Food Microbiology 27 (2010) 390e395 Contents lists available at ScienceDirect Food Microbiology journal homepage: www.elsevier.com/locate/fmPropionic acid production by cofermentation of Lactobacillus buchneriand Lactobacillus diolivorans in sourdoughChonggang Zhang a, Markus J. Brandt b, Clarissa Schwab a, *, Michael G. Gänzle aa Department of Agricultural, Food and Nutritional Science, 4-10 Agriculture Forestry Centre, University of Alberta, Edmonton, AB T6E 4J8, Canadab Ernst Böcker GmbH & Co., KG, 32427 Minden, Germanya r t i c l e i n f o a b s t r a c tArticle history: Cooperative metabolism of lactobacilli in silage fermentation converts lactate to propionate. This studyReceived 2 July 2009 aimed to determine whether propionate production by Lactobacillus buchneri and Lactobacillus diolivoransReceived in revised form can be applied for bread preservation. Propionate formation was observed in cofermentation with27 November 2009 L. buchneri and L. diolivorans in modiﬁed MRS broth as well as sourdough with low, medium and high ashAccepted 30 November 2009Available online 3 December 2009 contents. 48 mM of propionate was formed in sourdough with medium ash content, but only 9 and 28 mM propionate were formed in sourdoughs prepared from white wheat ﬂour or whole wheat ﬂour, respec- tively. Acetate levels were comparable in all three sourdoughs and ranged from 160 to 175 mM. SourdoughKeywords:Sourdough fermented with L. buchneri and L. diolivorans was used in breadmaking and its effect on fungal spoilage wasPropionate compared to traditional sourdough or propionate addition to straight doughs. Bread slices were inoculatedLactobacillus buchneri with Aspergillus clavatus, Cladosporium spp., Mortierella spp. or Penicillium roquefortii. The use of 20%Lactobacillus diolivorans experimental sourdough inhibited growth of three of the four moulds for more than 12 days. The use of 10% experimental sourdough deferred growth of two moulds by one day. Bread from traditional sourdough with added acetate had less effect in inhibiting mould growth. In conclusion, cofermentation with L. buchneri and L. diolivorans represents a process to increase antifungal capacities of bread. Ó 2009 Elsevier Ltd. All rights reserved.1. Introduction activity. Speciﬁc strains of lactobacilli additionally produce anti- fungal compounds such as phenyllactate, hydroxyl-fatty acids, and Growth of moulds on bread surface is considered one of the cyclic dipeptides (Schnürer and Magnusson, 2005). However, onlymost important causes for bread spoilage. Most moulds isolated from few lactobacilli were evaluated with respect to their antifungalspoiled bread are Penicillum spp. (Legan, 1993). To prevent bread effect in baking applications (Lavermicocca et al., 2000; Ryan et al.,spoilage, food preservatives such as lactate, acetate and propionate 2008, 2009).are added to inhibit growth of moulds and thus improve food safety Propionate formation by Lactobacillus buchneri and Lactobacillusas well as extend shelf-life (Gould, 1996). The antifungal activity of diolivorans during growth in silage was suggested to contributepropionate is higher than that of lactate and acetate. In addition to the to preservation (Driehuis et al., 1999; Krooneman et al., 2002).effect of propionic acid on the cytoplasmic pH, propionate is converted Lactate is converted into 1,2-propanediol by L. buchneri (Fig. 1, Oudeto propionyl-CoA, which inhibits pyruvate dehydrogenase and thus Elferink et al., 2001), and 1,2-propanediol is further converted intoinhibited metabolism of glucose in mould (Brock and Buckel, 2004). propionate by L. diolivorans (Fig. 1, Krooneman et al., 2002).Propionate at a level of 0.03% was effective in inhibiting growth Concentrations of propionate and acetate that are accumulatedof Aspergillus spp. and Penicillium corylophilum at pH 4.5, 25 C, and by co-cultures of L. buchneri and L. diolivorans have to date not beenan aW of 0.85 (Guynot et al., 2005). At pH 5, the minimum inhibitory characterized in either food or feed fermentations. Nevertheless,concentration