Food Poisoning: National Health Service 0 – uBiome 1


The culprit

Food poisoning is not a lot of fun. Especially when your immune system is dysfunctional as mine is from ME/CFS.

It was my mother’s birthday in August, and we took her for a meal in a nice village pub near Bath, England. I ordered a very tasty looking quail for the starter. It was a char-grilled piece and around half way through I turned it over and noticed that the other half was barely cooked. In fact the middle was completely raw. I had already eaten a great deal of it without really paying attention. Stupid!

Anyway, I thought it was likely to give me food poisoning but there wasn’t much I could do but wait and see. I didn’t have to wait long. Around four or five hours later my stomach was very angry. I felt extremely nauseous, though strangely did not vomit. It was terrible at the other end though…that’s all I’m going to say about symptoms. This went on for three days in which time I didn’t eat anything else. Then I started to improve a bit and thought the ordeal was over. Not so quick.

I started eating some mild foods, and then some stronger foods after a couple of days, but after about three or four days later I began feeling worse again. It got so bad I had to stop eating altogether for a second time. This time for a little over two days. I repeated the same process again but after around a week I started getting ill for a third time!! What the fuck!?!

By this time I had been to the doctors half a dozen times. My stomach in particular was so painful and I had provided two stool tests which came back normal. I was told that sometimes they couldn’t culture things very well. This NHS had no plan B for dealing with this, it’s good that the doctors were sympathetic but what I really needed was an avenue we could go down to figure things out, and there simply wasn’t one. Great. It seems ridiculous to me, to be trying to grow the bug in a dish when we are so rubbish at it! With ME/CFS I am used to having to doctor myself, so I sat down and gave it some thought. Could uBiome help, I wondered?
16S rRNA sequencing should pick up whatever is in the sample, it doesn’t need to be alive, it doesn’t need to be grown. I had a spare uBiome kit lying around so I sent off a sample.

I then had another four days without any food. Even in the days between when I had had food it had been very bland and not very nutritious. I was very weak, had lost a lot of weight and was pretty desperate and so I held out as long as I could in the hope of killing whatever it was off. Fasting for four days is not easy, especially when you are sick with food poisoning, ulcerative colitis and ME/CFS. But I managed it. I couldn’t do any longer, I had to eat. After this I was well enough to eat bland food for two weeks, then move onto more advanced food. I am still not right now, a month and a half on. I’d say 80% better. Progress is very slow.

My uBiome results arrived yesterday. It didn’t take me long to spot the problem: Campylobacter ureolyticus.
This was not present in any of my previous uBiome tests, and Campylobacter is a very common cause of food poisoning, especially via poultry; the avian intestinal tract having been identified as the main environmental niche. It is also a pathogen that was originally identified via molecular, rather than classical culture techniques, suggesting we aren’t great at growing it…The two most common species of Campylobacter recorded to cause gastroenteritis are C. jejuni and C. coli but there is acknowledgement that C. ureolyticus might be under reported, with some studies suggesting C. ureolyticus (24.4%) occurs more frequently than C. coli (6.7%). The most common species identified being C. jejuni (72.4%).

I think that is a pretty awesome result. The NHS couldn’t figure it out, but uBiome could. Maybe one day the NHS will catch up?

Further reading, for geeks:
O’Donovan D et al. Virulence. 2014;5. Campylobacter ureolyticus: a portrait of the pathogen.

Gut Wars: Diversity

Following on from my recent guest blog for uBiome, I decided to take a closer look specifically at my gut diversity. Here are the counts for comparison between my first test in March 2014, shortly after antibiotic treatment, and a year later in March 2015:

Interesting, I’m sure you’ll agree.

This has me wondering what normal is? I mean, uBiome’s norm when looking across all their samples, I wonder what that looks like. I don’t know the answer to that at the moment, and it may not be that relevant for the average person, but if you’ve been on antibiotics or had some other loss of gut diversity then it would certainly be useful to know. Though even some results from just a handful of “healthy” people would give a fair indication, I’d bet.

It is almost certainly an improvement that I now have twice the level of species diversity I had a year ago (leaving aside which species these actually are, as that is also likely important), but how much further have I to go? Is the norm 100, 150, or 500?

If you’re reading this and you have taken a uBiome test then I’d love it if you posted your stats for comparison. It’s quick and easy.

The easiest and quickest way to find this info is to click “Raw Taxonomy” in your uBiome dashboard. Then Ctrl+F to run a search for “phylum (beware that Subphylum and Superphylum will be counted if you don’t specify the leading inverted commas in your search) and then “class – “order – “family – “genus – one at a time and see how many hits you get in each case. I suggest using Chrome browser rather than Firefox which won’t return more than 100 hits.

Alternatively you can put your results into a spreadsheet, great instructions for which can be found on Richard Sprague’s blog. This option makes it easy to get those counts but also to manipulate your data in any other way you fancy going forward.

