Sugar-Free Juice

I want to try extracting some things from juice - like organic acids and pigments and polyphenols - and I think this will be easier if I get rid of the sugar. But how does one get rid of the sugar in juice? Here's my plan.

If you aerate yeast, it's more likely to use aerobic respiration than fermentation. This is called the Pasteur effect. So let's put an aquarium pump and some yeast in bottle of juice and see what it does? I don't think that will quite work; lots of yeasts will still ferment sugars when aerated if there's a lot of sugar around - like in juice. This is called the Crabtree effect.

So we could dilute the juice to reduce its sugar content, but I estimate you have to dilute by something like 1000 times to prevent the Crabtree effect when using saccharomyces cerevisiae on juice. And that's not a home scale experiment. 

Another option is that we could find a food-grade yeast that doesn't have the Crabtree effect. Kluyveromyces species fit this bill. Kluyveromyces species are pretty good at doing just respiration on carbohydrates when aerated, and they're found in live-culture kefir.

I don't know the composition of yeast and bacteria in the kefir itself (the fermented milk) as well as the composition in the kefir grains; the latter is the thing that's analyzed in detail in biology publications. Here are some quotes about the grains:

"Kefir grain is composed of a diverse spectrum of species and genera including lactic acid bacteria (Lactobacillus, Lactococcus, Leuconostoc), yeasts (Kluyveromyces, Candida, Saccharomyces, and Pichia) and sometimes acetic acid bacteria (Acetobacter) in a symbiotic association."

"A complex and highly variable symbiotic community can be found in these grains, which can include acetic acid bacteria (such as Acetobacter aceti and A. rasens), yeasts (such as Candida kefyr and Saccharomyces cerevisiae) and a number of Lactobacillus species, such as L. parakefiri, L. kefiranofaciens (and subsp. kefirgranum), L. kefiri, etc."

"Many different species of Saccharomyces have been isolated from kefir, however, S. cerevisiae and S. unisporus are the most common and present in many varieties (Angulo et al., 1993; Marquina et al., 2002; Latorre-García et al., 2007; Diosma et al., 2014). Kluyveromyces make up the majority or entirety of the lactose utilizing yeast population, with K. marxianus and K. lactis being the two most common species (Simova et al., 2002; Latorre-García et al., 2007; Diosma et al., 2014). The Candida population is made up of a wide range of species with C. holmii and C. kefyr being the most prevalent (Angulo et al., 1993; Marquina et al., 2002). Outside of these three genera, only Pichia has been identified with any regularity and in each case the species was identified as Pichia fermentans."

"As fermentation progresses the proportions of some yeast species change with non-lactose fermenting yeasts, such as Saccharomyces, decreasing, whereas lactose utilizing K. marxianus and K. lactis show a similar distribution between grain and kefir (Simova et al., 2002)."

Based on the last quote, I'm hopeful that I can use a well fermented kefir's liquid as a purer source of Kluyveromyces yeast than the grains would be.

Once source, "Survivability of Kluyveromyces marxianus Isolated From Korean Kefir in a Simulated Gastrointestinal Environment" says that "Kluyveromyces marxianus accounts for > 90% of the yeast population of kefir", so our isolation is already well in progress.

I've seen a procedure for isolating Kluyveromyces that used kefir directly (the fermented milk) rather than kefir grains. You take a wire loop and dip it in the kefir and scratch it over an agar plate enriched with a growth medium favorable to yeasts and then grab a sample of the the right yeast by visual discrimination once some spots have grown on the plate. 

I think that would work fine, but I don't want to do it. I'm not really up to mixing a growth medium, or making an agar plate, or waiting for cultures to grow on agar. For this to be an experiment worth my time, it has to be a lot less time consuming than that. But I think I have a solution!

Kluyveromyces species are famously heat tolerant as yeasts. Some quotes:

"The thermotolerance of different K. marxianus strains has already been documented, where it has been observed to grow at temperatures of up to 52°C and to ferment ethanol at temperatures between 38 and 45°C (Choudhary et al., 2016)."

