International Journal of Food Microbiology 171 (2014) 136–146

Contents lists available at ScienceDirect

International Journal of Food Microbiology

journal homepage: www.elsevier.com/locate/ijfoodmicro

Review

Ecological parameters influencing microbial diversity and stability of

traditional sourdough
Fabio Minervini ⁎, Maria De Angelis, Raffaella Di Cagno, Marco Gobbetti
Dipartimento di Scienze del Suolo, della Pianta e degli Alimenti, Università degli Studi di Bari Aldo Moro, via Amendola 165/a, 70126 Bari, Italy

a r t i c l e i n f o a b s t r a c t

Article history: The quality of some leavened, sourdough baked goods is not always consistent, unless a well propagated sour-
Received 11 July 2013 dough starter culture is used for the dough fermentation. Among the different types of sourdough used, the
Received in revised form 20 November 2013 traditional sourdough has attracted the interest of researchers, mainly because of its large microbial diversity,
Accepted 22 November 2013
especially with respect to lactic acid bacteria. Variation in this diversity and the factors that cause it will impact
Available online 1 December 2013
on quality and is the subject of this review.
Keywords:
Sourdough microbial diversity is mainly caused by the following factors: (i) sourdough is obtained through spon-
Microbial ecology taneous, multi-step fermentation; (ii) it is propagated using flour, whose nutrient content may vary according to
Traditional sourdough the batch and to the crop, and which is naturally contaminated by microorganisms; and (iii) it is propagated
Lactic acid bacteria under peculiar technological parameters, which vary depending on the historical and cultural background and
Yeasts type of baked good. In the population dynamics leading from flour to mature sourdough, lactic acid bacteria (sev-
Back-slopping eral species of Lactobacillus sp., Leuconostoc sp., and Weissella sp.) and yeasts (mainly Saccharomyces cerevisiae
Flour and Candida sp.) outcompete other microbial groups contaminating flour, and interact with each other at differ-
ent levels. Ecological parameters qualitatively and quantitatively affecting the dominant sourdough microbiota
may be classified into specific technological parameters (e.g., percentage of sourdough used as inoculum, time
and temperature of fermentation) and parameters that are not fully controlled by those who manage the prop-
agation of sourdough (e.g., chemical, enzyme and microbial composition of flour).
Although some sourdoughs have been reported to harbour a persistent dominant microbiota, the stability of
sourdough ecosystem during time is debated. Indeed, several factors may interfere with the persistence of species
and strains associations that are typical of a given sourdough: metabolic adaptability to the stressing conditions of
sourdough, nutritional and antagonistic interactions among microorganisms, intrinsic robustness of microorgan-
isms, and existence of a stable house microbiota.
Further studies have to be performed in order to highlight hidden mechanisms underlying the microbial struc-
ture and stability of sourdough. The comprehension of such mechanisms would be helpful to assess the most
appropriate conditions that allow keeping a given traditional sourdough as a stable microbial ecosystem, thus
preserving, during time, the typical traits of the resulting product.
© 2013 Elsevier B.V. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
2. Production of sourdough . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
3. Microbial dynamics from flour to mature sourdough . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
4. Interactions between lactic acid bacteria and yeasts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
5. Ecological parameters in the sourdough ecosystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
5.1. Specific technological parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
5.2. Not fully controllable parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
6. Factors influencing microbial stability of traditional sourdough . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
6.1. Metabolic adaptability to the stressing conditions of sourdough . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
6.2. Nutritional interactions among microorganisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141

⁎ Corresponding author at: via Amendola 165/a, 70126 Bari, Italy. Tel.: +39 080 5442948; fax: +39 080 5442911.
E-mail address: [email protected] (F. Minervini).

0168-1605/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ijfoodmicro.2013.11.021
 F. Minervini et al. / International Journal of Food Microbiology 171 (2014) 136–146 137

6.3. Intrinsic robustness of microorganisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142

6.4. Antagonistic interactions among microorganisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
6.5. Existence of a stable house microbiota . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
7. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144

1. Introduction each fermentation step. Back-slopping is also applied later, for propagat-
ing mature sourdough over time (De Vuyst et al., 2009).
Sourdough is a mixture of flour (mainly wheat or rye) and water, Dough is a nutrient-rich ecosystem. Complex carbohydrates (starch,
fermented with lactic acid bacteria (LAB) and yeasts, which are respon- above all) are present, but their partial hydrolysis to di-saccharides
sible for its capacity to leaven a dough, while contemporarily and (maltose, above all) and mono-saccharides (fructose and glucose),
unavoidably acidifying it (De Vuyst and Neysens, 2005; Gobbetti, by flour and microbial amylases, rapidly takes place. Nitrates, am-
1998; Vogel et al., 1999). In the modern bakery technology, sourdough monia, and proteins constitute the nitrogen sources for microbial
represents an alternative to the use of baker's yeast (although bakers growth. During dough fermentation, proteins are hydrolysed to
often use a combination of both leavening agents) to manufacture a more easily usable nutrients (peptides and free amino acids, FAA)
variety of products such as bread, crackers, snacks, pizza and sweet by flour and microbial proteinases. The values of water activity
baked goods, because it offers many advantages over baker's yeast: (aw), ranging from 0.96 to 0.98, do not limit the growth of the ma-
enhanced flavour (Hansen and Schieberle, 2005), prolonged shelf-life jority of contaminant microorganisms. The pH is sub-acid, although,
(Chavan and Chavan, 2011), improved dough structure (Arendt et al., during dough incubation and as the number of back-slopping steps
2007) and increased nutritional value (Gobbetti et al., 2013; Poutanen increases, it tends to become acid (values around 4.0) (Table 1 and
et al., 2009) of the leavened baked good. Although liquid (type II) and Fig. 1). The redox potential gradually decreases during dough for-
dried (type III) sourdoughs, produced at industrial level, are fairly wide- mation and incubation from positive to negative values (Hammes
spread among bakers because they are easy to be used (Brandt, 2007), et al., 2005). Taking into account the above physic-chemical param-
traditional (type I) sourdough tickles researchers' curiosity mainly for eters, sourdough allows LAB and yeast to outgrow other microbial
its large microbial diversity (De Vuyst et al., 2009). This feature of tradi- populations (Fig. 1). Specific influences of the sourdough ecosystem
tional sourdough is mainly caused by the use of a spontaneous multi- and its production on the microbial ecology of fermentation will be
step fermentation, needed for obtaining a sourdough (“mature” sour- discussed in later sections.
dough) with a constant leavening and acidifying capacity (Hammes
and Gänzle, 1998), and by the use of back-slopping as a tool (almost) 3. Microbial dynamics from flour to mature sourdough
daily applied for propagating sourdough (De Vuyst et al., 2009). For
this reason, most of the studies dealing with microbial ecology of sour- At the beginning of the first fermentation, the microbial population
dough focused on traditional sourdough. of dough reflects that of the flour, consisting of LAB, Gram-positive
The microbial ecology of cereal fermentation has been reviewed by (e.g., Bacillus sp.) and Gram-negative (e.g., Pseudomonas sp.) aerobic
Hammes et al. (2005). Ecological determinants of sourdough microbiota bacteria, Enterobacteriaceae, yeasts and moulds (Fig. 1). Each microbial
were examined as part of the review by De Vuyst et al. (2009). Other group is present at cell numbers generally not exceeding 5 log CFU/g
reviews focused either on general aspects (Chavan and Chavan, 2011) (Onno and Roussel, 1994; Rocha and Malcata, 2012; Stolz, 1999).
or on features of sourdough other than microbial ecology (Arendt Through bacterial 16S rRNA pyrosequencing, it has been recently
et al., 2011; Moroni et al., 2009; Yao et al., 2013). Since 2009, various found that, before the first fermentation, several bacterial phyla
studies have significantly advanced our knowledge about the microbial (e.g., Bacteroidetes, Cyanobacteria, Firmicutes, and Proteobacteria)
ecology of sourdough fermentations and have inspired the justification occur in the dough. However, the majority of these phyla either indicate
for this review. Therefore, the objectives of this review are: (i) to give an the presence of a non-active population in the flours or are outcompeted
overview about how LAB and yeasts become the dominant microbial by Firmicutes already after the first fermentation (Ercolini et al., 2013),
groups in traditional sourdough and how they interact with each which is consistent with already-known patterns in the microbial ecology
other; (ii) to examine factors affecting microbial diversity of traditional of fermented foods (Humblot and Guyot, 2009; Jeong et al., 2013; Jung
sourdough in a systematical way; and (iii) to discuss about parameters et al., 2013). Upon addition of water to flour, redox potential of the
influencing the microbial stability of traditional sourdough. dough decreases (Hammes et al., 2005), favouring the growth of faculta-
tive anaerobes (Enterobacteriaceae and yeasts) and of LAB (Fig. 1), most of
which are aerotollerant anaerobes. Because carbohydrate metabolism
2. Production of sourdough of LAB is highly adapted to mono- and di-saccharides (Gänzle and
Gobbetti, 2013), lactic and acetic acids are produced leading to a
Traditional sourdough originates from multiple steps of fermenta- decrease of pH of the dough. Such a decrease, usually becoming evi-
tion. In the first step a dough, usually composed of just flour and water, dent after the second fermentation step, may inhibit the growth of
is spontaneously fermented. Then, the fermented dough is used as inoc- Enterobacteriaceae, while it is well tolerated by yeasts. Consequently,
ulum for fermenting newly prepared dough, which, in turn, will be used as the number of fermentation steps increases, LAB and yeasts become
as inoculum for a subsequent step of fermentation. Additional ingredi- more and more adapted to the environmental conditions of sourdough
ents, such as grape juice/must, honey, hop, overripe fruit, salt, sugar, (Fig. 1), until they dominate the mature sourdough (Hammes and
vinegar may be used in early fermentation steps, in order to include in Gänzle, 1998), at numbers ranging from 6 to 9 log CFU/g and from 5 to
the dough immediately available nutrients and pro-technological micro- 8 log CFU/g, respectively (Lattanzi et al., 2013; Minervini et al., 2012a).
organisms. A protocol for the production of a mature “French style” sour- Actually, time (intended as the consecutive fermentation steps) is the
dough is given in Fig. 1 (Onno and Roussel, 1994). The figure also shows variable that mostly affects the structure of the sourdough microbiota
the typical trend of cell density of different microbial groups during this (Rocha and Malcata, 2012; Weckx et al., 2010a). For instance, it has
process. Apart from the first fermentation, the operation named “back- been recently found that a gradual succession between the active popula-
slopping” (or “refreshment”), consisting in the inoculation of flour and tions of Proteobacteria and Firmicutes occurs from the beginning of the
water with an aliquot of previously fermented dough, is repeated before first fermentation to the end of the second fermentation of rye-based
138 F. Minervini et al. / International Journal of Food Microbiology 171 (2014) 136–146

