8 Sourdough breads and related products

W.P. HAMMES AND M.G. GANZLE

S.l Introduction

Grinding of cereals and addition of water results in the formation of a

dough which, after some time, will turn into a sourdough characterized by
acid taste, aroma and increased volume due to gas formation. This fermen-
tation event may have been one of the first microbial processes employed
by man and led to the use of sourdough for breadmaking. Baking of leav-
ened bread can be traced back to Egypt in 1500 Be, and the study of the
microbiology of sourdough has a history of nearly 100 years. Not all
sourdoughs are subjected to baking. More fluid soured doughs are con-
sumed in various parts of the world. Boza in Turkey and Mageu in Africa
are examples of a group of raw foods that had once also a tradition in
Europe. For example, in Scotland these were known as sawens or flum-
meries (Fenton, 1974). There are even smooth borderlines to beer-like
beverages. These are considered as products of alcoholic fermentation
performed by yeasts and require the digestion of the starch by amylases.
Without specific technological precautions, however, it will always be, as in
sourdough, a lactic acid bacteria (LAB)-yeast association that primarily
develops in the cereal-derived substrates. An example is provided by
'Berliner WeiBe' beer, which is characterized by a deliberate LAB-yeast
fermentation and by the strong acid taste of the beer.
The key role of LAB and yeasts in sourdough was recognized by the
pioneers in the study of sourdough microbiology such as Holliger (1902),
Beccard (1921) and Knudsen (1924). As sourdough is an intermediate but
not an end-product, the microbial activities in the dough have to be judged
on the basis of their impact on the quality of the baked goods that are
produced with its aid. These are characterized by their flavour, nutritional
value and texture, i.e. the size and distribution of pores and the elasticity of
the crumb. To some extent these characteristics can be achieved without the
involvement of LAB by application of yeast and/or chemicals, but the
traditional process and various new modifications rely on the metabolic
activities of LAB in the sourdough. By one definition (Anon., 1994)
sourdough is described as 'a dough whose microorganisms (e.g. LAB and
yeasts) originate from sourdough or a sourdough starter and are metabo-
lically active or can be reactivated. Upon addition of flour and water they
continue to produce acid'.

B. J. B. Wood (ed.), Microbiology of Fermented Foods

© Thomson Science 1998
200 MICROBIOLOGY OF FERMENTED FOODS

An overwhelming multitude of baked goods are produced with the aid of

sourdough. These include above all breads from wheat, rye and mixtures
thereof as well as the well-known Italian products such as Panettone,
Colomb a, Pandoro and different types of brioches. Another sourdough
application was described for production of soda crackers (Sugihara, 1985).
Taking into account that the major part of rye bread and a substantial share
of that made from wheat (30% in Italy; Ottogalli et al., 1996) is made from
sourdough, an impressive role of this fermentation product becomes
evident.

8.2 Microbial ecology of sourdough

Factors influencing microbial growth and activity in sourdough are shown in

Figure 8.1 (Hammes et al., 1997). The effective endogenous parameters are
determined by the cereal substrate which contains the carbohydrates, nitro-
gen (N)-sources, vitamins and minerals. The fermentable carbohydrates

Temperature
Dough yield (a.)
Oxygen (Redox potential)
Femlentation time
Number of propagation steps

Product quality:
flavour, texture, shelflife, nutritional value

Figure 8.1 Factors affecting growth and metabolic activity of the sourdough microfiora and the
quality of sourdough bread. (Modified according to Hammes & Vogel, 1997.)
 SOURDOUGH BREADS AND RELATED PRODUCTS 201

play an essential role in the fermentation process. Their concentrations in

wheat and rye flours are presented in Table 8.1 (Hammes & Vogel, 1997).
The addition of water activates the enzymes in the flour where-
by further fermentable carbohydrates are continuously released from the
polysaccharides. In model studies it was shown by Rocken & Voysey (1995)
that, under aseptic conditions and by application of chemical acidification,
the concentration of glucose, maltose and fructose accumulated to 2.5%,
1.8% and 0.5%, respectively. The activity of the hydrolytic enzymes de-
pends strongly on biological and technological factors and is controlled in
the bakery by factors such as pH, temperature or addition of sodium
chloride.
Exogenous parameters effective in sourdough are controlled by the pro-
cess condition, such as temperature, redox potential and water activity,
which is determined by the dough yield (i.e. dough weight x 100/fiour
weight) and/or by addition of sodium chloride. Finally, the fermentation
time and the use of a more or less defined inoculum exert further effects. The
use of sourdough starters is state-of-the-art in bakeries working at an indus-
trial scale, whereas artisan and household technology rely to some extent
(especially for wheat doughs) on indigenous processes performed by the
fortuitous microorganisms. These may originate from the cereal itself, from
contaminants of baker's yeast, or from the environment at milling or in the
bakery. The LAB frequently isolated from wheat and rye flour are listed in
Table 8.2. These species have also been reported to be present in sourdough
(Ottogalli et at., 1996). The numerous species provide the impression that a
more or less undefinable multitude of species perform the fermentation
process. However, this is only true for indigenous processes. When analysing
sourdoughs that had been propagated for some time, it becomes evident that

Table 8.1 Fermentable sugars and amino acids in wheat and

rye whole grain or whole grain flour

Compound Rye (% DM) Wheat (% DM)

Maltose 0.03' 0.04'

Glucose O.OSb Ob
-
, O.72g
Glucfrun
Gluj-fru3 0.26g
Glucfru2 0.23 b,0.4g
Sucrose 0.8 b, 1.3' 0.6 b, 0.63', 0.26 g
Fructose O.OSb 0.042b, 0.04g
Raffinose -
, O.17 b
Free sugars 2.5'
Free amino acids 18mmoVkg doughd 0.7 g/kg dough'.!

