Potential of Lactobacillus reuteri from Spontaneous Sourdough as a Starter Additive for Improving Quality Parameters of Bread

Fermented bakery products diff er from region to region. The production of the majority of these products is based on traditional methods and their uniqueness depends on the raw materials used for sourdough fermentation (1). Sourdough used in bread making plays an important role in improving the texture and fl avour of the product through metabolically active yeasts and lactic acid bacteria (LAB). The LAB developing in sourdough may originate from the raw materials (2).


Introduction
Fermented bakery products diff er from region to region. The production of the majority of these products is based on traditional methods and their uniqueness depends on the raw materials used for sourdough fermentation (1). Sourdough used in bread making plays an important role in improving the texture and fl avour of the product through metabolically active yeasts and lactic acid bacteria (LAB). The LAB developing in sourdough may originate from the raw materials (2).
LAB are the biological basis for the production of a multitude of fermented foods. During the fermentation their metabolic activity determines the food quality. Furthermore, LAB strains may produce a wide range of antimicrobial metabolites (3) that have a potential to inhibit food pathogens (4). Thus, the strains producing antimicrobial substances can be used as biopreservatives, extending the shelf life and enhancing food safety (5)(6)(7)(8)(9).
In the bread industry, antimicrobial properties of LAB are particularly important in ensuring the stability of the sourdough. A microbially stable sourdough of good consistency has a signifi cant impact on bread quality. Another important reason for the use of antimicrobially active sourdough is the potential to suppress bread defects such as ropiness or bread moulding (10). The sourdough ISSN 1330-9862 original scientifi c paper doi: 10.17113/ft b.54.03. 16.4143 fermentation applied for the production of rye and wheat bread prolongs the freshness of the bread and enhances its microbial safety. Physical, chemical and biotechnological means are applied for bread antistaling and microbial spoilage (11)(12)(13). Preservatives of chemical origin are not natural tools, but their usage in bread increases its microbiological safety and extends its shelf life. Recently, physical methods, i.e. various sterilisation methods, are used to protect bread from mould spoilage on its surface (14,15). Antimicrobial packaging of bread has also become popular. Additives from the packages diff use slowly into the environment between the inner package layer and the bread during storage. This mechanism is eff ective in protecting the bread from mould spoilage or other defects of microbial origin (16). Antimicrobial agents that migrate from the packaging material to the food product prevent fungal outgrowth and extend the shelf life of the product, compared to conventional packaging (17). However, the physical and chemical tools for retarding bread staling are not natural and may have a negative impact on consumer health (18).
In contrast, Lactobacillus is a natural and eff ective tool for retarding bread staling and microbial spoilage. It is characterized by antimicrobial activity against microscopic fungi and spore-forming bacteria. This would appeal to health-conscious consumers looking for natural foods without chemical preservatives. This study was undertaken to identify LAB cultures from spontaneous sourdough and use them to improve the quality and prolong the shelf life of bread.

Composition and preparation of spontaneous sourdough
The spontaneous sourdough fermentation was done by mixing sift ed rye fl our type 1370 according to LST 1481:2004 (19) and tap water in a ratio of 1:1.5. The mixture was incubated at 27 °C under anaerobic conditions for 24 h by stirring periodically. Aft er incubation, 10 % of ripe sourdough was used to inoculate a fresh fl our and water mixture under the same conditions as described above. The re-inoculation procedure was carried out until moisture content reached 55 %. The obtained spontaneous sourdough was used for the isolation of naturally occurring lactic acid bacteria (LAB).