Fig. 1. Flow chart resuming the protocol of production of mature sourdough, according to the “French style” (adapted from Onno and Roussel, 1994), flanked by a graphical representation
of the typical dynamic of different microbial groups occurring therein. For each microbial group, the number of dots in the bars is directly proportional to the cell density at a given stage of
the protocol.

sourdough (Ercolini et al., 2013). A mature sourdough, i.e. a sourdough depending on the type of flour (Ercolini et al., 2013; Van der Meulen
with constant cell densities of LAB and yeasts, acidification and leavening et al., 2007; Weckx et al., 2010a, 2010b).
capacities, may be achieved in a number of days that varies (5–7 days) Within LAB, a typical population dynamic, referred to as “the three-
phase evolution”, has been found, regardless of the type of flour. These
Table 1 three phases are: (i) dominance of LAB species belonging to the genera
Ecological parameters and main relative effects on sourdough microbial ecosystem. Enterococcus, Lactococcus and Leuconostoc; (ii) increasingly important
Parameter Main effect(s) presence of sourdough-specific LAB, such as species belonging to the gen-
Specific technological parameters
era Lactobacillus, Pediococcus and Weissella; and (iii) dominance of well-
Dough yield Ratio lactic acid bacteria/yeasts; balance between adapted sourdough strains, belonging to obligate heterofermentative
homofermentative and heterofermentative lactic acid bacteria species (e.g., Lactobacillus sanfranciscensis, Lactobacillus fermentum) and
% sourdough used Ratio lactic acid bacteria/yeasts; balance among different lactic to Lactobacillus plantarum (Van der Meulen et al., 2007; Weckx et al.,
as inoculum acid bacterium genera
2010a, 2010b), although strains of Leuconostoc sp. are sometimes en-
NaCl Ratio lactic acid bacteria/yeasts
Redox potential Balance between homofermentative and heterofermentative countered in the sourdough ecosystem (Corsetti et al., 2001; Minervini
lactic acid bacteria et al., 2012a, 2012b; Rocha and Malcata, 1999; Zotta et al., 2008). This
Fermentation time Balance among differently stress-resistant lactic acid bacteria succession of LAB is mainly driven by different tolerance to acidic
Fermentation Ratio lactic acid bacteria/yeasts; balance between conditions and to different adaptation mechanisms related to car-
temperature homofermentative and heterofermentative lactic acid bacteria
bohydrate and nitrogen metabolism (Gänzle et al., 2007). Although
pH Ratio lactic acid bacteria/yeasts; balance among different lactic
acid bacterium genera different genera of LAB may inhabit the sourdough ecosystem,
Number of back- Increased selective pressures; increased disturbance the most frequently encountered genus is Lactobacillus (De Vuyst e
slopping steps Neysens, 2005). Among this genus, and taking into account only
Storage Cold stress as a cause of selective pressure
identifications obtained through genotypic methods or polyphasic
temperature
approach, the most common species isolated from sourdough are
Not fully controllable parameters Lactobacillus brevis, L. fermentum, Lactobacillus reuteri, Lactobacillus rossiae
Flour Introduction of contaminant microorganisms; slight and L. sanfranciscensis (obligate heterofermentative); Lactobacillus
modification of nutrient composition alimentarius, Lactobacillus paralimentarius and L. plantarum (faculta-
House microbiota Microbial stability of sourdough over time
tive heterofermentative); Lactobacillus amylovorus and Lactobacillus
 F. Minervini et al. / International Journal of Food Microbiology 171 (2014) 136–146 139

delbrueckii (obligate homofermentative) (for more details, refer to On the other hand, lactobacilli contribute to proteolysis during sour-
Huys et al., 2013). dough fermentation, increasing the concentration of aliphatic, dicarbox-
Although a great variety of yeast species have been found in sour- ylic and hydroxyl amino acid groups, most of which are used by yeasts
doughs (Hammes et al., 2005), Saccharomyces cerevisiae, Kazachstania (Gobbetti et al., 1994b).
exigua (formerly Saccharomyces exiguus, anamorph Candida holmii) Interactions between LAB and yeasts are often mediated by products
and Candida humilis (synonym Candida milleri) are those most frequent- of microbial metabolism. Obligate heterofermentative LAB synthesize
ly encountered, followed by Pichia kudriavzevii (formerly Issatchenkia acetic acid following heterolactic fermentation. In late stage of sour-
orientalis, anamorph Candida krusei) (Garofalo et al., 2008; Huys et al., dough fermentation (pH ≈ 4.0), most of acetic acid is undissociated
2013; Iacumin et al., 2009; Vogelmann et al., 2009). and therefore it may cross cytoplasmic membrane and enter the cell.
Apart from LAB and yeasts, sourdough may occasionally harbour Due to this, the growth of some yeasts (e.g., strains of S. cerevisiae) is
acetic acid bacteria, such as Acetobacter sp. (Minervini et al., 2012b; inhibited. However, tolerance of other yeasts to acetic acid may influ-
Scheirlinck et al., 2008; Vogelmann et al., 2009), possibly affecting ence microbial associations in sourdough (Suihko and Makinen,
dough acidification. However the origin and role of these Gram- 1984). For instance, tolerance shown by K. exigua to acetic acid may be
negative aerobic bacteria need to be elucidated. one of the causes of the stable association between this yeast and
L. sanfranciscensis in the traditional sourdough used for making San
4. Interactions between lactic acid bacteria and yeasts Francisco bread (Kline and Sugihara, 1971). On the other hand, the
antagonism of yeasts against LAB has not to be neglected (Viljoen,
In traditional sourdough, LAB and yeasts frequently interact mainly 2006), although, to our knowledge, it has not been reported for sour-
through metabolism of carbohydrates and of nitrogen sources, and dough so far. This interaction is strain-dependent, at both the yeast
through production of stimulatory or inhibitory compounds. The fre- and bacterium level and is related to production of inhibitory short
quent association between L. sanfranciscensis and C. humilis and/or chain fatty acids (e.g. hexanoic, octanoic, decanoic), sulphur dioxide,
K. exigua is based on commensalism (De Vuyst et al., 2009). Indeed, peptides, zymocidal proteins, and ethanol (Fleet, 2003).
L. sanfranciscensis preferentially uses maltose which, after being intro- Besides products of carbohydrate metabolism, microorganisms can
duced in the cytoplasm, is hydrolysed by maltose phosphorylase release a wide array of secondary metabolites (Keller and Surette,
to glucose-1-phosphate and glucose. This enables metabolization of 2006). The exposure of S. cerevisiae LBS and L. sanfranciscensis LSCE1
glucose-1-phosphate without expenditure of ATP, while glucose cells to oxidative, acid or osmotic sub-lethal stress gives rise to release
is exported outside the cell and may be metabolized by maltose- of ethyl esters of some unsaturated long-chain fatty acids (Guerzoni
negative yeasts (Neubauer et al., 1994). L. sanfranciscensis, as well as et al., 2007). These molecules may be regarded as stress markers and
other obligate heterofermentative LAB, may gain extra ATP through have a possible physiological function in the cell–cell communication
the use of additional electron acceptors, such as fructose, oxygen, and pathways (Black and Di Russo, 2006; Verstrepen et al., 2003). Further-
extra pyruvate generated from citrate. Indeed, in the presence of addi- more, under oxidative stress, L. sanfranciscensis releases two 2(5H)-
tional electron acceptors, the intermediate acetyl-phosphate, instead furanones (Guerzoni et al., 2007), which meet several criteria to be
of being reduced to ethanol, is converted into acetate, with concurrent included into cell–cell communication molecules (Ndagijimana et al.,
production of extra ATP (Gobbetti, 1998). The generation of fructose 2006).
from some flour oligosaccharides by specific enzymes of C. humilis
causes an ecological advantage for the lactobacilli capable of using fruc-
5. Ecological parameters in the sourdough ecosystem
tose as additional electron acceptor (Gobbetti et al., 1995; Stolz et al.,
1995), and reduces the competition for carbohydrates between LAB
In order to understand the combined effects exerted on the microbi-
and yeasts (Gobbetti and Corsetti, 1996). Competition for maltose and
al growth by different ecological parameters, these are traditionally
glucose may occur when obligate heterofermentative lactobacilli are
classified into endogenous (e.g., pH) and exogenous (e.g., temperature)
associated with the maltose-positive S. cerevisiae. Indeed, this yeast
parameters. However, this classification does not fit well to traditional
species consume those flour carbohydrates at a higher rate than
sourdough, because complex interactions, resulting from manual oper-
heterofermentative lactobacilli (Collar, 1996). Therefore, a decrease in
ations and microbial activities, make it a peculiar microbial ecosystem.
the metabolism of heterofermentative lactobacilli is expected when
Thereby, it should be preferred to distinguish between specific techno-
associated with maltose-positive yeasts (De Vuyst and Neysens,
logical parameters (e.g., percentage of sourdough, pH, temperature of
2005). However, traditional sourdough processes rarely lead to the
fermentation, etc…) and parameters that are not fully (or at all) under
depletion of fermentable carbohydrates. Indeed, maltose is continuously
the control of those who daily manage the propagation of sourdough,
supplied by the activity of flour amylases and it never gets exhausted in
namely flour used in the propagation and role of the so-called “house”
few hours (3–8 h) of fermentation (De Vuyst and Neysens, 2005). The
microbiota. All of these parameters and their combination are involved
steady presence of maltose and other flour carbohydrates, along with
in the selection of the typical microbiota of a given sourdough. Further-
the similar values of time of generation found for L. sanfranciscensis,
more, most of them also play a key-role in maintaining or disturbing the
C. humilis and S. cerevisiae, has been recently proposed as the cause of
microbial structure of a given sourdough over time.
the stable and non-competitive association between these microbial
species in traditional sourdoughs (Venturi et al., 2012).
In co-culture model systems, growth of L. sanfranciscensis and 5.1. Specific technological parameters
L. plantarum is stimulated by K. exigua and S. cerevisiae. This may be
related to both the lack of competition for the nitrogen source and to Specific technological parameters and their relative main effects on
the excretion of stimulatory compounds by yeasts (Gobbetti et al., sourdough microbiota are listed in Table 1 (De Vuyst et al., 2009;
1994a). In the co-presence of organic and inorganic nitrogen sources, Hammes et al., 2005). The influence of the most studied parameters
yeasts preferentially use the latter (e.g., ammonia), whereas LAB prefer on the dominant sourdough microbiota is discussed below. Obviously,
to use FAA and, above all, small peptides. During growth or as a conse- to consider one parameter at a time should not overlook the fact that
quence of accelerated autolysis, yeast cells excrete essential and/or microbial growth is influenced by multiple combinations of different
stimulatory amino acids for lactobacilli (Alexandre and Guilloux- parameters. Indeed, it is the continuous use of the same technological
Benatier, 2006; Gobbetti et al., 1994a; Velasco et al., 2004). A small parameters that ultimately leads to the selection of microbial strains
peptide (Asp–Cys–Glu–Gly–Lys), identified in freshly prepared yeast that are best adapted to the applied process conditions (Scheirlinck
extract, stimulates the growth of L. sanfranciscensis (Berg et al., 1981). et al., 2009).
140 F. Minervini et al. / International Journal of Food Microbiology 171 (2014) 136–146

