inhibited by using starter cultures with high ferment ing capacity.

Acknowledgements

The authors wish to thank Dr M.Th. Smith (CBS Yeast Division, Identification Service, Delft, The Netherlands), Mr F.H.C.N. Bastiaens, Mr P.P.H. Hennissen, Ms M.J.M. Offenberg, Ms B.S. Pohle, Mr B.G.A. Reincr.nk and Ms M.A. van Til tbr their contributions to this research programme. This study was made possible with a grant from Johma Holding International B.V., Losser, The Netherlands.

References

Adams. M.R.. Baker. T.A. :rod Mrocek. P.R. (IL}Sg) Salt and organic ackl interactions in the inhibition of .~iacchar,my('('s (:~/guus. J. Appl. Bacteriol. 67. xxvi.

Banks. J.G. and [h)ard. R.G. (1987) Some fiictor~ infh:encing the recovery of yeasts and moulds from chilled fl~ods. Int. J. Food Microbiol. 4. 1~7-20~.

Baumgart. J. (1~}77) Zur mikrobiologischen Stabilit~it yon Feinkostsalaten. Alimcnta 16, 197-21HL l]~mestr~a), M.II.. Kusters, B.J.M., Wit, J.(', dc and Rombouts, F.M. (l~tJ2)Glucose lind sucrose

fermenting capacity of homofermcntati~c lactic aod bacteria used as starters in fermented salads, Int. J. Fta~d Microbiol. 15, 365-376,

Brocklchurst. T.F., While, CA. and Dennis. C. (It}S3)The microflora of stored coleslaw and factors affecting the growth of spoilage yeaMs in coleslaw. J, Appl, Bactcriol. ,~5, 57-63.

Brocklchurst. T.F. and Lund, B.M. (ItJ:q4) Microbiological changes in ma.vonmfise-bascd s:llads during storage. F~a~d Microbiol. I, 5-12.

Buick, R.K. and Damoglou, A.P. (IL,~8~}) Effect of modified atmosphere packaging on the microbial development and visible shelf life of a mayonnaise-based vegetable salad. J. Sci. F~a~d Agric. 46. 33~-347.

Condole. S. (H)87) Responses of lactic acid bacteria to oxygen. FEMS Microbiol. Rcv. 46. 269-280. Debcverc. J.M. (1987) The use of buffered acidulant systems to improve the microbiological stability of

acid fimds. F~a~d Microbiol. 4. 105-114. Farbcr. J.M. (1991) Microbiological aspects of modified-atmosphere packaging technology--A review.

J. Fraud Prof. 54, 58-70.

320

(:;romzik. J. ( lq( / l ) llaltbarkeitsfristeu bei Feinkostsalaten. Reisehwirtschaft "/I, 470-474. I lartog. BJ. and Jansen, J.T. (19~6) A sun'ey of the mierobioh)gieal quality of raw-vegetable salads. De

Ware(n) Chemicus 16. "/~-(~). King Jr.. A.I).. Magnuson. J.A.. l'iiriik. T. ;rod Goodman. N. (I(,'t)l) Microhi:d flora and storage quality

()1 partially processed lettuce. J. Food Sei. 56. 459-461. Kirsop, B.N. and Brocklehurst. T.F. (1982) The spoilage of mayonnaise based delicatessen salads.

Process i~:ioehem. 17. g-I I. 14. Langeveld. L.P.M. and Bt)lle. A.C. (1989)Oxygen availability, carbon dioxide concentration and growth

¢)f yeasts and moulds in (ermented milk products: implications fi)r growth during cold storage and fi)r rapid ::nrichment. Neth. Milk Dairy .I. 43. 407-422.

Lund. B.M.. Brocklehurst. T.F. and Geeson. J.D. ( Ig84)The shelf-life of delicatessen salads and collage cheese varicti,:s. AFRC. Food Research Institute Norwich. Biennial Rei'~)~t 1983/1984. pp. 42-47.

Middelhoven. WJ. and van Baalen. A.II.M. (1988) Development of the yeast Ih)ra of whole-crop maize during ensiling and during subsequent aerobiosis. J. Sei. Food Agrie. 42. 190-2117.

Moon. N.J. (1983) Inhibition of the growth of acid t()lerant yeasts by acetate, lactate and propionate and their synergistic mixtures. ]. Appl. Bacteriol. 55: 453-46(}.

Phaff. It.J. and Slarmer. W.T. (1'090) Yeasts associated with plants, insects and soil. In: A.N. Rose and ,I.S. I larrison (Eds.). The Yeasts. Vol. I. Biology of Yeasts. Academic Press. London. pp. 123-181).

Scheffers. W.A. (1987)The role of oxygen in the metabolism of yeasts. De Ware(n)-Chemicus 17. 8-14. Smittle. R.B. (1977) Microbiology of mayonmlise and salad dressing: a review. J. Ftx)d Prof. 411.

415-422. Smittle. R.B. and Flowers, R.S. (1982) Acid tolerant micro-organisms involved in the spoilage of salad

dressings. J. Food Pr()t. 45. 977-9~3. T(ir~ik. T. and King Jr.. A.D. ( Iggl ) Comparative study on the identification of fi)od-borne yeasts. Appl.

Environ. Mien)biol. 57, 12117-1212.