of propionate for ﬁve fungi, including Penicillium propionate formation by lactobacilli may be exploited for breadroquefortii, was around 50 mM (Lind et al., 2005). preservation. Sourdough fermented with heterofermentative lactic acid It was the aim of this project to characterize lactate conversionbacteria accumulates lactate and acetate with limited antifungal to propionate by cofermentation of L. buchneri and L. diolivorans during sourdough fermentation, and to determine whether the use of * Corresponding author. propionate-producing cultures can replace the addition of propionate E-mail address: firstname.lastname@example.org (C. Schwab). as a preservative.0740-0020/$ e see front matter Ó 2009 Elsevier Ltd. All rights reserved.doi:10.1016/j.fm.2009.11.019
C. Zhang et al. / Food Microbiology 27 (2010) 390e395 391 2.3. Conversion of lactate to 1,2-propanediol by L. buchneri NAD+ NADH NAD+ NADH in mMRS and conversion of 1,2-propanediol to propionate by L. diolivorans Lactaldehyde Lactate Pyruvate NADH NAD+ + ADP The conversion of lactate and 1,2-propanediol was initially NAD+ NADH +ATP analyzed using single cultures of L. buchneri or L. diolivorans. Overnight 1,2-Propanediol Acetate cultures of L. buchneri were washed once in mMRS and 5 mL washed L. buchneri cells were used to inoculate 1 mL mMRS, pH 5.7, with additional 50 mM lactate and/or 5 mM maltose and/or 15 mM fructose. Cultures were incubated for 7 days at pH 5.7 and 30 C. Overnight cultures of L. diolivorans were washed once in mMRS and 5 mL washed cells were used to inoculate 1 mL mMRS, pH 5.5, containing 50 mM lactate and/or 5 mM maltose and/or 15 mM fructose. Cultures were incubated at 30 C for ﬁve days. Metabolites were analyzed using HPLC as 1-Propanol described below. Experiments were performed in duplicate. NAD+ H2O ADP ATP NADH 2.4. Cofermentation of L. buchneri and L. diolivorans 1,2-Propanediol Propionate in mMRS medium L. diolivorans NAD+ NADH Overnight cultures of L. buchneri and L. diolivorans were inocu- lated in mMRS medium containing either 10 mM maltose andFig. 1. Metabolic pathway for conversion of lactate to propionate in cofermentation of 45 mM lactate (mMRSA), or 23 mM maltose, 11.5 mM fructose andL. buchneri and L. diolivorans (Oude Elferink et al., 2001; Sriramulu et al., 2008). 45 mM lactose (mMRSB) at equal optical densities (OD600 ¼ 0.014). The pH was adjusted to pH 5.5. Cultures were incubated at 30 C for 7 days. Metabolites were analyzed using HPLC as described below. Experiments were performed in duplicate.2. Methods 2.5. Cofermentation in L. buchneri and L. diolivorans in sourdough2.1. Strains, culture and growth conditions Sourdoughs with a dough yield of 200 were prepared using L. diolivorans DSM14421 was obtained from the Deutsche Samm- wheat ﬂours Type 550 (white wheat ﬂour), Type 1050 (medium ashlung von Mikroorganismen und Zellkulturen GmbH (Braunschweig, content) and whole wheat ﬂour. Overnight cultures of L. buchneriGermany). L. buchneri FUA 3252 was isolated from ting, a fermented and L. diolivorans were inoculated at a cell density of 107 CFU kgÀ1.sorghum product in Botswana (Sekwati-Monang and Gänzle, 2007). Sourdoughs were fermented at 30 C for 14 days to ensure theL. buchneri and L. diolivorans were grown in modiﬁed MRS (mMRS) formation of sufﬁcient amounts of propionic acid. Samples werecontaining 0.5% beef extract, 0.5% yeast extract, 1% peptone, 0.3% taken during the fermentation period to analyse cell counts, pH, andammonium chloride, 0.4% K2HPO4, 0.26% KH2PO4, 0.01% MgSO4$7H2O, the concentration of metabolites. Colony morphology was deter-0.005% MnSO4$4H2O, 0.1% Tween 80, 0.05% L-cysteine hydrochloride, mined to ensure that the two strains used as inoculum dominatedand 0.0002% each of vitamin B1, B2, B6, B5, B12, and B9. Maltose, the fermentation microﬂora throughout the fermentation, andfructose, lactate and 1,2-propanediol were added after autoclaving as to selectively enumerate L. buchneri and L. diolivorans. At 14 daysdescribed in the following sections. The pH of the media was adjusted of