So do me a favour and post your diversity counts.

I’ll just finish this post off with a quick look at which of my bacteria went extinct in the last year. Given my diversity has vastly increased, I’m not expecting many. Here’s the species level data (courtesy of Richard Sprague’s analyser):


The disappearance of C. butyricum is interesting, it can be pathogenic but it is also widely used as a pro-biotic in Asia as a treatment for it’s misbehaving cousin C. diff. And as you can guess from its name it is a butyrate producing bacteria. Last year C. butyricum made up more than 1% of my total gut flora, which is much more than usual. Now it is gone. In fact, many of my clostridium have been kicked out, half of these now-extinct species are clostridium, many of which are normal in the human gut but also can be pathogenic. Last year the clostridium genus made up 8.63% of my gut whereas now it is at a normal level (1.56%), so probably good on the whole.

B. catenulatum is gone too, which seems a shame. I guess it has been out competed by the other species of bifidobacterium that I introduced over the last year.

L. lactis, often present in dairy products such as cheese, is gone. Why? No idea.

I’m pissed to see R. bromii is gone. It’s considered a cornerstone species and I was pleased that after my antibiotics, when I first tested, there was a tiny amount still holding on. But now it is gone. Why are my ruminococcus abandoning me? I’ll be looking at which cornerstone species are missing from my gut next. R. bromii is not the only one, and these key species being absent does not bode well for my future gut health.

That aside for now, when you consider that these 19 species pictured above were missing from my second sample, it makes my increase in species level diversity even more impressive! I started with 55 species, lost 19 of these and gained 73 new species. I must be doing something right – or at least, More things right, than wrong!

Bacteria: Dishing the Dirt on Soil

I recently posted about the different probiotic supplements available on the market, and the frustrating reality that much of what we might want isn’t actually available in supplement form. A few soil-based probiotic supplements are available, and might be an avenue worth pursuing, but these are almost exclusively not bacterium that are normal members of our gut flora – so I am yet to be sold on their benefit.

But soil is probably where the answer lies. It is absolutely chock-a-block with the stuff, including some genera that are just as at home in the human gut. So why don’t we have loads more soil-derived probiotics?

The Amiga 500 arrived in 1987. So did our latest antibiotics.

The Amiga 500 arrived in 1987. So did our latest antibiotics.

The problem, is that we are rubbish at growing it in the lab. This is also part of the reason why we haven’t had new antibiotics coming onto the market since 1987, as this is where most of our antibiotics were discovered in the first place. This is what a brand new personal computer looked like in 1987. Cell-phones were in their infancy and didn’t fit in your pocket and the internet wasn’t there. We are fighting today’s bacterial diseases with ludicrously outdated technology. We’re in the stone age.

So it’s positive that a new study reported in Nature earlier this week reports a new method for growing bacteria by burying them in soil. The technique isolates different bacterial cells into separate spaces, and then grows them while buried in the soil. This method looks very promising and could be used to develop new antibiotics. The researchers have already identified a new antibiotic which they have termed Teixobactin, which looks especially promising as it breaks down the cell wall of Gram-positive bacteria, something which would be effective while also likely being hard for bacteria to evolve resistance against. One also wonders if this technique might also help us culture tricky pathogens that already infect us but which we find difficult to grow from samples.

So that soil in your backyard contains all this useful stuff; it is teeming with bacteria, other micro-organisms and the things they produce. Plant’s living in the soil have their own equivalent of our gut biome, known as the rhizosphere which is in the soil around its roots. This has been known about for donkey’s years and probiotics for the soil have long been in use. If you are into your vegetable gardening then you may have even used some yourself. (They are particularly popular for growing beans and peas as certain bacteria allow the plants to fix nitrogen from the air). And as well as the Human Microbiome Project there is an equivalent Earth Microbiome Project.

As we are not born with anything like a full compliment of gut bacteria, but have adult-like populations in our gut by around age two, it seems a reasonable idea that we pick them up in those first couple of years from our environment. Part of that environment is soil.

We still have a lot to learn about the microbiome, whether it is in an animal, in the soil, or elsewhere. The closer we look the more we realise that there is transfer between the soil, ourselves, and other animals. And different bacteria also swap handy genes between these biomes (bacteria don’t really care much for our compartmentalized classifications of where they each belong). When it comes to soil bacteria though, something I find especially interesting is earthworms.


The soil is home to all kinds of creatures, not just microscopic bacteria and other microscopic organisms like fungi but invertebrates also. There are lots of studies looking at the effect earthworms have on bacteria in the soil. In addition to their indirect impact on the structure of soil, they provide, through their worm castings, a ready supply of organic substrates to the microbial community within the soil.

They also have loads of bacteria in their guts and some of this is shed in their castings into the soil. Plants love this stuff and some people grow worm castings (vermicompost) as a soil amendment, which has been shown to increase yields (Arancon NQ et al).