'Data not available; bSouci et al. (1994); 'Lorenz & Kulp

(1991); dSpicher & Nierle (1988); 'Collar et al. (1992); fGobetti
et al. (1994); gSaunders et al. (1972).
Modified from Hammes & Vogel (1997).
202 MICROBIOLOGY OF FERMENTED FOODS

Table 8.2 Lactic acid bacteria in wheat flour

Rods Cocci

Homofermentative Heterofermentative Homofermentative Heterofermentative

Lb. coryniformis Lb. cellobiosus Pd. pentosaceus Le. paramesenteroides
Lb. plantarum Lb. brevis Pd. parvulus
Lb. casei Lb. fermentum Pd. acidilactici
Lb. salivarius Ec. faecalis
Lb. curvatus Lc. lactis

Lb, Lactobacillus; Pd, Pediococcus; Ec, Enterococcus; Lc, Lactococcus; Le, Leuconostoc.
From Hammes & Vogel (1997).

a selection takes place which leads to the establishment of usually one or two
species at numbers three or four orders of magnitude above those of the
fortuitous flora (Hammes et ai. 1996; Hammes & Vogel, 1997).
Little is known about the implicit ecological factors in sourdough. Basi-
cally, all fermentation processes are endangered by bacteriophage attacks.
Bocker et al. (1990) investigated the microbiology of a commercial
sourdough starter preparation ('Reinzuchtsauer') which has been in use for
over 60 years. It is a continuously propagated sourdough which is distri-
buted in small portions to bakeries. Disturbance of the production process
was never observed and it was even possible to show that the two character-
istic strains of Lactobacillus sanfrancisco and one strain of L. pontis
were present in the preparation for at least 10 years. Thus, there was no
indication of an effect of bacteriophages in sourdough fermentation. On the
other hand, phage were isolated from a 'madre' that was lytic to L.
fermentum and L. brevis, respectively (Ottogalli et ai., 1996). The authors
reported that the presence of phage in the propagated sourdough can exert
deleterious effects on the fermentation process.
LAB strains present in sourdough may also produce bacteriocins which
might provide a selective advantage for the producer strain. Larsen et ai.
(1993) isolated a strain of L. bavaricus from an indigenous sourdough which
produced a bacteriocin. Ganzle et ai. (1995) described the formation of
bacteriocin by sourdough isolates allotted to L. sanfrancisco and L. reuteri.
The antimicrobial activity of the latter strain was studied in more detail and
it was revealed that the spectrum matched with competing organisms in the
dough. For example, the active compound reutericin 64 inhibited among
others strains of L. sanfrancisco, Bacillus cereus and B. subtilis. Inhibition of
B. subtilis is considered useful in sourdough, as strains of this species may
grow in bread and cause ropiness (Rocken & Spicher, 1993). Remarkably,
the bacteriocin of the above-described L. bavaricus matched in its activity
the antimicrobial spectra of meat-associated LAB such as L. sake or L.
curvatus (Tichaczek et ai., 1992). Bacteriocins from L. sanfrancisco were
also detected by Corsetti et ai. (1996). The role of bacteriocin in sourdough
under practical conditions remains to be determined.
 SOURDOUGH BREADS AND RELATED PRODUCTS 203

Based on the technology applied for their production, sourdoughs have

been grouped into three types by Bocker et al. (1995). Type I are traditional
doughs characterized by continuous (daily) propagation to keep the
microftora in an active state. A three-stage fermentation process is most
appropriate to achieve this condition. To ensure highest microbial activity it
was empirically found that each of the steps requires adjustments to a
specific dough yield, temperature (23-30°C) and fermentation time. The
final pH in the third-stage dough is 3.9 (Rocken & Voysey, 1995). Parts of
this sourdough are used as inocula that can be defined as multiple-strain
starter cultures (Hammes, 1991). These are known as 'Anstellgut' in Ger-
many, 'mother sponge' in the San Francisco area, 'Ie chef' in France, Omasa
madre' in Spain, and 'madre' or 'capolievito' in Italy. It has become evident
that in sourdoughs made from rye, wheat or mixtures thereof Lactobacillus
sanfrancisco is by far the most predominant LAB and L. pontis may
be found in some doughs. In addition to these organisms the hetero-
fermentative species L. !ructivorans, L. !ermentum and L. brevis were
found in some doughs in relevant numbers (Ottogalli et al., 1996). There-
fore, these species can be considered to be best adapted to traditional sour
doughs. Their phylogenetic relatedness to other species of importance (es-
pecially in sourdough) is shown in Figure 8.2. For unequivocal identification

L. sanfrancisco

L. alimentarius

L. delbrueckii

W.kandleri

Figure 8.2 Phylogenetic relationship of heterofermentative lactobacilli occurring in

sourdough. The bar indicates 5% estimated sequence divergence. (From Hammes & Vogel,
1997.)
204 MICROBIOLOGY OF FERMENTED FOODS

of sourdough, LAB 16S rRNA targeted gene probes have been developed
(Vogel et al., 1994). The sourdough LAB are associated with yeasts, the vast
majority of which have been allotted to the species Candida milleri, C.
holmii, Saccharomyces exiguus and S. cerevisiae. A great variety of addi-
tional yeast species have been isolated from sourdoughs. As described for
LAB, these are commonly contaminants originating from flour or the envi-
ronment. With the exception of S. cerevisiae, none of the species utilizes
maltose, which is the preferred carbohydrate fermented by L. sanfrancisco.
The tight association of specific LAB and yeasts is not restricted to
sourdoughs used in the western world, but finds parallels in tropical areas.
For example, Hamad et at. (1992) investigated the microbiology of
sourdoughs used in The Sudan for Kisra production. These are made from
sorghum flour, ferment at 30-40°C, and make use of a starter inoculum that
has been continuously propagated. These sourdoughs contained L.
fermentum, L. reuteri and L. amylovorus. Candida krusei was identified as
the corresponding yeast. The presence of L. amylovorus is remarkable, as
this species can hydrolyse starch.
The LAB occurring in type I doughs require for their effective cultivation
special media such as homohiochii (Kitahara et at., 1957) or sanfrancisco
medium (Bocker et al., 1990). The organisms have in common that they are
rather sensitive to preservation by drying. Therefore, dried starter cultures
containing strains of L. sanfrancisco have become commercially available
only recently. The organisms are also sensitive to low pH; thus, when
sourdough is kept at ambient temperature, acidification continues and these
LAB die off. Consequently, at some stage more acid-resistant species will
then dominate in the sourdough.
This description of an ecological change upon extended fermentation and
acidification leads on to and describes the nature of type II doughs. These
serve essentially to provide acidification and flavour to doughs. Leavening
is performed by baker's yeast that has to be added to the dough, in contrast
to traditional three-stage fermentation processes performed with type I
doughs. The need for yeast addition is also essential when less time-
consuming one-stage processes are employed which require a single fer-
mentation period of 15-20 hours. These are most common in German
bakeries and rely on the use of 'Anstellgut'-starters, which are either type I
doughs or other commercially available starter preparations. In these proc-
esses, however, gas formation by the LAB is strongly reduced. Type II
doughs can be produced in large volumes, are stored up to one week in
silos, and are taken in working-day portions for the production of
sourdough breads.
In these modern processes that have been modified with regard to tem-
perature (usually >30°C, to speed up the process), dough yield (suitable for
pumping), fermentation time or the nature of ingredients (bread can be
added to the sourdough), other LAB species than those of type I doughs are
more competitive and are employed in starter preparations. For example, a
 SOURDOUGH BREADS AND RELATED PRODUCTS 205

