As you know, here at drkarafitzgerald.com we like to push the boundaries of what we know, and sometimes this means being a little contrarian. In this blog, our Nutrition Director, Romilly Hodges CNS takes us on a tour of some of the emerging research on Th17 prompted by a new paper in the journal Cell, that postulated a reduction in Bifidobacteria may mediate some of the benefits of a ketogenic diet for epilepsy by reducing Th17.
Yes – we found it a head-turner too.
Since we use supervised ketogenic diets for a wide range of conditions with some frequency in the clinic, our interest was piqued – just as yours may be. We needed to take a little dive. Enjoy this fun read… and let us know if you come to the same conclusions. ~DrKF
The interest in ketogenic diets as a potential therapeutic tool continues to increase. Yet, one of the criticisms that has always been leveled at ketogenic diets is that the lower amounts of available carbohydrate substrate for the gut microbiota must have detrimental effects on the microbiome. Especially over time. This is a valid concern and one which leads us to continue to assess the gut health of our patients/clients who follow short/medium term or intermittent ketogenic diets.
The logic makes sense of course – take away the food for healthy microbial species and we must naturally end up reducing diversity and beneficial kinds of microbes. Add to that the confusion generated by animal research using laboratory high-fat diets which links high fat consumption to metabolic endotoxemia due to microbial dysbiosis created by the diet. What is not always well-communicated is that these diets are often high in refined carbohydrates such as syrup solids – not at all a ketogenic diet. There are other problems with these highly processed ‘rodent chow’ diets mass produced for the scientific research industry – chemical and heavy metal contamination, poor ingredient quality, lack of fiber and phytonutrients. Craig Warden of UC Davis came to talk to discuss these issues with our team a couple of years ago and changed how we investigate and interpret those high fat diet headlines (recording and materials available here).
Incidentally, our clinical approach to ketogenic diets, which we only use selectively and always carefully implemented and supervised, is less bacon-eggs-cream, and more mono- and polyunsaturated fats from fish, nuts, seeds, avocados and olives. Some good quality saturated fats, but in balance. Phytonutrients feature heavily in our version of the diet – abundant low-starch vegetables, herbs and spices that are rich in soluble fibers and polyphenols that are known to nourish healthy bacteria. Probiotic foods and supplements also feature.
A new report of reduced Th17 with ketogenic diet
It’s a newly published study that brings this topic right back front and center for our team: (Cell, May 2020) Ketogenic Diets Alter the Gut Microbiome Resulting in Decreased Intestinal Th17 Cells (paywall).
The ketogenic diet used in this study’s human participants (5/80/15 CHO/fat/pro) appeared to reduce levels of T helper cells belonging to the Th17 class in the small intestine compared with a baseline diet (50/35/15). This was evaluated by inoculating germ free mice with the keto-fed human fecal samples. In their paired animal models on a ketogenic diet, Th17 levels were also observed to fall in adipose tissue as well as the small intestine, suggesting (though of course not confirming) a systemic effect.
This is a pretty intriguing finding … after all, Th17 cells are strongly implicated in promoting systemic inflammation and autoimmunity. We do already know that ketogenic diets can help dial down inflammation because due to several mechanisms including the inhibitory effect of beta-hydroxybutyrate on the NLRP3 inflammasome. But this potential Th17 connection is new.
Microbiome changes don’t immediately look good
This apparent anti-inflammatory effect of reduced Th17 cells occurred despite a reduction in Bifidobacteria (a genus we normally consider to be helpful) and increase in Fusobacteria and Escherichia (with more pathogenic species) seen with the ketogenic diet intervention.
In further animal investigations (still under this same publication), the researchers studied different versions of high fat diets with carbohydrate amounts ranging from 0 to 15 percent of calories, still within the range of ‘ketogenic’ for humans (although actually, 0 percent carbohydrate is unrealistic for humans; 5-15 percent is possible). Actinobacteria (which include Bifidobacterium species) declined steadily as carbohydrate levels dropped off. Bacteroidetes (also considered good guys) increased progressively.
Incidentally, this study found that beta-hydroxybutyrate (BOHB), the main ketone body, seemed to have direct antimicrobial activity against Bifidobacterium. Could this seemingly-disadvantageous reduction in Bifidobacteria be the key…
Potential Role of certain Bifidobacterium in intestinal Th17 skew
The authors in the same study above report that certain human gut Bifidobacteria, such as B. adolescentis, promote the production of Th17 cells. They observed this in their study by introducing the probiotic into germ free mice. Subsequently, this increase in Th17 was also observed to be blunted by the most strict rodent ketogenic diet (0 percent CHO) but not the other rodent diets leading to the conclusion that ketone bodies may mediate the reduction in intestinal Th17 by reducing colonization levels of Bifidobacterium. Furthermore, when stool samples from the human part of the trial were inoculated into germ-free mice, only the human ketogenic diet microbiota, with lower Bifidobacteria, lowered Th17 levels.
We went looking for more research to see if other studies had pointed to this Th17 connection before…
It turns out that there is some other limited evidence to this effect: B. bifidum strains have been shown to induce IL-17, a TH17-derived cytokine in vitro. A group of researchers, most from Harvard Medical School, also reported in 2016 that they could induce Th17 transcription in mouse guts by administering B. adolescentis and Th17 responses in vitro with several Bifidobacterium species. They also noted they were able to demonstrate that this probiotic also drove “gut-distal Th17-cell-associated disease progression” in their autoimmune arthritis mouse model. Their paper was written up in Nature Reviews Rheumatology.