Isolation and identifi cation of LAB
To isolate LAB, 10 g of spontaneous sourdough was homogenised in 90 mL of sterile physiological solution and then suitable serial dilutions were plated on MRS agar (Biolife, Milano, Italy) and incubated at 30 and 37 °C (according to the optimal growth temperature of typical LAB species) under aerobic and anaerobic conditions. LAB colonies were selected according to their appearance, and the microscopic preparations of purifi ed isolates were observed by optical microscope Laboval 4 (Carl Zeiss Jena GmbH, Jena, Germany) using 1000× magnifi cation.
Aft er culturing in MRS broth (Biolife) at 30 °C for 18 h, the isolates were centrifuged at 900×g for 15 min and bacterial pellets were used for automated DNA extraction with QIAcube using a QIAamp DNA mini kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. PCR reaction and pyrosequencing were perf ormed following the manufacturer's instructions using a 3B Blacklight sepsis bacterial (3B Blackbio Biotech, Madrid, Spain) kit. PCR conditions were as follows: initial denaturation at 94 °C for 5 min, followed by 35 cycles at 94 ° C for 20 s, at 54 °C for 20 s and at 72 °C for 30 s, with a fi nal extension step at 72 °C for 5 min. PCR products were stored at -20 °C until use. Pyrosequencing was performed on the PyroMark Q24 automatic pyrosequencing instrument (Qiagen) using a Gold24 (Qiagen) kit with dATPαS, dCTP, dGTP, dTTP, enzyme mixture and substrate mixture. Nucleotide sequence analysis was performed using the PyroMark Q24 (Qiagen) soft ware and the National Center for Biotechnology Information Basic Local Alignment Search Tool (20).

Determination of antimicrobial activity
The pure LAB isolates inoculated into MRS broth were incubated for 72 h at 37 °C under anaerobic conditions and centrifuged at 6000×g for 15 min. The obtained cell-free supernatant was used for investigations. The antimicrobial activity of LAB isolates from spontaneous sourdough was tested by agar well diff usion assay. The reference bacterial cultures were pre-cultivated on plate count agar (PCA; Liofi lchem, Roseto degli Abruzzi, Italy) slants for 24 h at 30 or 37 °C and target LAB strains were pre-cultivated on MRS agar slants for 24 h at 30 or 37 °C. The cultures from the slants were transferred into 10 mL of 0.9 % NaCl solution to obtain the inoculum density equivalent to a McFarland standard no. 0.5. Adjusted suspension (1 mL) was transferred to 100 mL of PCA or MRS agar dissolved and cooled to 45 °C. Isolated micromycetes and yeast cultures were cultivated for 24-48 h at 25 °C, respectively on the malt extract agar (MEA; Liofi lchem) and Sabouraud dextrose agar (SDA; Liofi lchem) slants. An inoculum of 10 6 CFU/mL in 0.9 % NaCl solution was prepared. Aft er that, 1 mL of the target culture was added to 100 mL of the melted MEA or SDA cooled to 45 °C. The prepared mixture (10 mL) was poured into Petri dishes. A control solution (MRS broth), cell-free supernatant, neutralised cell-free supernatant and neutralised cell-free supernatant treated with enzymes (50 μL) were added to 7.5 mm-diameter wells made in the solidifi ed agar. Antibacterial activity was assessed aft er 24 h of incubation at 30 or 37 °C, while antifungal activity was determined aft er 48 h of incubation at 25 °C. Antimicrobial activity was determined according to the diameter of inhibition zones surrounding the wells aft er incubation.

Bread-making technology
The eff ectiveness of experimental sourdough containing Lactobacillus reuteri used for rye, wheat, and rye with wheat bread making was evaluated. Lactococcus reuteri for preparation of sourdough for the experimental bread and Lactococcus lactis ssp. lactis 140/2 for preparation of sourdough for the control bread were cultivated in 100 mL of MRS broth at 30 °C for 24 h. The MRS broth was then centrifuged at 5000×g for 5 min, washed twice in 0.9 % NaCl solution and the whole precipitate, approx. 10 g (10 6 CFU/g), was added into the fi rst mixture of fl our and water. The sourdough was prepared as described in the fi rst paragraph of the Materials and Methods section. Bread was made without adding any bread quality-enhancing additives (Table 1). Both the experimental and control brea d dough were prepared by mixing fl our with water (95-98 °C) and leaving it to rest for 1.5-2.0 h. Then the experimental and control sourdough (containing respectively L. reuteri and L. lactis ssp. lactis 140/2) were added to the mixture and further fermentation was carried out for 12 h at 27 °C until total titratable acidity of the obtained mixture reached 11°. Aft er fermentation, the remaining components were mixed and homogenised for 15 min in a spiral mixer SP-100A/NH-B (Metos, Kerava, Finland) and the dough was placed into the i ncubator for 1.5 h at 35 °C. Dough loaves (500 g) were baked for 20-25 min at 220 °C.
Aft er baking, the bread loaves were cooled at room temperature for 2 h and subsequently sealed in polyethylene bags.