heterofermentative species was isolated from long-fermented sourdough by

Wiese et al. (1996) and was described as Lactobacillus panis sp. nov. Strains
of this species are components of a starter preparation used for a defined
sourdough fermentation process (Anon., 1990). Included in this type II
group of doughs are also 'in-house' cultures used for the production of
dried baking aids. In these doughs Bocker et al. (1995) have identified as
dominating components strains of L. reuteri, L. johnsonii, L. sanfrancisco
and L. pontis.
Type III doughs are dried preparations of sourdough. The LAB present
in these preparations are resistant to drying and survive in that status.
Bocker et at. (1995) have identified in type III doughs Pediococcus
pentosaceus, L. plantarum and L. brevis at numbers ranging from 107 to
109 cfu/g.
A fermentation by LAB was also found to take place in indirectly pro-
duced yeast doughs. LAB have been isolated from pre-doughs employed
for the production of soda crackers and baguettes. It has been shown that
LAB present as contaminants from either the yeast or the flour grow in the
doughs to high cell counts and contribute significantly to the quality of these
baked goods (Sugihara, 1985; Wiese et al., 1995).
Starter cultures for the sourdough fermentation in the traditional mean-
ing have been introduced as 'Anstellgut', etc. These can be compared with
mixed-strain cultures used in milk fermentations and are still in use. They
are sourdoughs per se and thus have been obtained under non-aseptic
conditions. As novel developments, single- and multiple-strain starter
preparations have been introduced into practice. An overview of commer-
cially available cultures was provided by Spicher & Stephan (1987). More
recent developments were reported by Budolfson-Hansen (1988), who de-
scribed dried preparations containing L. delbrueckii and L. brevis or L.
plantarum. Freeze-dried preparations containing L. sanfrancisco have re-
cently been developed. Finally, cultures containing L. plantarum, L. brevis
and L. fructivorans or L. brevis, L. pontis and S. cerevisiae are available (H.
Knauf, Fa. Muller, GieSen, personal communication).

8.3 Technological effects of sourdough lactic acid bacteria

The aims of sourdough fermentation for the production of breads made

from wheat or rye are listed in Table 8.3 (Hammes & Vogel, 1997). An
essential part of these aims is accomplished by reduction of the pH, does
not require the fermentative action of LAB, and can be achieved by addi-
tion to the dough of acetic, lactic, tartaric, phosphoric or citric acid. Some-
what different from wheat doughs are those doughs made from rye, which
depend on lowering the pH to achieve their suitability for baking. This
necessity results from the absence of gluten, which in wheat doughs pro-
vides the water-binding and gas-retaining properties. In rye, these functions
206 MICROBIOLOGY OF FERMENTED FOODS

Table 8.3 Effect of sourdough fermentation on properties of breads from rye and wheat
doughs

Effect Rye dough Wheat dough

Effects of low pH
Prerequisite to achieve suitability for baking +
Inhibition of amylase activity +
Improved swelling of pentosans +
Improved water-binding capacity of the dough + +
Improved nutritional value + +
Prevention of malfermentation and spoilage + +
Additional effects
Crumb elasticity and bread volume + +
Development of aroma and taste + +
Contribution to leavening +' +'
Delayed staling + +
'Effect depending on the sourdough process employed.
From Hammes & Vogel (in press).

are taken over by pentosans, whose solubility and swelling increases with
the decrease in pH and becomes optimal at pH 4.9. Furthermore, upon
acidification, the activity of rye amylase is inhibited, which is an important
effect, as the gelatinization of rye starch (55-58°e) takes place in the same
interval as the optimum temperature of the amylase activity (50-52°e). The
acidification also exerts positive effects on the structure of starch granules,
leading to an increased water-binding capacity.
Sourdough also affects the nutritional value of bread. It was reported that
the glycaemic response to baked goods from sourdough is lowered
(Liljeberg & Bjorck, 1994; Liljeberg et at., 1995). Furthermore, the avail-
ability of minerals in sourdough bread is increased, as a major part of the
minerals in flour is complexed by phytate and thus is not available for
human nutrition. The low pH in sourdough leads to the solubilization of the
phytate complex, which is not accessible to enzymatic hydrolysis. Phytate is
subsequently degraded by the endogenous phytases of the flour that act
optimally at pH values of 5.0-5.5. Lactic acid bacteria isolated from
sourdough are unlikely to contribute significantly to phytate hydrolysis in
doughs as they exhibit only weak phytase activity (W. Hammes and M.
Ganzle, unpublished results). These findings are consistent with the results
described by Salovaara and Goransson (1983) and Larsson and Sandberg
(1991). These authors have found that phytate is degraded to the same
extent in either chemically or microbiologically acidified rye- or wheat
doughs.
Finally, the acidification is of importance for prevention of
malfermentation and bread spoilage. At the pH of sourdough, the growth
and activity of spoilage organisms such as Bacillus subtilis or clostridia,
which cause ropiness, can be suppressed. Furthermore, as reported by
 SOURDOUGH BREADS AND RELATED PRODUCTS 207

Briimmer (1974), Salovaara & Valjakka (1987) and Barber et al. (1992), the
mould-free shelf-life of the sourdough breads is increased in comparison
with breads made from yeast-leavened or chemically acidified doughs.
To obtain leavened dough, gas formation by microorganisms is required.
In sourdough, yeasts and LAB form CO2 , and the contribution of each
group to the overall gas volume differs with the type of starter and dough
technology applied (vide supra). Remarkably, in traditional sourdough the
contribution by heterofermentative LAB is substantial and may even be
decisive. As the metabolic activity of a cell is proportional to its surface area
rather than its volume, a yeast cell will produce about 10-20 times more
CO 2 than a bacterial cell of the size and shape of L. sanfrancisco. In
sourdoughs where yeasts represent 1 % of the total cell count, it can be
estimated that these provide about 15-20% of the total CO 2 formed, when
the nature of the metabolic pathways as well as the logarithmic growth of
the microftora are taken into account. Indeed, the sole activity of L.
sanfrancisco does suffice to warrant leavening of bread without any yeast.
By addition of a logarithmically growing culture of L. sanfrancisco to
dough, leavened bread was obtained that had gained a volume only slightly
smaller than that of bread produced with both LAB and sourdough
yeasts (Ganzle et ai., 1997). Gas formation by the sourdough microftora is
only of minor importance if baker's yeast is additionally applied in dough
preparation.

