Production of wheat bread without preservatives using sourdough starters

In order for the beneficial effects of sourdough application in breadmaking to take place a proper selection of lactic acid bacteria species and strains, an appropriate technology and effective control of the purity and activity of the selected cultures. Four symbiotic starters for sourdough for the production of bread were developed and probated in a production laboratory using the selected strains Lactobacillus brevis LBRZ7, L. buchneri LBRZ6, L. plantarum X2, L. paracasei RN5, L. sanfranciscensis R and L. fermentum LBRH10 and the probiotic strain Propionibacterium freudenreichii ssp. shermanii NBIMCC 327. The starter sourdoughs that include Propionibacterium freudenreichii ssp. shermanii NBIMCC 327 had greater antimicrobial activity against saprophytic microorganisms: Bacillus subtilis, B. mesentericus, Aspergillus niger, Penicillium sp. and Rhizopus sp., but none of them inhibited the growth of bakery yeasts Saccharomyces cerevisiae. It was established that in order to prevent bacterial spoilage 10% of the selected starter sourdoughs had to be added in the breadmaking process, while for prevention of mold spoilage the necessary amount of starter sourdough had to be between 15% and 20%.The application of the developed starters for the production of wheat bread guarantees longer shelf life and no adverse alterations in the features of the final bread.


Introduction
According to the European Commission Concerted Action on Functional Food Science in Europe (FuFoSE) and the International Life Sciences Institute (ILSI), Europe 'a food that beneficially affects one or more target functions in the body beyond adequate nutritional effects in a way that is relevant to either an improved state of health and well-being and/or reduction of risk of a disease' is called a functional food.
Not all strains of the genus Lactobacillus can be included in the composition of functional foods. For each type of product there are selected strains of microorganisms that carry out their metabolic processes, which contribute to the formation of the flavour and define the terms of storage of the final products.
Bakery products have a very short shelf life. Their quality depends on the time interval between baking and consumption. [1] Spoilage of bakery products is mainly due to the growth of moulds, the main species belonging to the genera Aspergillus, Fusarium and Penicillium, as well as to the roping of the bread, caused by Bacillus sp., especially B. subtilis and B. licheniformis. [2] The freshness of bread depends on the flavour, appearance and crispness of the crust, the hardness of the crumb and the volume of the bread. The taste of the bread, however, is considered the most important feature for consumers, as a criterion for eligibility of the products. [3] During storage, reduction in the freshness of the bread along with the increase of the hardness of the crumb result in loss of acceptable appearance to consumers, a process known as staling. [4] There are several physical methods for food preservation: heat treatment, cold storage, modified atmosphere storage, drying and freeze-drying. [5] The protection of bakery goods from mould spoilage is achieved primarily through inactivation of contaminating spores using (1) IR and microwave radiation; (2) fungal inhibitors such as ethanol and propionic, sorbic, benzoic and acetic acid and their salts; (3) appropriate packaging techniques such as modified atmosphere packaging; and (4) addition of sourdough. [6À8] The addition of sourdough is the best technique to keep the bread from spoilage meeting consumer's demand for natural food without additives. [7] Sourdough is a mixture of flour (from wheat, rye, rice, etc.) and water, which is fermented by the action of lactic acid bacteria and yeasts. [7] These microorganisms usually come from flour, dough ingredients or the environment.
The objectives to be achieved by the use of sourdough are a significant increase in the shelf life and the nutritional value of bread and improvement of the organoleptic properties of bread. The increase in the retention time of the sourdough bread is due to the higher levels of acidity and the higher concentration of produced organic acids in comparison to the commercial bread produced by using only yeasts. Moreover, the addition of sourdough increases the bioavailability of minerals in bread as a result of phytate hydrolysis. The improvement of the organoleptic properties of bread is due to the presence of non-volatile and volatile compounds that improve the flavour of the bread. However, the production of bread using sourdough is a very sensitive method that depends on various parameters which need to be controlled. The most important parameters of fermentation are pH during fermentation, the temperature of fermentation and the careful selection of starter cultures for obtaining sourdough with specific and desired properties. [3,9] There are a number of benefits of the application of sourdough in bread making: improvements in the volume of the bread and the structure of the crumb, [10,11] the flavour, [12] the nutritional value [13,14] and shelf life, [1,15À18] due to the delay of the process of staling and the prevention of mould and bacterial spoilage. [19,20] These positive effects are associated with the metabolic activity of the selected pure cultures of yeast and homoand heterofermentative lactic acid bacteria in the composition of the sourdough, e.g. lactic acid fermentation, proteolysis, exopolysaccharide production and synthesis of volatile and antimicrobial compounds. [1,7,21] The fermentation of sourdough can affect intestinal health through several mechanisms: (1) modulation of the complex dietary fibres and the subsequent pattern of fermentation, (2) production exopolysaccharides with prebiotic properties, and (3) possible transfer of metabolites from the fermentation of lactic acid bacteria that affect the intestinal microflora. [22] In order for those beneficial effects to take place a proper selection of lactic acid bacteria species and strains, an appropriate technology and effective control of the purity and activity of the cultures are required. The selection of pure cultures consists of using a species or a combination of species specific to the technological process, fully adapted to the environment of sourdough and to the applied fermentation conditions. [23,24] There is a considerable diversity of lactic acid bacteria isolated from sourdough: L. acidophilus, L. delbrueckii, L. farciminis (obligate homofermentative), L. plantarum, L. homohiochii, (facultative heterofermentative), L. brevis, L. buchneri, L. fermentum, L. hilgardii, L. sanfranciscensis, L. viridiscens, L. panis and L. pontis (obligate heterofermentative). [7,25,26] With the inclusion of starter cultures, the pH falls very quickly, so the whole manufacturing process is accelerated, which leads to economic benefits for the producer. The secondary effects of the acidification and the acceleration of the fermentation time include changes in the activity of the enzymes of the cereal substrates or the bacterial strains. [1] The greater part of the starter cultures are natural isolates of the desired microorganisms normally found in cereal substrates. [27,28] The aim of the present study was to develop starter symbiotic combinations for wheat sourdough and to determine the percentage of starter sourdough to be used in bread making that would extend the shelf life of bread without preservatives without any adverse alterations of bread quality.