"The thermotolerant yeast Kluyveromyces marxianus, growing at high temperature (45℃) , showed stronger survival under heat shock at 50℃ than the brewing yeast Saccharomyces cerevisiae, which was unable to grow at 45℃. The survival rate of K. marxianus decreased to 10% during heat shock at 50℃ for 20 min, and to less than 0.01% at 60℃ for 20 min. Cells with damaged cellular membranes were infrequently observed at 50℃ and had decreased significantly from heat shock at 60℃. The metabolic activity of K. marxianus was retained at 50℃, whereas that of S. cerevisiae was not."

So my plan is get some Kluyveromyces from kefir, and heat it to 50℃ for long enough to reduce other yeasts (and hopefully bacteria) without killing the Kluyveromyces, i.e. perhaps somewhat less than 20 minutes so that I don't decrease it all the way down to 10%.

I think this temperature and exposure length will kill Lactococcus and Leuconostoc and Acetobacter, but not all Lactobacillus.

So that's where I'm stuck at the moment. I think I can kill the other yeasts, but I want to also kill of the other bacteria. And then I'll maybe build up the Kluyveromyces culture a little by feeding it some sugar and add it to a container of juice with an air pump. Maybe there will be a HEPA filter or two somewhere for sterility.

Instead of using Kluyveromyces, I've also put some thought into protocols that remove sugar from juice by fermenting the sugar to alcohol using the standard Saccharomyces and then removing the alcohol, perhaps simultaneously with the ferment or in several alternating stages. I'm not optimistic about that, but I haven't given up on it entirely.

...

Anyway, the next task is selection for Kluyveromyces over Lactobaccilus. I think Kluyveromyces might be a little more tolerant of low pH, but I'm not sure and even if I'm right, I don't think it's a big enough effect that I'd want to rely on it for selection. I've heard of using a protein called lysozyme as an anti-bacterial agent against gram-positive bacteria, and Lactobaccilus species are gram-positive, but apparently their response to lysozyme is strain dependent and I don't have any to begin with. Perhaps I could find a food that Kluyveromyces can metabolize but Lactobaccilus can't? That might be hard: lactobaccilus species can produce at least lactase, proteases, fructanases, amylases, bile salt hydrolases, phytases, and esterases. I hear most Lactobaccilus species are disparaged by the presence of hops in beer, but Lactobacillus brevis isn't.

...

You know, if I can't separate the two, maybe it's not the biggest problem? I'll still get juice that's free of simple sugars and alcohol. It might have extra lactic acid. Sad. Also, I'm a little worried that if I have a yeast and a bacteria, then that's a lot of metabolic machinery, and other things in juice that I wanted to extract might get eaten - no more esters, no more weird polysaccharides, no more variety. 

...

Hm! Here's an interesting quote: "Because they are unable to synthesize cytochromes, Streptococci cannot carry out oxidative phosphorylation. They are able to ferment sugars, but the end product is always lactic acid. Therefore, Streptococci are very acid tolerant and count among the lactic acid bacteria order."

Cytochromes are proteins that can perform electron transport because they have a heme group, a little tetrapyrole with an iron atom in the middle. So I looked into Streptococcus thermophilus, a lactic acid bacterium which is also a streptococcus, and which is used in dairy fermentation. And I got some different quotes from a source that I trust more, "Metabolic Profiles of Carbohydrates in Streptococcus thermophilus During pH-Controlled Batch Fermentation":

"S. thermophilus ferment sugars by the Embden-Meyerhoff-Parnas (EMP) pathway to pyruvate, which is converted into lactic acid by lactate dehydrogenase (LDH)."

"Streptococcus thermophilus is a typical homofermentative organism but also produces other metabolites, such as acetic acid, formic acid, acetaldehyde, and ethanol."

"Mixed-acid metabolism, a type of homofermentation, is characterized by the production of formate, acetate, ethanol, and/or CO2 in addition to lactate. And the homolactic metabolism can be shifted to a mixed-acid metabolism under certain conditions (carbon limitation, carbon excess of slowly metabolized sugars) (Mayo et al., 2015). However, the mechanisms underlying the shift from homolactic fermentation to mixed-acid fermentation have been the subject of considerable controversy (Papagianni et al., 2007). The primary metabolites and genes of carbohydrate metabolism in LAB have been studied."

If I'm reading table 1 correctly, it looks like S. thermophilus produces about 5000 times as much lactic acid as ethanol, so maybe that would be fine. Maybe I could take some juice, and turn the sugar into lactic acid, and then neutralize that with a base to get at all the other non-sugar and non-acid things.

...