Empiricism Must Not Be Denied.

Michael Gänzle’s name should be a household one for home bakers.  He has done some very important work in linking process-parameters to the growth of Lb sanfranciscensis in response to process parameters in rye and wheat sourdoughs.

His work suggests Lb SF grows best at 5 – 20% inoculation and at 32ºC, but, as any astute person might guess, bacteria genetically adapt to their environment at a much quicker pace than larger organisms with a longer life-cycle.  Why do I mention this?  Because it seems that many traits that belong to Lb SF, as well as the whole sourdough microfloral canon, are strain-specific, not species.

Some of the best real-world research into sourdough has come out of Italy, not France, and there are many discoveries that run counter to traditional thinking.  The biggest one, at least for me, is that Lb SF will remain the dominant bacterial presence from inoculations 5 – 40%.  Why this is, I do not know, as I doubt this is true the world over.  What’s more, many of these studies are conducted on sourdoughs that are maintained in either wheat (Triticum aestivum) and durum wheat (Triticum durum).  Another generalisation can be made about these Italian sourdoughs:  for particular Italian strains of Lb SF to be dominate, the key temperature range seems to be between 25 - 30 ºC.

So, does this mean we, as bakers, can use our backslopping parameters to actively evolve the optimal conditions of our microflora over time?  The answer is, of course, yes.  But it must be at incrementalised and optimised conditions.

A Hypothesis for the Hypothesis.

First, we must discover the first time and place man began interacting with wheat and/or rye.  Turkey and northern Syria are the likely locale, and I suspect the date is much older than is currently speculated and/or uncovered from current paleo-archaeological research.

Second, we must discover the common forms of fermentations occurring in that area at that time, both spontaneous and as they relate to man.

My hypothesis:  Lb SF evolved from a set of milk-based fermentative LAB microflora (e.g., one in which man was involved) to a grain-based one through a rapid series of reductive evolutionary steps that resulted in massive gene-death (inactivation) and deletion (total decrease in genome size) as well as an efficient genomic design that allowed it to “exploit” (increase metabolic co-existence) several species, including rye and wheat grasses, humans, those mammals commonly associated with humans, insects, lactic-acid bacteria, yeasts, and bifidobacterium found in the colon, all at the time and place mentioned above.

Part of the above hypothesis would indicate several pre-conditions:

1.  The lactic-acid bacteria involved in human-based milk fermentations have evolved to the human digestive tract, while also actively modifying the digestive tract’s own workings for a net gain for both host and symbiont by acting as both prebiotic and probiotic organisms at the same time. (More on this later.)

2.  There is a common evolutionary pathway seen in LAB-human fermentations, with a certain set remaining ‘generalists’ and another ‘specialists,’ and yet they are all working toward the same, meta-biotic goal:  increasing and maintaining co-existence (i.e., biodiversity).

3.  Lb SF’s evolution resulted in its ability to play a prebiotic role for those LAB based in the human colon (and hence also rectum, feces, and so on).  The likely scenario would have involved a human “noticing” spontaneous fermentation in wheat or rye grains from the human-based LAB mentioned in #1.  “Backslopping” was likely already in existence as a method by early Homo sapiens as a method for milk-based fermentations.  Conditions were not “wash-your-hands” sterile, obviously, when backslopping.  Lb SF likely inherited, through lateral-gene transfer, the intracellular and intraspecies molecular-signalling genes (i.e., the ones specific to the human, and likely other higher, associated mammal, colon) already present in these “specialist” LAB present in these milk-based fermentations.  Lb SF-based sourdough bread is abundant in prebiotic substrates for these colonically-based LAB to this day.

Dear Debra Wink & Peter Reinhardt, Drink the Pineapple Juice.

“Two type 0 Triticum aestivum flours were used . . . Sourdough production and propagation were established based on the most diffuse and traditional used for brad making the central and southern areas of Italy . . . A control sourdough, without starter, was also produced under the same conditions.  The 10 sourdoughs (9 with individual L. sanfranciscensis strains and the control) were incubated in sterile plastic beakers at 30 ºC for 8h.  After fermentation, sourdoughs were stored at 10ºC for about 16 hours and further used for propagation.

            “Each sourdough was then propagated for 10 days by daily back-slopping.  Cell densities after each sourdough fermentation [were recorded] . . . The lowest value was found after the first day of propagation . . .