8.4 Physiology of lactic acid bacteria in sourdough

The utilization of carbohydrates by LAB in sourdough and the formation of

acid, gas and aroma precursors are the most important achievements for
practical purposes. The carbohydrate metabolism of the most characteristic
species, L. sanfrancisco and L. pontis, has been reviewed by Hammes
et ai., 1996 and Hammes & Vogel (1997). As compared with L. sanfrancisco,
strains of L. pontis have a higher demand for anoxic conditions, a greater
tolerance of low pH, and a greater requirement for electron acceptors as co-
substrates. Both species are equally well adapted to maltose utilization;
however, rare strains exist that do not utilize maltose.
In the original description of L. sanfrancisco , Kline & Sugihara (1971)
found that the strain isolated from San Francisco sourdough exclusively
utilized maltose. The strain was associated with a yeast that was later
identified as Candida milleri (Sugihara, 1985), which in its turn did not
utilize maltose but fermented glucose preferentially. We have investigated
about 80 strains of L. sanfrancisco isolated from German and Italian wheat
and rye sourdoughs and observed that these exhibited a high heterogeneity
in their pattern of fermentable carbohydrates. While 25% of the isolates
(Table 8.4, group II) were in accordance with the original description of the
208 MICROBIOLOGY OF FERMENTED FOODS

type strain, the species contains strains that can utilize up to eight different
sugars, including even ribose and the trisaccharide raffinose (Table 8.4,
group XV). Remarkably, the type strain originally described by Kline &
Sugihara (1971) was later shown also to ferment glucose, though only after
an extremely extended adaptation phase of >120 hours (Stolz et at., 1993).
These cells immediately resumed fermentation upon transfer to media
containing maltose. The species L. sanfrancisco contains strains that can
utilize up to eight different sugars, including ribose and raffinose. Strain-
dependent properties were also observed with regard to arginine hydrolysis
and the stereo-configuration of the lactate isomers. Some 61 out of 84
strains hydrolysed arginine, and 18 of 84 strains formed more than 95% L
(+) lactate.
A similar heterogeneity is present within the strains of L. pontis. As
shown in Table 8.5, the pattern of fermentable sugars varies between two
(one strain of this type did not utilize maltose) and 14, including even
lactose. This heterogeneity shows that the various strains have not all
adapted similarly to the man-made dough substrate and leaves open the
question as to the natural habitat of the species. One interesting observa-
tion was made by S.J. Botha (University of Pretoria, personal communica-
tion), who has isolated strains of L. sanfrancisco from the teeth of
pre-school children.
In sourdough, maltose is the most abundant fermentable carbohydrate
and its metabolism by sourdough LAB has been studied in more detail. An
overview of the carbohydrate metabolism and related reactions taking
place in these organisms is presented in Figure 8.3. It was observed by Stolz
et al. (1993) that with L. sanfrancisco the cell yield was increased 1.5-1.7-
fold when media were employed containing maltose instead of glucose
(double molarity). Maltose-adapted cultures of L. sanfrancisco as well as L.
pontis excreted intermediarily up to 8mmol/1 glucose into media containing
20mmolll maltose. In the course of extended growth the accumulated glu-
cose became utilized. Resting cells of these species accumulated up to
1 mmol of glucose per mole of maltose utilized. It was revealed by Stolz et
al. (1996) that these observations are consistent with an energy-saving
mechanism of maltose utilization. In accordance with reaction 1 in Figure
8.3, a constitutive maltose phosphorylase catalyses the phosphorolytic
cleavage of maltose. In cells growing exponentially in maltose-containing
media, hexokinase is virtualliy absent and glucose becomes excreted.
As shown by Neubauer et at. (1994), maltose is taken up actively via a
maltose/H+ symport system using the proton motive force. This system was
constitutive and not affected by glucose, thus indicating transport by two
independent systems for maltose and glucose, respectively. The transport of
glucose into the cell is mediated without the expenditure of metabolic
energy. The excretion of glucose into the media was shown to be catalysed
by a uniport system which is inducible by maltose, but not by glucose.
Table 8.4 Grouping of strains of Lactobacillus sanfrancisco according to their physiological characteristics

Group II III IV V VI VII VIII IX X XI XII XIII XIV XV

Lactic acid DL,L DL,L DL DL DL DL,L DL DL DL DL DL DL,L DL DL DL,L
configuration
% of strains 21.0 25.0 1.3 3.9 5.2 2.6 3.9 5.2 2.6 2.6 1.3 9.2 3.9 1.3 10.5
Maltose + + + + + + + + + + + + + +
D-glucose + + + + + + + + + + +
D-fructose + + + + + + + + + +
Sucrose + + + + + + + + + +
D-raffinose + + + + +
Galactose + + + +
Melibiose + +
Ribose +
Gluconate +
a-methyl-D-glucoside + + +
Arbutin +
Salicin +
The following compounds were not used by any of the strains: glycerol, erythritol, arabinose, xylose, adonitol, l3-methyl-xyloside, D-mannose, L-sorbose,
rhamnose, dulcitol, inositol, mannitol, sorbitol, a-methyl-D-mannoside, N-acetyl glucosamine, amygdalin, esculin, cellobiose, lactose, trehalose, inulin,
melezitose, starch, glycogen, xylitol, l3-gentiobiose, D-turanose, D-lyxose, D-tagatose, fucose, arabitol, 2-keto-gluconate, 5-keto-gluconate.
210 MICROBIOLOGY OF FERMENTED FOODS

Table 8.5 Physiological characteristics of Lactobacillus pontis

V) V) N '<t t-
'"ao
"'
<') ao t- t-
t- V) V) V) V)
,....; N <') <') N
::c: ::c: ::c: ::c: ::c:
ti ti ti ti E-<
....:I

