Fermented Hot Sauce: Keep Kahm, Ferment On

A growing number of us at the lab are experimenting at work and at home with fermented foods. It's a natural extension of our work with beer and brewing yeasts, since once you've learned one fermentation, many of the skills are transferrable to another type. As a result we have staff experimenting with makgeolli, sauerkraut, tempeh, enormous quantities of sourdough bread (the people run on about as much carbs as the yeast), in an ever expanding list of bubbling jars and "what the heck is that in the incubator". One of the goals of all of this is to discover new microbes we can apply to new contexts, like taking a Lactobacillus from sourdough and using to to make a sour beer.

Available now to brewery customers ;-)

One of the biggest food fermentation projects of the year is our fermented hot sauce, which we have now called E.L. Fuego Fermented Hot Sauce. Due to demand last year, we scaled up production this year and produced several cases of bottles, with two different blends. The peppers are fermented in a 2% salt brine in a food-safe bucket, using weights to keep the peppers submerged. The bucket is fitted with a lid and and airlock. It's essentially a scaled-up version of the mason jar + airlock combo seen far and wide in home fermentation.

Plant Manager Evan Graham moonlights as hot sauce co-conspirator.
Check out that personal protective equipment!

Due to timing, one batch of peppers was fermented several weeks prior to the remaining two batches. In this case, the fermentation pH dropped to ~3.3. Seeing the strong performance of the first batch, we saved some of the brine to inoculate into the subsequent batches of pepper ferments to encourage a healthy and rapid ferment. Those also dropped pH quite aggressively, between pH 3.2 and 3.4. In all cases the pepper ferments were done within 2-3 weeks.

This time, we did notice something interesting: some of the ferments had formed what we would call a 'pellicle', and what the home fermentation world calls 'kahm yeast'. This is a thin, white-ish layer of microbial growth on the surface of a ferment. I guess the lid didn't fit tight enough on those, and let some oxygen in.

Habanero pepper ferment, with some evidence of 'kahm yeast' formation. 

I've always wondered about whether beer pellicles and kahm yeast are the same thing, so I started to do what any qualified scientist does. That's right, I started Googling. A quick search of Google Scholar for the term 'kahm yeast' reveals surprisingly little in terms of information. It seems thatover the course of time this phenomenon has been referred to as many different terms: 'mycoderma' (literally fungus-skin), 'flowers of wine' (I like this one), 'pellicle', 'biofilm', and 'kahm yeast'.

It seems like these various terms are all used to refer to the same common phenomenon in ferments: when a liquid fermentation containing certain microbes is exposed to some degree of oxygen (from the air), those microbes form a film on top of the liquid where they presumably have better access to oxygen. Many of these microbes need oxygen for critical metabolic activities, so there's a big benefit to them in being able to float on top of the liquid.

Kahm yeast/pellicle on a ferment. Some look like this, others 'flaky' or 'bubbly' but all are formed by fermentation microbes.

And that's a really important point to make with pellicles/kahm yeast: yes, it indicates the presence of oxygen, but it also indicates that certain microbes are being fed and are growing. This could have either positive or negative consequences for your fermented product, depending on factors like what the specific microbes are, and timing.

As an example, a common pellicle-forming organism in wild beers is Pichia membranifaciens, a yeast which can produce a pellicle in as little as two days. It is a regular player in Lambic fermentation and present through most of the fermentation and aging process. Given that this yeast is present in some of the most sought-after beers in the world, we definitely can't discount that it might be doing something special, although science has yet to catch up. However, if this yeast sees too much oxygen for too long, it can metabolize the ethanol in a beer or wine and produce acetic acid (vinegary) or ethyl acetate (solventy), both considered off-flavours when there is too much.

A sample of kahm/pellicle, under the microscope (400x) reveals that in this instance it mostly contains yeast, which in this case are mostly apiculate (sausage-shaped)

Enhance! The same sample under 1000x magnification. Here we can even see the vacuoles (the yeast cell's recycling factories) as circles within the cells.

This is presumably why kahm yeast is often regarded as a negative in fermentation circles, since it usually also signals a fermentation that has sat for too long exposed to air where there is a greater potential for off flavours. I would, however, like to counter that the microbes that produce the film do have potential to add flavour and complexity to a ferment, like with any mixed culture.

It all comes down to timing. In our ferments, a bucket of sweet peppers which was initially fairly boring transformed into something totally different, with notes of smoke and light funk. Even more magical, a bucket of habaneros (above, in all their kahm-y glory) ended up smelling like a basket of guavas, the fermentation enhancing the fruitiness of these peppers.

This led me to wonder a couple things: why is kahm yeast so vilified, and what the heck is actually in there?

Agar plating revealed several different bacterial and yeast colony types.

As you can see in the microscope images and agar plate above, there was both yeast and bacteria present in the sample. In the case of the agar plate, I plated out the brine itself, hoping to also find the bacteria, which were not obviously visible in the straight kahm/pellicle sample.

I was able to obtain pure cultures of four different yeasts and three different bacteria, which seemed to be the dominant players in the pepper ferments. There could have been other microbes, but this was a bit of a side project from my usual gig of selling yeast and solving yeast problems.

Luckily, our lab techs were also interested in this project and took the pure culture samples and identified them using ITS sequencing (yeast) and 16s sequencing (bacteria). Here are the dominant players in the hot sauce ferments:

• Hot Sauce Bacteria 1 - Lactobacillus plantarum/paraplantarum
• Hot Sauce Bacteria 2 - Lactobacillus brevis
• Hot Sauce Bacteria 3 - Lactobacillus brevis/parabrevis
• Hot Sauce Yeast 1 - Pichia kluyveri
• Hot Sauce Yeast 2 - Pichia anomala
• Hot Sauce Yeast 3 - Brettanomyces custersianus
• Hot Sauce Yeast 4 - Zygoascus hellenicus/meyerae

Whoa! So we see a lot of the same Lactobacillus that often show up in sour beer, such as L. plantarum and L. brevis. A follow-up wort acidification trial with these showed terminal pH of 3.4-3.7 (the L. plantarum strain seemed to perform the best) indicating they may be useful in sour beer production in the future. Flamin' Hot Lacto, anyone?

The yeasts are also interesting. There's Pichia again — two different species, both of which commonly show up in fermentation environments. P. kluyveri has also been used in non alcoholic beer development. Also fascinating was the isolation of Brettanomyces custersianus, a species of Brett that shows up in ferments with less frequency than B. bruxellensis and B. anomalus, but which does appear to be present in some food ferments including spontaneously fermented beers. What can this strain do? I'm eager to find out. The last one (unclear whether it is Z. hellenicus or Z. meyerae from the available data) is a bit of an oddball, but does appear to be present in some wild-fermented wines and some spoiled wines.

I guess that's the whole point here, isn't it? One person's 'wild' is another person's 'spoiled'. I for one am really excited about what we've learned here, and I am hoping that we can apply what we've learned about vegetable ferments and the microbes involved to enhance positive flavours and avoid off-flavours. I'm also hoping we can start to see more microbial cross-pollination across fields, as this is a way to enhance product diversity. I'm excited by the prospect of using our hot sauce microbes to make a sour beer, or using a fruity Lacto strain to brighten flavour in a vegetable ferment.

Recommended Reading

• Wild Fermentation by Sandor Katz
• The Noma Guide to Fermentation by René Redzepi and David Zilber