Vermicomposting toilets are also used by some and have been found superior when it comes to mass reduction, pathogen destruction, and compost quality, compared to latrine style microbial composting toilets (Hill GB et al). As well as pathogen destruction there appears to be a bottleneck effect on bacteria, “favouring the existence of a reduced but more active microbial population” as reported in another study looking at earthworm digestion of animal manures (Gomez-Brandon M et al).

You can encourage worms into your garden soil in a variety of ways. The simplest is to have a wormtower, pit/trench compost or utilize a no-dig method of produce growing in which you lay down cardboard to suppress weeds and the worms love it (they use it as bedding as it rots down). I’ve been using the no-dig method recently and it seems to be working well. I also plan to build a wormtower and I’ll blog about it when I do.


Manure obviously has lots of bacteria present and if you add it to your garden then you will be introducing new bacteria into your soil. This is probably a good thing in general, if we assume that what is good for plants is also good for us and with the help of our friendly earthworms we should have some protection against pathogens. But if you fertilize your garden with manure then you ought to be picky about where it comes from, as animals treated with antibiotics may cause an increase in antibiotic resistant bacteria within the soil (Udikovic-Kolic N et al).

All this talk of worm castings, and manures, reminds me of an interesting post over on VegetablePharm about naked mole-rats being fed faeces when nursing. Elephants are also known to feed dung to their young. I’m not suggesting we eat each other’s crap, but it probably points to the use of animal manure for growing veg as a positive thing.

What’s in your soil?

Let's go play in the mud

Let’s go play in mud

There are lots of factors that influence which bacteria are present in the soil and you may able to get some data for free with a few clicks. In the UK, what used to be The Cranfield Soil and AgriFood Institute (CSAI), incorporating the National Soil Resources Institute (NSRI), is a centre within Cranfield University, which provides data on what is in the soil across England and Wales (and I think there may be some data for Scotland and Northern Ireland too).

Seriously cool maps!

The maps allow you to look at any place and see various layers of data displayed as you choice. Want to know the level of lead in your soil? No problem. Arsenic? Sure thing. Minerals, drainage, fertility – it’s all there. But it gets better: they have data on bacteria!!

First though, the type of soil where I live. My soil is described as “Lime-rich loamy and clayey soil with slightly impeded drainage. Highly fertile. Topsoil carbon is low.” (means it would benefit from some leafmold). This is all spot on by the way; it describes my soil perfectly.

Under general cropping guidance it says “Suited to autumn sown crops and grass but shortage of soil moisture can restrict yield, and timeliness with field work is important to avoid structural damage particularly in spring” and this is true, when it’s wet it has a tendency to mud, so you have to avoid stepping on it much of the year. But otherwise it is excellent soil.

Nearby, about a mile away, is also “Shallow lime-rich soils over chalk or limestone” and this is also correct as nearby are the limestone quarries from which Bath Stone comes. In fact directly under my feet is the old cold-war bunker, where the UK government and Royal Family would have sat out any nuclear war. It was decommissioned some years ago at which point Burlington Bunker came to the public attention for the first time. It was built in the miles of old quarry tunnels that run underneath the area. But back to bacteria!! Three ecology indices are available. Here they are with the scores in my area:

CS Topsoil Microbes – Bacterial Simpson Diversity is 0.945-0.950 (that’s very high)

Bacterial Shannon diversity is 3.92 which is the highest bracket on the scale.

Bacterial community structure is 0.32 which again, is high.

So how meaningful are these summary measures? Well, they are a quite rough measures in comparison to 16S rRNA testing of bacterial communities. But still, they suggest that my soil will include a diverse abundance of bacterial types and will be evenly spread – they are all about mathematical probability of equal results in the dataset, so a high number means that the dataset (usually species, but can be genus or family) is diverse. How consistent these sampling methods are across sample areas, I have no idea. But this data is just egging me on, now I want to find out exactly what is in my soil.

There may be similar survey data available in other countries. I looked for something similar in the US, and although the United States Department of Agriculture has a Web Soil Survey with some interesting data, I can’t see any data on bacteria unfortunately.

I’m gonna uBiome my vegetable patch

This year, once the soil warms up, and I have more growing in it, I will be uBioming my garden soil. So we’ll not only see what progress I’ve had in my gut recovering from antibiotics but we’ll also know what is in my the soil where I grow my veggies and how relevant this exposure might actually be to human gut health – what exactly might we be exposing ourselves to if we re-embrace soil in our daily lives.

Image credit: Wikipedia Commons. Create Commons Licence.

Prostaglandin and COX-2: ME/CFS, Ulcerative Colitis and Gut Permeability

When I developed ulcerative colitis, I had been on several months of treatment as prescribed by Prof Kenny De Meirleir, an myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) researcher and clinician. The treatment included antibiotics to treat a suspected chronic Bartonella infection. I also took a bunch of supplements and injections as part of his protocol.