            “After the third day of propagation, the control sourdough, without starter added, reached almost the same stable cell numbers . . . Only three of the nine starters used dominated throughout the 10 days of propagation carried out under rigorously standardized conditions.  The others were outcompeted by autochthonous population of the wheat flour and disappeared progressively starting from the first day of propagation . . .One autochthonous strain of L. sanfranciscensis was found to be dominant in all sourdoughs . . .

            “Although this study was carried out under laboratory conditions that might have differed from bakery environmental conditions, the following was shown about wheat flour:  (i) it is the source of the autochthonous lactic acid bacteria that can associate with or outcompete starter lactic acid bacteria and (ii) it plays a key role in establishing the stable microbial consortia within a short time . . . Apart from the starter used, all sourdoughs propagated by using the two types of wheat flour harbored L. sanfranciscensis strains.”

 “Taxonomic Structure and Monitoring of the Dominant Population of Lactic Acid Bacteria during Wheat Flour Sourdough Type I Propagation Using Lactobacillus Sanfranciscensis Starters,” Siragusa, di Cagno, Ercolini, Minervini, Gobbetti & De Angelis.  Appl. Environ. Microbiol. 2009.

What's in a Relationship?

“Another possible form of commensalism takes place in starters for Gouda cheese, where PrtP- L. lactis strains benefit from the peptides that are released from milk protein through the action of extracellular proteases (PrtP) produced by PrtP+ strains while the PrtP+ strains do not seem directly affected.  In milk, PrtP+ strains produce more biomass than their isogenic PrtP- variants lacking plasmids containing the protease gene.  In pure cultures of PrtP+ strains grown in milk, PrtP- variants rapidly occur.  The outcome of the long-term propagation of PrtP+ and PrtP- strains in a protein-containing medium like milk is that the strain that makes the least use of the resources in the medium, namely, the PrtP- strain, will become dominant.  In this case, the immediate gain for the PrtP- strain is traded for the long-term community benefit.  This particular example is also known as the ‘prisoner’s dilemma’ in evolutionary game theory.  The population dynamics of PrtP+ and PrtP- isolates are highly dependent on the growth conditions that influence the costs and benefits of proteolytic phenotype.”

Sieuwerts, M. De Bok, Hugenholtz, et al., “Unraveling Microbial Interactions in Food Fermentations:  from Classical to Genomics Approaches,” Applied and Environmental Biology, Aug 2008.

Lb sanfranciscensis is PrtP- and prefers the uptake of peptides, while most other lactic-acid bacteria recovered from wheat- and/or rye-based fermentations are PrtP+.  Notice the relationship?

It Got Startered.

“During evolution, due to the low energy gain by fermentation and to the harsh life condition in an ‘oxidant’ world (like the one in which we now live), LAB have been compelled to specialise their metabolism rather toward stress defence than to acquire strategic biosynthetic abilities.  Therefore, they developed symbiont/parasite relationships with plants and animals which can supply the vitamins, proteins, and amino acids.  It is worth noting that LAB had acquired the ability to recognise several sugars, such as for instance xylose, cellobiose, ribose, arabinose, glucose, and fructose, before they developed the ability to ferment lactose to lactate, which was made possible only after mammal’s expansion on Earth.  Therefore, they firstly colonised fruits and vegetable ecological niches, and later cheese, wine, and especially milk, which constitute their election habitat being rich in lactose.”

Enrica Pessione, “Lactic-Acid Bacteria Contribution to Gut Microbiota Complexity:  Lights and Shadows, 2012.

Some Useful Tips for Maintaining a Lactobacillus-Sanfranciscensis-Based Sourdough Starter.

Keep your culture in log every day.

Do not pay attention to any external clues until your starter is established.  Pretend you’re Helen Keller.  Ignore all visual, tactile, or olfactory input until your starter is established.  (“It smells like acetone.” or “It’s not doubling.”)

A culture can change completely in only 3 – 5 generations under optimal conditions.

If you use Saccharomyces cerevisae commonly, there’s a good chance (more than 50-50%) it will end up as a permanent, stable cast-member in your starter’s microflora, usually as a mutant strain that, as a result of random genetic mutation, has decided to go native.

