Brettanomyces is a yeast loved by sour beer brewers and feared by wineries. It causes billions of dollars in losses annually to the wine, soda, and fruit juice industry. It also is the foundation of most sour beer styles. Despite this broad importance, we do not know where while Brettanomyces lives in the wild. Which makes it pretty hard to both control it when we don’t want it, and to harvest it when we do. Any brewer whose tried to cool ship a beer has felt this pain, as we rarely capture wild Brettanomyces. A recent and seemingly unpublished paper (thanks Dan for digging it up) found the same in traditional Lambic brewing – beer leaving coolships lacked Brettanomyces, which instead appears in the barrels. In the spring of 2022 I came, by chance, across some data in a scientific article that claimed to find wild Brettanomyces in the rhizosphere – a group of organisms that live in association with the roots of plants. These organisms are often ‘fed’ by the plant, and in turn provide the plant with other nutrients, or protect the plant from pathogens.
A deeper dive into the scientific literature and found a number of other studies that supported this idea, and which identified wild Brettanomyces on the roots of numerous organisms including corn, dill, beans, rice, and other plants. I was so excited about this finding that I made a video on it, going over the science as it stood in June of 2022. I’ve embedded the video below if you want a quick overview of what I found.
This led in the summer of 2022 to a deliberate attempt to sample the rhizosphere across my farm to see if I could find wild Brettanomyces in this environmental niche. My exploration of this possibility did lead to the isolation of some wild Brettanomyces, culminating in some disappointing fermentation tests. The full series of blog posts on those experiments can be found below:
- Part I: These Are Not The Yeast You Are Looking For
- Part II: These ARE The Yeast Your Are Looking For
- Part III: Wild Brettanomyces – a Needle in a Haystack
- Part IV: Fermentation characteristics
The short version is that while I was able to isolate Brettanomyces from plant roots, it was extremely difficult to do so. Largely because a different yeast – Debaryomyces hansenii would take over the culture. Which brings me to the spring of 2023 and the next part of this project. Rather than stretching things out over multiple posts, I will instead summarize the summers work in here.
Designing Selective Media
In my 2022 isolation attempts I generated a simple selective medium which I hoped would help isolate Brettanomyces from the thousands of species of bacteria, yeasts, and fungi that live in the soil. To try and selectively grow out wild Brettanomyces I created a selective medium that would kill most bacteria (due to the presence of two antibiotics) and most yeast/fungi (due to the presence of cycloheximide). Unfortunately, Brettanomyces isn’t the only cycloheximide-resistant yeast out there, with Debaryomyces being similarly resistant, but also much faster growing. Thus, while I’d see some Brettanomyces early into the isolation process, it would often get out-grown by Debaryomyces.
My goal in 2023 was to refine this approach, in an attempt to generate a medium that would more effectively enrich for Brettanomyces. And, as any good science does, this started with a deep dive into the scientific literature. Specifically, in a search for chemicals that could be added that would be toxic to Debaromyces, but not to Brettanomyces.
This was not an easy search, but eventually I did find a few things to try. Firstly, while many pof the yeasts and fungi I was trying to select against were cycloheximide-sensitive, Debaryomyces and a few other genera (Kluveromyces, as one example) are fairly resistant to it. But I did find a few papers that indicated that Debaromyces is less resistant to cycloheximide than Brettanomyces, with the former struggling at concentrations above 50 mg/L, while the latter are resistant to over 100 mg/L.
Another interesting chemical compound that may be selective is acetic acid – yes, as in vinegar! Brettanomyces grows just fine up to 120 mM (~0.7%) acetic acid, while Debaromyces (and many other yeasts) are killed at doses from 25 mM to 40 mM. One of the nice things about vinegar is that it is cheap, easily available, and unlike cycloheximide, is not horrifically toxic to humans!
My previous selective medium contained antibiotics + cycloheximide, and did a generally good job of suppressing fungi and bacteria, so I continued to use it as a base medium. Based on my new findings, I formulated a couple of media recipes to try. I made all of these as agar plates, but they are easily made as liquid media by simply excluding the agar:
Base Medium: Prepare and auotclave. Cool to ~60C then add the selective agents and cast as plates. A total of 250 mL will be prepared
Original Medium: 12.5 mg each ampicillin, chloramphenicol and cycloheximide (50 mg/L final concentration). Dissolve in 20 mL of sterile water.
Chx Medium: Same as original medium, but using 25 mg (100 mg/L) cycloheximide.
Ethanol Medium: Same as Chx medium, except dissolve antibiotics in 20 mL of 40% ethanol (vodka) to give a final concentration of ~3.2% ABV.
Vinegar Medium: Same as Chx medium, except dissolve antibiotics in 15 mL of 5% distilled (white) vinegar + 5 mL of sterile water.