Glycerol
L-arabinose +
Ribose + + + + +
D-xylose +
~-methyl-xyloside +
Galactose + + +
D-glucose + + +
D-fructose + + + + +
D-mannose
Rhamnose
Mannitol + +
Sorbitol
a-methyl-D-mannoside
a-methyl-D-glucoside + +
N-acetyl glucosamine
Amygdalin
Arbutin +
Esculin +
Salicin
Maltose + + + +
Lactose + + +
Melibiose + + +
Sucrose + + +
Trehalose
D-raffinose + + + +
Starch
Glycogen
~-gentiobiose

D-turanose + +
Gluconate +

The following compounds were not used: erythritol, D-

arabinose, cellobiose, melezitose L-xylose, adonitol, L-
sorbose, dulcitol, inositol, inulin, xylitol, D-lyxose, D-tagatose,
D-fucose, L-fucose, D-arabitol, L-arabitol, 2-keto-gluconate,
5-keto-gluconate.

The characteristics of glucose and maltose metabolism by L. sanfrancisco

and L. pontis are in accordance with the competitiveness of these organisms
in sourdough. The preferred utilization of the abundantly available maltose
provides an advantage over most yeasts, the vast majority of which do not
use this substrate. The excretion of glucose by the LAB suppresses maltose
utilization by competing yeast such as S. cerevisiae (Stolz et at., 1993). Thus,
maltose remains available for the exclusive use of the sourdough LAB.
In the heterofermentative LAB, glucose-I-phosphate is further metabo-
lized in accordance with the 6-phosphogluconate/phosphoketolase pathway
 SOURDOUGH BREADS AND RELATED PRODUCTS 211

to yield lactate, ethanol and CO 2, In dough, however, a number of modifi-

cations occur as a result of co-metabolic interactions. For example, the
presence in dough of citrate, malate, fructose or oxygen gives rise to the
formation of additional metabolites such as mannitol, succinate, glycerol
and acetate (reactions 17-26 and 29, in Figure 8.3). These reactions have
been reviewed in detail by Hammes et al., 1996. Generally such co-
metabolic pathways result in the gain of additional ATP and acetate via the
acetate-kinase reaction (Figure 8.3, reaction 17). For this purpose, reduc-
tion equivalents, usually as NADH, have to be transferred to co-
metabolites or intermediates of their turn-over (e.g. reactions 13, 18, 19,20,
28 and 29). As a result of maltose co-metabolism with the various substrates
in the dough, L. sanfrancisco and L. pontis achieve higher growth rates and
cell yields. Furthermore, the acetate formed is of major importance for the
development of flavour and the microbial stability of the bread. The
molar ratio of lactate to acetate in bread is defined as the 'fermentation
quotient' and is considered optimum in the range between 2.0 and 2.7.
Lower values are desired when sourdough is subjected to drying processes
to serve as a baking aid (vide supra) in accelerated direct fermentation
processes.
The study of the metabolic properties of LAB provide rational means to
control the fermentation quotient by technological measures. For example,
Rocken et al. (1992a,b) have shown that addition of fructose or invert sugar
to the dough increases the acetate content; a similar effect was observed by
Gobbetti & Corsetti (1996) on addition of citrate.
Breads made from sourdough containing L. sanfrancisco were found to
be the most flavourful (Spicher et al., 1980). As discussed by Rothe (1974)
and Martinez-Anaya (1996), the aroma compounds which contribute to
bread flavour originate mainly from non-enzymatic browning during bak-
ing, fatty acid oxidation and products of microbial metabolism. The concen-
tration of dough volatiles was found by Hansen et al. (1989a,b) to be
influenced by the microflora of sourdough. Damiani et al. (1996) have
shown that doughs prepared with strains of L. sanfrancisco had a similar
and characteristic profile of volatiles as compared with doughs prepared
with other lactic acid bacteria. The main compounds identified were alco-
hols (ethanol, propanol, 2-methyl-1-pentanol, 1-heptanol and l-octanol),
aldehydes (3-methyl-1-butanal, heptanal, trans-2-heptanal, octanal and
nonanal), acetic acid, and comparatively little ethyl acetate. These com-
pounds may result from either microbial metabolism or fatty acid oxidation
by flour enzymes (Martinez-Anaya, 1996); a clear identification of the
aroma value of these compounds is, however, still missing. Damiani et al.
(1996) as well as Hansen & Hansen (1994) could furthermore demonstrate
that the profile of volatiles produced in dough by sourdough yeasts such as
C. milleri differs from that of baker's yeast.
Further contributions to bread flavour are provided by amino acids and
peptides acting in the dough as flavour precursors. It has been found that
212 MICROBIOLOGY OF FERMENTED FOODS

Glucose Maltose H+
I I A
~niPort~mpo_
~ ~
Maltose H+
Cytoplasm
""-
Glucose Glucose-l-P

AT~ / 27
Ribose ADP GIUr:.;;,.< F ructose-6-P
ATP
31 4 :,\,. 7 ~
ADP '" NADH+H+
'I( Erythrose-4-P
Ribose-5-P 32
6-Phosphogluconate
28 V NADH+H+
5 NAD+ I Pi7NAD+
Glycerol
co
- 2 'I(
~
NADH+H
+
Erythritol
Pi ::r NAO+ ~Ribulose-5-P
+ 6 t
29 ~ ~~e-5-p

