Interesting Correspondence on Sourdough (Long)

Skip to first unread message

Darrell Greenwood

Feb 15, 1997, 3:00:00 AM2/15/97

Dan Wing was kind enough to forward to me the correspondence below together
with permission to post. I did a small amount of copy editing. Hopefully I
didn't affect any nuances of meaning but just got rid of a few typos.

My thanks to Michael and Dan for very graciously sharing their correspondence.




Date: Thu, 13 Feb 1997 10:49:32 +0100 (MEZ)
From: Michael Gaenzle
Subject: sour dough microbiology
MIME-Version: 1.0

Dear Daniel Wing!

Your letter to Prof. Hammes has reached Hohenheim, and Prof. Hammes has
asked me to take care of the communication. I am a Ph.D. candidate in
Hammes' lab working on the physiology of sour dough lactobacilli.

Please feel welcome to address questions to us concerning sour dough
microbiology and technology! I will mail two recent publications or our lab
concerning the physiology of sour dough lactic acid bacteria by mail, but
as they may take a week or longer to reach you, I will give a few comments
on the questions in your letter:

- yeasts do not produce appreciable amounts of either lactic or acetic
acids, their main metabolites are ethanol and CO2. If acidification of the
dough is desired of required (e.g. if rye flour is used), lactic acid
bacteria or organic acids (most commonly lactic or citric acids) are added.

- homefermentative lactic acid bacteria do produce solely lactic acid from
maltose or glucose under anaerobic conditions (as they are prevailing in
sour dough fermentations). Thus, doughs acidified with homofermentative
lactic acid bacteria (LAB) contain but little acetic acid. As
homofermentative lactic acid bacteria do not produce CO2, yeast must be
added to ensure leavening of the dough.

- In sour doughs with a tradition of continuous propagation (such as the
San Francisco French Bread Sour Dough process, German rye sour doughs or
sour dough employed in Pannettone production in Italy), heterofermentative
lactobacilli, especially L. sanfrancisco, are dominating the fermentation.
Heterofermentative lactobacilli produce lactate, ethanol, and CO2 from
hexoses (most strains do not ferment pentoses), HOWEVER, if additional
substrates are present that serve as electron acceptor to balance, acetate
is produced instead of ethanol. I do not know whether or not you are
familiar with the concept of the "redox balance": Degradation of hexoses
via the pentose-phosphate pathway as employed by heterofermentative LAB
results in phosphorylation of ADP to ATP, and in the reduction of NAD to
NADH. As there is no use for NADH, it must be oxidized to NAD again. In the
absence of other substrates, acetyl-Phosphate is reduced to ethanol, with
two NADH becoming oxidized to HAD in the process. If either fructose,
oxygen, citrate or malate are present, these become reduced to mannitol,
H2O, lactic and acetic acid, and succinate, respectively, and acetyl-P is
dephosphorylated to acetate. (This explanation may not be very
straightforward, I hope we did a better job in the publications I`m about
to send you; these also include a diagram showing the metabolic pathways of
L. sanfrancisco). The consequence for the molar ration of lactate:acetate
(fermentation quotient, FQ) in sour dough fermentations is, that acetate in
produced only if one or more of the above mentioned co-substrated is
present. Oxygen is present only in the beginning of the fermentation, and
the amounts of oxygen are too low to result in significant amounts of
acetic acid, though, in principle, it is possible to increase the acetate
content by aeration of dough. Fructose is present in sucrose and other
glucofructans with higher molecular weights. Fructose is released from
these compounds by cereal or dough enzymes (many strains of L. sanfrancisco
don`t even cleave sucrose) and consequently reduced to mannitol by L.
sanfrancisco. The ration of mannitol : acetate in sour dough fermentation
is approximately 2:1, suggesting that fructose is the most important
electron acceptor. Furthermore, citrate and malate are present in the dough
in amounts < 10 mmol/kg, these are utilized also.

Thus, the effect of substrates and oxygen on the FQ is nicely explained by
the metabolic characteristics of the dominating fermentation organisms.
Dough yield (=kg dough per 100 kg flour) and temperature also influence the
FQ. Spicher reports that softer doughs lead to an increased FQ; an increase
in temperature results in higher amounts of lactic acid, while the amount
of acetic acid remains more or less the same, thus, the FQ is increased
again. I do not have a straightforward explanation for these phenomena, but
changes in dough yield and temperature will result in changes in buffering
capacities of the dough, modified activities of cereal and microbial
enzymes, as well as a changed ration of yeasts : lactobacilli counts, all
of which are likely to influence the FQ.


