Microbial Diversity of Type I Sourdoughs Prepared and Back‐Slopped with Wholemeal and Refined Soft (Triticum aestivum) Wheat Flours

The fermentation of type I sourdough was studied for 20 d with daily back‐slopping under laboratory and artisan bakery conditions using 1 wholemeal and 2 refined soft wheat (Triticum aestivum) flours. The sourdough bacterial and yeast diversity and dynamics were investigated by plate counting and a combination of culture‐dependent and culture‐independent PCR‐DGGE approach. The pH, total titrable acidity, and concentration of key organic acids (phytic, lactic, and acetic) were measured. Three flours differed for both chemical and rheological properties. A microbial succession was observed, with the atypical sourdough species detected at day 0 (i.e. Lactococcus lactis and Leuconostoc holzapfelii/citreum group for bacteria and Candida silvae and Wickerhamomyces anomalus for yeasts) being progressively replaced by taxa more adapted to the sourdough ecosystem (Lactobacillus brevis, Lactobacillus alimentarius/paralimentarius, Saccharomyces cerevisiae). In mature sourdoughs, a notably different species composition was observed. As sourdoughs propagated with the same flour at laboratory and artisan bakery level were compared, the influence of both the substrate and the propagation environment on microbial diversity was assumed.     

Utilization of African grains for sourdough bread making

Abstract: Acha and Iburu flours were singly subjected to sourdough fermentation with previously selected autochthonous starters. Sourdoughs were used (30%, wt/wt) as aroma carriers and acidifiers during short time fermentation with the addition of baker's yeast. Acha and Iburu sourdough breads were compared to wheat sourdough bread started with the same strains and to breads made with the same formula but using baker's yeast alone. During Acha and Iburu sourdough fermentations, starter lactic acid bacteria reached almost the same cell density found in wheat sourdoughs. Acidification was more intense. Iburu sourdough bread had the highest total titratable acidity, the lowest pH, and contained the highest levels of free amino acids and phytase activity. The values of in vitro protein digestibility did not differ between Acha sourdough and wheat sourdough breads, while Iburu sourdough bread showed a slightly lower value. Acha and Iburu sourdough breads showed lower specific volume and higher density with respect to wheat sourdough breads. Nevertheless, Acha and Iburu sourdough breads were preferred for hardness and resilience. As shown by sensory analysis, Acha and especially Iburu sourdough breads were appreciated for color, acid taste and flavor, and overall acceptability. Practical Application: This study was aimed at evaluating the technological and nutritional properties of the African cereals Acha and Iburu. Sourdough fermentation and the use of selected starters increased the nutritional and sensory qualities and the potential application for bakery industry.     

The microbial stability of two bakery sourdoughs made from conventionally and organically grown rye

In order to determine the microbial stability of sourdoughs, two bakery sourdoughs made from conventionally and organically grown rye, produced on a large production scale under defined conditions, were studied at regular intervals over a period of 7 months. The sourdoughs contained lactic acid bacteria at levels of 8·43–9·14 log₁₀cfu g⁻¹and yeasts at levels of 6·00–8·04 log₁₀cfu g⁻¹. The number of lactic acid bacteria seemed to increase and the number of yeast to decrease with increasing fermentation time. Four types (A–D) of lactic acid bacteria dominated in both sourdough systems. Their frequencies varied during the sampling period, but seemed to follow the same trend in both systems. Identification based on API 50CH assimilation profiles, classical identification methods and 16S ribosomal DNA sequencing showed that the lactic acid bacteria contained in the two sourdough systems not clearly belonged to any known species of the genus Lactobacillus. Saccharomyces cerevisiae was the only yeast species isolated from the sourdoughs, represented by different phenotypes. In the conventional sourdough system one phenotype dominated throughout the sampling period. In the organic sourdough the phenotype changed after 3 months.     