Dough yield (DY) is expressed by the ratio between dough weight 2008). Overall, low values of initial pH favour yeasts over LAB. When
and flour weight, multiplied by 100. Dough weight results from the sourdough fermentation began under sub-acid conditions (initial pH
sum of flour, water, starter inoculum and other ingredients, such as ranging from 5.6 to 5.8), L. sanfranciscensis showed a higher multiplica-
salt. For instance, the values of DY in the first and in the second fermen- tion factor than C. humilis. The opposite occurred at an initial pH value of
tation steps illustrated in Fig. 1 are: 5.0 or lower, with lactobacilli being completely inhibited at pH 4.1
(Brandt et al., 2004) (Table 1).
(First fermentation) DY = [(600 g + 300 g + 3 g)/600 g] ∗ 100 = The temperature of fermentation affects the microbial composition
150.5; of sourdough. Overall, this parameter is inversely correlated to time of
fermentation. Sourdough LAB have a growth temperature optimum
(Second fermentation) DY = [(300 g + 300 g + 130 g + 1.5 g)/
of 30–40 °C, higher than that of yeasts (25–27 °C) (Brummer and
(300 g + 200 g)] ∗ 100 = 146.3.
Lorenz, 1991; Gänzle et al., 1998; Spicher and Stephan, 1999). This
Because water is, along with flour, the principal ingredient of could be the rational for the so-called “baker's rule”, according to
the dough, DY is mainly related to the amount of water used in dough which (relatively) low temperatures (20 to 26 °C) during sourdough
formulation: the higher is the amount of water, the higher is the value fermentations are better for yeast growth than higher temperatures
of DY. In most of cases, traditional sourdoughs are firm doughs, charac- (Spicher and Stephan, 1999) (Table 1). Vrancken et al. (2011) highlight-
terized by a value of DY of ca. 150–160. However, some traditional sour- ed a modulation of the microbiota as a function of time and temperature
doughs, such as those produced according to the “American style” of back-slopping, going from almost exclusively L. fermentum at 30 °C
(Kulp, 2003), may reach DY values of up to 225, corresponding to aw and 37 °C, with back-slopping every 24 h, to a codominance of
value of ca. 0.98 (Hammes et al., 2005). Such values of aw are not limit- L. fermentum and L. plantarum at 30 °C and 48-h back-slopping, to the
ing for either LAB or yeasts. Yet, when a relatively high DY is combined complete absence of these species at 23 °C. The latter fermentations
with long (24–48 h) fermentation time, LAB are favoured over yeasts were dominated exclusively by Leuconostoc citreum. Furthermore,
(Decock and Cappelle, 2005). Furthermore, relatively high values of stable numbers of yeasts and LAB were obtained in all the fermentation
DY, combined with higher temperatures (ca. 35–37 °C), favour the conditions, except for the fermentation at 37 °C and 24 h. Under these
growth of homofermentative LAB (Decock and Cappelle, 2005; Onno conditions, yeasts were almost completely absent, possibly reflecting
and Roussel, 1994) (Table 1). Firmer sourdoughs (DY of ca. 160) are an increased competitiveness of LAB compared to the yeasts or a growth
easily colonized by yeasts (minimal aw for most of yeasts is 0.88), but limitation of the yeasts caused by the high temperature (Vrancken et al.,
represent a more selective environment for LAB, allowing the domi- 2011).
nance of halotolerant strains (Hammes et al., 2005). Sometimes NaCl When considering the number of back-slopping steps, as it in-
is used (1–3% on the total dough weight) during back-slopping creases, the environmental conditions become more and more selective,
(Fig. 1). While a low level of NaCl (up to 0.7%) stimulates growth of resulting in the dominance of a very low number of species, such as
LAB, higher levels (1.6–3.2%) decrease their growth to a much greater L. sanfranciscensis (Lattanzi et al., 2013). Conversely, at each back-
extent than yeasts. Indeed, with increasing level of NaCl, the cell density slopping new variables (e.g., batch of flour) may be introduced, causing
ratio LAB/yeasts changed from 20:1 to 1:1 (Simonson et al., 2003) modifications of the sourdough ecosystem in a relatively short time
(Table 1). (Ottogalli et al., 1996) (Table 1).
Redox potential is mainly influenced by the level of oxygen incorpo-
rated during dough kneading. The presence of oxygen may represent an 5.2. Not fully controllable parameters
ecological advantage for those lactobacilli that are able to use it as exter-
nal electron acceptor. Indeed, through the activity of enzymes such as Flour used for propagating traditional sourdough may strongly influ-
NADH oxidase, NADH peroxidase, phosphotransacetylase, pyruvate ence microbial diversity, because it provides sourdough microorgan-
dehydrogenase, and pyruvate oxidase, heterofermentative lactobacilli isms with nutrients and, being non-sterile, is a vehicle of contaminant
may generate additional energy by the activity of acetate kinase, microorganisms, which, through daily and continuous back-slopping,
which allows the recycling of NAD+ without the need of ethanol could have the chance to become dominant (De Vuyst et al., 2009).
formation (De Vuyst and Neysens, 2005) (Table 1). Accordingly, Cereal flours have a highly heterogeneous nutrient composition, which
increased acidification was found when a sourdough started with offers the possibility for the simultaneous occupation of specific ecological
L. sanfranciscensis CB1 had been insufflated with air. Furthermore, cell niches by different microbial species and strains (De Vuyst et al., 2009).
viability of the starter was not affected during 24 h of fermentation The capacity of some sourdough lactobacilli (e.g., L. plantarum) to fer-
(De Angelis and Gobbetti, 1999). ment all wheat flour carbohydrates (pentoses included) may reduce
The percentage of sourdough used as inoculum during daily back- the metabolic competition with yeasts (Corsetti et al., 2001), although
slopping usually ranges from 10 to 40% of the total dough weight. As catabolite repression by glucose could also occur (Stolz et al., 1993).
this parameter increases, the initial pH of the dough decreases, because L. paralimentarius was isolated from Apulian (Southern Italy) sour-
of intrinsic acidity of sourdough. In this way, the percentage of sour- doughs with a relatively high frequency and this could be related to
dough influences growth rates of LAB (Brandt et al., 2004). Percentage the capacity, by all the isolates allotted to that species, to ferment the
of sourdough lower than 2% favoured growth of L. sanfranciscensis four main flour soluble carbohydrates (Corsetti et al., 2001). A compar-
over that of C. humilis. On the opposite, the use of high (approaching ative study of LAB and yeasts dominating 19 sourdoughs used for
to 50%) percentages of sourdough inhibited the growth of lactobacilli, manufacturing traditional/typical Italian breads, and of nutrient (solu-
but not that of yeasts, probably because of the low initial pH of the ble carbohydrates and FAA) composition of the different flours used
dough and of inhibitory concentrations of undissociated organic acids for propagation, highlighted that the type of flour (Triticum durum or
(Brandt et al., 2004) (Table 1). Triticum aestivum), as well as the concentration of nutrients, may have
Traditional sourdoughs have a pH range (3.5–4.3) that usually meets a key-role for selecting the population of LAB (Minervini et al., 2012a).
the growth requirements of the dominant sourdough microorganisms The main distinguishing features of T. durum flour were the higher
(De Vuyst and Neysens, 2005). Overall, within LAB, lactobacilli domi- concentrations of maltose, glucose, fructose and FAA. Compared to
nate this ecosystem, also because of their adaptation to low pH. Howev- the T. aestivum-based sourdoughs, the T. durum-based sourdoughs
er, genera of LAB present in the cereal kernels and flour, such as were characterized by the sole or main presence of obligate
Enterococcus, Lactococcus, Leuconostoc, Pediococcus and Weissella heterofermentative LAB (mainly L. sanfranciscensis, Leuconostoc sp.,
(Corsetti et al., 2007a), may dominate sourdoughs characterized by Weissella cibaria and L. brevis), the lower number of facultative
higher pH (Corsetti et al., 2007b; Rocha and Malcata, 1999; Zotta et al., heterofermentative LAB, and the lower cell density of yeasts
 F. Minervini et al. / International Journal of Food Microbiology 171 (2014) 136–146 141