Prior to treatment I had undergone a barrage of immune system tests, including prostaglandin E2 (PGE2) and interleukin 8 (IL-8) both of which were highly elevated, especially PGE2. I had loads of other markers involved with inflammation that were very high. Over the course of treatment, the others reduced, while PGE2 and IL-8 did not. They went up even further.

One of the supplements I was on was berberine, which has a bunch of effects on the gutflora, which aren’t necessarily all good though it’s often touted as a cure-all (but that’s a whole other topic). Something else that Berberine does is reduce PGE2 synthesis (several studies showing that).

Well it didn’t work for me. Here are my results:

Prostaglandin E2 (normal, 0.17-6.4)
Jan 2013 (prior to any treatment): 28.3
August 2013 (after a few months on treatment): 68.95
Dec 2014 (after eight/nine months of treatment): 37.68

Prior to treatment, and before I developed ulcerative colitis (UC), I had significantly elevated PGE2. I am unaware of anyone measuring PGE2 in ME/CFS patients and reporting it as a finding, though I know others who have been tested and have elevated PGE2 and Maes et al. have reported elevated cyclo-oxygenase-2 (COX-2) in ME/CFS patients (which leads to production of PGE2 – more on that in a minute).

Although the cause of UC is also unknown, several pathogenic bacteria have been put forward as culprits, and some are known to induce the expression on PGE2 during mucosal bacterial infection. (Mallory Agard et al.). I developed UC in September 2013 as a direct result of the antibiotics I was taking which significantly disrupted my gut flora. In November I started taking Pentasa (mesalazine). Although the Dec 2013 result was still very high (higher than pre-treatment, in fact) it did drop between August and Dec 2014. Did the mesalazine cause that drop?

COX enxymes (also known as prostaglandin-endoperoxide synthase) catalyze the conversion of arachidonic acid into prostaglandins, including PGE2. Mesalazine has been shown to down regulate the COX-2/PGE2 axis in inflammatory cells in the colonic mucosa of UC patients (Collier HO et al. and Sharon P et al.) So, mesalazine seems to reduce PGE2 via COX-2 inhibition.

Even now, with my UC much better, if I forget a dose of mesalazine I can feel my bodies inflammation ramp right up and it stays that way until I take the mesalazine and it gets to work (slow release, so often this is up to a day). So maybe this is relevant to some people with ME/CFS, not just UC.

And although I’m not going to focus on it in this blog post, it’s interesting to note that activated B-cells highly express COX-2, and that inhibition of COX-2 dramatically reduces B-cell antibody production (Elizabeth P. Ryan et al). Could this be part of why Rituximab shows such great promise in the treatment of ME/CFS (Fluge Ø et al.) and is now in phase III trials?

Prostaglandin Receptors and Gut Permeability

PGE2 is produced by a bunch of tissues including those of the gastrointestinal tract and it interacts with the epithelial cells via specific cell receptors, namely, EP1, EP2, EP3, EP4, each resulting in different biological consequences. So it is not PGE2 alone that determines the biological effect, the receptor type matters. When the colonic mucosa is inflamed, the EP2 receptor on cells are over-expressed (V. Takafuji et al.) and mucosal PGE2 release appears to increase in correlation with disease activity in UC (DS Rampton et al.).

A new study published in PLoS One by Lejeune M et al. suggests that these prostaglandin cell receptors are involved in gut permeability, with the epithelial layer being more permeable — as measured by transepithelial electrical resistance (TER) — in cells that express less EP2 receptors, resulting in a significant loss of the tight junction protein claudin-4 through proteosomal degradation. The study authors managed to show that it was the receptor that was responsible, not prostaglandin itself (as suggested previously by Rodríguez-Lagunas MJ et al.). Now let me dumb that down for when I’m reading this back to myself in a few day’s time: Loss of EP2 receptors on the colonic epithelium can decrease barrier integrity by degrading a particular protein responsible for tight junctions. But it’s the prostaglandin receptor which is important, not the prostaglandin.

PGE2 is also known to induce IL-8 through its affinity for the EP4 receptor (I Dey et al.). That may explain why my IL-8 also remained high while the others did not, and further demonstrates why PGE2 may be important.

The problem is that PGE2 has all kinds of impacts through these four cell receptor types that it interacts with. We are understanding better what these receptors do thanks in no small part to studies using knock down mice (for which we can thank Nobel Prize winner Mario Capecchi who introduced the concept. Wonderfully, he is now involved in the End ME/CFS Project) but there is still a lot we do not know.