~ADP -- l ~
Glyceraldehyde-3-P \ Acetyl-P
Pi AD+ AT ~ CoASH
14 ~ p. CO2
'I( I

1,3-Di hosphoglycerate Acetyl-CoA Fructose ~ Pyruvate Fumarate

ADP 17 I: V
AOH+H+ ~ AOH+~NADH+H~NAOH+~
9 I ~ ATP ~ AD+ ~ 18 NAD+ ~ 19NAD+ ~ 13NAD+ ~ 25
'I( CoASH'I( 'I( 'I( 'I(

3-Phosphoglycerate Acetaldehyde Mannitol HP2 Lactate Succinate

10'1(
I ADH
+H ~
ADH+H+
16 ~ + 20
2-Phosphoglycerate 'I( 'I( NAD NAD+ 'I( 26
Acetate EthanQI 2H20
II H 0
'I( 2 ADP ATP NADH+H+ NAD+
Phosphoenolpyruvate ~:> Pyruvate ~:> Lactate
12 13
 SOURDOUGH BREADS AND RELATED PRODUCTS 213

the proteolytic activity in sourdoughs is higher than in yeast doughs and

therefore more precursors are formed. The main proteolytic activity can be
attributed to the endogenous enzymes of the flour (Kratochvil & Holas,
1988; Spicher & Nierle, 1988). The proteolytic activity of L. sanfrancisco
was studied by Gobbetti et al. (1996a). The enzyme was characterized as a
cell-wall-associated serine protease. Its activity was in agreement with the
adaptation of the organism to the dough environment, as it exhibited a
higher activity on gliadine than on <x'c and ~-caseins and maintained a rela-
tively high activity under conditions prevailing in fermenting dough, i.e. pH
4.0-5.0 and 30°C. The activities of aminopeptidase, tripeptidase and
iminopeptidase were located in the cytoplasm, whereas the dipeptidase was

Figure 8.3 Carbohydrate metabolism of L. pontis and L. sanfrancisco. The enzymatic activities
and organisms have been described by Stolz et al., 1995 (modified):
1 Maltose phosphorylase: L. sanfrancisco ATCC 27651, LTH 1729, LTH 2581, L. pontis LTH
2587, LTH 1731, LTH 1735, L. reuteri LTH 3120, L. fermentum LTH 3125.
2 Phosphoglucomutase: L. sanfrancisco ATCC 27651, LTH 1729, LTH 2581, L. pontis LTH
2587, LTH 1731, LTH 1735, L. reuteri LTH 3120, L. fermentum LTH 3125.
3 Hexokinase: all lactic acid bacteria.
4-6 Enzymes of the phosphogluconate pathway: all heterofermentative lactic acid bacteria.
7 Phosphoketolase: all heterofermentative and facultative heterofermentative lactic acid
bacteria.
8-13 Enzymes of glycolysis: all lactic acid bacteria.
14-16 Acetyl transferase: all heterofermentative lactic acid bacteria with the exception of
strains of L. buchneri and L. brevis.
17 Acetate kinase: all heterofermentative and most facultative heterofermentative lactic acid
bacteria.
18 Mannitol dehydrogenase: L. sanfrancisco ATCC 27651, LTH 1729, LTH 2581, L. pontis
LTH 2587, LTH 1731, LTH 1735, L. fermentum LTH 3125.
19 NADH-HzOz-oxidase: L. sanfrancisco ATCC 27651, LTH 1729, LTH 2581.
20 NADH-peroxidase: L. sanfrancisco ATCC 27651, LTH 1729, LTH 2581.
21 Citrate lyase: L. sanfrancisco ATCC 27651, LTH 1729, L. amylovorus LTH 3122, L.
fermentum LTH 3125.
22 Oxaloacetate decarboxylase: L. sanfrancisco ATCC 27651, LTH 1729, L. amylovorus LTH
3122, L. fermentum LTH 3125.
23 Malic enzyme: not found.
24 Fumarase: L. sanfrancisco ATCC 27651, LTH 1729, L. pontis LTH 1731, L. reuteri LTH
3120, L. amylovorus LTH 3122, L. fermentum LTH 3125.
25 Succinate dehydrogenase: L. pontis LTH 1731, L. reuteri LTH 3120, L. amylovorus LTH
3122, L. fermentum LTH 3125.
26 Malolactic enzyme: L. sanfrancisco ATCC 27651, LTH 1729, LTH 2581.
27 Glucosephosphate-isomerase: all lactic acid bacteria.
28 Erythritol dehydrogenase and erythrose-4-P-phospho-transferase: L. sanfrancisco LTH
1729, LTH 2581.
29 Glycerol3-P dehydrogenase and glycerol kinase: L. sanfrancisco LTH 1729, LTH 2581, L.
pontis LTH 1735.
30 Malate dehydrogenase: L. sanfrancisco CB1, L. pentosus DSM 20314.
31 ATP:D-ribose 5-phosphotransferase: L. sanfrancisco LTH 4447.
32 D-ribose 5-phosphate ketol-isomerase: L. sanfrancisco LTH 4447.
ATCC, American Type Culture Collection, Maryland, USA;
LTH, strain collection of the Dept. of Food Technol., Universitiit Hohenheim, Stuttgart;
DSM, Deutsche Sammlung von Mikroorganismen und Zellkulturen, Braunschweig.
214 MICROBIOLOGY OF FERMENTED FOODS

membrane-associated. These activities were especially high when com-

pared with those of other sourdough LAB.
The knowledge of the genetics of sourdough LAB is still in its infancy.
The presence of plasmids in strains of L. sanfrancisco was shown by Lanner
et al. (1990). No information is available concerning the information en-
coded on the extra-chromosomal elements. Strains of L. sanfrancisco have
also been subjected to genetic modification. Gobbetti et al. (1996b) have
introduced into L. sanfrancisco CB1 by electroporation the gene coding for
a-amylase from Bacillus stearothermophilus. The gene was located on a
vector (pC194amy) and three further plasmid vectors could be used for
transformation and remained stably maintained for at least 140 generations.
Note added in proof. The epithets of L. sanfrancisco and L. sake have
been corrected to L. sanfranciscensis and L. sakei, respectively, according
to the International Code of Nomenclature of Bacteria (Triiper & de Clari,
1997).