(Minervini et al., 2012a). Obligate heterofermentative LAB were domi- addition, when sourdough is stored, LAB may face with nutritional lim-
nant in only three (Pane Casereccio di Reggio Calabria, Pane Casereccio itation (Gänzle and Gobbetti, 2013). The response to stress may be spe-
del Molise and Bozza Pratese) out of nine T. aestivum-based sourdoughs. cific, i.e. following exposure to a given stress, LAB over-synthesize
Because these three sourdoughs were based on T. aestivum flours char- certain proteins which are not over-synthesized following exposure to
acterized by the highest concentration of fermentable carbohydrates a different stress. On the contrary, it may happen that different stresses
and total FAA, it was hypothesized that obligate heterofermentative (e.g., low temperature, NaCl) induce the over-synthesis of proteins,
LAB are less competitive than the two other metabolic groups in sour- some of which are over-synthesized irrespective of the kind of stress
dough based on flours containing such nutrients at (relatively) low (Hörmann et al., 2006).
concentrations (Minervini et al., 2012a). Despite this, the role of the To overcome deleterious effects caused by low temperature, bacteria
type of flour in the process leading to the formation of the microbial have to develop a transient adaptive cold–shock response (Graumann
structure of mature sourdough deserves further studies. Indeed, it has and Marahiel, 1998). Sourdough LAB may continue to grow at a reduced
been recently found that the number of shared Operational Taxonomic rate after a temperature decrease of ca. 20 °C below their optimum
Units between sourdoughs based on different flours (rye, durum and (De Angelis and Gobbetti, 2004). L. sanfranciscensis, L. plantarum,
soft wheat) increases as the process goes on (Ercolini et al., 2013). L. brevis, Lactobacillus hilgardii, L. alimentarius and Lactobacillus
This means that initial differences of microbial community among fructivorans grew in wheat flour hydrolysate at 15 °C by increasing the
doughs based on different flours tend to disappear as the number of lag phase (from ca. 2 to 5 h) and the generation time (from ca. 10 to
back-slopping steps increases. This results in the definition of a core 18 h). An array of 14–18 (depending on the species) cold-shock pro-
microbiota (consisting of LAB) which is shared between mature sour- teins (CSPs) were over-synthesized by L. plantarum DB200, L. brevis
doughs, regardless of the type of flour (Ercolini et al., 2013). H12, L. plantarum 20B and L. sanfranciscensis CB1 upon cold adaptation
Although a given sourdough is commonly propagated over time (15 °C, 2 h) and subsequent freezing. The higher level of synthesis of
using the same type of flour, seasonality characterizing crops affects CSPs was related to the cell recovery after freezing, which was higher
nutrient composition of flour (Table 1). Because the capability to than for cells directly (i.e. with no adaptation) subjected to freezing
adapt to a specific substrate is highly strain-specific, even small changes (De Angelis and Gobbetti, 2004).
of substrate quality will have effects on the microbiota (Vogelmann Survival under acidic conditions is positively affected by the adapta-
et al., 2009), with the exception of intrinsically robust microorganisms tion to low pH, a mechanism known as acid-tolerance response (Foster
(Minervini et al., 2010; Siragusa et al., 2009). and Hall, 1991). After growth at a constant pH of 6.4, the survival of
Spontaneous sourdough fermentations carried out at laboratory mid-exponential phase cells of L. sanfranciscensis CB1 decreased dra-
level under semi-aseptic conditions indicate that flour used for propaga- matically when suddenly subjected to pH 3.2–3.4, as set by lactic acid
tion is one of the possible sources of LAB (Van der Meulen et al., 2007) or by a mixture of lactic and acetic acids. Higher survival could be ob-
and yeasts (Vrancken et al., 2010). In these conditions, if a given micro- tained upon adaptation to pH 5.0 for 1 h, prior to exposure to low pH,
bial species is not detected in the flour, it will not be in the sourdough. and was attributed to 15 acid-shock proteins over-synthesized by
On the other hand, under the practical conditions of sourdough propa- L. sanfranciscensis CB1 (De Angelis et al., 2001). Specific metabolic path-
gation (at the bakery), the influence of the “house” microbiota seems ways of utilisation of glutamine and arginine may help LAB to overcome
to be of outmost importance (Scheirlinck et al., 2009). “House” microbi- acidic stress. Strains of L. sanfranciscensis and of L. reuteri are able to con-
ota is the term used to indicate those microorganisms contaminating vert glutamine into glutamate through glutaminase, generating ammo-
the bakery setting and equipment. This parameter is not fully controlla- nia that limits the decrease of intracellular pH (Vermeulen et al., 2007).
ble by those who manage sourdough propagation, unless a thorough Several lactobacilli possess all the enzymes of the arginine deiminase
sanitization plan is daily scheduled (Table 1). Because house microbiota pathway (De Angelis et al., 2002; Rollan et al., 2003; Thiele et al.,
may affect microbial stability of traditional sourdough (Scheirlinck et al., 2002; Vrancken et al., 2009). In this pathway arginine is converted
2009), this parameter will be discussed later. into ornithine and ammonia, with the consumption of two protons,
which allows microbial cells to spare ATP required for proton extrusion
6. Factors influencing microbial stability of traditional sourdough (Konings, 2002). Both glutamate and ornithine contribute to the flavour
of the product (De Vuyst et al., 2009).
Some traditional sourdoughs are characterized by a microbial Compared to cold and acid stress adaptation, nutrient limitation
composition that remains stable over years (Scheirlinck et al., 2009; faced by microorganisms has received scarce attention. Carbon catabo-
Venturi et al., 2012) and, sometimes, over decades (Böcker et al., lite repression could cause nutrient limitation to some microorganisms
1990; Gänzle and Vogel, 2003). On the contrary, some studies highlight- in the sourdough ecosystem. Indeed, glucose excreted by some
ed a relative instability of the sourdough ecosystem, even in a relatively lactobacilli (e.g., L. sanfranciscensis), following utilisation of maltose,
short time (Minervini et al., 2010, 2012b; Siragusa et al., 2009). Overall may induce catabolite repression in other LAB and yeasts. When glucose
in traditional sourdough a stable microbiota is only achieved when the is exhausted, these microorganisms could start to metabolize other
multiplication factors (expressed by the ratio between cell density at carbohydrates (e.g., maltose), provided that these latter have not been
the end of fermentation and cell density at the beginning of fermenta- totally consumed by the abovementioned lactobacilli. Stress from nutri-
tion) of the different microorganisms are very similar (Brandt et al., ent limitation causes phenotypic responses encompassing the use of
2004). The use of specific technological parameters having constant external electron acceptors, the preferential and/or simultaneous
values during time is undoubtedly a prerequisite for guaranteeing the use of unconventional energy sources (e.g., amino acids, ribose and de-
microbial stability of a given sourdough (Vogelmann and Hertel, oxyribose), and the interaction with exogenous and flour enzymes
2011). Yet, besides that, microbial stability is affected by the following (Gänzle and Gobbetti, 2013).
factors: (i) metabolic adaptability to the stressing conditions of sour-
doughs; (ii) nutritional interactions among microorganisms; (iii) intrin- 6.2. Nutritional interactions among microorganisms
sic robustness of microorganisms; (iv) antagonistic interactions among
microorganisms; and (v) existence of a stable house microbiota. Overall, it is obvious that the absence of competition for carbohy-
drates, as well as for other nutrients, may be a key-factor in the micro-
6.1. Metabolic adaptability to the stressing conditions of sourdough bial stability of traditional sourdough. Differences in the use of flour
soluble carbohydrates may result in a non-competitive association of
During propagation and, above all, storage of sourdough, LAB are different species of LAB, such as in the case of the stable association
exposed to sub-optimal or critical values of temperatures and pH. In between L. sanfranciscensis and L. plantarum. The former species
142 F. Minervini et al. / International Journal of Food Microbiology 171 (2014) 136–146

preferentially utilizes maltose and is generally unable to ferment fruc-

tose, whereas the latter species preferentially ferments glucose and
fructose with maltose metabolism being subject to carbon catabolite
repression (Corsetti et al., 2001; Gobbetti, 1998).
As discussed previously, stable, non-competitive associations exist
between the maltose-negative, acid-tolerant K. exigua or C. humilis and
the maltose-positive L. sanfranciscensis in traditional sourdoughs
(Gobbetti, 1998; Hammes et al., 1996). The association between
L. sanfranciscensis and C. humilis has been recently confirmed as sta-
ble, regardless of technological parameters applied during propaga-
tion and of the presence of potentially competing microorganisms
(L. fermentum, Lactobacillus helveticus, Lactobacillus pontis, L. reuteri,
Lactobacillus johnsonii, S. cerevisiae, I. orientalis) deliberately added
(Vogelmann and Hertel, 2011).