Materials and methods Microorganisms
The studies in this work were conducted using six strains of the genus Lactobacillus, isolated by the authors and currently included in the collection of microorganisms of Department 'Microbiology' at the University of Food Technologies, Plovdiv, Bulgaria and Propionibacterium freudenreichii ssp. shermanii NBIMCC 327. The six strains were as follows: Lactobacillus brevis LBRZ7 and L. buchneri LBRZ6 (isolated from fermented cabbage), L. plantarum X2, L. paracasei RN5 and L. sanfranciscensis R (isolated from naturally fermented sourdough), L. fermentum LBRH10 (of human origin) (unpublished data). These strains had been identified by physiological, biochemical and molecular-genetic methods. [29,30] Media Sterile skimmed milk with titratable acidity 16À18 T (Scharlau) MRS-broth (medium of Man, Rogosa, Sharpe) (Scharlau) MRS-agar. Composition (g/dm 3 ): MRS-broth (Scharlau), agar À 20.

Cultivation and storage of the studied microorganisms
The studied strains of microorganisms were cultured in a liquid medium (MRS-broth) and on agar medium (MRSagar) at 30 C for L. brevis LBRZ7, L. buchneri LBRZ6 and L. sanfranciscensis R and L. plantarum X2, and at 37 C for L. paracasei RN5 and L. fermentum LBRH10 for different periods of time. Propionibacterium freudenreichii ssp. shermanii NBIMCC 327 was cultured in skimmed milk at 30 C. All tested Lactobacillus strains had been isolated from a single colony and were cultured in MRS-broth medium for 24 hours.

Development of starters for sourdough for wheat and rye bread
First, the Basic Combination, containing L. paracasei RN5, L. plantarum X2, L. brevis LBRZ7 and L. fermentum LBRH10 in a ratio of 1:2:1:1 was prepared.
The four new combinations were subcultured in MRSbroth daily for the duration of 96 hours and the changes in the titratable acidity and the number of viable cells of the strains of Lactobacillus sp. and Propionibacterium sp. were monitored. The LAB counts were determined by appropriate 10-fold dilutions and spread plating on coloured LAPTg10-agar medium. The number of propionic acid bacteria was determined by appropriate 10-fold dilutions and pour plating on elective medium for Propionibacterium sp. A standard method [31] was used for measurement of the titratable acidity.