Another interesting connection point and area of research on this topic is in epilepsy…
Epilepsy, the gut microbiome, Th17 and… autoimmunity?
Epilepsy, and specifically intractable, non-drug-responsive epilepsy, is where ketogenic diets have their longest beneficial track record – nearly 100 years of use now. However, the mechanisms of its effects have only been speculative. In the past 2-3 years there has been a burst of scientific study on the microbiome as a previously un(der)recognized, now potentially-important, mediator. In fact, pre-treatment of the gut microbiome with antibiotics or germ-free rearing, has been shown (in mice) to completely revoke dietary benefits. I.e. the gut microbiota seem to be a required mediator.
In patients with intractable epilepsy, it has now also been observed that circulating Th17 cells and their cytokines are higher and that they can be partially lowered by following a ketogenic . If the gut microbiota is a required mediator of the beneficial effects of the diet, could this be in part via the Bifidobacteria-Th17 connection?
Along this line of thinking, within the last decade, some have highlighted the connection between autoimmune diseases and refractory epilepsy. The prevalence of epilepsy among those with autoimmunity has been estimated at 17.5 percent, compared with 0.4 percent of the general population. Immunotherapy (including IV methylprednisolone and IV immune globulin) typically used in autoimmune diseases have been reported as beneficial in humans. Given the implied role of Th17 cells in autoimmune diseases there is logic there that we can follow.
We should also mention, too, that ketogenic diets and their associated impacts on the gut microbiome have been linked to increased GABA and GABA:glutamate ratios in the brain. Another potential contributing mechanism.
Some wrinkles in the theory
In recent studies showing the effects of KD for epilepsy on microbiota composition, however, not all studies report the same decrease in Bifidobacterium species. As you can see, there’s variability in effects:
- Lindefeldt et al., 2019: KD for epilepsy associated with decreased Actinobacteria and Bifidobacterium, increased Proteobacteria
- Zhang et al. 2018: KD-associated benefits for epilepsy also corrected with decreased Actinobacteria and Firmicutes relative to Bacteroidetes.
- Xie et al., 2017: KD for epilepsy correlated with reduced Proteobacteria and increased Bacteroidetes, Prevotella and Bifidobacterium
Review authors have postulated that the lack of consistency in which microbial taxa are affected may potentially derive from length of ketogenic diet, specific ketogenic regimens, subtypes of epilepsy etc. I would add to that – baseline microbial status.
Another study of microbiome alterations in multiple sclerosis patients on a ketogenic diet, Swidinski et al., 2017, extended the thinking: The authors argued that a ketogenic diet can be restorative for dysbiotic guts specifically (rather than healthy guts). I think it’s worth reading twice and considering the implications thereof. They also found in their study that bacterial richness and diversity decreased in the initial phases of a ketogenic diet, but after 12 weeks the bacterial populations began to recover back towards baseline. And after 23-24 weeks, bacterial concentrations were above those at baseline.
So, we really don’t know how the microbiome would have continued to reshape in the original Cell study above.
Will this change our use of ketogenic diets?
A definitive no. After all, our clinical experience with various conditions from refractory autoimmunity, to severe cardiometabolic dysregulation, to neurological disorders, to certain cancers has been overwhelmingly good. It’s a highly important tool. And there is an increasing abundance of clinical data available now on the benefits of using ketogenic diets in these conditions and more.
We will, however, continue to:
- Track microbiota diversity and changes in our patients/clients on ketogenic diets
- Include keto-friendly prebiotics, fermented foods and probiotics in keto diets
- Remain outcomes based and up-to-date with the literature
Will we stop using Bifidobacterium probiotic species for autoimmune disorders (or in individuals at risk for AI) given the Th17 association?
Again, no. As with using probiotics, including Bifidobacterium species in autoimmune and other diseases has been overwhelmingly positive. And other studies report beneficial effects on Treg/Th17 ratios from Bifidobacterium administration – here and here – demonstrating that the conclusions drawn above are neither clear cut nor reproducible across all models.
And, while we’d love to see more research, there are positive outcomes seen with Bifidobacterium use in clinical trials for multiple sclerosis and rheumatoid arthritis. Human studies of anti-inflammatory activity mechanistically and for other conditions are also compelling. Infant microbiota that are supplemented with Bifidobacterium infantis are less inflamed and carry fewer antibiotic-resistance genes.
And for one more, crucial reason…
A final twist – just how “bad” are Th17 cells?
As usual, science likes to throw us a few curve balls. While much FxMed attention has justifiably been on the understanding that excessive Th17 responses are a major contributor to autoimmune diseases, there may be dual roles for these T helper cells. Scientists studying IL-17A, the major Th17 cytokine have found that it also has beneficial effects in the gut, including:
- Promoting enterocyte proliferation
- Tight junction formation
- Epithelial barrier integrity
- Increased expression of antimicrobial peptides
- Promotes homeostasis rather than inflammation
So, Th17 is a tale of two potential outcomes. We don’t yet know how that’s regulated, but we can certainly speculate that the increased Th17 seen following Bifidobacterium supplementation is in most cases beneficial rather than pathological.
A postscript, too interesting to cut – effects of ketogenic diet on the mucosal layer
Citing prior work, the authors of the new Cell study above that started this whole conversation noted that they might expect to see a decline in mucosal health with a ketogenic diet because the microbiota, in the absence of dietary food sources, may scavenge polysaccharides from that mucus layer. However, in the animal portion of the study the mucous layer was maintained without notable changes (human mucosal changes were not assessed, unsurprisingly). This is interesting for us Functional Medicine and Nutrition practitioners interested in preserving mucosal (and by extension immunological and barrier) health.