6.3. Intrinsic robustness of microorganisms

The frequent predominance of L. sanfranciscensis as the key-lactic

acid bacterium of traditional sourdough is possibly the result of the
selective pressures that arise through the environmental conditions
pertinent to the typical technological parameters (relatively low
fermentation temperature, continuous back-slopping) applied during
propagation of sourdough (Corsetti et al., 2001; Foschino et al.,
1999; Kline and Sugihara, 1971). Recently, the genome analysis of
L. sanfranciscensis TMW 1.1304 showed that, although the genome
size is the smallest within the lactobacilli (ca. 1.3 Mbp), ribosomal
RNA operons are present at the highest density among all known
genome of free-living bacteria. This feature could allow this bacteri-
um to respond quickly to favourable conditions of sourdough, and to
initiate suddenly fermentative metabolism and fast growth (Vogel Fig. 2. Persistence of Lactobacillus sanfranciscensis and Lactobacillus plantarum strains in
et al., 2011). However, the capacity of this species to persist in sour- sourdough after continuous 10 day-long propagation. In (a) the LS numbers refer to the
different strains of L. sanfranciscensis. In (b) the alpha-numerical codes refer to the different
dough seems to be strain-dependent. The structure and stability of the strains of L. plantarum (adapted from Siragusa et al., 2009 and from Minervini et al., 2010).
dominant bacterial population were investigated during 10 day-long
continuous propagation (at laboratory level, under semi-aseptic condi-
tions) of nine different sourdoughs singly started with as many strains diacetyl and hydrogen peroxide (showing antimicrobial activity),
of L. sanfranciscensis (Siragusa et al., 2009). Only three starters dominat- and especially synthesis of bacteriocins (Gobbetti et al., 2005), such
ed throughout ten days of continuous back-slopping, whereas the other as plantaricins. Plantaricins were active also during sourdough
strains progressively decreased to less than 3 log CFU/g (Fig. 2a). fermentation (Settanni et al., 2005) and they inhibited growth of
Weissella confusa, L. sanfranciscensis, L. plantarum, L. rossiae, L. brevis, other L. plantarum strains (Todorov et al., 1999). Some strains of
Lactococcus lactis ssp. lactis and Pediococcus pentosaceus were identified L. plantarum synthesize plantaricin A (plnA), which acts as a pheromone
as the dominant bacterial species in the flour used for back-slopping. At in the mechanism of inter-species cell–cell communication (Di Cagno
the end of propagation, one strain of L. sanfranciscensis, belonging to the et al., 2010). Under conditions that mimicked the sourdough fermenta-
contaminant microbiota of the flour, was found in all the sourdoughs. tion, the level of plnA depended on the microbial partner and plnA
Persistent starters were found in association with other biotypes of acted as inhibitory compound, especially towards some strains of
L. sanfranciscensis and with W. confusa or L. plantarum. The three persis- L. sanfranciscensis (Di Cagno et al., 2010).
tent L. sanfranciscensis strains were further used as single starters for the
production of as many sourdoughs, which were propagated, under the 6.4. Antagonistic interactions among microorganisms
same conditions as above, by using a different flour, whose lactic acid
bacterium population in part differed from the previous one. Also in Several sourdough LAB may synthesize antimicrobial compounds,
this case all the three starters persisted during propagation (Siragusa like bacteriocins, bacteriocin-like inhibitory substances (BLIS) and anti-
et al., 2009). A subsequent study was set up with a similar approach, biotics (Corsetti et al., 1996; Olsen et al., 1995). Bacteriocins are small
aiming to investigate the robustness of seven sourdough strains of peptides or proteins, ribosomically synthesized, endowed with bacteri-
L. plantarum. In this case, five out of seven strains maintained elevat- cidal or bacteriostatic activity especially towards bacteria not strictly
ed cell numbers (ca. 9 log CFU/g) throughout ten days of daily prop- correlated, from a phylogenetic point of view, to the producing bacteri-
agation. The other two strains progressively decreased to less than um (De Vuyst and Vandamme, 1994). BLIS are antimicrobial molecules,
5 log CFU/g (Fig. 2b) (Minervini et al., 2010). sharing some characteristics with bacteriocins, but not purified at
Overall, these two studies showed that, during daily propagation, homogeneity (Tagg, 1991). The first screening studies about antimicro-
intrinsic robustness of strains seems to play a key-role in the establish- bial activity of sourdough LAB trace back to the early '90. Bavaricin A,
ment of the peculiar microbiota of a given sourdough (Minervini et al., synthesized by Lactobacillus sakei, was characterized (Larsen et al.,
2010; Siragusa et al., 2009). The ability of some LAB and yeasts to 1993). Subsequently, BLIS C57, synthesized by L. sanfranciscensis C57
adapt to many different substrates underlies their intrinsic robustness (Corsetti et al., 1996) and plantaricin ST31, synthesized by L. plantarum
(Vogelmann et al., 2009). In particular, robustness of L. plantarum ST31 (Todorov et al., 1999) were described. Production of BLIS has
strains is attributed to several factors: ability to use the main flour car- been also reported for several L. plantarum, Lactobacillus pentosus and
bohydrates, as well as low concentration carbohydrates (e.g., pentoses) L. rossiae strains (Corsetti et al., 2004). Several studies report about
and alternative energy sources (sugar nucleotides, amino acids, etc…), LAB, such as L. amylovorus DCE 471 (Messens and De Vuyst, 2002) and
adaptation to environmental stresses, rapid acidification, synthesis of Lc. lactis M30 (Hartnett et al., 2002), isolated from sourdough, that
 F. Minervini et al. / International Journal of Food Microbiology 171 (2014) 136–146 143

showed interesting bacteriocinogenic properties during sourdough along with stable species, and largely differed between artisan
fermentation (Gänzle and Gobbetti, 2013). bakery and laboratory levels. On the other hand, some strains
Reutericyclin is a low molecular weight antibiotic, synthesized by (e.g., L. sanfranciscensis s1 in sourdough MT.A used for Pane di Matera
some strains of L. reuteri, which is active against a broad range of PGI and L. plantarum s15 in sourdough AM.B used for Pane di Altamura
Gram-positive bacteria (including some LAB), in concentrations of less PDO) showed intrinsic robustness, because they were among the domi-
than 1 mg/l (Gänzle et al., 2000). The reutericyclin-producing L. reuteri nant LAB regardless of the environment of propagation (Tables 2 and 4).
LTH2584 was isolated in 1988 from SER, an in house rye sourdough pre- Permutation analysis based on bacterial diversity, assessed through
pared for the production of a commercially available baking aid (Böcker culture-dependent and -independent methods, showed that in five
et al., 1995). Relevant cell counts of L. reuteri were observed over out of seven cases, sourdoughs propagated at artisan bakery and
a 10 year-long monitoring of the SER sourdough microbiota (Gänzle those propagated in the laboratory diverged. This may be explained
and Vogel, 2003). Synthesis of reutericyclin in dough inhibited the probably by incomplete control of relevant factors and by the influ-
growth of a reutericyclin-sensitive indicator strain of L. sanfranciscensis ence of house microbiota, whose level of contamination is supposed
(Gänzle and Vogel, 2003). Several other reutericyclin-producing strains to be much higher in the bakery than in the laboratory (Minervini
of L. reuteri, isolated in high cell counts from SER sourdough, showed a et al., 2012b).
two- to four-fold higher tolerance to reutericyclin than L. sanfranciscensis Regarding yeasts, differences among artisan- and laboratory-
strains and most other lactobacilli (Gänzle et al., 2000). These findings propagated sourdoughs confirmed the importance of the environment
highlight that production of reutericyclin provides a competitive advan- of propagation as primary source of S. cerevisiae (Vrancken et al.,
tage to the producer strains, thus contributing, along with other factors 2010). This could be related to the frequent use of baker's yeast,
(e.g., adaptation to available substrates, minimal pH and temperature re- which may become a common contaminant of the environment of
quired for growth), to the stable persistence of L. reuteri in the SER sour- propagation (De Vuyst and Neysens, 2005). However, the bakery envi-
dough over a period of 10 years (Gänzle and Vogel, 2003). ronment may also affect bacterial diversity and stability of traditional
sourdough. In order to test this hypothesis, flours, sourdoughs and
6.5. Existence of a stable house microbiota their environment of propagation were sampled from two artisan
Belgian bakeries (D01 and D10) (Scheirlinck et al., 2009). The
Previous introduction of flour into the bakery environment helps to sourdoughs produced at bakery D10 were mainly dominated by
build up a house microbiota that may serve as an important inoculum L. sanfranciscensis, which was detected in the air of the storage and
for subsequent sourdough fermentations (De Vuyst et al., 2009). In work rooms, as well as on benches and dough mixer. This finding dem-
order to highlight the influence of the environment of propagation on onstrates that this species circulates throughout the environment of
the diversity of the lactic acid bacterium and yeast microbiotas, seven propagation. Conversely, sourdough samples produced in bakery D01
traditional sourdoughs were daily back-slopped for 80 days at an arti- were characterized by a higher bacterial diversity (L. paralimentarius,
san bakery or in the laboratory, using the same batch of flour and apply- L. plantarum and Lactobacillus spicheri). Both L. plantarum and L. spicheri
ing the same technological parameters (Minervini et al., 2012b). While were detected on the hands of the baker, but of these two species only
cell density of LAB and related biochemical features (pH, total titratable L. plantarum was also detected in the flour and in the air of the bakery.
acidity and concentration of organic acids) were not affected by the These results indicate the importance of air as a potential carrier of
environment of propagation, the number of yeasts and the concentra- LAB in food-processing environments (Scheirlinck et al., 2009). Further-
tion of ethanol markedly decreased from artisan bakery to laboratory more, L. plantarum, L. spicheri and L. sanfranciscensis isolates coming
propagation. Denaturing Gradient Gel Electrophoresis showed that the from sourdough and bakery environment samples were genetically in-
DNA band corresponding to S. cerevisiae was no more detectable in distinguishable. This suggested that the sourdoughs and their corre-
four out of seven sourdoughs propagated in the laboratory. Twelve sponding environments of propagation were dominated by a single
species of LAB were variously identified through a culture-dependent strain of these species. Despite the use of different flour batches and
approach. All sourdoughs harboured a certain number of species and possible variations in flour characteristics during subsequent propaga-
strains, which were dominant throughout time and, in several cases, tion of the sourdoughs analysed, those strains appeared to persist in
varied depending on the environment of propagation (Tables 2–4). the doughs over at least 3 years of sampling. This persistence may be
Overall, L. plantarum, L. sakei and W. cibaria dominated some sour- the result of the continuous use of the same fermentation parameters
doughs propagated at artisan bakeries, whereas Ln. citreum seemed to and of significant contamination from the environment of propagation
be more persistent under laboratory conditions. Other LAB (Lactobacillus (Scheirlinck et al., 2009). Overall, the results of this study suggest that
curvatus, Lc. lactis and P. pentosaceus) were only temporarily revealed, bakery environment, because of its usually high level of microbial

Table 2
Species and strains of lactic acid bacteria isolated from traditional sourdough MT.A propagated at artisan bakery and laboratory levels for 1, 20, 40, 60, and 80 days. The dot indicates the
presence of strains. Data were obtained from Minervini et al. (2012b).

Bakery-day Bakery-day Bakery-day Bakery-day Bakery-day Laboratory-day Laboratory-day Laboratory-day Laboratory-day

1 20 40 60 80 20 40 60 80

Lactobacillus parabrevis s1 ●
Lactobacillus plantarum s1 ● ●
Lactobacillus plantarum s2 ● ●
Lactobacillus plantarum s3 ● ● ● ● ●
Lactobacillus plantarum s4 ●
Lactobacillus plantarum s5 ●
Lactobacillus sakei s1 ●
Lactobacillus sanfranciscensis s1 ● ● ● ● ● ● ● ● ●
Lactobacillus sanfranciscensis s2 ● ●
Lactobacillus sanfranciscensis s3 ● ●
Leuconostoc citreum s1 ●
Pediococcus pentosaceus s1 ●
144 F. Minervini et al. / International Journal of Food Microbiology 171 (2014) 136–146

Table 3
Species and strains of lactic acid bacteria isolated from traditional sourdough MT.C propagated at artisan bakery and laboratory levels for 1, 20, 40, 60, and 80 days. The dot indicates the
presence of strains. Data were obtained from Minervini et al. (2012b).