High Selection Medium: Reduce water in base medium to 215 mL. Dissolve the antibiotics as per the Chx medium in 20 mL vodka + 15 mL of 5% distilled (white) vinegar.
Testing the Media
Once I had prepared the various media I began testing their selectivity, using a strain of Brettanomyces bruxellensis I isolated from my solera, and a strain of Debaryomyces hansenii isolated from soil back in the summer of 2022. I streaked out each organism on each medium, and grew them for 14 days at room temperature. Results are scored as follows:
- Negative (-): no visible growth after 14 days
- Weak positive (+): small, scattered colonies after 14 days
- Medium positive (++): medium-sized colonies at modest density after 14 days
- Strong positive (+++): sufficient growth that identifying single colonies is difficult
Things look promising – Brettanomyces grew on everything, while the vinegar and high-selection medium suppressed Debaryomyces quite potentially. So the next step was to test these using rhizosphere soil.
First Rhizosphere Test
For my first test I decided to plate rhizosphere samples directly onto the vinegar and high-selection plates prepared above. This process was fairly simple:
- Shake as much dirt as possible off of roots.
- Place roots in sterile water and shake vigorously to remove rhizosphere soil.
- Allow the soil to settle out of suspension for 10 minutes.
- Spread-plate 2-3 drops of the water.
I did this using roots from corn and beans, both of which have been found to house Brettanomyces in the literature. But I forgot the lessons of 2022, and while I didn’t have the forsight to take photos, I ended up with plates similar to my first attempts back in 2022. Plates dominated by moulds.
Liquid Medium Tests
To limit mould growth, in 2022 I ran the samples through an enrichment culture in which I added the soil samples to liquid medium, and then capped the tubes with vegetable oil. The oil prevents oxygen from reaching the culture. Moulds are obligate aerobes, so this prevents them from growing, while the yeast are able to ferment the medium and thus expand their population. But before testing this with more soil samples, I tested the ability of my purified Brettanomyces and Debaryomyces cultures to grow in liquid medium. I prepared a fresh lot of each media, leaving out the agar this time to produce liquid medium. 5 mL of each medium was placed into tubes, and inoculated with a single colony of my test organisms. As before, these were grown for 14 days at room temperature, and scored as follows:
- Negative (-): no turbidity after 14 days
- Weak positive (+): faint turbidity after 14 days
- Medium positive (++): modest turbidity after 14 days
- Strong positive (+++): sufficient turbidity to prevent viewing objects through medium (image above)
Here things were not as selective, with the Deabaryomyces showing growth in all but the “High-Selection” medium. So I went forward with the high-selection medium for a test from soil.
Second Rhizosphere Test
For this test I generated a series of 5 mL sterile tubes of high-selection medium. These were capped tightly and shaken vigorously to oxygenate, and then 2-3 drops of rhizosphere suspension added to each tube (these were prepared in the same manner as the suspensions added to the plates). I sampled the roots of 20 plants from around my home, including everything from corn and beans, to barley, hops, grapes, dandelions, and other domesticated and wild plants. These were allowed to start growing for 24 hours and then capped with oil. 14 days later I assessed them all for growth…and every single was lacked any measurable growth.
Undeterred, I plated out each sample onto plain YPD agar; as in a yeast-friendly medium without anything to suppress bacteria, yeasts, or moulds…again, after 14 days there was no growth.
As a last-ditch attempt I took the now 28-day cultures and concentrated them by centrifugation, placing the samples under my microscope. While there were some debris, there was nothing that looked like bacteria or yeast.
Third Rhizosphere Test
At this point I was well into fall and was running out of time to work with living plants. So I made a third hail-Mary attempt to enrich wild Brettanomyces from some rhizosphere samples. I figured I had over-diluted the samples in attempt 2, so this time I took a different approach. Instead of trying to wash the organisms off of the rhizosphere, I instead placed lightly rinsed roots directly into medium. In theory, this should provide a fairly large inoculum of rhizospheric organisms. The photo’s below were taken yesterday (about 5 months later) and summarize the growth I observed:
In polite company I would describe the growth as “bupkis”, “nadda”, “nothing” and “none”.
On the bright side I prevented the growth of Debaryomyces and other problematic yeasts…on the other hand, I also killed everything in the culture.
I am uncertain why I do not have growth at this point. One possible explanation is that the Brettanomyces I am using as a test organism is brewery-adapted and therefore is more resilient to growth in liquid medium. If this were the case, using a less stringent medium may work better. We’re in the depths of winter, so it will be a while until I can restart tests, but I will need to try inoculating my original 5 test medium with rhizosphere samples to see which support growth, and what grows. With that information I may be able to generate a better selective medium and further refine my isolation procedures.