References

Anon. (1990) Sauerteig aus Isernhagen wird laufend iiberpruft. Biickerzeitung, 90, 4--6.
Anon. (1994) Bekanntmachung von weiteren Leitsiitzen des Deutschen Lebensmittelbuches.
Bundesanzeiger, 46, 7-8.
Beccard, E. (1921) Beitriige zur Kenntnis der Sauerteiggiirung. Zbl. F. Bakt. II. Abt., 54, 465-71.
Barber, B., Ortola, C., Barber, S. & Fernandez, F. (1992) Storage of packaged white bread. III.
Effects of sour dough and addition of acids on bread characteristics. Z. Lebensm. Unters.
Forsch., 194, 442-9.
Bocker, G., Vogel, R.F. & Hammes, W.P. (1990) Lactobacillus sanfrancisco als stabiles Ele-
ment in einem Reinzucht-Sauerteig-Priiparat. Getreide Mehl und Brot, 44, 269-74.
Bocker, G., Stolz, P. & Hammes, W.P. (1995) Neue Erkenntnisse zum Okosystem und zur
Physiologie der sauerteigtypischen Stiimme Lactobacillus sanfrancisco und Lb. pontis.
Getreide Mehl und Brot, 49, 370-4.
Brummer, I.M. (1974) Biickereitechnologische MaBnahmen zur Unterdruckung der Schim-
melanfiilligkeit von Brot. Getreide und Mehl und Brot, 28, 45-9.
Budolfsen-Hansen, G. (1988) Gefriergetrocknete Starterkulturen fUr gesiiuerte Backwaren.
Int. Ztschr. Lebensm.-Technol. U. Verfahrenstechnik, 39, 597-602.
Collar, c., Mascaros, AF. & Benedito de Barber, C. (1992) Amino acid metabolism by
yeasts and lactic acid bacteria during bread dough fermentation. Journal of Food Science, 57,
1423-7.
Corsetti, A, Gobbetti, M. & Smacchi, E. (1996) Antibacterial activity of sourdough lactic acid
bacteria: isolation of a bacteriocin-like inhibitory substance from Lactobacillus sanfrancisco
C57. Food Microbiology, 13, 447-56.
Damiani, P., Gobbetti, M., Cossignani, L., Corsetti, A, Simonetti, M.S. & Rossi, J. (1996) The
sourdough microflora. Characterization of hetero- and homofermentative lactic acid bacte-
ria, yeasts and their interactions on the basis of the volatile compounds produced. Lebensm.
Wiss u. Technol., 29, 63-70.
Fenton, A (1974) Sowens in Scotland. Folk-life, a Journal of Ethnological Studies, U, 41-7.
Giinzle, M.G., Hertel, C. & Hammes, W.P. (1995) Antimicrobial activity of lactobacilli isolated
from sourdough, in Beijerinck Centennial; Microbial Physiology and Gene Regulation:
Emerging Principles and Applications (eds W.A Scheffers & J.P. van Dijken), Book of
Abstracts, Delft University Press, The Hague, The Netherlands, pp. 380-l.
Giinzle, M.G., Hiiusle, S. & Hammes, W.P. (1997) Wechselwirkungen zwischen Laktobazillen
und Hefen des Sauerteiges. Getreide, Mehl und Brot, 51, 209-15.
 SOURDOUGH BREADS AND RELATED PRODUCTS 215
Gobbetti, M. & Corsetti, A (1996) Co-metabolism of citrate and maltose by Lactobacillus
brevis subsp. lindneri CB1 citrate-negative strain: effect on growth, end products and
sourdough fermentation. Z. Lebensm. Unters. Forsch., 203, 82-7.
Gobetti, M., Simonetti, M.S., Rossi, J., Cossignani, A & Damiani, P. (1994) Free D- and L-
amino acid evolution during sourdough fermentation and baking. Journal of Food Science,
59,881-4.
Gobbetti, M., Smacchi, E. & Corsetti, A (1996a) The proteolytic system of Lactobacillus
sanfrancisco CB1: Purification and characterization of a proteinase, a dipeptidase, and an
aminopeptidase. Applied and Environmental Microbiology, 62, 3220-6.
Gobbetti, M., Corsetti, A, Mortelli, L. & Elli, M. (1996b) Expression of a-amylase gene from
Bacillus stearothermophilus in Lactobacillus sanfrancisco. Biotechnology Letters, 18, 969-
74.
Hamad, S.H., Backer, G., Vogel, R.F. & Hammes, W.P. (1992) Microbiological and chemical
analysis of fermented sorghum dough for Kisra production. Applied Microbiology and
Biotechnology, 37, 728-31.
Hammes, W.P. (1991) Fermentation of non-dairy food. Food Biotechnology,S, 293-303.
Hammes, W.P. & Vogel, R.F. (1977) Sauerteig in G. MUller, in Mikrobiologie der
Lebensmittel, Lebensmittel pflanzlicher Herkunft (eds W. Holzapfel & H. Weber), Behr's
Verlag, Hamburg, pp. 201-85.
Hammes, W.P., Stolz, P. & Giinzle, M.G. (1996) Metabolism of lactobacilli in traditional
sourdoughs Advances in Food Science, 18, 176-84.
Hansen, A, Lund, B. & Lewis, M.J. (1989a) Flavour of sourdough rye bread crumb. Lebensm.
Wiss. u. Technol., 22,141-4.
Hansen, A, Lund, B. & Lewis, M.J. (1989b) Flavour production and acidification of
sourdoughs in relation to starter culture and fermentation temperature. Lebensm. Wiss. u.
Technol., 22, 145-9.
Hansen, B. & Hansen, A (1994) Volatile compounds in wheat sourdoughs produced by lactic
acid bacteria and sourdough yeasts. Z. Lebensm. Unters. Forsch., 198, 202-9.
Holliger, W. (1902) Bakteriologische Untersuchungen fiber Mehlteiggiirung. Zbl. F. Bakt. Il.,
9,305-12,361-71,473-83,521-37.
Kitahara, K., Kaneko, T. & Goto, O. (1957) Taxonomy studies on the hiochi-bacteria, specific
saprophytes of sake. I. Identification and classification of hiochi-bacteria. Journal of General
and Applied Microbiology, 3, 102-10.
Kline, L. & Sugihara, F.T. (1971) Microorganisms of the San Francisco sour dough bread
process. II. Isolation and characterisation of undescribed bacterial species responsible for
the souring activity. Applied Microbiology, 21, 102-10.
Knudsen, S. (1924) Uber die Milchsiiurebakterien des Sauerteigs und ihre Bedeutung
fUr die Sauerteiggiirung. Dissertation. Den. Kg!. Veterinaer-og Landbohoskoles
Arsskrift.
Kratochvil, J. & Holas, J. (1988) Use of a cereal protein substrate for characterization of
proteolytic enzymes in rye sourdough. Getreide Mehl und Brot., 42,166-9.
Larsen, AG., Vogensen, F.K. & Josephsen, J. (1993) Antimicrobial activity of lactic acid
bacteria isolated from sourdough: purification and characterisation of bavaricin A, a
bacteriocin produced by Lactobacillus bavaricus MI401. Journal of Applied Bacteriology,
75,113-22.
Larsson, M. & Sandberg, AS. (1991) Phytate reduction in bread containing oat flour, oat bran
or rye bran. Journal of Cereal Science, 14, 141-9.
Liljeberg, H. & Bjorck, I. (1994) Bioavailability of starch in bread products. Postprandial
glucose and insulin responses in healthy subjects and in vitro resistant starch content.
European Journal of Clinical Nutrition, 48, 151-63.
Liljeberg, H.G.M., Lanner, C.H. & Bjorck, I.M.E. (1995) Sourdough fermentation or addition
of organic acids or corresponding salts to bread improves nutritional properties of starch in
healthy humans. Journal of Nutrition, 125, 1503-11.
Lanner, C., Preve-Akesson, K. & Ahrne, S. (1990) Plasmid contents of lactic acid bacteria
isolated from different types of sourdoughs. Current Microbiology, 20, 201-7.
Lorenz, K.J. & Kulp, K. (1991) Handbook of Cereal Science and Technology, Marcel Dekker,
New York.
Martinez-Anaya, M.A (1996) Enzymes and bread flavour. Journal of Agriculture and Food
Chemistry, 44, 2469-80.
216 MICROBIOLOGY OF FERMENTED FOODS