Biodiversity of lactic acid bacteria and yeasts in spontaneously-fermented buckwheat and teff sourdoughs

In this study, four different laboratory scale gluten-free (GF) sourdoughs were developed from buckwheat or teff flours. The fermentations were initiated by the spontaneous biota of the flours and developed under two technological conditions (A and B). Sourdoughs were propagated by continuous back-slopping until the stability was reached. The composition of the stable biota occurring in each sourdough was assessed using both culture-dependent and -independent techniques. Overall, a broad spectrum of lactic acid bacteria (LAB) and yeasts species, belonging mainly to the genera Lactobacillus, Pediococcus, Leuconostoc, Kazachstania and Candida, were identified in the stable sourdoughs. Buckwheat and teff sourdoughs were dominated mainly by obligate or facultative heterofermentative LAB, which are commonly associated with traditional wheat or rye sourdoughs. However, the spontaneous fermentation of the GF flours resulted also in the selection of species which are not consider endemic to traditional sourdoughs, i.e. Pediococcus pentosaceus, Leuconostoc holzapfelii, Lactobacillus gallinarum, Lactobacillus vaginalis, Lactobacillus sakei, Lactobacillus graminis and Weissella cibaria. In general, the composition of the stable biota was strongly affected by the fermentation conditions, whilst Lactobacillus plantarum dominated in all buckwheat sourdoughs. Lactobacillus pontis is described for the first time as dominant species in teff sourdough. Among yeasts, Saccharomyces cerevisiae and Candida glabrata dominated teff sourdoughs, whereas the solely Kazachstania barnetti was isolated in buckwheat sourdough developed under condition A. This study allowed the identification and isolation of LAB and yeasts species which are highly competitive during fermentation of buckwheat or teff flours. Representatives of these species can be selected as starters for the production of sourdough destined to GF bread production.     

Adaptability of lactic acid bacteria and yeasts to sourdoughs prepared from cereals, pseudocereals and cassava and use of competitive strains as starters

The adaptability of lactic acid bacteria (LAB) and yeasts to sourdoughs prepared from cereals, pseudocereals and cassava was investigated using PCR-DGGE and bacteriological culture combined with rRNA gene sequence analysis. Sourdoughs were prepared either from flours of the cereals wheat, rye, oat, barley, rice, maize, and millet, or from the pseudocereals amaranth, quinoa, and buckwheat, or from cassava, using a starter consisting of various species of LAB and yeasts. Doughs were propagated until a stable microbiota was established. The dominant LAB and yeast species were Lactobacillus fermentum, Lactobacillus helveticus, Lactobacillus paralimentarius, Lactobacillus plantarum, Lactobacillus pontis, Lactobacillus spicheri, Issatchenkia orientalis and Saccharomyces cerevisiae. The proportion of the species within the microbiota varied. L. paralimentarius dominated in the pseudocereal sourdoughs, L. fermentum, L. plantarum and L. spicheri in the cassava sourdough, and L. fermentum, L. helveticus and L. pontis in the cereal sourdoughs. S. cerevisiae constituted the dominating yeast, except for quinoa sourdough, where I. orientalis also reached similar counts, and buckwheat and oat sourdoughs, where no yeasts could be detected. To assess the usefulness of competitive LAB and yeasts as starters, the fermentations were repeated using flours from rice, maize, millet and the pseudocereals, and by starting the dough fermentation with selected dominant strains. At the end of fermentation, most of starter strains belonged to the dominating microbiota. For the rice, millet and quinoa sourdoughs the species composition was similar to that of the prior fermentation, whereas in the other sourdoughs, the composition differed.     

Effects of several starter cultures on the anti-mold activity and sensory attributes of a traditional flat bread (Sangak) from Iran

Effects of 8 different sourdoughs and their replacement levels at 10, 20, and 30%(w/w) on the volume (of dough), crust hardness, organoleptic, and anti-mold properties of Iranian sangak bread were investigated. Saccharomyces cerevisiae, Lactobacillus plantarum, Lactobacillus acidophilus, and Leuconostoc mesenteroides were selected for the preparation of sourdough samples. The highest dough volume was achieved when sourdough (those having S. cerevisiae) was used at 30%. The highest taste scores were found with the bread sample using the above-mentioned starters at 30% sourdough replacement level. Considering the chewing, appearance, and overall quality of the new products, most of the samples maintained the favorite sensory aspects of sangak bread. Use of lactic acid bacteria and yeast strains as part of the sourdough formulation (followed by the use of sourdough in the dough formulation) resulted in improved crust properties and greater anti-mold activities.     