Bread analysis
The quality and sensory evaluation of control and experimental bread was performed 12 h aft er baking and during 7-day storage at room temperature. It was assumed that the bread was suitable for consumption for 7 days. The moisture content of bread crumbs was determined by drying the samples at 105 °C to the constant mass. Acidity was measured by the titration method, and porosity was determined according to method LST 1442:1996 (21).
Crumb fi rmness was measured using a universal testing machine Instron 3343 (Instron Engineering Group, High Wycombe, UK). Samples for analysis (50 mm×40 mm×2 6 mm) were prepared and slices of 13 mm thickness were compressed to 40 % of their original height at a crosshead speed of 1 mm/s (22).
Colour measurements of bread were carried out using a CR-400 Chroma Meter colourimeter (Konica Minolta Sensing Inc., Warrington, UK). The obtained resu lts were expressed as colourspace according to the CIELAB scale.
Sen sory properties of baked bread samples were measured using a quantitative descriptive analysis by creating sensory profi les for each bread sample. Assessors were seven panellists between 20 and 60 years old. All training and data collection sessions were held in the sensory analysis laboratory of Food Institute. Bread samples were kept in the constant climate chambers (Binder, Tuttlingen, Germany) before testing. They were presented to the panel at 21 °C in closed plastic boxes coded with three digit numbers. For the development of sensory profi les, a fully balanced randomised sample presentation plan with two repetitions was applied. A set of six bread samples (three samples of control bread (rye, wheat and rye with wheat) prepared with Lactococus lactis ssp. lactis 140/2 and three samples of experimental bread (rye, wheat and rye with wheat) prepared with Lactobacillus reuteri) was presented to all participants in duplicate. The participants' responses were recorded and collected using a computerised data system FIZZ (Biosystemes, Couternon, France). The 15 poin ts of numbered scales with indented anchors (left anchor 'low intensity/absent', right anchor 'high intensity') were used to evaluate each sensory att ribute. Evaluated att ributes included odour of bread starter, richness of taste, and intensity of non-typical odour and taste.

Determination of protective properties of spontaneous sourdough
The protective properties of spontaneous sourdough containing Lactobacillus reuteri were evaluated according to the microbiological parameters of a bread crumb aft er storage at room temperature for 7 days. The number of moulds was determined in Dichloran Rose Bengal Chloramphenicol (DRBC) agar (Liofi lchem) (23). The number of spores of mesophylic bacteria was determined by heat-  ing the dilutions for 10 min at 80 °C before plating on PCA (24). The total bacterial count was determined on PCA according to standard method (25).
L. reuteri isolated from spontaneous sourdough were tested for antimould activity by spraying the cell-free super natant on the surface of bread obtained in the local super market. The two control variants were the bread loaves sprayed and not sprayed with MRS broth (in order to determine the eff ect of MRS broth ingredients on the antimould activity). The bread loaves were stored for 8 days at 24-25 °C and 15-16 °C in polyethylene bags and the rind of loaves was examined externally for the presence of moulds according to the scale: -no moulds, + small colonies, and ++ large colonies (26). The LAB count in MRS broth for neutralised cell-free supernatant preparation was determined using a UV-Vis spectrophotometer Spectronic Genesys 5 (Thermo Fisher Scientifi c, Madison, WI, USA) at a wavelength of λ=600 nm and converted to a cell co unt (27).