fermentation, only the original inoculants L. buchneri and L. dio-to 5.7 and 6.0 for L. buchneri and for L. diolivorans, respectively. livorans were recovered from the sourdoughs excluding the occur- rence of microbial contamination. Metabolites were prepared and2.2. Mould isolation, characterization and spores preparation analyzed using HPLC as described below. One and three moulds were isolated from blue mould cheese 2.6. Analysis of metabolites with high performance liquidand bread, respectively. The bread was obtained in a local super- chromatography (HPLC)market and stored until mould growth was visible. For isolation,moulds grown on blue mould cheese and bread were streaked Organic acids, alcohols, and sugars were determined by HPLCon meat extract agar medium and incubated for one week. Pure with an Aminex HPX-87 column (300 mm Â 7.8 mm, Biorad, USA) atcolonies were obtained by repeated streaking on meat extract agar. a temperature of 80 C and a ﬂow rate of 0.4 mL minÀ1 with 5 mM Moulds were identiﬁed based on morphology on both meat H2SO4 as the eluent. The injection volume was 10 mL. Refractiveextract agar medium and Czapek agar (0.3% KNO3, 0.1% K2HPO4, index detector and UV detector (210 nm) were used for detection.0.05% MgSO4$7H2O, 0.05% NaCl, 0.001% FeSO4$7H2O, 3% sucrose, Concentration of maltose, fructose, lactate, acetate, ethanol, 1,2-1.5% agar). The blue mould cheese isolate was identiﬁed as propanediol, propionate, 1-propanol were determined usingP. roquefortii Thom. Fungal bread isolates were identiﬁed as Asper- external standards. For sample preparations, 3.7% perchloric acidgillus clavatus Desmazieres, Cladosporium spp. and Mortierella were added to supernatants of fermentation liquor or sourdoughspp. (Barnett and Hunter, 1998; Ramirez, 1982). and incubated at 4 C overnight. Precipitated protein was removed To obtain spores, moulds were inoculated into meat extract agar by centrifugation.medium and incubated for two weeks. Conidiospores were collectedby immersing in physiological solution containing 0.85% NaCl and 2.7. Baking experiments2% peptone, ﬁltered with Whatman No. 1 ﬁlter paper and transferredinto screwed tubes with 50% glycerol and stored at À80 C. Con- Sourdough was prepared using wheat ﬂour Type 1050 and fer-idiospore counts in suspension were measured by serial dilution and mented with L. buchneri and L. diolivorans for 14 days as describedspread plating. above. Bread dough was prepared as summarized in Table 1 with
392 C. Zhang et al. / Food Microbiology 27 (2010) 390e395Table 1 3. ResultsIngredients of sourdough bread for baking [g]. Ingredients Recipe [g] 3.1. Metabolism of L. buchneri and L. diolivorans in single Control 5% 10% 20% 10% 10% Propionate and cofermentation in mMRS SD SD SD BRS BRS-Acc Wheat ﬂour 1050 2000 1900 1800 1600 1800 1800 2000 L. buchneri produced 1,2-propanediol during growth in mMRS Experimental 0 180 360 720 0 0 0 (Table 2). In presence of maltose, higher concentration of 1,2-pro- sourdougha panediol was synthesized compared to absence of maltose. The highest BRSb 0 0 0 0 360 360 0 concentration of 1,2-propanediol was obtained when L. buchneri Water 1200 1120 1040 840 1040 1008 1200 Yeast 40 40 40 40 40 40 40 was grown in the in presence of maltose and lactate. The addition of Salt 36 36 36 36 36 36 36 fructose increased the formation of lactate, acetate and ethanol Acetate 0 0 0 0 0 32 0 but reduced formation of 1,2-propanediol. L. diolivorans converted Ca-propionate 0 0 0 0 0 0 8 1,2-propanediol into propionate (Table 3). Highest concentrations a Wheat 1050 sourdough was fermented with L. buchneri and L. diolivorans for 14 of propionate were formed when L. diolivorans was grown withdays at 30 C. 5%, 10%, or 20% of the experimental sourdough was added to bread 1,2-propanediol as sole carbon source. In absence of 1,2-propanediol,dough. production of propionate was very low. L. diolivorans synthesized b BRS, Böcker Reinzucht Sauerteig Rye, fermented for 18 h at 28 C. c BRS-Ac, Böcker Reinzucht Sauerteig Rye, fermented for 18 h at 28 C with the higher amounts