Michael Ganzle


Dear Michael Gaenzele

Thank you for sending one of the most gracious letters I have ever received
in response to any kind of an inquiry. Since I wrote to Prof. Hammes I have
been able to copy a number of articles from English language publications
by Drs. Brummer, Spicher, Vogel, and so forth. Unfortunately, some of them
have been in non-technical journals and were thus short on details, and
even the less technical ones were not as clearly and idiomatically written
as your letter. I DID have a hard time understanding what was meant by
Dough Yield, for instance, although I had figured it out before I got your
letter. I am still not sure I understand some of the statements those
authors made about the acid content of doughs (such as the units of
measurement), but I have been piecing things together by looking at all the
articles cumulatively. Your letter has clarified a great deal. I will put
stars next to my current questions to make THIS letter easier to answer.
Like this *****.

One problem for me was that I did not realize how predominant rye flours
were in German sourdough baking. I know that typical rye pentose is about
8% and that pentose viscosity is important in gas-trapping in rye doughs
(He and Hoseney, 1991) but I still don't know how an acidified rye dough
behaves differently from a more neutral one. ****Does it affect viscosity
somehow?*** He and Hoseney studied neutral doughs only.

I also do not understand why Brummer says "Anstellgut" is a
non-translatable term. *****What do you think it translates as?*****I take
it that this a very ripe starter, very acid, maintained at room temperature
at some infrequent rate of refreshment?***** Is it always rye based?****
Always a high-ash flour?**** How is it different from the type of French
and American wheat starters that are refreshed 1:1 every eight hours, or
1:4 every 12 hours?***** What is its consistency, pH, Total Titratable
Acid?***** My assumption is that my lack of understanding comes from the
German use of sourdough as primarily acidification, whereas here we look
for a little acidification, a good flavor, and good leavening power.**** Do
German bakers ever make wheat breads leavened with higher starter
percentages than those Brummer cites, for example 20% or 30% starter?****
Or do they acidify with very ripe starters and leaven with commercial

I am curious about the flavor/sensory aspects of the FQ: *****When a bread
is fairly sour (SF Sourdough, some rye breads) is the perceived sourness
mostly lactate, mostly acetate, or due to the pH or TTA of the bread?*****
Calvel brings this subject up, but does not resolve it to my understanding.

As for your answers to my previous questions, thank you-- I will look this
material over again, and let you know if I have questions.*****Do you mind
if I put the text of your letter (with attribution) on the internet as a
posting to the newsgroup Rec.Food.Sourdough? I will NOT put your address or
email address in the posting, unless you want me to. Please let me know, as
I think it might become part of the FAQ file there (Frequently Asked
Questions). I will forward your entire letter to a very few people in
academia here who have been helping me, so you might hear from one of them.

Dan Wing


Date: Fri, 14 Feb 1997 15:50:30 +0100 (MEZ)
From: Michael Gaenzle
To: Dan Wing <>
Subject: Re: sour dough microbiology
MIME-Version: 1.0

Dear Dan Wing!

I do not mind if the answer is posted to the I've also
been browsing in that newsgroup.

To answer a few of your questions:

I) There is no rye bread without acidification of the dough. Rye flour does
not contain gluten (or a different type of gluten that does not have the
gas-retaining properties), so that the structure of rye bread relies mainly
on gelatinized starch. Rye flour does have a higher amylase activity than
wheat flour, furthermore, the gelatinization temperature is a few degrees
lower than that of wheat starch. Thus, with the temperature optimum of rye
amylase being about 50 - 52C (with substantial activity up to temperatures
of 70C) and starch gelatinization starting at 55C, starch is degraded
during the baking process UNLESS the amylases are inactivated by lowering
the pH below 4.5. The situation is exacerbated if there was wet weather
during the harvest, as germinating rye has higher amylase activities and
the starch granules are damaged, thus facilitating hydrolysis.