Analysis of components produced by Lactobacillus reuteri
Neutralised cell-free supernatants were obtained by adjusting the pH to 6.5 using 1 mol/L of KOH. To confi rm the production of a proteinaceous compound produced by L. reuteri, neutralised cell-free supernatants were treated separately with catalase (Sigma-Aldrich, St. Louis, MO, USA) and proteinase K (Sigma-Aldrich) at a concentration of 20 mg/mL and incubated respectiv ely at 25 and 56 °C for 30 min. Antimicrobial activity of treated neutralised cell-free supernatants was determined by the agar diff usion bioassay as described above.
Fatt y acids secreted in the growth medium by L. reuteri were extracted from 3 mL of cell-free supernatant by mixing it with n-hexane and methylating with KOH, yielding methyl esters according to the standard method (28). Fatt y acid methyl esters were analysed with a Shimadzu GC-2010 (Shimadzu, Kyoto, Japan) gas chromatograph using a 120-metre column BPX-70 (29). Chromatographic peaks were identifi ed by comparison with retention times of a mixture of a Supelco 37 Component FAME Mix reagent kit (Supelco Analytical, Thermo Scientifi c, Bellefonte, PA, USA). Analytical conditions were as follows: column temperature was 60 °C for 2 min, then increased to 230 °C at 13 °C/min and maintained the same for 55 min; injector temperature was 250 °C and detector temperature was 270 °C; nitrogen was used as the carrier gas. The content of each relevant fatt y acid was calculated as a percentage of total fatt y acids in each sample.

Statistical analysis
Baking experiments were repeated three times, while mean values of bread parameters were estimated from three replicates of a single batch of bread. The comparison of microbiological tests was performed by a one-way ANOVA (Tukey's HSD) test using SPSS v. 16.0 soft ware (30). Sensory evaluation was performed in duplicate while instrument measurements were done in triplicate. All data were subjected to t-tests to assess the signifi cance of treatment mean values at the 5 % signifi cance level.

Dominant lactic acid bacteria in spontaneous sourdough
Twenty LAB cultures were isolated from spontaneous sourdough. The number of LAB reached 8.6·10 7 CFU/g. All isolates were rod-shaped. The sequences generated for the tested LAB isolated from spontaneous sourdough were the following: V1: CACTGGTGAT CCATCGTCAA TCAGGTGCAA GCACCATCAA TGA, and V2: GTGA-CTTTCT GCTTGGATAC CGTCACTGGC TGAACA and V3: GTCATTGCGT CCCCGAAGGG AACGCTTAT CTC-TAAGGTTA. This enabled all tested LAB isolates to be identifi ed as the same species among Lactobacillus spp. The obtained sequences clearly corresponded to L. reuteri when compared to GenBank sequences (KC700337.1 or KC561127.1).
L. reuteri was one of the dominant LAB species isolated from spontaneous sourdough. Molecular DNA-based methods like DNA sequencing and fi ngerprinting became the essential complement for identifi cation of yeasts and LAB. Compared to phenotyping methods, molecular DNA-based identifi cation methods off er a much higher taxonomic resolution at species up to strain level (31).
Several species belonging to the genera Leuconostoc, Weissella, Pediococcus, Lactococcus, Enterococcus and Streptococcus have been isolated from sourdough, while Lactobacillus strains were the most abundant (32). Yeasts such as Saccharomyces exiguous, Torulopsis holmii, Candida krusei, Pichia norvegensis and Hansenula anomala, are also present in sourdough, but Saccharomyces cerevisiae is frequently present or added (33). Only Lactobacillus reuteri isolated from spontaneous sourdough was considered to evaluate the antimicrobial properties, while yeasts were not taken into account in this study.