Diversity and Stability of Yeast Species During the Fermentation of Tarhana

Tarhana is a traditional cereal-based fermented food produced with a mixture of yoghurt and flour. The main microbiota in the fermentation of tarhana is yeast, together with lactic acid bacteria. In this study, the yeast microbiota of home-made tarhana (HMT) and plant-type tarhana (PTT) dough samples was evaluated and compared during fermentation. Culture-dependent LSU and ITS-5.8S rDNA sequence analysis of yeast isolates collected during the tarhana dough fermentation clarified 45 selected isolates representing different clusters. These yeast isolates displayed high homologywith species Pichia kudriavzevii (11), Candida glabrata (11), Candida humilis (10), Saccharomyces cerevisiae (7), Kluyveromyces marxianus (4), Kazachstania servazzi (1), and Kazachstania unispora (1). Additionally, both culture-dependent and PCR-Denaturated Gradient Gel Electrophoresis (PCR-DGGE) analyses showed that S. cerevisiae, P. kudriavzevii and K. marxianus were abundant in the fermentation of HMT dough samples whereas P. kudriavzevii, C. humilis, and C. glabrata dominated the PTT dough samples. It was concluded that tarhana fermentation was accomplished with the presence of a wide variety of yeast species that mainly included P. kudriavzevii in both HMT and PTT dough samples.     

Microbial re-inoculation reveals differences in the leavening power of sourdough yeast strains

A method based on microbial re-inoculation, or the so-called backslopping and subsequent proofing of rye bread dough simulating commercial one-stage sourdough process, was used for the screening of the leavening capacity of sourdough yeast strains. Two yeast strains were initially tested with seven Lactobacillus strains. Thereafter, 17 yeast strains, mostly of sourdough origin, were tested with a backslopping procedure with heterofermentative Lactobacillus brevis as an acidifying lactic acid bacteria (LAB). The highest leavening capacity was found in sourdoughs containing Candida milleri, in particular when it was accompanied by obligately homofermentative Lactobacillus acidophilus or facultatively heterofermentative Lactobacillus plantarum when it acted homofermentatively. The leavening capacity of the reference strain Saccharomyces cerevisiae was about half that of C. milleri in all sourdoughs tested. The re-inoculation procedure increased the differences found in the leavening capacity of the tested yeast strains during final proofing of rye bread dough. The backslopped sourdoughs containing a heterofermentative Lactobacillus strain were more suppressive than those containing a homofermentative strain. The highest leavening capacity was found in C. milleri strains. The use of one backslopping cycle before assaying the leavening capacity of a laboratory sourdough is recommended since it helps to differentiate between yeast strains to be tested for their leavening power in the final bread dough.     

Free D- and L-Amino Acid Evolution During Sourdough Fermentation and Baking

The evolution of free D- and L-amino acids in sourdoughs started with various lactic acid bacteria (LAB) and yeasts was studied. Lactobacillus brevis subsp. lindneri CB1 and Lactobacillus plantrum DC400 had high proteolytic activity. During sourdough fermentation, Saccharomyces cerevisiae 141 and Saccharomyces exiguus M14 sequentially utilized free amino acids produced by bacterial activity. Due to increased cell yeast autolysis, more S. exiguus M14 inocula caused more free amino acids which were partially utilized by LAB without causing hydrolysis of wheat flour protein. D-alanine, D-glutamic acid and traces of other D-isomers were observed in sourdoughs fermented with L. brevis subsp. lindneri CB1 and S. cerevisiae 141. Free total D- and L-amino acid content decreased by more than 44% after baking the sourdoughs. No abiotic generation of new D-amino acid isomers was detected in the baked sourdoughs.     

Effect of Select Parameters of the Sourdough Rye Fermentation on the Activity of Some Mixed Starter Cultures

The effect of select parameters (i.e., rye flour ash content, temperature, and dough yield) of the sourdough fermentation on the fermentation activity of different starter cultures (lactic acid bacteria Lactococcus lactis ssp. Lactis, Weissella confusa, Lactobacillus plantarum, Lactobacillus brevis, Lactobacillus helveticus, and yeast Kluyveromyces marxianus subsp. Marxianus) was determined. The major metabolic end products of fermentation (D, L-lactic acid, acetic acid, ethanol and glycerol) and the evolution of total phenolic content and folic acid during bread making were measured. Lactobacillus helveticus and Kluyveromyces marxianus allowed obtaining sourdoughs with the highest lactic acid/acetic acid ratios. The mixed starter culture with Lactococcus lactis and Saccharomyces cerevisiae generated the most important quantities of D/L lactic acid. The maximum values of ethanol concentration were obtained in case of the sourdoughs from whole rye flour fermented at lower temperature (30°C) with mixed starter cultures containing Sacchomyces cerevisiae. The fermentation process and type of starter culture are also tools to increase the bioactive compounds, enabling the increase of the phenolic content of the sourdough.     