Neubauer, H., Glaasker, E., Hammes, W.P., Poolman, B. & Konings, W.N. (1994) Mechanism
of maltose uptake and glucose excretion in Lactobacillus sanfrancisco. Journal of Bacteriol-
ogy,176,3007-12.
Ottogalli, G., Galli, A. & Foschino, R (1996) Italian bakery products attained with sourdough:
characterization of the typical sourdough flora. Advances in Food Science (CMTL), 18,
131-44.
Rocken, W. & Spicher, G. (1993) Fadenziehende Bakterien-Vorkommen, Bedeutung,
GegenmaBnahmen. Getreide, Mehl und Brot, 47, 30-5.
Rocken, W. & Voysey, P.A. (1995) Sourdough fermentation in bread making. Journal of
Applied Bacteriology (Supplement), 79, 38-48.
Rocken, W., Rick, M., Mack, H. & Briimmer, I.M. (1992a) Versuche zur Kompensierung der
Essigsaureverluste beim Trocknen von Sauerteigen. Getreide, Mehl und Brot, 46, 139-44.
Rocken, W., Rick, M. & Reinkemeier, M. (1992b) Controlled production of acetic acid in
wheat sour doughs. Z. Lebensm. Unters. Forsch., 195, 259-63.
Rothe, M. (1974) Aroma von Brot, Akademie Verlag, Berlin.
Salovaara, H. & Goransson, M. (1983) Breakdown of phytic acid during production of acidi-
fied and non-acidified rye bread. Naringsforskning, 27, 97-101.
Salovaara, H. & Valjakka, T. (1987) The effect of fermentation temperature, flour type and
starter on the properties of sour wheat bread. International Journal of Food Science and
Technology, 22, 591-7.
Saunders, RM., Ng, H. & Kline, L. (1972) The sugars of flour and their involvement in the San
Francisco sour dough French bread process. Journal of Cereal Science, 49, 86-91.
Souci, S.W., Fachmann, W. & Kraut, H. (1994) Food Composition and Nutrition Tables, 5th
edn, Medpharm Scientific Publishers, Stuttgart.
Spicher, G. & Nierle, W. (1988) Proteolytic activity of sourdough bacteria. Applied Microbiol-
ogy and Biotechnology, 28, 487-92.
Spicher, G. & Stephan, H. (1987) Handbuch Sauerteig: Biologie, Biochemie, Technologie,
Behr's Verlag, Hamburg.
Spicher, G., Schroder, R & Stephan, H. (1980) Die Mikroflora des Sauerteiges. X. Mitteilung:
Die backtechnische Wirkung der in 'Reinzuchtsauem' auftretenden Milchsaurebakterien
(Genus Lactobacillus Beijerinck). Z. Lebensm. Unters. Forsch., 171, 119-24.
Stolz, P. (1995) Untersuchungen des Maltosemetabolismus von Laktobazillen aus Sauerteig.
Dissertation an der Universitat Hohenheim.
Stolz, P., Backer, G., Vogel, RF. & Hammes, W.P. (1993) Utilisation of maltose and glucose
by lactobacilli isolated from sourdough. FEMS Microbiology Letters, 109, 237-42.
Stolz, P., Vogel, RE & Hammes, W.P. (1995). Utilization of electron acceptors by lactobacilli
isolated from sourdough. II. Lactobacillus pontis, L. reuteri, L. amylovorus, and L.
fermentum. z. Lebensm. Unters. Forsch., 201, 402-10.
Stolz, P., Hammes, W.P. & Vogel, RF. (1996) Maltose-phosphorylase and hexokinase activity
in lactobacilli from traditionally prepared sourdoughs. Advances in Food Science (CMTL),
18,1-6.
Sugihara, F.T. (1985) Microbiology of breadmaking, in Microbiology of Fermented Foods, Vol.
1 (ed. B.I.B. Wood), Elsevier Applied Science Publishers, New York, pp. 249-62.
Tichaczek, P., Nissen-Meyer, 1., Nes, I.F., Vogel, RF. & Hammes, W.P. (1992) Characteriza-
tion of the bacteriocins curvacin A from Lactobacillus curvatus LTH1174 and sakacin P
from L. sake LTH673. Systematic and Applied Microbiology, 15, 460-8.
Triiper, H.G. & de Clari, L. (1997) Taxonomic note: Necessary correction of specific epithets
formed as substantives (nouns) "in apposition". International Journal of Systematic Bacteri-
ology, 47, 908-9.
Vogel, RF., Backer, G., Stolz, P., Ehrmann, M., Fanta, D., Ludwig, W., Pot, B., Kersters, K.,
Schleifer, K.H. & Hammes, W.P. (1994) Identification of Lactobacilli from sourdough and
description of Lactobacillus pontis sp. nov. International Journal of Systematic Bacteriology,
44,223-9.
Wiese, B.G., Berger, RG. & Diekmann, H. (1995) The influence of lactic acid bacteria on
aroma components of wheat bread. Biospectrum, Special Issue, 3, 102 (PC 010).
Wiese, B.G., Strohmar, W., Rainey, F.A. & Diekmann, H. (1996) Lactobacillus panis sp. nov.,
from sourdough with a long fermentation period, International Journal of Systematic
Bacteriology, 46, 449-53.