The antimicrobial activity of Lactobacillus reuteri
When it was determined that all LAB isolates from spontaneous sourdough were identifi ed as one species of L. reuteri, three mixtures of L. reuteri isolates (all of them were made from fi ve randomly selected isolates) were used for analysis of antimicrobial activity. L. reuteri was characterised as active against the tested bacterial strains usually found in foods and foodborne micromycetes that are most commonly detected as spoilers of milk and bread products ( Table 2).
MRS broth contains some substances that in combination with microbial acidifi ers may exert antimicrobial activity. The control solution did not show antimicrobial activity against the tested microorganisms, therefore the observed eff ects of cell-free supernatant were att ributed to L. reuteri metabolites.
Gram-negative Escherichia coli ATCC 25922 was the most resistant to the antimicrobial activity of L. reuteri and the resulting inhibition zones were statistically smallest (p≥0.05) compared with the inhibition zones of the other tested bacterial strains. The inhibitory activity of neutralised cell-free supernatant aft er the addition of catalase indicates the possibility to produce bacteriocins that are antimicrobially active components of L. reuteri. However, the active components of L. reuteri in the neutralised supernatant aff ected and not aff ected by catalase inhibit-ed only the reference culture of Listeria monocytogenes. These data strongly support the observation that Lactobacillus reuteri isolated from spontaneous sourdough produce bacteriocins and the inhibitory activity of these compounds depends on the type of bacteria subjected for inhibition. The loss of the antimicrobial activity aft er treatment with proteinase K indicates that antimicrobial substances produced by the tested LAB have a proteinaceous nature. They might be bacteriocins because their protease sensitivity is a key criterion in the characterisation of bacteriocin-like inhibitory substances (34).
L. reuteri was found to have a synergistic eff ect with the other tested Lactobacillus and Lactococcus spp. The synergistic eff ect of a dominant LAB culture present in sourdough is important because LAB strains with antimicrobial properties could be used for the production of starters that prevent bread defects and achieve the good taste properties.
According to the results of antifungal activity of Lacto bacillus reuteri isolates, the cell-free supernatant was more active against Cladosporium herbarum, Penicillium chrysogenum and Scopulariopsis brevicaulis when compared with the activity of cell cultures of these strains. The same data that the supernatants of the tested LAB were more active against Penicillium nordicum than the cell culture were found by other researchers (35). The results suggest that biologically active substances in supernatants have a greater eff ect on the spore formation than on the growth of mycelium. It is known that Lactobacillus plantarum supernatants had a wide spectrum of activity against the Eurotium, Penicillium, Aspergillus and fungi of other genera (36).
Lactobacillus reuteri isolates did not show an antifungal activity against Candida parapsilosis, Debaryomyces hansenii, Kluyveromyces marxianus, Pichia guilliermondii and Yarrowia lipolytica. Some LAB produce fungicidal substances that act against Candida spp. (37,38). K. marxianus was most sensitive to the L. plantarum eff ect of the examined yeasts (39). On the other hand, the synergistic eff ect of the investigated L. reuteri with yeasts is useful, because the sourdough LAB are mainly responsible for its acidifi cation, whereas yeasts are very important for the production of fl avour compounds and for a balanced fl avour in combination with acids.
Although L. reuteri showed only antilisterial activity aft er neutralising the acidity of culture supernatant, they nevertheless could be att ributed to LAB demonstrating superior antimicrobial activity against micromycetes and high inhibition by cell culture against bacterial strains. In order to determine L. reuteri as a starter culture that improves bread quality, its application was investigated in rye, wheat and rye with wheat bread making.