Identification of the major yeasts isolated from high moisture corn and corn silages in the United States using genetic and biochemical methods

The objective of this study was to identify species of yeasts in samples of high moisture corn (HMC) and corn silage (CS) collected from farms throughout the United States. Samples were plated and colonies were isolated for identification using DNA analysis. Randomly selected colonies were also identified by fatty acid methyl esters (FAME) and by physiological substrate profiling (ID 32C). For CS, Candida ethanolica, Saccharomyces bulderi, Pichia anomala, Kazachstania unispora, and Saccharomyces cerevisiae were the predominant yeasts. Pichia anomala, Issatchenkia orientalis, S. cerevisiae, and Pichia fermentans were the prevalent species in HMC. The 3 identification methods were in agreement at the species level for 16.6% of the isolates and showed no agreement for 25.7%. Agreement in species identification between ID 32C and DNA analysis, FAME and ID 32C, and FAME and DNA analysis was 41.1, 14.4, and 2.2%, respectively. Pichia anomala and I. orientalis were able to grow on lactic acid, whereas S. cerevisiae metabolized sugars (galactose, sucrose, and glucose) but failed to use lactic acid. The yeast diversity in CS and HMC varied due to type of feed and location. Differences in species assignments were seen among methods, but identification using substrate profiling generally corresponded with that based on DNA analysis. These findings provide information about the species that may be expected in silages, and this knowledge may lead to interventions that control unwanted yeasts.     

Flavour of sourdough wheat bread crumb

This study investigates the effect of adding sourdough to wheat bread dough on the production of flavour compounds in wheat bread crumb. The sourdoughs were fermented with starter cultures of lactic acid bacteria alone and in combination with sourdough yeasts. The volatile compounds in the bread crumb were isolated by a dynamic headspace technique and extraction analysis, analysed by gas chromatography (GC), and identified on the basis of GC retention times for reference compounds and mass spectrometry (MS). The chemical analyses were combined with sensory evaluation. The volume of the loaves increased significantly when the doughs had 5–20% sourdough added compared with the control bread (bread without sourdough). In the sourdough bread, the content of acetic acid, 2-methylpropanoic acid, and 3-methylbutanoic acid was generally higher, and loaves made with the addition of sourdoughs fermented withLactobacillus plantarum, L. delbrueckii, orL. sanfrancisco had a higher content of 2- or 3-methyl-1-butanol than control bread. Interactions were seen between the starter cultures and the sourdough yeasts, and the production of the following compounds was increased depending on the starter culture used and on the sourdough yeast: ethanol, 2-methylpropanol, 2/3-methyl-1-butanol, 2-phenylethanol, benzyl alcohol, acetic acid, 2-methylpropanoic acid, and 3-methylpropanoic acid. Bread made with an addition of 5% to 15% sourdough fermented withL. sanfrancisco had a pleasant, mild and sour odour and taste.L. plantarum bread had a strong, sour and unpleasant odour and a metallic sour taste with a sour aftertaste, but when the sourdough was also supplemented with the sourdough yeastSaccharomyces cerevisiae, the bread attained a more aromatic wheat bread flavour, which may be caused, in part, by a higher content of 2/3-methyl-1-butanol, 2-methylpropanoic acid, 3-methylbutanoic acid and 2-phenylethanol.     

Development of Co-Culture Sourdough Systems for Improving Bread Quality and Delaying Staling

Six different blends of lactic acid bacterial cultures (Lactobacillus plantarum or Pediococcus cerevisiae) combined with Saccharomyces cerevisiae were used as co-culture for preparing sourdoughs. These prefermented sourdoughs were applied for producing pan breads. Acidification power, leavening ability, and microbial counts were conducted at different sour fermentation intervals. The highest acidification power in sourdoughs were recorded when 2% w_d/w_d (dry weight of cells/dry weight of flour) of either L. plantarum or P. cerevisiae was used in combination with 2% w_d/w_d S. cerevisiae. The leavening ability of sourdoughs was significantly (p < 0.05) enhanced by increasing the addition percentage of the studied LAB. According to staling rate, physical, and sensory characteristics, the best two treatments were obtained when the co-culture of 2% w_d/w_d S. cerevisiae with 4% w_d/w_d L. plantarum or 4% w_d/w_d P. cerevisiae were used. These two treatments were successfully applied in production of the flat bread. According to staling rate, physical and sensory characteristics, sourdough prefermented with 2% w_d/w_d S. cerevisiae combined with 4% w_d/w_d L. plantarum was the best treatment for preparing of flat bread. Overall the co-culture of L. plantarum and with S. cerevisiae might be a useful tool for preparing sourdough starter that was most effective for improving sensory and physical properties of both pan and flat bread. In addition it can help extend the shelf life by delaying bread staling and inhibiting mold spoilage.     