Bakery-day Bakery-day Bakery-day Bakery-day Bakery-day Laboratory-day Laboratory-day Laboratory-day Laboratory-day

1 20 40 60 80 20 40 60 80

Lactobacillus brevis s1 ●
Lactobacillus curvatus s1 ●
Lactobacillus plantarum s7 ● ● ●
Lactobacillus plantarum s8 ● ● ●
Lactobacillus plantarum s9 ● ● ● ● ●
Lactobacillus sanfranciscensis s4 ●
Leuconostoc citreum s4 ● ● ● ● ●

Table 4
Species and strains of lactic acid bacteria isolated from traditional sourdough AM.B propagated at artisan bakery and laboratory levels for 1, 20, 40, 60, and 80 days. The dot indicates the
presence of strains. Data were obtained from Minervini et al. (2012b).

Bakery-day Bakery-day Bakery-day Bakery-day Bakery-day Laboratory-day Laboratory-day Laboratory-day Laboratory-day

1 20 40 60 80 20 40 60 80

Lactobacillus casei s9 ●
Lactobacillus plantarum s15 ● ● ● ● ● ● ● ● ●
Lactobacillus plantarum s16 ●
Lactobacillus sanfranciscensis s9 ●
Lactobacillus sanfranciscensis s10 ●
Lactobacillus sanfranciscensis s11 ● ● ●
Lactobacillus sanfranciscensis s12 ●
Lactococcus lactis ssp. lactis s3 ●

contamination, may be the source not only of yeasts, but also of LAB One of the main issues of the bakers is that the variations of the
that, by virtue of their intrinsic capacities, may or may not dominate sourdough performances cannot be eliminated but, at most, can be at-
traditional sourdough. tenuated. Although this variability may be regarded as a trait guarantee-
ing the artisanal, irreproducible quality of sourdough-based leavened
baked goods, bakery industry obviously considers that as a weakness.
7. Conclusions Therefore, further studies have to be performed in order to highlight
hidden mechanisms underlying the microbial structure and stability of
The microbial ecology of traditional sourdough is currently a matter sourdough. The comprehension of such mechanisms would be helpful
worthy of further investigation. This is mainly due to the complexity of to assess the most appropriate conditions that allow preserving the typi-
microbiota, the influence of flour and environment of propagation. Sodi- cal traits of traditional sourdough over time.
um chloride and quality of water could affect the sourdough microbial
community, and this could be one of the targets for future research.
Overall, traditional sourdoughs are characterized by relative microbial References
stability, provided that their propagation is carried out applying
Alexandre, H., Guilloux-Benatier, M., 2006. Yeast autolysis in sparkling wine—a review.
constant technological parameters and in their usual environment. Aust. J. Grape Wine Res. 12, 119–127.
However, differences in the composition of flour nutrients, as well as Arendt, E.K., Ryan, L.A.M., Dal Bello, F., 2007. Impact of sourdough on the texture of bread.
possible bacteriophage infections (Foschino et al., 2005), may result in Food Microbiol. 24, 165–174.
Arendt, E.K., Moroni, A., Zannini, E., 2011. Medical nutrition therapy: use of sourdough
disturbance of the microbial community. Such modifications can be lim- lactic acid bacteria as a cell factory for delivering functional biomolecules and food
ited at the level of strains, i.e. a given sourdough stably harbours one or ingredients in gluten free bread. Microb. Cell Fact. 10, S15.
more microbial species, but different strains alternate (Minervini et al., Berg, R.W., Sandine, W.E., Anderson, A.V., 1981. Identification of a growth stimulant for
Lactobacillus sanfrancisco. Appl. Environ. Microbiol. 42, 786–788.
2012b). The reasons for this succession are not very clear and are Black, P.N., Di Russo, C.C., 2006. Yeast acyl-CoA synthetases at the cross roads of fatty acids
probably hidden in the myriad of sourdough ecological niches, whose metabolism and regulation. Biochim. Biophys. Acta 1771, 286–298.
dynamics are difficult to understand (De Vuyst et al., 2009). Modifica- Böcker, G., Vogel, R.F., Hammes, W.P., 1990. Lactobacillus sanfrancisco als stabiles Element
in einem Reinzucht–Sauerteig–Präparat. Getreide, Mehl und Brot 44, 269–274.
tions of the typical sourdough microbiota may occur even at the level
Böcker, G., Stolz, P., Hammes, W.P., 1995. Neue Erkenntnisse zum Ökosystem Sauerteig
of species. Indeed, it has to be taken into account that when sourdough und zur Physiologie des sauerteigtypischen Stämme Lactobacillus sanfrancisco und
is propagated, not only dominant but also sub-dominant microorgan- Lactobacillus pontis. Getreide, Mehl und Brot 49, 370–374.
isms (e.g., Enterococcus sp., Pediococcus sp.) are perpetuated. Therefore, Brandt, M.J., 2007. Sourdough products for convenient use in baking. Food Microbiol. 24,
161–164.
any disturbance against dominant LAB may allow subdominant to Brandt, M.J., Hammes, W.P., Gänzle, M.G., 2004. Effects of process parameters on growth
become prevailing (Corsetti et al., 2007b). and metabolism of Lactobacillus sanfranciscensis and Candida humilis during rye sour-
Succession of strains and species during sourdough propagation dough fermentation. Eur. Food Res. Technol. 218, 333–338.
Brummer, J.M., Lorenz, K., 1991. European developments in wheat sourdoughs. Cereal
could affect the technological performances of sourdough and this, in Foods World 36, 310–314.
turn, would influence the characteristics of the product. Furthermore, Chavan, R.S., Chavan, S.R., 2011. Sourdough technology-a traditional way for wholesome
during final fermentation, contribution of the sourdough microbiota to foods: a review. Compr. Rev. Food Sci. Food Saf. 10, 170–183.
Collar, C., 1996. Biochemical and technological assessment of the metabolism of pure and
the product quality is presumably affected by the use of ingredients, mixed cultures of yeast and lactic acid bacteria in breadmaking applications. Food Sci.
such as flour, water, (eventually) baker's yeast, and (in the case of Technol. Int. 2, 349–367.
sweet products such as Panettone) sucrose, eggs, and butter. For Corsetti, A., Gobbetti, M., Smacchi, E., 1996. Antibacterial activity of sourdough lactic acid
bacteria: isolation of a bacteriocin-like inhibitory substance from Lactobacillus
instance, if baker's yeast is used, the contribution of the sourdough
sanfrancisco C57. Food Microbiol. 13, 447–456.
yeasts to the leavening of the dough will probably be restricted. There- Corsetti, A., Lavermicocca, P., Morea, M., Baruzzi, F., Tosti, N., Gobbetti, M., 2001. Pheno-
fore, further studies about that matter are warranted. typic and molecular identification and clustering of lactic acid bacteria and yeasts
 F. Minervini et al. / International Journal of Food Microbiology 171 (2014) 136–146 145