The eff ect of Lactobacillus reuteri on bread quality
The experimental bread technological parameters complied with the requirements of standard bread quality (40) and matched the quality of all types of bread. Lactococcus lactis ssp. lactis 140/2 was chosen for control bread making because this Lactococcus is suitable for all types of bread and it retains its specifi c composition aft er many generations during sourdough preparation.
According to the obtained colour measurement data (data not shown), it was found that the surfaces of all types of bread loaves were darker than bread slices, the crusts of fresh bread were less red-toned, while the crumbs were more yellow-toned. The colour characteristics of experimental and control bread samples did not change during storage for 7 days at 18 °C. The yeasts and LAB present in the dough reduced a certain amount of sugar, while during baking, the remaining sugar is involved in the Maillard reaction, in which aromatic substances are formed and give a desirable fl avour to the fi nal bread and a yellowish brown tone to the bread crust. This  shows that L. reuteri present in spontaneous sourdough are closely involved in the fermentation process and do not aff ect the colour characteristics of experimental bread (41).
The results of the sensory analysis of the bread samples revealed that the overall quality of bread prepared using L. reuteri was not signifi cantly diff erent compared with control bread. Experimental bread samples were acceptable for all assessors (Fig. 1) and sensory att ributes did not change signifi cantly during the 7-day storage.
Although the results revealed the diff erences among the diff erent types of bread, these fi ndings are in agreement with the results of other researchers (42) who determined the eff ectiveness of using baking yeasts, kefi r or Lactobacillus casei immobilized on brewer's spent grains in wheat bread production without any signifi cant negative eff ect on bread taste or aroma.
The bread crumb textural profi le of samples (Fig. 2) showed a signifi cant eff ect of sourdough on bread crumb fi rmness compared with the control samples.
Freshly baked experimental rye bread was soft er than the control bread. An analogous trend was found aft er rye bread storage for 7 days. This suggested that the eff ect of sourdough on the bread fi rmness was statistically significant (p<0.05) and a lower crumb fi rmness of rye bread could be expected when using spontaneous sourdough for bread preparation. Experimental wheat bread and rye with wheat bread had a higher crumb fi rmness. Bread crumb fi rmness underwent changes during storage. The chemical structure of wheat and rye is quite similar, so only small diff erences could be observed in starch retrogradation of rye and wheat bread (43). The structure of A-type starch switches into hydrated crystalline form after bread baking and during storage, but fermented rye bread was less crystalline. The less intense increase of wheat bread crumb fi rmness than of rye bread demonstrated the dependence of bread fi rmness on the used fl our, but not on the used sourdough composition.
There is some literature data about the use of Lactobacillus reuteri for bread making. It was noticed that the fi rmness of bread made with L. reuteri LTH5448 was higher when compared with bread made with Weissella cibaria 10M (44). Regardless of the possibility that bread consistency is aff ected by L. reuteri, certain LAB strains may be added as specifi c LAB cultures into sourdough to improve some parameters of the fi nal bread. The use of a mixture of Kluyveromyces marxianus and Lactobacillus plantarum as a sourdough starter cultures provides the possibility to improve the specifi c volume of bread without decreasing its sensory quality or acceptability (45).
Results of the textural and sensory analysis of experimental rye, wheat and rye with wheat bread prepared with L. reuteri show a signifi cant reduction of bread fi rmness compared to the control rye bread prepared with L. lactis ssp. lactis 140/2 as a starter culture suggesting its wide application for the rye bread making.