There is some evidence that PGE2 is protective of the intestinal mucosa (Morteau O et al.) but remember, some pathogens are known to induce it’s production, so it clearly isn’t all good. But we’re getting a bit of a mixed message here. PGE2 is also ultimately responsible for inducing fever by acting on the hypothalamus which controls the bodies temperature setting. I don’t get fevers (people with ME/CFS usually don’t get colds/flu when exposed, for some unknown reason) so there must be more to it than just having high PGE2, else my high PGE2 would induce a constantly high temperature… I haven’t looked but I’m guessing this might have something to do with those different cell receptor subtypes again. Maybe a reader knows, and can enlighten me.

But let’s say our COX-2/PGE2 is high when it shouldn’t be (perhaps in ME/CFS and perhaps in IBD when we are in remission) then it may be a reasonable step to try and reduce it. If we wanted to try that then as well as mesalazine, a common but related drug, aspirin, is known to significantly inhibit COX activity, as do other NSAIDs (that’s how they work to reduce pain/inflammation). Generally, nonsteroidal anti-inflammatory drugs are sometimes considered a risk for IBD, both UC and Crohn’s because they are hard on the gut. Though aspirin seems not to have this negative impact in UC (Chan SS et al.)

A number of COX2 inhibitors came on the market and resulted in increased heart attacks and strokes, leading some to be withdrawn, and warnings to be added to traditional NSAIDs. COX2 is actually pretty essential, so drugs that inhibit it can be dangerous.

Mesalazine is generally considered to be very safe though – it’s the front-end treatment for ulcerative colitis and is taken long-term – why is that? I suspect it is because it acts at a very local level; the majority is not absorbed so it stays in the intestines (making it a great drug for treating inflammatory bowel diseases) without having too much systemic impact.

I guess what I’d really like to know is why PGE2 is elevated in UC in the first place (as we’ve known for a long time – Wiercinska-Drapalo A et al.) and seemingly also in some patients with ME/CFS (who commonly suffer gut dysfunctions, including leaky gut). Herpes viruses induce COX-2 (Kaul R et al.) so could it be that herpes viruses, long proposed as the pathogenesis of ME/CFS, are to blame? Who knows.

Speculation aside, I think if we want to reduce COX-2/PGE2 then our aim should not be to potently inhibit it but to lessen it if it is inappropriately high.

Let us take a look at what probiotics might do for us then. This study (Jan-Michel Otte et al.) showed that the probiotics VSL#3 and Mutaflor (E. coli Nissle 1917) “ameliorated induced COX-2 and PGE2 secretion.” That’s nice and might explain why these probiotics have proven helpful in IBD.

The study continues, “Lactobacillus acidophus, however, significantly increased COX-2 expression and PGE2 secretion.” So you want to avoid L. acidophus – something I already do, as it may reduce commensal E.coli. It’s hard to avoid entirely though as it is in many products, including VSL#3 which in this study, on balance, still reduced COX-2 and PGE2 through the other types of bacteria in the product.

Another study (R. Korhonen et al.) suggests that Lactobacillus rhamnosus GG induces COX-2. Maybe best to avoid that one too then. And yes, there are lots of foods and supplements out there which are known to reduce COX-2/PGE2 but that’s a whole other topic. But I can’t resist pointing you to one interesting study comparing the effects of Omega-3 and Omega-6 fatty acids (DiNicolantonio JJ et al.) on Cox-2/PGE2.

Which Genera of Bacteria are Covered by Probiotics, and Which Aren’t?

Probiotics: more holes than whole.

Probiotics: more holes than whole.

There are many brands of probiotics available, but the reality is that only a slim piece of the broad spectrum of commensal bacteria are available to buy as probiotic supplements. Different brands – but most are focused on the same types of bacteria.

So which friendly bacteria are actually available? And which aren’t?

It will take far less time for me to tell you what is available, than what isn’t!

Only a few genera of bacteria are available as probiotics, and although within this limited range there are many different species and strains sold by different brands, as most of the data we can actually get on what is present in our own gut is based on 16S rRNA testing (which predominantly tells you Genus level data), I figure that a list at the genus level is where we should start looking. Yes, different species and different strains do matter, but if we can’t get any species or any strain for a particular genera that we are lacking in, then that’s the real limiter.

So I’m starting with a list of which genera are covered by commercially available probiotics. I’ll be updating this list as more probiotics appear on the market. If you see something that is missing then please let me know! In time, I’ll maybe go and add species and strain level data.

For genera that are covered by many brands, I’m not listing the options because you can’t find them easily via a search engine or in a health store, but I’ve added links to the more tricky to find ones (but I haven’t tried any of them myself, so it’s not a recommendation).

Stuff not covered by the list:
Bacteria that you can buy but which are not listed as probiotics for consumption – for instance, starter kits for foods and beverages, or for veterinary use, or for research purposes – I am not listing these. I am also not listing any yeasts, just bacteria, and also absent are soil based bacteria that are not commonly found to colonize the human gut.