Understanding why a specific organism evolved the way it did will also allow the baker to reverse-engineer optimal parameters.

Grain-based microflora will always win out over every other organism, except one:  Lb sanfranciscensis.

Never refrigerate or freeze your starter.  Ever.  Throw it out.  Start a new one.  How?  Easy.  Just do this.

Lb sanfranciscensis does not have a latency period (i.e., it does not take two weeks to get a well-established culture up and running).  Don't believe me?  Read here.

A “mature” or “ripe” starter is one that has a pH of at least 4.0.

100% flour is the same as saying maximum microflora population.

For an Lb SF-based culture, only feed your starter wheat or rye, preferably as whole a grain as you can.  Stone-ground’s even better.

Weigh.  The margins of error are too small, and you may very well unintentionally change the entire nature of your starter if you do not.

Measure temperature.  See above.

Mix at the same time every day, preferably at some point during day time.  It is best to find a point that best meets your routine.

For the most flavour, maintain a starter with a hydration within the range of 50 – 75%.

The most effective way to increase total acetic-acid content is to oxygenate your starter.  I simply tear mine apart at regular intervals throughout the day and stick it back together.

Do not stress your culture out:  Adapt it.  How?  More on this later.

The Fastest, Easiest Way to Create a Lactobacillus-Sanfranciscensis-Dominant Sourdough Starter.

Creating a starter.

Refreshment 0.

20        g                      flour, rye, whole-grain

240     g                      water, 25ºC

Combine ingredients in a large Ziploc bag and seal, removing all excess air.  Fill a large, plastic container (such as a cooler) with water that is 37 ºC, ensuring the temperature is exact.  Drop the bag into the water, placing something heavy atop it to prevent it from floating.  Place the container’s lid on, and let the bag sit for 18 – 24h.  Check the water’s temperature often, adding more hot water as needed to maintain a constant 37 ºC.

Refreshment #1.

100     g                      flour, wheat, whole-grain

Remove the Ziploc bag from the water and add all of the whole-wheat flour.  Seal, once again removing all excess air.  Lay the bag on a flat surface, and, using your palms, squish the bag back and forth in order to create a homogenous liquid.  Place back into the plastic container, adding enough hot water to reach 32ºC.  As before, check the water’s temperature often, adding more hot water as needed to maintain a constant 32ºC.  Allow to sit for 18 – 24h.

Refreshment #2.

100     g                      flour, wheat, whole-grain

45        g                      water

30        g                      starter, from refreshment #1

Combine all three ingredients in a bowl, and then mix thoroughly until a homogenous texture is achieved.  Let ferment for 24h at 20 - 30ºC.

Final dough temperature.  30 ºC.

Refreshment #3 - ∞.

50        g                      flour, wheat, whole-grain

32.5    g                      water

4.125  g                      starter, from refreshment #2

Combine all three ingredients in a bowl, and then mix thoroughly until a homogenous texture is achieved.  Let ferment for 24h at 20 - 30ºC in an air-tight, plastic container.

Final dough temperature.  30ºC, if day's high temperature is below 29ºC, and 20ºC if day's high-temperature is above 30ºC.

Second Clarification for Varda

The second fundamental assumption I am challenging (and which has been confirmed in a variety of studies from the last five years) is the assumption that LB SF has a latency period.

This "myth" has been repeatedly echoed time and again.  If you're an observant reader, you'll notice that most bakers will tell you it takes two days to create a culture and then two weeks for LB SF to "drop" into the mix.  All of this has been based upon one finding by Stolz et al. years and years ago.

Sigh.

This is simply not true.

Under non-sterile conditions and under optimal processing-conditions (which I keep repeating and repeating until I turn blue in the face), LB SF will occur after one refreshment.  Problem with LB SF is, it requires a sub-dominant microflora to be present for it to really do well and become dominant.

Most schema that involve the creation of a starter do so at sub-optimal conditions, and the establishment of LB SF hence takes longer.  These "normal" way of creating and building a starter is inefficient because it assumes that (i) LB SF takes two weeks to get going (remember time means nothing to these organisms; only temperature does, which actively changes their implied energy-state and "sense" of time), and (ii) most recommend process-parameters that would decrease the presence of LB SF in a continually-maintained starter (over 3 - 5 generations).  That is, the temperature and inoculation percentage are usually much too low and much to high, respectively, to establish LB SF.