from wheat (species Triticum durum and Triticum aestivum) sourdoughs of Southern Guerzoni, M.E., Vernocchi, P., Ndagijimana, M., Gianotti, A., Lanciotti, R., 2007. Generation
Italy. Int. J. Food Microbiol. 64, 95–104. of aroma compounds in sourdough: effects of stress exposure and lactobacilli–yeasts
Corsetti, A., Settanni, L., Van Sinderen, D., 2004. Characterization of bacteriocin-like inhib- interactions. Food Microbiol. 24, 139–148.
itory substances (BLIS) from sourdough lactic acid bacteria and evaluation of their Hammes, W.P., Gänzle, M.G., 1998. Sourdough breads and related products, In: Wood,
in vitro and in situ activity. J. Appl. Microbiol. 96, 521–534. B.J.B. (Ed.), Microbiology of Fermented Foods, 2nd ed. Blackie Academic and Profes-
Corsetti, A., Settanni, L., Lopez, C.C., Felis, G.E., Mastrangelo, M., Suzzi, G., 2007a. A taxo- sional, an imprint of Thomson Science, London, UK, pp. 199–216.
nomic survey of lactic acid bacteria isolated from wheat (Triticum durum) kernels Hammes, W.P., Stolz, P., Gänzle, M., 1996. Metabolism of lactobacilli in traditional sour-
and non-conventional flours. Syst. Appl. Microbiol. 30, 561–571. doughs. Adv. Food Sci. 18, 176–184.
Corsetti, A., Settanni, L., Valmorri, S., Mastrangelo, M., Suzzi, G., 2007b. Identification of Hammes, W.P., Brandt, M.J., Francis, K.L., Rosenheim, M., Seitter, F.H., Vogelmann, S., 2005.
subdominant sourdough lactic acid bacteria and their evolution during laboratory- Microbial ecology of cereal fermentations. Trends Food Sci. Technol. 16, 4–11.
scale fermentations. Food Microbiol. 24, 592–600. Hansen, A., Schieberle, P., 2005. Generation of aroma compounds during sourdough
De Angelis, M., Gobbetti, M., 1999. Lactobacillus sanfranciscensis CB1: manganese, fermentation: applied and fundamental aspects. Trends Food Sci. Technol. 16, 85–94.
oxygen, superoxide dismutase and metabolism. Appl. Microbiol. Biotechnol. 51, Hartnett, D.J., Vaughan, A., van Sinderen, D., 2002. Antimicrobial-producing lactic acid
358–363. bacteria isolated from raw barley and sorghum. J. Inst. Brew. 108, 169–177.
De Angelis, M., Gobbetti, M., 2004. Environmental stress responses in Lactobacillus: a Hörmann, S., Scheyhing, C., Behr, J., Pavlovic, M., Ehrmann, M., Vogel, R.F., 2006. Compar-
review. Proteomics 4, 106–122. ative proteome approach to characterize the high-pressure stress response of
De Angelis, M., Bini, L., Pallini, V., Cocconcelli, P.S., Gobbetti, M., 2001. The acid–stress Lactobacillus sanfranciscensis DSM 20451T. Proteomics 6, 1878–1885.
response in Lactobacillus sanfranciscensis CB1. Microbiology 147, 1863–1873. Humblot, C., Guyot, J.P., 2009. Pyrosequencing of tagged 16S rRNA gene amplicons for
De Angelis, M., Mariotti, L., Rossi, J., Servili, M., Fox, P.F., Rollan, G., Gobbetti, M., 2002. rapid deciphering of the microbiomes of fermented foods such as pearl millet slurries.
Arginine catabolism by sourdough lactic acid bacteria: purification and characteriza- Appl. Environ. Microbiol. 75, 4354–4361.
tion of the arginine deiminase pathway enzymes from Lactobacillus sanfranciscensis Huys, G., Daniel, H.-M., De Vuyst, L., 2013. Taxonomy and diversity of sourdough yeasts
CB1. Appl. Environ. Microbiol. 68, 6193–6201. and lactic acid bacteria. In: Gobbetti, M., Gänzle, M.G. (Eds.), Handbook of sourdough
De Vuyst, L., Neysens, P., 2005. Biodiversity of sourdough lactic acid bacteria. Trends Food biotechnology. Springer Science + Business Media, New York, NY, pp. 105–154.
Sci. Technol. 16, 43–56. Iacumin, L., Cecchini, F., Manzano, M., Osualdini, M., Boscolo, D., Orlic, S., Comi, G., 2009.
De Vuyst, L., Vandamme, E.J., 1994. Antimicrobial potential of lactic acid bacteria. In: Description of the microflora of sourdoughs by culture-dependent and culture-
De Vuyst, L., Vandamme, E.J. (Eds.), Bacteriocins of Lactic Acid Bacteria. Blackie independent methods. Food Microbiol. 26, 128–135.
Academic and Professional, London, UK, pp. 91–142. Jeong, S.H., Jung, J.Y., Lee, S.H., Jin, H.M., Jeon, C.O., 2013. Microbial succession and metab-
De Vuyst, L., Vrancken, G., Ravyts, F., Rimaux, T., Weckx, S., 2009. Biodiversity, ecological olite changes during fermentation of dongchimi, traditional Korean watery kimchi.
determinants, and metabolic exploitation of sourdough microbiota. Food Microbiol. Int. J. Food Microbiol. 164, 46–53.
26, 666–675. Jung, J.Y., Lee, S.H., Lee, H.J., Jeon, C.O., 2013. Microbial succession and metabolite changes
Decock, P., Cappelle, S., 2005. Bread technology and sourdough technology. Trends Food during fermentation of saeu-jeot: traditional Korean salted seafood. Food Microbiol.
Sci. Technol. 16, 113–120. 34, 360–368.
Di Cagno, R., De Angelis, M., Calasso, M., Vicentini, O., Vernocchi, P., Ndagijimana, M., De Keller, L., Surette, M., 2006. Communication in bacteria: an ecological and evolutionary
Vincenzi, M., Dessì, M.R., Guerzoni, M.E., Gobbetti, M., 2010. Quorum sensing in sour- perspective. Nat. Rev. Microbiol. 4, 249–258.
dough Lactobacillus plantarum DC400: induction of plantaricin A (PlnA) under co- Kline, L., Sugihara, T.F., 1971. Microorganisms of the San Francisco sourdough bread pro-
cultivation with other lactic acid bacteria and effect of PlnA on bacterial nd Caco-2 cess. II. Isolation and characterization of undescribed bacterial species responsible for
cells. Proteomics 10, 2175–2190. the souring activity. Appl. Microbiol. 21, 459–465.
Ercolini, D., Pontonio, E., De Filippis, F., Minervini, F., La Storia, A., Gobbetti, M., Di Cagno, Konings, W.N., 2002. The cell membrane and the struggle for life of lactic acid bacteria.
R., 2013. Microbial ecology dynamics during rye and wheat sourdough fermentation. Antonie Van Leeuwenhoek 82, 3–27.
Appl. Environ. Microbiol. 79, 7827–7836. Kulp, K., 2003. Baker's yeast and sourdough technologies in the production of U.S. bread
Fleet, G.H., 2003. Yeast interactions and wine flavour. Int. J. Food Microbiol. 86, 11–22. products. In: Kulp, K., Klaus, L. (Eds.), Handbook of dough fermentations. Marcel Dekker,
Foschino, R., Terraneo, R., Mora, D., Galli, A., 1999. Microbial characterization of sour- New York, NY, pp. 97–142.
doughs for sweet baked products. Ital. J. Food Sci. 11, 19–28. Larsen, A.G., Vogensen, F.K., Josephsen, J., 1993. Antimicrobial activity of lactic acid bacteria
Foschino, R., Venturelli, E., Picozzi, C., 2005. Isolation and characterization of a virulent isolated from sourdough: purification and characterization of bavaricin A, a bacteriocin
Lactobacillus sanfranciscensis bacteriophage and its impact on microbial population produced by Lactobacillus bavaricus MI401. J. Appl. Bacteriol. 75, 113–122.
in sourdough. Curr. Microbiol. 51, 413–418. Lattanzi, A., Minervini, F., Di Cagno, R., Diviccaro, A., Antonielli, L., Cardinali, G., Cappelle, S.,
Foster, J.W., Hall, H.K., 1991. Inducible pH homeostasis and the acid tolerance response of De Angelis, M., Gobbetti, M., 2013. The lactic acid bacteria and yeast microbiota of
Salmonella typhimurium. J. Bacteriol. 173, 5129–5135. eighteen sourdoughs used for the manufacture of traditional Italian sweet leavened
Gänzle, M.G., Gobbetti, M., 2013. Physiology and biochemistry of sourdough lactic acid baked goods. Int. J. Food Microbiol. 163, 71–79.
bacteria. In: Gobbetti, M., Gänzle, M.G. (Eds.), Handbook of sourdough biotechnology. Messens, W., De Vuyst, L., 2002. Inhibitory substance produced by Lactobacilli isolated
Springer Science + Business Media, New York, NY, pp. 183–216. from sourdough—a review. Int. J. Food Microbiol. 72, 31–43.
Gänzle, M.G., Vogel, R.F., 2003. Contribution of reutericyclin production to the stable per- Minervini, F., De Angelis, M., Di Cagno, R., Pinto, D., Siragusa, S., Rizzello, C.G., Gobbetti, M.,
sistence of Lactobacillus reuteri in an industrial sourdough fermentation. Int. J. Food 2010. Robustness of Lactobacillus plantarum starters during daily propagation of
Microbiol. 80, 31–45. wheat flour sourdough type I. Food Microbiol. 27, 897–908.
Gänzle, M.G., Ehman, M., Hammes, W.P., 1998. Modeling of growth of Lactobacillus Minervini, F., Di Cagno, R., Lattanzi, A., De Angelis, M., Antonielli, L., Cardinali, G., Cappelle,
sanfranciscensis and Candida milleri in response to process parameters of sourdough S., Gobbetti, M., 2012a. Lactic acid bacterium and yeast microbiotas of 19 sourdoughs
fermentation. Appl. Environ. Microbiol. 64, 2616–2623. used for traditional/typical Italian breads: interactions between ingredients and
Gänzle, M.G., Höltzel, A., Walter, J., Jung, G., Hammes, W.P., 2000. Characterization of microbial species diversity. Appl. Environ. Microbiol. 78, 1251–1264.
reutericyclin produced by Lactobacillus reuteri LTH2584. Appl. Environ. Microbiol. Minervini, F., Lattanzi, A., De Angelis, M., Di Cagno, R., Gobbetti, M., 2012b. Influence of
60, 4325–4333. artisan bakery- or laboratory-propagated sourdoughs on the diversity of lactic acid
Gänzle, M.G., Vermeulen, N., Vogel, R.F., 2007. Carbohydrate, peptide and lipid metabo- bacterium and yeast microbiotas. Appl. Environ. Microbiol. 78, 5328–5340.
lism of lactic acid bacteria in sourdough. Food Microbiol. 24, 128–138. Moroni, A.V., Dal Bello, F., Arendt, E.K., 2009. Sourdough in gluten-free bread-making: an
Garofalo, C., Silvestri, G., Aquilanti, L., Clementi, F., 2008. PCR-DGGE analysis of lactic acid ancient technology to solve a novel issue? Food Microbiol. 26, 676–684.
bacteria and yeast dynamics during the production processes of three varieties of Ndagijimana, M., Vallicelli, M., Cocconcelli, P.S., Cappa, F., Patrignani, F., Lanciotti, R.,
Panettone. J. Appl. Microbiol. 105, 243–254. Guerzoni, M.E., 2006. Two 2 (5H)-furanones as possible signalling molecules in
Gobbetti, M., 1998. Interactions between lactic acid bacteria and yeasts in sourdoughs. Lactobacillus helveticus. Appl. Environ. Microbiol. 72, 6053–6061.
Trends Food Sci. Technol. 9, 267–274. Neubauer, H., Glaasker, E., Hammes, W.P., Poolman, B., Konings, W.N., 1994. Mechanism
Gobbetti, M., Corsetti, A., 1996. Co-metabolism of citrate and maltose by Lactobacillus of maltose uptake and glucose excretion in Lactobacillus sanfrancisco. J. Bacteriol.
brevis subsp. lindneri citrate-negative strain: effect on growth, end-products and 176, 3007–3012.
sourdough fermentation. Z. Lebensm. Unters. Forsch. 203, 82–87. Olsen, A., Halm, M., Jakobsen, M., 1995. The antimicrobial activity of lactic acid bacteria
Gobbetti, M., Corsetti, A., Rossi, J., 1994a. The sourdough microflora. Interactions between from fermented maize (kenkey) and their interaction during fermentation. J. Appl.
lactic acid bacteria and yeasts: metabolism of amino acids. World J. Microbiol. Bacteriol. 79, 506–512.
Biotechnol. 10, 275–279. Onno, B., Roussel, P., 1994. Technologie et microbiologie de la panification au levain. In:
Gobbetti, M., Simonetti, M.S., Rossi, J., Cossignani, L., Corsetti, A., Damiani, P., 1994b. Free De Roissart, H., Luquet, F.M. (Eds.), Bactèries Lactiques. Lorica, 11. Uriage, France,
D- and L-amino acid evolution during sourdough fermentation and baking. J. Food Sci. pp. 293–320.
59, 881–884. Ottogalli, G., Galli, A., Foschino, R., 1996. Italian bakery products obtained with sour
Gobbetti, M., Corsetti, A., Rossi, J., 1995. Maltose-fructose co-fermentation by Lactobacillus dough: characterization of the typical microflora. Adv. Food Sci. 18, 131–144.
brevis subsp. lindneri CB1 fructose-negative strain. Appl. Microbiol. Biotechnol. 42, Poutanen, K., Flander, L., Katina, K., 2009. Sourdough and cereal fermentation in a nutri-
939–944. tional perspective. Food Microbiol. 26, 693–699.
Gobbetti, M., De Angelis, M., Corsetti, A., Di Cagno, R., 2005. Biochemistry and physiology Rocha, J.M., Malcata, F.X., 1999. On the microbial profile of traditional Portuguese sour-
of sourdough lactic acid bacteria. Trends Food Sci. Technol. 16, 57–69. dough. J. Food Prot. 62, 1416–1429.
Gobbetti, M., Rizzello, C.G., Di Cagno, R., De Angelis, M., 2013. How the sourdough Rocha, J.M., Malcata, F.X., 2012. Microbiological profile of maize and rye flours, and sour-
may affect the functional features of leavened baked goods. Food Microbiology dough used for the manufacture of traditional Portuguese bread. Food Microbiol. 31,
37, 30–40. 72–88.
Graumann, P., Marahiel, M.A., 1998. A superfamily of proteins that contain the cold–shock Rollan, G., Lorca, G.L., Font de Valdez, G., 2003. Arginine catabolism and acid tolerance re-
domain. Trends Biochem. Sci. 23, 286–290. sponse in Lactobacillus reuteri isolated from sourdough. Food Microbiol. 20, 313–349.
146 F. Minervini et al. / International Journal of Food Microbiology 171 (2014) 136–146