Fundamental study on the effect of hydrostatic pressure treatment on the bread-making performance of oat flour

The use of sourdough in wheat and rye breads has been extensively studied; however, little is known about its potential effect when baking oat bread. Consequently, the impact of sourdough on oat bread quality was investigated. Two different sourdoughs were prepared from wholegrain oat flour without the addition of starter cultures, by continuous propagation at 28 (SD 28) or 37 °C (SD 37) until the composition of the lactic acid bacteria remained stable. The dominant LAB were identified by sequence analysis of the 16S rDNA isolated from pure cultures. LAB from SD 28 belonged to the species Leuconostoc argentinum, Pedicoccus pentosaceus and Weissella cibaria, while Lactobacillus coryniformis dominated SD 37. The isolated LAB were further used as starter cultures for the production of oat sourdoughs. Fundamental rheology revealed softening of the sourdoughs compared to non-acidified and chemically acidified controls, which could not be attributed to proteolytic activity. Incorporation of oat sourdough into an oat bread recipe resulted in significantly increased loaf-specific volume as well as improved texture, independent of addition level or sourdough type. Overall, the results of this study show that sourdoughs containing lactic acid bacteria isolated from oats have the potential to enhance oat bread quality.     

Parameters of rye,wheat, barley,and oat sourdoughs fermented with Lactobacillus plantarum LUHS135 that influence the quality of mixed rye–wheat bread,including acrylamide formation

A Lactobacillus plantarum strain was used for the production of rye, wheat, barley, and oat sourdoughs, and the influence of different sourdoughs on mixed rye–wheat bread quality parameters and acrylamide formation was evaluated. L. plantarum LUHS135 demonstrated versatile carbohydrate metabolism, good growth and acidification rates, and the ability to excrete amylolytic and proteolytic enzymes in various cereal sourdoughs. The same starter and different cereal substrates allow to produce sourdoughs showing different characteristics. The type of sourdough and its quantity had significant influence on acrylamide content in bread (P ≤ 0.0001), and using 5% or 10% of wheat sourdough, 5%, 15%, or 20% of barley sourdough, and 5% or 15% of oat sourdough, it is possible to reduce acrylamide content in bread. Thus, manufacturers need to take into account application of apparent technological approaches for acrylamide in bread reducing.     

Detection and identification of wild yeast in Koumiss

Koumiss is a slightly alcoholic fermented mare's milk beverage, originally obtained by using a natural mixed starter of lactic acid bacteria and yeasts. Yeast is an important component of Koumiss processing which can affect the aroma, texture, as well as the nutrients beneficial to human health, but few reports have examined the yeast ecology of local ecosystems. The purpose of this study was to isolate and identify the yeast present in Koumiss from three representative regions of China using a polyphasic method. A total of 655 yeast isolates were obtained from 96 Koumiss samples collected from three regions in China. Koumiss harbored yeast populations at 5-7 log CFU/ml. Twelve different yeast species belonging to nine genera were detected in the Koumiss samples tested, including Candida pararugosa, Dekkera anomala, Geotrichum sp., Issatchenkia orientalis, Kazachstania unispora, Kluyveromyces marxianus, Pichia deserticola, Pichia fermentans, Pichia manshurica, Pichia membranaefaciens, Saccharomyces cerevisiae and Torulaspora delbrueckii. Kluyveromyces marxianus, Kazachstania unispora and Saccharomyces cerevisiae were the dominant species present in this traditional fermented dairy product. This study is the first to identify the yeast communities associated with Koumiss in China. The results enrich our knowledge of yeast in Koumiss, give us a more complete picture of the microbial diversity in Koumiss and can be used to promote the development of the local dairy industry.     