Let us arrive at the truth and make better bread.

Elaboration for Varda.

Let me explain in simpler terms.

There are two kinds of sourdough starters:  those which are spontaneous, say, whenever a substrate and water are mixed together; and then there are those that are maintained.

The difference between the two is obvious.  The first would be similar to those species in situ (think: wolf, fox, dingo), while the other, via whatever process parameters are used, actively selects and modifies the species present (over time, you get specialisation, like Chihuahuas, Golden Retrievers, Huskies, and so on).

When flour and water is mixed, there are an abundant of organisms that actively contaminate it.  The shift from a spontaneous microflora (that which is sort of just naturally present at first, from whatever source) to a "maintained" and "selected" microfloral culture generally involves three steps (for any basic fermentative process, too):

1.  Presence of atypical fermentative microorganisms.

2.  Growth and eventual dominance of microorganisms more frequently associated / typical with whatever fermentation is occurring, with a decrease in the atypical species.

3.  Growth and eventual dominance of typical but more highly-evolved species to whatever fermentative process is being used.

In sourdoughs that are continuously maintained, the species from #3 become the dominant microflora while those from either #1 or #2 become the sub-dominant.

The pineapple-juice method presupposes that those organisms from the first step (Leuconostoc, etc.) are undesirable.  They are not.  They are generally recovered from sourdoughs in cooler climates as the sub-dominant microflora(Belgium, France, the Netherlands, etc.).

Here's a list of the LAB species that have permanently colonise the human intestinal tract:  Lbs acidophilus, brevis, casei, crispatus, delbrueckii, fermentum, fructivorans, gasseri, paracasei, plantarum, rhamnosus, ruminis, sakei, salivarius, and vaginalis.

All of these are from the #2 category, and all have been recovered as dominant and/or sub-dominant organisms in wheat- and/or rye-based sourdoughs.

Only one species represents #3 for wheat- and/or rye-based sourdough fermentations:  LB SF.

There are other truly undesirable species that occur in #1 (true entero-pathogens, etc.), but they will be outcompeted every time by all the above mentioned species in a continuously maintained sourdough.

So, here's what the pineapple juice method says:  Let's skip #1 and go straight to #2 so as to rid ourselves of undesirable organisms.  Problem is, some of those in #1 are, in fact, desirable, depending upon the process-parameters (maintenance conditions) chosen, and those from #1 that ARE truly dangerous will lose out in time, every time to those desirable organisms from #s 1 -3 in a continually fed system.

Those LABs from #2 create more flavour, better fermentative conditions, etc.; hence, why we select them.

Those from #2 are more acid-tolerant than the one species from #3.  Wink et al. assume that adding pineapple juice will acidify the substrate and automatically bring about the presence of those from #2 while eliminating all those from #1.  NOT TRUE.  Many of the bad entero-pathogens will disappear, but many species she identified as "bad" can and do survive such acidified conditions.

The best way to establish those microflora from #2, instantly, is temperature.  Two of those organisms, fermentum and plantarum, are also recovered from raw wheat and rye grains, so no cross-contamination from humans is necessary.  All one needs is the raw flour that contains these organisms (basically EVERY rye and/or wheat flour ever tested in the world), water (necessary for life activity) and the right temperature.  Outside of substrate, temperature is the most important processing condition to select for or against any of these species.

Adding a mixture of whole rye and/or wheat flour with an unusually high amount of water (most of these LAB are non-motile) PLUS the right temperature will instantly activate and guarantee the presence of plantarum and/or fermentum, which are two of the most desirable "secondary" microflora.  They can out-compete ANYTHING above 37 / 35 degrees Celsius, respectively, including LB SF.

My method involves a Ziploc bag, water and grain; a large plastic container (like a cooler) that has a lid; then filling the container with 37-degree water (say from the bath-tub); and then dropping the baggie into the water for 12 - 24 hours, adding extra hot water to maintain a constant 37 degrees.  The ensures #2 is established.  Upon second refreshment, I switch to an inoculation percentage and temperature (32) more favourable to LB SF.  I arrive at a stable culture using this method in 3 - 5 inoculations (under optimised conditions), which is sort of the magic number for completely changing and/or establishing ANY sourdough culture.