Preparation of cellular suspensions for the inoculation of flour/water mixture
For the production of sourdough with the four developed multistrain starters the 24-hour cultural suspensions of the Lactobacillus strains, included in the starter, were mixed according to the proportions given in 'Development of starters for sourdough for wheat and rye bread' and homogenized. The lactobacilli cells were harvested by centrifugation at 5000 £g for 15 minutes, washed twice with PBS-buffer and the biomass sludge was resuspended to the initial mixed suspension volume with sterile saline solution. Then the cellular suspension was mixed in the respective ratio with the 24À48 hour cultural suspension of P. freudenreichii ssp. shermanii NBIMCC 327 in milk and the obtained homogenized mixture was used for inoculation of the flour/water mixture for the production of multistrain sourdough.

Preparation of sourdoughs with multistrain starters
The cellular suspensions of the four combinations were obtained using the procedure described above and used to inoculate the flour/water mixtures. The changes in the concentration of viable cells of lactic acid bacteria, yeasts and moulds and in the titratable acidity of the sourdoughs were monitored by passaging every 24 hours over a period of 96 h of cultivation at 30 C according to the following scheme: (1) first day À 44% flour: 56% tap water and 10% of the mixed cellular suspension; (2) second to fifth day: 25% sourdough from the previous day: 75% fresh mixture flour/water. The fresh mixture was prepared in a ratio: 44% flour/ 56% water.
The number of lactic acid bacteria, propionic acid bacteria, yeasts and moulds was determined by appropriate 10-fold dillusions and spread plating on coloured LAPTg10-agar medium for the enumeration of lactic acid bacteria or pour plating on elective medium for Propionibacterium sp. for the enumeration of propionic acid bacteria. Total titratable acidity is determined by a standard method. [31] Determination of the antimicrobial activity against saprophytic microorganisms The agar diffusion method was used to determine the antimicrobial activity of the four prepared starter sourdoughs. A dilution in a ratio of 1:1 of a sourdough:saline solution of each of the sourdoughs was prepared. The antimicrobial activity was tested against the following saprophytic test microorganisms: bacteria À B. subtilis, B. mesentericus; yeasts À Saccharomyces cerevisiae, moulds À Aspergillus niger, Penicillium sp., Rhizopus sp. A suspension of each of the test microorganisms (10 6 À10 7 CFU/cm 3 ) was used to inoculate a Petri dish with agar medium and after the hardening of the agar wells (7 mm) were prepared. 0.06 cm 3 of the dillusions were pipetted in the wells of the plates and the plates with the test microorganisms were incubated at 37 C for 24À48 hours, and then the inhibition zones in mm were reported.
Approbation of the starter sourdoughs in the production laboratory 'Mother' doughs with 10%, 15% or 20% of the 96-hour starter sourdoughs were prepared. Each dough was prepared with 1.5% NaCl, 2% yeast starter, the respective percentage of sourdough and tap water (the amount of water was determined by the water absorption of the type of flour). The dough was kneaded in a mixer: slow kneading (1000 rpm) for 4 min and fast kneading (1400 rpm) for 10 min, after which the dough was rested for about 10 min in a proofer in order for its elastic properties to be improved. Loaves were formed and placed in the forms. Then followed leavening for about 40À45 min at 30 C and 80 § 5 RH. In the production laboratory wheat bread with sourdough as well as control bread (bread without sourdough with starter) were baked, cooled and evaluated. Baking was carried out at 225 § 5 C for 30 min in a deck oven. Loaves were allowed to cool for 120 min at room temperature.

Determination of bacterial spoilage of baked bread
The determination of the appearance of bacterial spoilage was performed by 10 judges in the production laboratory and was evaluated according to a scale of IÀIV, with each of the degrees corresponding to the following descriptions: I barely noticeable (pleasant fruity odour); II weak (change in the odour À distinct); III medium (moisty, sticky crumb, sharp odour); IV strong (unpleasant odour, brown-yellow crumb).

Determination of mould spoilage of baked bread
The determination of the appearance of mould spoilage was performed by 10 judges in the production laboratory and was evaluated according to the appearance of single mould colonies.