of propionate in mMRS with maltose and 1,2-pro-addition of 0.1% acetate. panediol compared to mMRS with maltose, fructose and 1,2-pro- panediol. However, less lactate, acetate and ethanol was produced. Propionate formation during cofermentation of L. buchneri andaddition of 5%, 10% or 20% of sourdough. For comparison, breads L. diolivorans was observed only in mMRSB and resulted in thewere prepared containing 10% one-stage BRS (Böcker Reinzucht production of lactate, 43.6 Æ 2.3 mM LÀ1; acetate, 28.5 Æ 1.5 mM LÀ1;Sauerteig Rye, fermented for 18 h at 28 C) or 10% one-stage BRS ethanol, 32.0 Æ 3.6 mM LÀ1; propionate, 5.4 Æ 0.2 mM LÀ1; andand 0.1% acetate (BRS-Ac). For control, breads were also prepared 1-propanol, 4.9 Æ 0.5 mM LÀ1.from straight dough and from straight dough with the additionof 0.4% Ca-propionate. Ingredients were mixed in a spiral kneader 3.2. Kinetic analysis of L. buchneri and L. diolivorans(SPA 12-4, Diosna, Osnabrück, Germany) for 5 min, shaped after cofermentation in sourdough15 min dough resting time, at 25e26 C and proofed for 50e60 minat 32 C and 85% r.m. in pans. Prooﬁng time for the dough prepared L. buchneri and L. diolivorans grew to high cell counts in all threewith 20% addition of sourdough fermented with L. buchneri and sourdoughs prepared with ﬂours of different ash content withinL. diolivorans was extended to 240 min because of obvious inhibi- two days of incubation and acidiﬁed the sourdoughs to a pH of 3.6tion of yeast activity. Bread was baked in a multi-deck oven set at (Fig. 2). In sourdough prepared with Type 550 and Type 1050 ﬂours,a temperature of 220 C (upper plate) and 210 C (lower plate) for cell counts of L. buchneri decreased slightly between day three45 min. and 14 whereas cell counts of L. diolivorans dropped to levels below 105 CFU mLÀ1 after 14 days. In wheat 550 sourdough, propionate2.8. Antifungal activity was detectable after day ﬁve. Between day ﬁve and 14, levels of acetate increased by around 35 mM. In contrast, amounts of lactate Antifungal activity was determined according to Dal Bello et al. and 1-propanol both decreased by about 5 mM. In wheat 1050(2007) with slight modiﬁcations. Bread was cut into 1.5 cm wide sourdough, propionate was detectable by day two and increasedslices. A ten fold dilution of mould stock was applied to both sides to 48 mM by day 14. At day ﬁve, levels of acetate started to increase,of the bread slices by spraying ﬁve spots on each side. Around whereas lactate and 1-propanol decreased.4.8 Â 103, 8 Â 102, 8 Â 102, and 8 Â 101 CFU conidiospores of In wheat whole grain ﬂour, cell counts of both L. buchneri andA. clavatus Desmazieres, Cladosporium spp., Mortierella spp., and L. diolivorans remained above 108 CFU kgÀ1 throughout 14 days ofP. roquefortii Thom were sprayed to each spot of bread slices, fermentation. Lactate concentrations were higher than in the otherrespectively. Bread slices were wrapped in non-oxygen permeable sourdoughs (approx. 170 mM kgÀ1). In contrast, the amount of propi-plastic bags to prevent moisture loss and stored at room tempera- onate was lowest (12 mM kgÀ1) compared to the other two ﬂours.ture in a temperature controlled room. Mould growth was moni-tored daily for a duration of 12 days. Growth was recorded by 3.3. Antifungal activitycounting the number of the total of 10 spots per slice that had visiblegrowth of moulds. A sample was considered spoiled if growth was Baking experiments were performed to evaluate the effect ofvisible on one of the 10 spots. propionate and acetate formation by L. buchneri and L. diolivorans onTable 2Change of metabolite concentration after fermentation by L. buchneri in mMRS for 7 days at 30 C. Substrates [mM] Metabolitesa [mM] Maltose Lactate Fructose Maltose Fructose Mannitol Lactate Acetate Ethanol 1,2-POH [5 mM] [50 mM] [15 mM] D metabolite þ þ þ À4.3 Æ 0.1 À15.3 Æ 0.0 9.3 Æ 0.6 7.9 Æ 0.1 11.8 Æ 0.8 12.6 Æ 1.4 0.8 Æ 0.2 concentrationa þ þ À À4.4 Æ 0.1 À À À0.8 Æ 0.9 9.3 Æ 0.7 9.8 Æ 0.6 1.6 Æ 0.1 þ À þ À4.0 Æ 0.1 À15.0 Æ 0.0 8.0 Æ 0.6 12.6 Æ 1.2 12.2 Æ 1.7 15.3 Æ 0.4 0.9 Æ 0.6 À þ þ À À15.4 Æ 0.0 8.4 Æ 0.2 À0.1 Æ 1.8 11.2 Æ 0.3 3.6 Æ 0.0 0.2 Æ 0.1 À þ À À e e À4.5 Æ 1.6 5.2 Æ 1.6 6.4 Æ 1.6 0.2 Æ 0.11,2-POH ¼ 1,2-propanediol. a The difference of metabolite concentrations after seven days to the initial concentrations. Positive numbers indicate the production of metabolites; negative numbersindicate the consumption of the metabolites.