Above each genus is a heading stating which Phylum, Class, and Order it belongs to.

Firmicutes > Bacilli > Lactobacillales
These are the bacteria commonly referred to as lactic acid bacteria (LAB). There are several different genera from different families of bacteria grouped together as LABs and these make up the vast majority of probiotic supplements on the market. Many can be gained naturally by eating various non-pasteurized fermented foods.

Pedicoccus (NutriLots)
Enterococcus (Dr. Ohhira’s Formula OR Symprove OR Bifilac OR Threelac OR BIO-THREE)
Oenococcus (as the name suggests, this bacteria is involved in wine making and several sites sell this for wine production, though not strictly as a probiotic supplement)
Weisella (Oral Diet)

Actinobacteria > Bifidobacteriales > Bifidobacteriaceae

As with the lactobacillales above, these Bifidobacterium are sometimes referred to as lactic acid bacteria (LAB) and can be found in many fermented foods as well as being commonly produced as a probiotic supplement. However, they actually belong to a different phylum of bacteria than the Lactobacillales above.


Firmicutes > Bacilli > Bacillales >
Bacillus (Prescript Assist OR Microflora OR Bifilac OR Threelac OR BIO-THREE)
Staphylococcus (not available as a supplement but you can get it from Dawadawa – fermented locust beans)

Firmicutes > Clostridia > Clostridiales
Clostridium (Bifilac OR AOR Probiotic-3 OR BIO-THREE OR Miyarisan which you can buy outside of Japan on ebay)

Proteobacteria > Gammaproteobacteria > Enterobacteriales
Escherichia (Mutaflor)

Proteobacteria > Beta Protecobacteria > Burkholderiales
Alcaligenes (not available as a supplement but you can get it from Gari – fermented cassava)

Actinobacteria > Actinobacteria > Actinomycetales
Corynebacterium (not available as a supplement but you can get it from Gari – fermented cassava, or Ogi – fermented maize)
Propionibacterium (not available as a supplement but you can get it from Swiss type cheese such as Emmental cheese, Gruyere, and Leerdammer)

That’s it folks – I hope you weren’t expecting a big list!
If like me, you were hoping for some other Clostridiales such as Roseburia, Coprococcus or Ruminococcus, or anything from the Bacteroidetes phylum, then I’m sorry to disappoint you will this list. It’s clear to see why some people turn to faecal transplants as an answer. We’ll talk about that further down the line.

The Science of Sauerkraut

Sauerkraut-wikipediaFermented foods are often promoted as a good way to get beneficial bacteria into your digestive tract if you have bowel problems like ulcerative colitis, Crohn’s disease or IBS, or one of the many other conditions that come bundled with free bowel dysfunction, such as ME/CFS. And it’s true; gross-tasting fermented food can provide you with much needed friendly bacteria. But it is not a panacea.

The reality is that only certain bacteria are produced in the making of sauerkraut and other fermented foods. If you already have these in your gut and are lacking something else, then sauerkraut isn’t going to fix your problems. You might have endured eating it for no good reason!

So it might be worth a go, but if you want to be really scientific about it then you can first find out what bacteria are in your gut with a test through a company like uBiome or AmericanGut and compare your results to see whether the types of bacteria present in sauerkraut are already present in your gut, or not. You may find there is no need and thus avoid the misery of eating this rank ‘food’.

How does the fermentation process work?
There are a bunch of factors that influence which bacteria are in your sauerkraut, pH, length of fermentation, temperature, salinity, etc. and there is a large turnover of different types of bacteria as the fermentation progresses with the dominant species changing as the fermentation environment becomes more acidic.

What happens is that lactic acid bacteria (LABs) metabolise the glucose and fructose within the cabbage and produce lactic acid, and to a lesser degree acetic acid and mannitol, as by-products. Lactic acid overtakes glucose in concentration after just 7 days of fermentation. As the environment changes, so do the species of bacteria that are more suited to the conditions present.

As the name suggests, LABs are a group of related bacteria that all produce lactic acid, and the production of this acid helps to inhibits pathogenic bacteria and other organisms that might cause spoilage. LABs are generally considered to be safe, and occur in many foods such as cheese, beer and wine. They are present in healthy guts and often absent in autoimmune conditions. They belong to the Order Lactobacillales and if I look at my uBiome results, it seems that mine are much lower than average:


And this is not explained by my having been on long-term antibiotics. Many Lactobacillales are antibiotic resistant [2] and looking at my result compared to the antibiotic comparison group, the normal range is actually a lot higher than other groups, as high as 14.3% (because other types of bacteria are killed while Lactobacillales survived) which is double that of healthy omnivores. Mine are probably low for some other reason; either because of disease, or due to some other unknown factor. Increasing my lacrobacillales is unlikely to cure me of anything, but it may help a bit.