Selection of strains
In order to be included in the composition of the starters for sourdough for bread a mandatory selection lactic acid bacteria strains is required. It is crucial to select lactobacilli strains that accumulate high concentrations of viable cells in short time in order for a targeted fermentation process to be conducted. Therefore, the reproduction ability of the lactobacilli strains and the produced amount of lactic or lactic and acetic acids were examined. The ability of the strains L. paracasei RN5, L. plantarum X2, L. brevis LBRZ7, L. fermentum LBRH10, L. buchneri LBRZ6 and L. sanfranciscensis LSR to grow in flour/water environment and to accumulate high concentrations of viable cells and acids was examined. The selected lactobacilli strains grew very well in the flour/water environment, reaching 10 14 À10 15 CFU/cm 3 viable cells by the 96th hour and the acidity of the resulting sourdoughs increased to values above 10 N. [32]

Development of symbiotic combinations
Based on the results in our previous studies (unpublished data) on starter cultures for sourdough for wheat and rye bread, the starter with the best performance was determined to be the combination of L. paracasei RN5:L. plantarum X2:L. brevis LBRZ7:L. fermentum LBRH10 in a ratio of 1:2:1:1 (called Basic Combination). Four novel combinations À two, in which the L. buchneri LBRZ6 was included (for wheat bread), and two, in which L. sanfranciscensis LSR was included (for rye bread), were designed.  Table 1).
The results for the titratable acidity and the concentration of viable cells upon passaging clearly indicated a symbiotic relationship between the strains in the combinations.