II) "Anstellgut" is more or less the same as the continuously propagated
wheat starters of the SF sour dough bread, so no harm is done if it is
translated as "starter sponge" or something like. German sourdoughs usually
are rye based for two reasons: 1) Due to the climatic conditions in
Germany, especially in the northern and eastern parts that make it
difficult to grow wheat, rye flour is just as important for bread
production as wheat flour. 2) As these is no necessity to acidify wheat
flour (though it enhances the flavor), most bakers do not use sour dough to
produce wheat bread. Starter sponges are not necessarily propagated
separately. If the dough is taken care of according to traditional methods,
it is re-inoculated three times to produce bread dough (reading Bruemmer
and Spicher, you probably have already encountered the "three stage sour
dough method." A part of the bread dough is used to prepare the sour dough
for the next day. This makes 3 - 4 inoculations a day, the ratio of sour
dough to fresh dough being approximately 1:3. One has to make a point of
it: there is no typical sourdough without continuous propagation! The
microflora of these rye starters is actually the same as for wheat starter
in SF or Italy: Lactobacillus sanfrancisco and Candida milleri or
Saccharomyces exiguus. The pH of a ripe sour dough will be between 3.6 and
4.0 (L. sanfrancisco does not grow below pH 3.6). The total titrable
acidity (TTA) depends on the flour employed: as the lactobacilli acidify to
pH 3.6, flours with high buffering capacity (amount of acid required to
lower the pH), e.g. whole flours, have a higher TTA than white flours with
a low buffering capacity. Furthermore, if "hard" water with high
concentrations of Me2+ CO3- is used, the TTA will be higher.

3) Acidification vs. leavening: As mentioned above, rye flour or mixtures
of rye and wheat flours containing more than 20% rye must be acidified in
order to get bread. As the propagation of sour dough is very time consuming
if the full leavening capacity of the organism is to be obtained, quite a
few processes have been developed in Germany that ensure that the dough is
acidified (or that the sour dough added to the bread dough contains enough
acid to bring the pH of the bread dough below ca. 4.5), but no leavened by
the sour dough microflora. Leavening is achieved by bakers yeast.
Basically, there are three possibilities: 1) Dried sourdough with a high
TTA (>20) is added to the bread dough, there are no lactobacilli involved
in the fermentation (sometimes they are present in the dried sour dough
preparation anyway, as in Germany, something called sour dough must contain
viable lactic acid bacteria. The dried dough is sold much more readily if
it can be called sourdough). 2) A sour dough is kept at room temperature
for up to one week. The TTY of that dough is high enough to use it for
baking, but as the organisms are rather stressed in such an environment,
they will not contribute to the leavening of the dough. Such doughs do not
contain lactobacillus sanfrancisco, but other lactobacilli that are more
acid tolerant (the ph of such a dough reaches 3.4 - 3.6 after one day, and
stays there for the four or five more days that the dough is kept). 3) One
stage or two stage processes with starter sponges. One or two stage
processes usually do not ensure that the lactobacilli in the dough are
fully metabolically active if the bread dough is prepared, thus, the
leavening capacity is rather poor, but enough acid has been produced. As
far as I know (I never made a survey, though), only few bakers make bread
with traditional processes without bakers yeast added to leaven the dough.
Acidification of the bread dough with sour dough is rather common, and the
sensory quality of such bread is quite close to that of bread made without
bakers yeast. Straight processes with bakers yeast and chemical
acidification (citric, lactic, and acetic acid, or mixtures thereof) are
also quite common to produce rye bread.

4) Lactic acid and acetic acid will change taste and flavor of bread beyond
the decrease of pH: the taste buds (sour, bitter, sweet, salty) are on the
tongue, any other aroma is perceived with the nose; therefore, the aroma
compounds must be volatile. Acetic acid is more volatile than lactic acid,
thus, it's impact on the flavor is more pronounced than that of lactic
acid. Spicher says that a ratio of 20 acetate to 80 lactate is optimal. It
must also be taken into account, that the lowering of the pH influences the
formation of other aroma compounds during the baking process. The acetic
acid is furthermore important as growth of spoilage organisms such as molds
or rope causing bacilli (Bacillus subtilis) is inhibited by high acetic
acid concentrations.

I hope that I could answer your questions

With kind regards

Michael Ganzle

Darrell Greenwood, Vancouver, BC
My web homepage...

Reply all
Reply to author
0 new messages