Organic and conventional flours produce different sourdough fermentations

It’s difficult to connect the dots throughout our complex food system. Although it is rarely demonstrated scientifically, we generally accept that what happens on farms impacts the quality of our food. For microbial foods, the raw materials we use in fermentation can introduce different microbes depending on how those materials were produced. A recent study in Italy of sourdough fermentation demonstrated that organic vs. conventional farming can affect the quality of sourdough bread. This exciting new research highlights the role that microbes play in shaping food quality as it moves along the path from farm to fork.

In this study published in May in the journal Applied and Environmental Microbiology, researchers in Italy grew durum wheat at the same farm using four different farming practices: conventional (standard fertilizers and pesticides), organic with cow manure, organic with green manure, and no inputs (no fertilizer or pesticides). The organic treatments were applied in accordance to EU Regulations. They then looked at how the farming practices impacted not only the quality of the flour that was milled from the wheat, but also the composition of the microbes in the flour before and after being used for sourdough fermentation. A key aspect of this study is they linked all of these variables to the final product, the bread, by measuring everything from biochemical properties to “chromaticity coordinates of the bread crust” (crust aesthetics!).

To determine the abundance and composition of microbes in the flour before and after fermentation, the researchers used several approaches. To determine which species of bacteria were present, they used a standard DNA-sequencing approach called 16S amplicon sequencing that can identify bacteria to the genus-level (Lactococcus, Leuconostoc, Pseudomonas etc.). They also used a DNA fingerprinting approach called DGGE to identify the composition of yeasts present. They were particularly interested in the identities of lactic acid bacteria since they play a significant role in sourdough fermentation, and therefore plated out samples of dough before and after fermentation on a special medium to identify specific strains of lactic acid bacteria.

Before the dough was fermented, there were clear differences in the composition of microbes across the four farming system treatments. For example, conventional flour had a higher amount of lactic acid bacteria from the genus Leuconostoc (19%) compared to other farming systems and the two organic treatments had high amounts of Pseudomonas. After fermentation, there were still differences among the four farming systems. The green manure organic system had the highest amount of Lactobacillus and the no inputs treatment had the greatest amount of Leuconostoc. The yeasts were not impacted by the farming treatment, and several species of yeast (Saccharomyces cerevisiae, Candida humilis/Kazachstania barnettii, Saccharomyces bayanus/Kazachstania sp.) were found across all treatments.

Impact of farming system on the microbiology of sourdough. 7 groups of bacteria, represented by the green rows, were the most abundant groups detected across the fresh dough and the fermented dough. Intensity of color indicates how abundant the different bacteria were in the farming system treatments. Note that this is an approximation of the detailed data presented in the paper. See Figure 2 of paper cited below for more details.


So did these microbial differences created on the farm and amplified during fermentation translate into impacts on sourdough appearance and flavor? The researchers found that the organic breads were superior in terms of specific volume (measured with this crazy gadget), crumb structure, and crust color compared to the conventional farming system. In a surprising twist, the farming system where no inputs were added had the best bread structure. The authors noted that despite the high quality of bread from the no input wheat, the overall productivity of those fields was so low that it wouldn’t make economic sense to grow wheat in that manner.

Previous studies have shown that while the bread-making environment can be a source of sourdough microbes, the flour used to make bread is a major source for the microbes that ferment sourdough. But it was unclear if wheat production practices had an impact on the composition of microbes in the flour. This study clearly demonstrates that how the wheat is produced can have trickle-down effects through a food system all the way to the color of the bread’s crust!

This study opens up many unanswered questions. Why are the microbes different across the farming systems? Is it because the farming systems promote the growth of different types of bacteria on the wheat which then make it into the flour? Could the organic treatments, such as application of manure, actually serve as a source of microbes that make it to the sourdough? Or maybe the farming systems change the chemistry (nutrients available for microbes) of the plants and that difference in chemistry impacts which microbes grow during the fermentation. How much of the observed effects come from changes in the composition of the flour itself irrespective of the microbes that are present?