Approbation of starter cultures for wheat bread in a production laboratory
The four combinations were kneaded every 24 hours for the duration of 96 hours in a production laboratory. The changes in the acidity and the aroma of the sourdoughs in passaging as well as the concentration of viable cells of the Lactobacillus strains at the 0th and at the 96th hour were determined. The results of these studies are summarized in Table 2.
For 96 hours of passaging in the production laboratory it was clear that the sourdoughs with the new starters were stabilized in terms of acidity. The biggest change in the concentration of viable cells was observed in Starter 3 (5 logN), followed by Starter 2 (4 logN), Starter 4 (3 logN) and Starter 1 (2 logN).
Each of the four 96-hour starter sourdoughs was diluted in saline solution in a ratio of 1:1 and the prepared diluted suspension was used to determine the antimicrobial activity of the sourdoughs against saprophytic microorganisms (Table 3). All sourdoughs had inhibitory activity against B. subtilis, B. mesentericus, A. niger, Rhizopus sp. and Penicillium sp. Neither of the four sourdoughs inhibit the bakery yeast strain Saccharomyces cerevisiae. A comparison between the starters with L. buchneri LBRZ6 (Starter 3 and Starter 4) showed that Starter 4 (with the participation of P. freudenreichii ssp. shermanii NBIMCC 327) had higher antimicrobial  activity against the saprophytes included in the study. The similar conclusions could be drawn from the comparison between Starter 1 and Starter 2, the two starters with L. sanfranciscensis LSR, which was clear evidence of the inhibiting effect of P. freudenreichii ssp. shermanii NBIMCC 327 against saprophytes. Bread was baked with different percentage of the four sourdoughs (10%; 15% and 20%) with the four 96-hour starter sourdoughs in order to determine the best starter combination for wheat and rye bread, as well as the optimum percentage of starter sourdough to be incorporated in the preparation of the 'mother' dough to prevent mould and Bacillus growth without adversely affecting the organoleptic characteristics of the final bread.
Highly contaminated flours containing high concentration of Bacillus spores (over 10 2 CFU/g) were used for the determination of bacterial and mould spoilage of the baked breads. The baked breads with 10%; 15% or 20% of the 96-hour starter sourdoughs were incubated in non- aseptic conditions in parallel experiments at room temperature and in a thermostat at 37 C for 96 hours for the determination of bacterial spoilage and for 120 hours at 30 C and at room temperature for determination of mould spoilage. It was established that bacterial spoilage due to the growth of representatives of the genus Bacillus occurred earlier in the control loaf incubated at 37 C, than in that incubated at room temperature. The earliest appearance of bacterial spoilage was in the control bread À on the 48th hour after taking the loaves out of the oven at 37 C and on the 72nd hour at room temperature. According to the standard requirements there should be no signs of bacterial decay up to the 48th hour. The control bread did not meet the standard requirements for microbial safety of bakery products. Upon addition of 10% of sourdough there were signs of bacterial spoilage on the 72nd hour both at 37 C and at room temperature. If the percentage of sourdough addition rose to 15%, in the obtained sourdough bread with Starter 1 or Starter 3 bacterial decay became visible on the 96th hour, while variants baked with sourdough with Starter 2 or Starter 4 (they both included the probiotic strain P. freudenreichii ssp. shermanii NBIMCC 327) showed no signs of bacterial spoilage even on the 96th hour both at 37 C and at room temperature. Upon addition of 20% of sourdough no bacterial spoilage was established except for the variants prepared with 20% of sourdough with Starter 1 or Starter 3 and were stored at 37 C (Table 4).
The earliest appearance of mould spoilage was in the control bread À on the 72nd hour both at 30 C and at room temperature. Upon addition of 10% of sourdough the first signs of spoilage became noticeable on the 96th hour, the degree of mould spoilage being the least in the variants made by the inclusion of sourdough with Starter 2 or Starter 4 both at 30 C and at room temperature. If the percentage of incorporation of starter sourdough in the bread-making process was increased to 15%, mould spoilage became visible on the 120th hour in all variants both at 30 C and at room temperature, except for those prepared with sourdough with Starter 2 or Starter 4 (they both included the probiotic strain P. freudenreichii ssp. shermanii NBIMCC 327) that were stored at room temperature in which there was no mould spoilage even on the 120th hour. Upon addition of 20% of sourdough there was no mould spoilage up to the 120th hour (Table 5). Experimental data suggest accelerating of the fermentation. The doughs obtained applying the new starters were tougher, more elastic, and the pieces of bread are higher. As in previous studies with sourdough, the quality of wheat bread was improved by the addition of sourdough. Addition of sourdoughs yielded breads with greater specific loaf volumes when compared to the control. Crumb firmness values showed the opposite trend, with sourdough breads showing a softer and lighter crumb than the control, which was in compliance with previously published studies.
[33À35] The results confirmed the results obtained by Clarke and Arendt [36] demonstrating that the metabolic products of the yeasts and LAB improve the properties of the flour, as well as aroma, taste, nutritive value and shelf life of the bread.
The incorporation of 15% or more sourdough in the bread-making process inhibited the growth of both bacterial and mould spores and ensured long shelf life of the baked bread, [2] and while the increase in acidification might be necessary for the optimal swelling and baking of bread, for the control of enzymatic activities, elasticity and suitability of the crumb, and for prolonging shelf life [37]; in contrast, excessive acidification has a deleterious effect on some rheological parameters. [37,38] The sourdoughs with the developed starters, described in the present article, were incorporated in a ratio of 15% in non-sterile (nonaseptic) conditions, in contrast to the experiments, described by Mentes et al. [2] which were conducted in aseptic conditions. The prevention of bacterial and mould spoilage during incubation of the baked bread at room temperature (25 À 30 C) was also demonstrated in the present article.
Currently the two starter sourdoughs (sourdough with Starter 2 for rye bread and sourdough with Starter 4 for wheat bread) are still being passaged daily in the production laboratory.

Conclusion
Four symbiotic starter combinations were developed and probated in a production laboratory using the selected Lactobacillus strains that were able to grow in flour/water environment accumulating high concentrations of viable cells: L. brevis LBRZ7, L. buchneri LBRZ6, L. plantarum X2, L. paracasei RN5, L. sanfranciscensis R and L. fermentum LBRH10 and the probiotic propionic acid bacterial strain P. freudenreichii ssp. shermanii NBIMCC 327. It was established that in order to prevent bacterial spoilage 10% of the selected starter sourdoughs (sourdough with Starter 2 for rye bread and sourdough with Starter 4 for wheat bread, both starters including P. freudenreichii ssp. shermanii NBIMCC 327) had to be added in the bread-making process while for prevention of mould spoilage the necessary amount of starter sourdough was between 15% and 20%. The developed sourdoughs would allow consumers to prepare and consume delicious and safe bread with extended shelf life without the addition of preservatives.