Scheirlinck, I., Van der Meulen, R., Van Schoor, A., Vancanneyt, M., De Vuyst, L., Vermeulen, N., Gänzle, M.G., Vogel, R.F., 2007. Glutamine deamidation by cereal-
Vandamme, P., Huys, G., 2008. Taxonomic structure and stability of the bacterial com- associated lactic acid bacteria. J. Appl. Microbiol. 103, 1197–1205.
munity in Belgian sourdough ecosystems assessed by culturing and population Verstrepen, K.J., Van Laere, S.D.M., Vanderhaegen, B.M.P., Derolelinckx, G., Dufour,
fingerprinting. Appl. Environ. Microbiol. 74, 2414–2423. J.-P., Pretorius, I.S., Winderickx, J., Thevelein, J.M., Delvaux, F.R., 2003. Expression
Scheirlinck, I., Van der Meulen, R., De Vuyst, L., Vandamme, P., Huys, G., 2009. Molecular levels of the yeast alcohol acetyltransferase genes ATF1, Lg-ATF1, and ATF2 control
source tracking of predominant lactic acid bacteria in traditional Belgian sourdoughs the formation of a broad range of volatile esters. Appl. Environ. Microbiol. 69,
and their production environments. J. Appl. Microbiol. 106, 1081–1092. 5228–5237.
Settanni, L., Massitti, O., Van Sinderen, D., Corsetti, A., 2005. In situ activity of a bacteriocin- Viljoen, B.C., 2006. Yeast ecological interactions. Yeast–yeast, yeast–bacteria, yeast–fungi
producing Lactococcus lactis strain. Influence on the interactions between lactic acid interactions and yeasts as biocontrol agents. In: Querol, A., Fleet, G.H. (Eds.), Yeasts
bacteria during sourdough fermentation. J. Appl. Microbiol. 99, 670–681. in Food and Beverages. Springer-Verlag, Berlin Heidelberg, pp. 83–110.
Simonson, L., Salovaara, H., Korhola, M., 2003. Response of wheat sourdough parameters Vogel, R.F., Knorr, R., Müller, M.R.A., Steudel, U., Gänzle, M.G., Ehrmann, M., 1999. Non
to temperature, NaCl and sucrose variations. Food Microbiol. 20, 193–199. dairy lactic fermentations: the cereal world. Antonie Van Leeuwenhoek 76, 403–411.
Siragusa, S., Di Cagno, R., Ercolini, D., Minervini, F., Gobbetti, M., De Angelis, M., 2009. Tax- Vogel, R.F., Pavlovic, M., Ehrmann, M.A., Wiezer, A., Liesegang, H., Offschanka, S., Voget, S.,
onomic structure and monitoring of the dominant population of lactic acid bacteria Angelov, A., Böcker, G., Liebl, W., 2011. Genomic analysis reveals Lactobacillus
during wheat flour sourdough type I propagation using Lactobacillus sanfranciscensis sanfranciscensis as stable element in traditional sourdoughs. Microb. Cell Fact. 10, S6.
starters. Appl. Environ. Microbiol. 75, 1099–1109. Vogelmann, S.A., Hertel, C., 2011. Impact of ecological factors on the stability of microbial
Spicher, G., Stephan, H., 1999. Handbuch Sauerteig: Biologie, Biochemie, Technologie 5th associations in sourdough fermentation. Food Microbiol. 28, 583–589.
ed. Behr's Verlag, Hamburg, Germany. Vogelmann, S.A., Seitter, M., Singer, U., Brandt, M.J., Hertel, C., 2009. Adaptability
Stolz, P., 1999. Mikrobiologie des Sauerteiges, In: Spicher, G., Stephan, H. (Eds.), Handbuch of lactic acid bacteria and yeasts to sourdoughs prepared from cereals,
Sauerteig: Biologie, Biochemie, Technologie, 5th ed. Behr's Verlag, Hamburg, Germany, pseudocereals and cassava and use of competitive strains as starters. Int. J. Food
pp. 35–60. Microbiol. 130, 205–212.
Stolz, P., Böcker, G., Vogel, R.F., Hammes, W.P., 1993. Utilisation of maltose and glucose by Vrancken, G., Rimaux, T., Wouters, D., Leroy, F., De Vuyst, L., 2009. The arginine deiminase
lactobacilli isolated from sourdough. FEMS Microbiol. Lett. 109, 237–242. pathway of Lactobacillus fermentum IMDO 130101 responds to growth under stress
Stolz, P., Böcker, G., Hammes, W.P., Vogel, R.F., 1995. Utilization of electron acceptors by conditions of both temperature and salt. Food Microbiol. 26, 720–727.
lactobacilli isolated from sourdough. Z. Lebensm. Unters. Forsch. 201, 91–96. Vrancken, G., De Vuyst, L., Van der Meulen, R., Huys, G., Vandamme, P., Daniel, H.M., 2010.
Suihko, M.L., Makinen, V., 1984. Tolerance of acetate, propionate and sorbate by Yeast species composition differs between artisan bakery and spontaneous laborato-
Saccharomyces cerevisiae and Torulopsis holmii. Food Microbiol. 1, 105–110. ry sourdoughs. FEMS Yeast Res. 10, 471–481.
Tagg, J.R., 1991. Bacterial BLIS. ASM News 57, 611. Vrancken, G., Rimaux, T., Weckx, S., Leroy, F., De Vuyst, L., 2011. Influence of temperature
Thiele, C., Gänzle, M.G., Vogel, R.F., 2002. Contribution of sourdough lactobacilli, yeast, and and backslopping time on the microbiota of a type I propagated laboratory wheat
cereal enzymes to the generation of amino acids in dough relevant for bread flavor. sourdough fermentation. Appl. Environ. Microbiol. 77, 2716–2726.
Cereal Chem. 79, 45–51. Weckx, S., Van der Meulen, R., Allemeersch, J., Huys, G., Vandamme, P., Van Hummelen, P.,
Todorov, S., Onno, B., Sorokine, O., Chobert, J.M., Ivanova, I., Dousset, X., 1999. Detection De Vuyst, L., 2010a. Community dynamics of bacteria in sourdough fermentations as
and characterization of a novel antibacterial substance produced by Lactobacillus revealed by their metatranscriptome. Appl. Environ. Microbiol. 76, 5402–5408.
plantarum ST31 isolated from sourdough. Int. J. Food Microbiol. 48, 167–177. Weckx, S., Van der Meulen, R., Maes, D., Scheirlinck, I., Huys, G., Vandamme, P., De Vuyst,
Van der Meulen, R., Scheirlinck, I., Van Schoor, A., Huys, G., Vancanneyt, M., Vandamme, P., De L., 2010b. Lactic acid bacteria community dynamics and metabolite production of rye
Vuyst, L., 2007. Population dynamics and metabolite target analysis during laboratory sourdough fermentations share characteristics of wheat and spelt sourdough fermen-
fermentations of wheat and spelt sourdoughs. Appl. Environ. Microbiol. 73, 4741–4750. tations. Food Microbiol. 27, 1000–1008.
Velasco, I., Tenreiro, S., Calderon, I.L., André, B., 2004. Saccharomyces cerevisiae Aqr1 is an Yao, G., Li, H., Gao, P., Bao, Y., Wang, X., Zhang, H., 2013. The study and application of
internal-membrane transporter involved in excretion of amino acids. Eukaryot. Cell 3, lactic acid bacteria in fermented sourdough. J. Chin. Inst. Food Sci. Technol. 13,
1492–1503. 163–170.
Venturi, M., Guerrini, S., Vincenzini, M., 2012. Stable and non-competitive association of Zotta, T., Piraino, P., Parente, E., 2008. Characterization of lactic acid bacteria isolated from
Saccharomyces cerevisiae, Candida milleri and Lactobacillus sanfranciscensis during sourdoughs for Cornetto, a traditional bread produced in Basilicata (Southern Italy).
manufacture of two traditional sourdough baked goods. Food Microbiol. 31, 107–115. World J. Microbiol. Biotechnol. 24, 1785–1795.