One big question for me: what about the bread flavor? The authors measured quite a few chemical aspect of the breads, from phenolic compounds to free amino acids. They noted that there were some differences and these differences were correlated with the types of microbes present, but this part of the paper was a bit fuzzy to me. More importantly, they didn’t use sensory analysis to see what type of bread people preferred based on taste.

Are there practical implications of this study? It’s really hard to know at this point until we better understand answers to some of the questions mentioned above. Many sourdough bread producers swear that a specific type of flour gives them their best breads. Part of the reason may be how the wheat was grown. And perhaps microbes are also involved.

Could similar impacts of how raw materials are produced be observed in other fermented foods? As we described in a previous post, there is some evidence from fermented vegetables that initial raw materials can vary in their microbiomes and this variation can lead to impacts on the final fermented product. It’s likely that we’ll see more studies like this that explore microbial farm-to-ferment connections.

Post written by Benjamin Wolfe

For more details on this study, please check out the full article here: Rizello et al. “Organic cultivation of Triticum turgidum subsp. durum is reflected in the flour-sourdough fermentation-bread axis” Applied and Environmental Microbiology 81 (2015): 3192–3204. 

There are 7 comments on this article

  • Nicole says:

    Very interesting! I would be curious to also see how the milling method affected the fermentation – stone milled flour vs impact milled. Thank you for sharing, this was a great read!

    Reply to this comment
    • john downes says:

      I have been a commercial sourdough baker, and I know from experience that roller milled flour produces poor leavens on every measurable parameter…as soon as i change back to stoneground biodynamic wholemeal, the leavens are sweeter, a lot more active and produce better bread by sensory analysis, which is how bread should be evaluated.

      Reply to this comment
  • Very interesting! I’ve observed from my own experiences with my sourdough starters that the type of flour I use does have an impact on the final bread. I’ve also noticed that the character of my starter changes according to the flour that I use. I would be interested in being involved in sensory experiments of this kind-a creative way to put my wine tasting expertise to use!

    Reply to this comment
  • Eric says:

    Awesome find. Excellent read. This highlights the *complexity* of wild fermented breads and other foods. It’s that very complexity that we savor and seek, and it underscores that by seeking efficiency of production (in the raw materials as well as the value-added products) we often reduce the complexity of a product and thus the overall flavors.

    Reply to this comment
  • David Chenelle says:

    Interesting but this “study” is full of holes in many ways. Just to name a few. It is not a blind study. Therefore the authors bring in their own bias and prejudices to the table. Another glaring fault is all of this was produced in one farm. Big huge mistake although I thank them for their honesty on this one study. It takes hundreds of studies just to prove a theory. This needs to happen on several farms with scientist that are not privy to what is grown from where and from locations that have different weather patterns. Finally there ism’t a sensory application to a FOOD experiment. Good grief, It is like spending all kinds of time studying a car and not driving it.

    Reply to this comment
    • Hi David –

      Thanks for your thoughts on this study. I agree that there are limitations to this study (as with all scientific studies). But this was the first of its kind in sourdough research. Now that this initial study at one farm was completed, other future studies can see how results hold up across many different farms or farming regions. I also hope that there will be some followup work to look at how all the variables measured actually impact how the bread tastes.

      In terms of biases, it would be interesting to know more details on how the samples were processed by the researchers. Perhaps they did in fact have some way of making the sample origins unknown as they were processing them to make sure they weren’t biased. We do this all of the time in my lab when we are processing samples. Unfortunately, the scientists didn’t report in their methods if they did this.

      Thanks for reading and posting a comment!


      Reply to this comment
  • Interesting read. I also wonder about the microbes found ie is one better than another?

    Reply to this comment

Leave a Reply to Susan Fairbairn Cancel reply

Your email address will not be published. Required fields are marked *