C. Zhang et al. / Food Microbiology 27 (2010) 390e395 393Table 3Change of metabolite concentrations after fermentation by L. diolivorans in mMRS for 5 days at 30 C. Substrates [mM] Metabolitesa [mM] Maltose 1,2-POH Fructose Maltose Fructose Mannitol Lactate Acetate Ethanol 1,2-POH Propionate 1-POH [5 mM] [30 mM] [15 mM] D metabolite þ þ þ À4.6 Æ 0.2 À15.1 Æ 0.0 2.6 Æ 0.2 16.8 Æ 0.2 18.3 Æ 0.2 5.9 Æ 0.6 À34.1 Æ 0.0 7.0 Æ 0.1 19.9 Æ 2.9 concentrationa þ þ À À4.4 Æ 0.2 e e 1.8 Æ 0.5 15.5 Æ 0.1 1.2 Æ 0.1 À35.8 Æ 0.0 8.3 Æ 0.2 20.1 Æ 2.8 þ À þ À4.4 Æ 0.2 À14.9 Æ 0.0 4.9 Æ 0.4 16.0 Æ 0.3 12.9 Æ 0.3 9.9 Æ 1.0 0.0 Æ 0.0 2 Æ 0.9 0.1 Æ 0.0 À þ þ À0.3 Æ 0.1 À16.5 Æ 0.0 1.8 Æ 0.3 4.4 Æ 1.2 19.6 Æ 1.6 1.3 Æ 0.1 À37.5 Æ 0.0 7.7 Æ 0.1 18.2 Æ 1.3 À þ À À0.5 Æ 0.1 e e 0.8 Æ 1.0 2.7 Æ 0.1 0.7 Æ 0.1 À37.8 Æ 0.0 19.2 Æ 1.8 11.3 Æ 1.11,2-POH ¼ 1,2-propanediol; 1-POH ¼ 1-propanol. a The difference of metabolite concentrations after 7 d to the initial concentrations. Positive numbers indicate the production of metabolites; negative numbers indicate theconsumption of the metabolites.mould growth on sourdough bread. Sourdough prepared from aerobic microorganism (Oude Elferink et al., 2001). L. buchneriwheat ﬂour Type 1050 was used for the baking experiments because produces acetate and 1,2-propanediol from lactate, which is furtherthe use of ﬂour with medium ash content resulted in the highest metabolized to 1-propanol and propionate by L. diolivorans (Holzerpropionate concentrations (Fig. 2). The addition of 10% sourdough et al., 2003). This study provides the characterization of cooperativefermented with L. buchneri and L. diolivorans did not inhibit metabolism of L. buchneri and L. diolivorans in mMRS and foodP. roquefortii but deferred the growth of A. clavatus, Cladosporium fermentations, and evaluated the effects of acetate and propionatespp., and Mortierella spp. by one day one compared to the control on mould growth on bread.(Table 4). The addition of 20% L. buchneri and L. diolivorans sour- The strain of L. buchneri used in this study was isolated fromdough inhibited growth of Saccharomyces cerevisiae at the dough an African cereal product (Sekwati-Monang and Gänzle, 2007).stage (data not shown), inhibited growth on bread of A. clavatus, The cereal isolate L. buchneri converted lactate to 1,2-propanediol inCladosporium spp., and Mortierella spp. for more than 12 days and mMRS. Lactate was obtained by the fermentation of maltose ordeferred growth of P. roquefortii by two days compared to the supplied to the growth medium. Higher amounts of 1,2-propanediolcontrol. The addition of acetate to commercial BRS sourdough bread was detected when the strain was grown in the presence of maltose.deferred the growth of Cladosporium spp. by one day; A. clavatus, The alternative metabolites from lactate, acetate and 1,2-propane-Mortierella spp. and P. roquefortii were not inhibited. Bread prepared diol were not produced in a ratio of 1:1 as reported by Oude Elferinkwithout sourdough but with the addition of propionate deferred et al. (2001), which may be attributable to the more complexgrowth of A. clavatus by three days compared to the control. Growth carbohydrate composition of the media employed in our study.of Cladosporium spp. and Mortierella spp. was deferred by one and The maize silage derived L. diolivorans produced the highestﬁve days respectively, and P. roquefortii was not affected. amounts of propionate in presence of 1,2-propanediol as sole carbon source (Krooneman et al., 2002). The ratio of 1-propanol to4. Discussion propionate was close to the ratio of 1.5e1 as reported previously if 1,2-propanediol was the sole substrate (Krooneman et al., 2002). The cofermentation of L. buchneri and L. diolivorans defers Substantially higher ratios of propanol to propionate were observedaerobic deterioration of maize silage caused by the growth of in the presence of hexoses as additional substrates. 