Microbiological and technological characterization of sourdoughs destined for bread-making with barley flour

The aim of the present study was the microbiological and technological characterization of laboratory- made sourdoughs for use in barley-flour-based bread-making. A defined multi-strain starter culture consisting of selected lactic acid bacteria (LAB) and yeasts from wheat sourdoughs was inoculated into three flour–water mixtures, composed of: (i) 100% wheat flour (ii) 50% wheat flour and 50% hull-less barley flour (composite flour); (iii) 100% hull-less barley flour. After two months of continuous propagation, the chemical characteristics of the three sourdoughs were investigated by measuring: pH, total titratable acidity and concentrations of various microbial metabolites by HPLC (i.e. lactic, acetic, phenyllactic and butyric acids and diacetyl). The microbial traits were studied through viable counts, isolation and typing of LAB and yeasts and PCR-DGGE analyses. Only Saccharomyces cerevisiae and Lactobacillus plantarum were detectable in the sourdoughs together with other lactobacilli species which were different depending on the type of flour blend used. The molecular typing of the isolates highlighted that only a few strains among those initially inoculated prevailed. The volume increases of the three types of sourdough were also investigated and a correlation was seen between an increase in the barley flour content and a reduction in the dough volume.     

Prevalence,Genetic Diversity,and Technological Functions of the Lactobacillus sanfranciscensis in Sourdough: A Review

The use of sourdough as a leavening agent in bread making is a very old method that can be traced back to ancient times. Sourdoughs harbor a complex microbiota that is affected by multiple factors including factors related to cereal plants, grains, and sourdough processing techniques. Lactobacillus sanfranciscensis is the key autochthonous bacterium of the traditional sourdough microbiota and it is said to be “sourdough adapted” species. Despite the great dominance of this bacterium in sourdoughs, the origin of this species still remains unclear. Lactobacillus sanfranciscensis positively influences all aspects of sourdough and fermented foods. However, the positive influence of this species on sourdough is a strain‐dependent characteristic. The first purpose of this review was to discuss factors affecting the microbiota of sourdoughs with particular emphasis on reasons behind the remarkable prevalence of L. sanfranciscensis in this ecological niche. The second objective was to discuss the genotypic and phenotypic classification of L. sanfranciscensis strains and the influence of this species on technological and functional characteristics of sourdough including its influence on rheological properties of dough and bread characteristics, texture, aroma, and shelf‐life through the inhibition of fungal growth.     

Predominant Bacteria Diversity in Chinese Traditional Sourdough

The purpose of this study was to identify the major bacteria in Chinese traditional sourdough (CTS). Five CTS samples (Hn‐87, Sx‐91, Gs‐107, Hf‐112, and Hr‐122) were collected from different Chinese steamed breads shops or private households. The total bacterial DNA was extracted from sourdough samples and sequenced using Illumina Hiseq 2000 system. Illumina tags were assigned to BLASTN server based on 16S rRNA libraries to reveal a genetic profile. Phylogenetic analysis revealed that the bacteria in traditional sourdough samples were dominated by the genera Leuconostoc and Lactobacillus. Beta diversity analysis, principal component analysis, and cluster analysis compared the bacterial differences in traditional sourdough samples. The results showed that Leuconostoc, Lactobacillus, and Weissella were the predominant genera among the 5 samples. This differentiated the sourdoughs into 3 typologies, namely, 1) Gs‐107 and Sx‐91, 2) Hr‐122 and Hn‐87, and 3) Hf‐112. This study identified 3 unique major bacteria genus in CTS bread ecosystems.     

Description of the microflora of sourdoughs by culture-dependent and culture-independent methods

Four types of sourdoughs (L, C, B, Q) from artisanal bakeries in Northern Italy were studied using culture-dependent and culture-independent methods. In all samples, the yeast numbers ranged from 160 to 10⁷ cfu/g, and the numbers of lactic acid bacteria (LAB) ranged from 10³ to 10⁹ cfu/g. The isolated LAB were sequenced, and a similarity was noted between two samples (C, Q), both in terms of the species that were present and in terms of the percentage of isolates. In these two samples, Lactobacillus plantarum accounted for 73% and 89% of the bacteria, and Lactobacillus brevis represented 27% and 11%. In the third sample (B), however, the dominant LAB isolate was Lb. brevis (73%), while Lb. plantarum accounted for only 27%. The fourth sourdough (L) was completely different from the others. In this sample, the most prominent isolate was Weisella cibaria (56%), followed by Lb. plantarum (36%) and Pediococcus pentosaceus (8%). In three out of four samples (L, C and Q), all of the yeasts isolated were identified as Saccharomyces cerevisiae, yet only Candida humilis (90%) and Candida milleri (10%) were isolated in the fourth sample (B). The microbial ecology of the sourdoughs was also examined with direct methods. The results obtained by culture-independent methods and DGGE analysis underline a partial correspondence between the DNA and RNA analysis. These results demonstrate the importance of using a combined analytical approach to explore the microbial communities of sourdoughs.