Exactly which bacteria are in the sauerkraut?
There are some limitations to the studies that have tried to answer this question. As fermentation progresses and the environment becomes more acidic, some bacteria that are present at the start of the process may be dead by the end. Some test methods would identify both the living and dead types without being able to say which ones were alive at the end. Other test methods get around this by culturing living bacteria though this also has limitations as some types of bacteria do not culture well.

That said, A 2007 study by Plengvidhya et al [1] looked at cultured isolates and also 16S rRNA gene sequence analysis in commercial samples and reported a rapid increase in the numbers of LABs and a rapid decrease in populations of Enterobacteriaceae in the first week of fermentation.

They conclude that “Under normal conditions, the fermentation is essentially complete within 2 weeks, with the most-acid-tolerant species, Lactobacillus. plantarum, predominating.”

L. plantarum was by far the dominant species, making up between 80% and 100% of the population after 14 days in different samples. Sadly, a small population of bacteria, less than 10%, was classified as “Unknown” at 14 days, followed by relatively small populations of Lactobacillus. brevis and Lactobacillus. paraplantarum.

Before this, at 7 and 9 days you also see these other LABs present that do not survive to 14 days:
Leuconostoc argentinum, Lactobacillus curvatus, Lactobacillus coryniformis, Leuconostoc fallax, Pediococcus pentosaceus. I suppose, if you felt that having these present in your gut were a good thing, then you could consume your sauerkraut at this earlier point.

Leuconostoc mesenteroides, Weissella species and Leuconostoc citreum are present earlier in the process in the first few days and are key for kicking off the fermentation process.

How much bacteria is present? I could just take some pills which taste a lot nicer…
It varies over the course of the fermentation but we are talking about billions of colony forming units per gram. L. plantarum and L. brevis have been well studied and have shown a wide variety of health benefits. You can buy some of the above as pro-biotics, if you prefer.

Fine, I want some of this nasty stuff. How do I make it?

fermenting crock pot

fermenting crock pot

Commercially it is made in big tanks, or in wooden barrels, and you can buy some and save yourself all the effort if you like. Just make sure you buy unpasteurized sauerkraut, or else it’s just sauerkraut with dead bacteria, and unless you like the way it tastes (hardly likely) then there’s little point eating it.

At home, it is typically made one of two ways: in a fermenting crock, or in a sealed jar. In either case, one of the main things you’re trying to do is remove all the oxygen from the equation. Lactic acid bacteria do not need it and many pathogenic bacteria do. I have not tried making it in a jar yet; it appears to be a bit hit and miss while using a fermenting crock makes it easy to keep the oxygen out. If you are going to try it in a jar then use a one way valve – this will allow any oxygen out but not let new oxygen back in.

How do fermenting crocks work?
It is just a container with a special top with a moat in it that you poor water in, and the lid sits in this moat, meaning that air cannot get in. Air from inside (the original air and the gasses produced by the bacteria) are allowed to escape via a couple of small nooks in the lid; they work to let air out but not in.

The other thing it comes with are some weights, usually ceramic which help to hold the cabbage under the salt water. This again helps to keep things anaerobic.

You don’t need a starter or anything like that, the bacteria naturally on the cabbage itself are what will multiply in the fermenting process. The only ingredients are cabbage (obviously), water and salt. Last time I used a standard cabbage, this time it is an organic one. I think either should work fine.

So, fermenting with a crock. They can be quite expensive. I bought mine, a second-hand 5 litre one made by Gairtpf for £20 on ebay. It was open but never actually used. They are heavy so you can get them for a bargain if you are local to the seller.

Things you’ll need
Cabbage (go figure)
Any other veg if you want, such as carrot or onion. I plan on trying this next time.
Water (I use natural mineral water but tap water would be okay)

First, before you begin, make sure everything is really clean, but also rinsed well to ensure there are no antibacterial agents left about, especially inside the container, as they may interfere with what we’re trying to do here.

My cabbage weighed 1.6lb. I used two cabbages last time, which turned into three or four large jars at the end of the process. But I didn’t eat it all (did I mention that I don’t like this stuff because it tastes like sour cabbage?….oh). Take two big outer leaves off the cabbage and set them aside. Shred the rest of the cabbage. It doesn’t matter how finely.

sauerkraut2Once you have shredded the cabbage, you add it to the crock in layers with the salt. Once it is all added you are supposed to pound the cabbage to get it to release its liquid from inside. You need approximately one tablespoon of salt per 1.5lb of cabbage. I did this but because I have ME/CFS I have little strength, so I just do a little and it seems to work fine even though it is not well pounded. The salt pulls water out of the cabbage, through osmosis. 24 hours later (or right away if you get knackered like I did and just want the ordeal to be over) you add the two leaves that you set aside, on top (this helps to keep the shredded cabbage from floating to the top) you place the weights on top of everything and then add enough water to just cover the cabbage and the weights.