6.5 6.5 A B C Cells count [CFU g ] -1 9 6.0 6.0 10 5.5 5.5 8 pH 10 5.0 5.0 4.5 4.5 7 10 4.0 4.0 6 3.5 3.5 10 D E F Concentration [mM] 160 120 80 40 0 0 4 8 12 0 4 8 12 0 4 8 12 Time(days)Fig. 2. Kinetic measurement of dough pH, cells counts and main metabolites during sourdough fermentation with wheat ﬂour Type 550 (low ash content, panels A, D), wheat ﬂourType 1050 (medium ash content, panel B, E) and whole wheat ﬂour (high ash content, panels C, F) for 14 days at 30 C. Panels A, B and C show cells counts of L. buchneri (C) andL. diolivorans (B) and dough pH (7) during fermentation of sourdough. Panels D, E and F indicate concentrations of lactic acid (C), acetic acid (B), ethanol (7), propionic acid (6),and 1-propanol (-) produced during fermentation.
394 C. Zhang et al. / Food Microbiology 27 (2010) 390e395Table 4 The addition of 20% experimental sourdough is not suitable forTime in days to visible growth of moulds on sourdough bread stored for 12 days. practical applications as high levels of propionate and acetate in A. clavatus Cladosporium sp. Mortierella sp. P. roquefortii bread dough inhibited the activity of bakers yeast and are expected Control 3a 3 6 6 to inﬂuence sensory properties (Patterson and van Holy, 2001). 5% SDb 3 3 6 6 Preliminary sensory trials indeed indicated strong acidic ﬂavour and 10%SD 4 4 6 6 taste (data not shown). However, a 10% addition of experimental 12 12 12 sourdough, corresponding to 5 and 17 mM kgÀ1 propionate and 20% SD 8 10% BRSb 3 3 6 6 10% BRS-Ac 3 4 6 6 acetate, respectively, delayed growth of two moulds by one day Propionate 6 4 12 6 relative to the control breads prepared from straight dough or with a Bread was cut into 1.5 cm wide slices and conidiospores were applied on ten spots traditional sourdough without adverse effects on yeast activity ofper slice to transfer about 4.8 Â 103, 8 Â 102, 8 Â 102, 8 Â 101 conidiospores of bread quality. Experimental mould contamination of 400e100 000A. clavatus, Cladosporium sp, Mortierella sp, and P. roquefortii per spot, respectively. spores per slice exceeds mould contamination observed in industrialBread slices were wrapped in plastic bags and stored aerobically at room temperature. practice by far. Nevertheless, sourdoughs cofermented with L. dio-Mould growth was monitored visually daily for a duration of 12 days. Data correspond livorans and L. buchneri exhibited increased antifungal propertiesto the day when mould growth was visible on all 10 spots. b SD, experimental sourdough; wheat ﬂour Type 1050 fermented fro 14 days with in comparison with other studies on antifungal starter cultures forL. buchneri and L. diolivorans. BRS, Böcker Reinzucht Sauerteig Rye, fermented for bread production that reported no difference between experimental18 h at 28 C; BRS-Ac, BRS sourdough with the addition of 0.1% acetate. and traditional sourdoughs (Dal Bello et al., 2007; Moore et al., 2008) or a one day difference between experimental sourdough with L. buchneri and L. diolivorans grew to high cell counts in wheat, antifungal starter culture and traditional sourdough with L. san-rye and buckwheat sourdoughs (data not shown). In wheat sour- franciscensis (Ryan et al., 2008) when sourdoughs were added at 20%.doughs, both strains reached cell counts comparable to sourdough Elongated delay of mould growth was only achieved when breadstarter cultures growing in wheat, rye, or gluten-free cereals and was fully fermented with sourdough starter cultures, however,pseudocereals (Vogelmann et al., 2009). Cell counts of L. buchneri this process is not applied in baking practice (Gerez et al., 2009;were slightly higher in all three sourdoughs. The pH ranged from Lavermicocca et al., 2000).pH 3.6 to pH 3.7, comparable to levels observed in traditional wheat In conclusion, cofermentation with L. buchneri and L. diolivoranssourdoughs (Hammes et al., 1996; Gänzle et al., 1998). allows the fermentative accumulation of propionate and acetate in Cofermentation of L. buchneri and L. diolivorans formed lactate, sourdough and represents a process to increase antifungal capacities ofacetate and ethanol in sourdough. Levels of lactate production bread. Industrial production of sourdough for use as baking improverincreased in the order wheat ﬂour type 550, wheat ﬂour type 1050 applies fermentation time of up to seven days. Modiﬁcations to currentand whole wheat ﬂour, in keeping with the increased buffer capac- processes may enable the fermentation of propionate enrichedities of the ﬂours (Hammes et al., 1996). Remarkably, the levels of sourdoughs at industrial scale.acetate were around three to ﬁve times higher compared to tradi-tional sourdough inoculated with Lactobacillus sanfranciscensis due Acknowledgmentsto lactate conversion to acetate by L. buchneri (Gänzle et al., 1998).During forage preservation by L. buchneri, acetate was partially We are grateful to ﬁnancial support of Ernst Böcker GmbH,gained from degradation of lactate (Weinberg and Muck, 1996). Minden, Germany. We acknowledge Elena A. Dlusskaya for help Independent of the type of ﬂour used, a decrease of lactate on identiﬁcation of mould based on morphology and Franz-Josefcontent during fermentation correlated to an increase in propionate. Wiedenhorst for bread baking.Propionate was only detectable after two or more days of fermen-tation, indicating that propionate was not produced during expo- Referencesnential growth. Formation of 1,2-propanediol by L. buchneri does notsupport cell growth and requires low pH. The conversion of lactate Barnett, H.L., Hunter, B.B., 1998. Illustrated Genera of Imperfect Fungi, fourth ed. APSto propionate was optimal at pH 3.8 compared to pH 4.3 and 5.8 Press, St. Paul, Minnesota, pp. 218ff. Brock, M., Buckel, W., 2004. On the mechanism of action of the antifungal agent(Oude Elferink et al., 2001). propionate propionyl-coA inhibits glucose metabolism in A. nidulans. Eur. 90% of bread spoilage is caused by Penicillium spp.; Aspergillus, J. Biochem. 271, 3227e3241.Cladosporium, Mucor, Monilia, Endomyces, Fusarium and Rhizopus Dal Bello, F., Clarke, C.I., Ryan, L.A.M., Ulmer, H., Schober, T.J., Ström, K., Sjörgen, J.,spp. have also been identiﬁed as bread spoilage fungi (Legan, 1993). van Sinderen, D., Schnürer, J., Arendt, E.K., 2007. Improvement of the quality and shelf life of wheat bread by fermentation with antifungal strain Lactoba-The use of sourdough delays fungal spoilage due to the formation of cillus plantarum FST1.7. J. Cereal Sci. 45, 309e318.organic acids by the lactic acid bacteria starters. Breads containing Driehuis, F., Oude Elferink, S.J.W.H., Spoelstra, S.F., 1999. Anaerobic lactic acid degra-sourdoughs fermented with various lactobacilli delayed growth of dation during ensilage of whole crop maize inoculated with L. buchneri inhibits yeast growth and improves aerobic stability. J. Appl. Microbiol. 87, 583e594.Aspergillus, Fusarium, and Penicillium spp. (Gerez et al., 2009; Moore Gänzle, M.G., Ehmann, M., Hammes, W.P., 1998. Modeling of growth of Lactobacilluset al., 2008; Ryan et al., 2008). Furthermore, the addition of sour- sanfranciscensis and Candida milleri in response to process parameters of thedough allowed the reduction of calcium propionate addition (Gerez sourdough fermentation. Appl. Environ. Microbiol. 64, 2616e2623. Gerez, C.L., Torino, M.I., Rollán, G., Font de Valdez, G., 2009. Prevention of breadet al., 2009; Ryan et al., 2008). In the European Union, propionic acids mould spoilage by using lactic acid bacteria with antifungal properties. Foodcan be added to a maximum level of 0.3%, corresponding to about Control 20, 144e148.40 mM propionate. Up to 48 mM kgÀ1 propionate and 175 mM kgÀ1 Gould, G.W., 1996. Methods for preservation and extension of shelf life. Int. J. Food Microbiol. 