The third refreshment also takes place in a bag, and uses a temperature profile that establishes a considerable yeast presence (28 degrees).

The method was developed for home-bakers without access to immersion circulators (I own many).  It's what David Chang calls "ghetto-vide."

Do you follow my logic?

The problem with Wink's "research" is that it misunderstands the way in which most of these desirable sourdough organisms out-compete undesirable (pathogenic) organisms.  She assumes it's via pH conditions, but we know this simply isn't true (LB SF dominates at conditions much, much more alkaline than any other LAB).  Acidification is only ONE variable for establishing a culture, but not THE, especially for an LB-SF-based culture.

E-mail me for more details.

Once I have a good internet connection, I'll do a step-by-step process with the exact amounts necessary to complete this.

A hypothesis.

My hypothesis:  LB SF plays an important role in the human digestive process by directly transforming its preferred substrates to metabolites (inulin-type fructans, pyruvate) that survive after baking and that are in turn used by the bifidobacterium in the human colon as substrates for their own metabolism.  This relationship would increase the digestion of undigested and/or undigestable polysaccharides.  This would imply that LB SF evolved to an extraordinary set of different species (humans, other mammals, insects, wheat and rye, the sourdough environment, other bacteria and fungi, and, especially, the bifidobacterium found in our colon, rectum, anus and even feces), and has likely developed the ability to “communicate” with all of the species mentioned.  LB SF, therefore, positively impacts every species it has come into contact with.  What’s more, it’s genomes suggests an evolutionary path widely divergent from most gut-based LAB (via reductive processes), and plays a role more closely resembling an entero-pathogen.  There is no latency.  All you need to do to create a starter is use non-sterile (i.e., non-laboratory) conditions, and establish from the beginning the parameters necessary for its growth.  Other experiments have confirmed this:  a reasonable LB SF population occurs and even dominates after just one refreshment.

Why is the pineapple-juice method ineffective?  Because, like most “starter” schemes, it delays the conditions necessary for a large LB SF population to develop.  This means:  ignore the previous ratios you’ve been taught for establishing a new starter (1:1:1 and then onto 1:2:2 or even 1:4:4).  Beginning with whole-grain rye (100% hydration) above 35 degrees Celsius will ensure, after just 12 – 18 hours, a sufficient drop in pH to establish the dominance of those LAB species (especially pontis, fermentum, and plantarum).  Please understand that heavy acidification is not necessary for LB SF to occur.  Rather, LB SF will dominate in a wheat- or rye-based matrix over every species if the pH is between 5.0 to 5.6.  This is fact.  The reason for using a high-temperature in the beginning (one above at which LB SF grows) is to establish a more desirable sub-dominant sourdough microflora.  Again, let me repeat:  LB SF will dominate every time between the above-mentioned pH range.  We cultivate these other lactobacilli because, well, they make a tastier end-product than other non-acidifying species!  Put another way, all the species Debra Wink's "research" has identified as undesirable are, in fact, not, and are commonly recovered from sourdoughs in northern European countries where the starters are maintained at a lower-temperature range (> 20ºC and < 22º).  Funnily enough, they occur as the sub-dominant species to LB SF, and produce quite good bread.  Why?  LB SF's role, as we'll find out in more detail later, is more as a mediator or conductor during sourdough fermentation, setting the standard by which all other microflora follow.

Because LB SF evolved to work in concert with all other species involved in getting it into the sourdough matrix (this is its catch-22:  it cannot become dominant or even be evolutionarily successful without the sourdough environment, so it has evolved to use just about any path available to it to ensure it ends up in sourdough), LB SF becomes dominant after just one refreshment if and when the following are in place:  the inoculation amount is between 5 - 20%; rye- and/or wheat-based substrates are used; and when the temperature is < 32ºC and greater than 20ºC.  All other methods will delay these conditions being fulfilled, such as using the (empirically-wrong) assumption that high-acid conditions are necessary for LB SF to become dominant.  This is just plain, stupidly wrong (and also the reason LB SF is never recovered from type II sourdough processes).  Try it out yourself.  I’ll be doing a post soon, with photographs and formulas, for a sure-fire way I have developed for getting a fully-active starter by the time the fourth or fifth refreshment occurs.