Now you cover the Fermentation crock and fill the moat with water. Remember to check the moat daily as the water will evaporate and you will need to top it up. If you are making it in a jar then put the lid on it. Place it somewhere coolish (ideally, 18 degrees C, but room temperature is fine).

How do you know when it is done?
pH. It generally takes 2-4 weeks and as discussed above, the bacteria in the pot changes over time. So you will get different bacteria depending on when you decide to tuck in, and perhaps this will influence when you consider yours ready.

Last time I made sauerkraut it took 3 weeks. I tested the pH at two weeks and that told me it wasn’t ready (though it would have been safe to eat). After three weeks I was down around 3.6 which is where you want to be if you want completely cured sauerkraut. So get yourself some litmus paper.

Once it’s ready it keeps really well. You put it in a jar with a lid and cover it in the brine. It does not have to be refrigerated but it is a good idea and will extend its shelf life. You do, perhaps, start to get used to it once you eat it, and I suspect that if you grew up eating it, rather than say custard doughnuts, chocolate or pretzels, then you may even have convinced yourself that you like it. Personally, I think the bacteria ate the good bit and left me the pungent nastiness that is sauerkraut because they hate me. But I’m still gonna eat it, not for my sake, but for my gut’s.

If you want to help people like me who suffer from ME/CFS who have waited a long time for world class research into our gut microbiome, then please check out the Microbe Discovery Project and consider helping out.

1. Plengvidhya V et al. Appl Environ Microbiol. Dec 2007. DNA Fingerprinting of Lactic Acid Bacteria in Sauerkraut Fermentations
2. Zhou N et al. J Dairy Sci. Sep 2012. Antibiotic resistance of lactic acid bacteria isolated from Chinese yogurts
Image credit: sauerkraut in jar – wikipedia commons.

uBiome Raw Data: Taxonomy

When you have a sample tested by uBiome, uBiome provide the results via an interactive web interface. It’s designed for normal people, not technically-minded science boffins so you should get on with it whatever your level of knowledge.

It is currently a beta release so some displays are incomplete. The tree view for instance — which allows you to drill down through the hierarchy of a bacteria’s classification, all the way from phylum down to genus — might be missing some data, particularly when you get down toward the genus level. For this reason, I currently prefer to use the ‘Compare’ interface where you look at one rank at a time, but which seems to have all the data correctly listed.

The obvious limitation of this is that you cannot visualize each bacteria’s hierarchy; the taxonomic ranks it belongs to. For this reason you might find it useful to look at the raw data underneath it all. uBiome make this easily available.

You login to the beta web interface and click on “Dashboard” and you’ll see “Raw Taxonomy” and “Raw Reads” links appear. If you click on Raw Reads then you can download a zip of all your sequence reads in a .fastq format. (I’m not that familiar with this format yet but I know there are various programs you can run the data through – more on that in the future, maybe).

This post is about the “Raw Taxonomy” data. Click this and a few seconds later your data appears in raw form in your browser; a couple of hundred lines. All bacteria are genetically related and this data describes the family tree for the bacteria in your sample.

Let’s look at an example that many will be familiar with: Bifidobacterium.
If I search my raw data for that, I find this row:

{“taxon”:”1678″, “parent”:”31953″, “count”:”549″, “count_norm”:”11041″, “avg”:null,
“tax_name”:“Bifidobacterium”, “tax_rank”:”genus”, “tax_color”:null},

This is telling us that the genus “Bifidobacterium” has been assigned the taxon identifier “1678”. It also tells us that it’s parent is identified by the taxon ID “31953”. If you search for that ID in your raw data it will predictably take you to the row of data for “Bifidobacteriaceae”, the family that Bifidobacterium belongs to. This also has a parent, and if you follow that it will take you to the order, “Bifidobacteriales” which in turn gives a parent, which if you look up will take you further up the tree to the subclass “Actinobacteridae” and if you follow that up another level you’ll see the class “Actinobacteria” and above that one final level up you reach “Bacteria”! in row 2 of the raw data file.

Going back to the genus “Bifidobacterium” row, it also gives a “count” which is the number of sequence reads of this bacteria in my sample. The “count_norm” value is just a normalization of your result so that it can be compared to other samples on a level playing field. This has been calculated as follows:

(counts taxon X)x(1,000,000)/(counts taxon 2)

Where taxon 2 represents the total sequence reads of all the bacteria in your sample.

If you want to know what percentage of the whole any given taxon is you can take the “count_norm” value and divide it by 10,000. So, how abundant is Bifidobacterium in my sample? 11,041 / 10,000 which is 1.1%.

The other neat thing that I can see in the raw data that doesn’t yet appear in the web interface is some species level information. I don’t know how this is worked out and not each species is identified, but some are there, which is a nice bonus!

It turns out, most of my Bifidobacterium is species B. catenulatum.