33, 51e64.acetate were formed by cofermentation of L. buchneri and L. dio- Guynot, M.E., Ramos, A.J., Sanchis, V., Marín, S., 2005. Study of benzoate, propionate,livorans in wheat ﬂour Type 1050 sourdough. Bread prepared with and sorbate salts as mould spoilage inhibitors on intermediate moisture bakery20% sourdough accordingly contained about 10 mM kgÀ1 propionate products of low pH (4.5e5.5). Int. J. Food Microbiol. 101, 161e168.and 35 mM kgÀ1 acetate. The addition of 20% sourdough almost Hammes, W.P., Stolz, P., Gänzle, M.G., 1996. Metabolism of lactobacilli in traditional sourdoughs. Adv. Food Sci. 18, 176e184.completely inhibited growth of moulds. It is remarkable that 20% Holzer, M., Mayrhuber, E., Danner, H., Braun, R., 2003. The role of Lactobacillusaddition of experimental sourdough was more effective than either buchneri in forage preservation. Trends Biotechnol. 21, 282e287.control bread with propionate addition or traditional sourdough Krooneman, J., Faber, F., Alderkamp, A.C., Oude Elferink, S.J.H.W., Driehuis, F., Cleenwerck, I., Swings, J., Gottschal, J.C., Vancanneyt, M., 2002. Lactobacillus dio-bread prepared with additional acetate, which conﬁrms synergistic livorans sp. nov., a 1,2-propanediol-degrading bacterium isolated from aerobicallyactivity of sourdough with propionate (Ryan et al., 2008). stable maize silage. Int. J. Syst. Evol. Microbiol. 52, 639e646.
C. Zhang et al. / Food Microbiology 27 (2010) 390e395 395Lavermicocca, P., Valerio, F., Evidente, A., Lazzaroni, S., Corsetti, A., Gobbetti, M., 2000. Ryan, L.A.M., Dal Bello, F., Czerny, M., Köhler, P., Arendt, E.K., 2009. Quantiﬁcation of Puriﬁcation and characterization of novel antifungal compounds from the sour- phenyllactic acid in wheat sourdough using high resolution gas chromatogra- dough lactobacillus plantarum strain 21b. Appl. Environ. Microbiol. 66, 4084e4090. phyemass spectrometry. J. Agric. Food Chem. 57, 1060e1064.Legan, J.D., 1993. Moulds spoilage of bread: the problem and some solutions. Int. Schnürer, J., Magnusson, J., 2005. Antifungal lactic acid bacteria as biopreservatives. Biodeterior. Biodegrad 32, 33e53. Trends Food Sci. Technol. 16, 70e78.Lind, H., Jonsson, H., Schnürer, J., 2005. Antifungal effect of dairy propionibacteria e Sekwati-Monang, B., Gänzle, M.G., 2007. Microbiological characterisation of ting, contribution of organic acids. Int. J. Food Microbiol. 98, 157e165. a sorghum-based, gluten free fermented cereal products from Botwana. In:Moore, M.M., Dal Bello, F., Arendt, E.K., 2008. Sourdough fermented by Lactobacillus Presentation at the First International Symposium on Gluten-Free Cereal plantarum FST1.7 improves the quality and shelf life of gluten free bread. Eur. Products and Beverages, Cork, Ireland. Food Res. Technol. 226, 1309e1316. Sriramulu, D.D., Liang, M., Hernandez-Romero, D., Raux-Deery, E., Lünsdorf, H.,Oude Elferink, S.J.W.H., Krooneman, J., Gottschal, J.C., Spoelstra, S.F., Faber, F., Parsons, J.B., Warren, M.J., Prentice, M.B., 2008. Lactobacillus reuteri DSM Driehuis, F., 2001. Anaerobic conversion of lactic acid to acetic acid and 1,2- 20016 produces cobalamin-dependent diol dehydratase in metabolosomes propanediol by Lactobacillus buchneri. Appl. Environ. Microbiol. 67, 125e132. and metabolizes 1,2-propanediol by disproportionation. J. Bacteriol. 190,Patterson, T.L., van Holy, A., 2001. Effect of selected natural antimicrobials on Bakers 4559e4567. yeast activity. Lett. Appl. Microbiol. 33, 211e215. Vogelmann, S.A., Seitter, M., Singer, U., Brandt, M.J., Hertel, C., 2009. Adaptability ofRamirez, C., 1982. Manual and Atlas of the Penicillia. Elsevier Biomedical Press, lactic acid bacteria and yeasts to sourdoughs prepared from cereals, pseudo- Amsterdam, New York, Oxford, pp. 874. cereals and cassava and use of competitive strains as starters. Int. J. FoodRyan, L.A.M., Dal Bello, F., Arendt, E.K., 2008. The use of sourdough fermented by Microbiol. 130, 205e212. antifungal LAB to reduce the amount of calcium propionate in bread. Int. J. Food Weinberg, Z.G., Muck, R.E., 1996. New trends and opportunities in the development Microbiol. 125, 274e278. and use of inoculants for silage. FEMS Microbiol. Rev. 19, 53e68.