Protective properties of Lactobacillus reuteri
Following the determination of antimicrobial activity of L. reuteri isolated from spontaneous sourdough, it is important to estimate the eff ectiveness of this activity under natural fermentation conditions and heat treatment during baking. The protective properties of spontaneous sourdough fermented by L. reuteri were assessed by determining the number of extraneous microorganisms in baked bread samples. It was found that the fresh samples were not contaminated by microscopic fungi and aerobic mesophylic bacteria; only some bacterial colonies were detected (Fig. 3).
The positive protective eff ect of spontaneous sourdough was revealed during bread storage, because the total bacterial count was 1-20 times lower in the experimental bread than in the control bread prepared with Lactococcus lactis ssp. lactis 140/2 aft er 7 days of storage at room temperature. The fact that spontaneous sourdough possesses antimicrobial activity is also demonstrated by the absence of microscopic fungi and aerobic spore-forming bacteria in bread samples during storage over the course of the experiment.
LAB strains producing antimicrobial active substances secrete them into the growth environment, so anti-  Spraying a culture supernatant on bread slices is not a suitable approach for bread preservation since it was prepared in the MRS broth, which is composed of non-food grade chemicals. However, this method allows the assessment of the more detailed eff ect of metabolites produced by L. reuteri on fungal spoilage. Results of spoilage determination of bread loaves sprayed with MRS broth were coincident with those of unsprayed loaves and it suggested that only metabolites of L. reuteri were eff ective against moulding, not the components of MRS. Cell-free supernatant protected the commercial bread against mould during storage for 15 days, while the commercial bread that was not sprayed with the tested solution became mouldy aft er 8 days of storage at 25 °C and aft er 12 (wheat bread) and 15 days (rye bread) during storage at 15 °C (Table 3).
Under certain conditions, some lactobacilli and lactococci possessing lipolytic activities may produce significant amounts of fatt y acids (46). The unsaturated fatt y acids are active against Gram-positive bacteria, and the antifungal activity of fatt y acids is dependent on the chain length, concentration and pH of the medium (47). Fatt y acid analysis (Table 4) showed that L. reuteri produced 42 % of saturated, 31 % of monounsaturated, 26 % of polyunsaturated, 0.3 % of trans-fatt y acids, 5 % of omega-3 and 15 % of omega-6 fatt y acids. According to the inhibition zones (Table 2), L. reuteri had a greater impact on the growth inhibition of Gram-positive bacteria (Bacillus subtilis ssp. spizizenii, B. cereus, Listeria monocytogenes and Staphylococcus aureus) than Gram-negative (Escherichia coli). This could happen because of high total amount of unsaturated fatt y acids (57.3 %) produced by L. reuteri.
LAB suspension on the bread surface can be used for inhibition of the undesirable microorganism growth and for prevention of bread moulding (22). The same researchers applied the spraying of Pediococcus acidilactici and P. pentosaceus cell suspensions on the bread surface and a positive result of prolonging the shelf life of bread up to 8 days was determined. They proposed the LAB count of 10 4 -10 5 CFU/cm 2 for spraying the surface of bread. In our research, cell-free supernatant obtained by decanting approx. 1.8·10 9 cells (according to A 600 nm =2.3) was used for spraying bread loaves. The appearance of mould colonies was observed from 4 to 8 days later on the bread surfaces than on the control bread samples (not sprayed). Meanwhile, baked bread prepared with spontaneous sourdough containing L. reuteri was characterised by good antifungal properties and the antimould activity lasted during the storage for 15 days, when the small colonies of moulds appeared on the loaves of control rye    w(fatt y acid)/% butyric with wheat bread in the end of experiment. The mould contamination may be weaker in experimental bread prepared under laboratory conditions than in bread prepared under manufacturing conditions. Spores of microscopic fungi that get into the sourdough from the raw materials or from the air are killed during baking, while the spores may re-enter the surface of the bread during cooling, packaging, transportation or storage of the product and cause crumb damage through crust cracking.

Conclusions
Lactobacillus reuteri was isolated and identifi ed as lactic acid bacteria present in spontaneous sourdough. It expressed a good inhibiting eff ect against reference bacterial strains Bacillus subtilis ssp. spizizenii ATCC 6633, Bacillus cereus ATCC 11778, Listeria monocytogenes ATCC 19111, Esherichia coli ATCC 25922, Staphylococcus aureus ATCC 25923 and foodborne micromycetes Cladosporium herbarum SR-11, Penicillium chrysogenum SR-12, Scopulariopsis brevicaulis Mi-Gr-5, Aspergillus brasiliensis Mi-G-21 and Aspergillus versicolor Mi-Pr-4. The use of spontaneous sourdough for rye bread making results in signifi cantly lower fi rmness of bread in comparison with wheat and rye with wheat bread. The less intense increase of wheat bread crumb fi rmness than the fi rmness of rye bread during storage demonstrated the dependence of bread fi rmness on the used fl our, but not on the used sourdough composition. Antimicrobial substances produced by L. reuteri were observed to have a signifi cant antimould activity on the bread surfaces during 12-15 days of storage. The use of antimicrobial active sourdough could prevent bread defects such as ropiness or moulding and prolong the shelf life of bread.