Scientific Research (1/5: Gut Flora)

I have somewhat of a love/hate relationship with science: on one hand, it provides a uniquely privileged (and fascinating) look into the mechanisms of reality. On the other, the practice of science is often distorted by institutional and financial factors, which can warp what gets studied and who gets to study it away from the ideal and make life pretty miserable for would-be scientists.

Still, I’m drawn to science like a fish to water, and as a hobby I sometimes plan out lines of scientific inquiry I would pursue if I were a scientist and/or rich, particularly on issues which might not come up in normal science funding cycles. Every week in July I’ll be posting about one such idea. It’ll be a grab bag. Sit back and enjoy it.
What I’d do with a research lab, part 1:
Study the effects of diet on gut flora, and the effects of gut flora on physiology and psychology

Working hypothesis: Diet significantly and predictably influences gut flora, and gut flora has a significant influence on physiology, psychology, and in aggregate, perhaps even national character.

Gut flora, as anyone who’s read any yogurt labels nowadays knows, refers to the bacteria in our guts which help us digest food and absorb nutrients. The way this is commonly put is that, just as cows have cellulose-eating bacteria in their guts which help them digest grass, we have our own bacteria, or gut flora, which help break down hard-to-digest foods like Twinkies and Big Macs. This sounds simple enough– but what common wisdom misses about these bacteria in our guts is just how widely important and frighteningly under-studied they are. We know these bacteria not only play a key role in health[1][2]- they may be the primary external factor in immune system function- but can actually influence perception[3]. We also know literally almost nothing[4] about what lives in our guts. So, while the rest of the body is fairly well mapped-out, we have this huge question mark on gut flora, with a scribbled-in note that

1. Whatever does live in our guts outnumbers the ‘human’ cells in our body about 10-to-1,
2. gut flora is terribly important to body function (perhaps on the scale of a major bodily organ),
3. gut flora likely has some influence on nearly every part of our physiology, and
4. lots of things can go wrong. But we have few reliable metrics to figure out if someone’s gut flora has gone wrong, let alone how.

A big frontier in health science is figuring out what sorts of things influence this gut flora we carry around. Scientific wisdom has us getting much of our initial flora as a baby from our mothers, largely from breast milk and the birthing process, and these initial gut ecosystems are thought to be at least minimally stable. We also get bacteria from the food we eat, both from obvious sources such as yogurt but also from bacteria that naturally grow on foods such as grains and produce. Gut flora and mouth flora are linked, as are one’s genes and gut flora (e.g., genes influence gut flora, but gut flora also influences gene expression[5]). Taking a cycle of antibiotics is thought to sort of “reroll the dice” on gut flora: most bacteria die off and the survivors must scramble to repopulate the gut before their competitors do. There is, of course, a lot of randomness in all of this.

The nuances and relative contributions of these factors are several frontiers in themselves. Scientists are also starting to explore the indirect contributions of food to our gut flora, or whether different diets stochastically give rise to different biosystems by virtue of changing the competitive landscape for bacteria in our guts. One issue that I think is on the edge of researchers’ minds is whether a significant part of America’s systemic health problems is not just that we as a nation tend to eat food that’s bad for us, but we’re eating food (e.g., high-fructose corn syrup, food additives, preservatives (particularly sulfates), low-level food-borne antibiotics and pesticides) that makes the wrong bacteria thrive in our guts.

I think that’s probably the case and, getting further afield, what I’d like to do is attempt to look into whether the changes in gut ecosystems caused by changes in eating habits and food manufacturing trends- aggregated over the 300 million people who live in America- could contribute to a stochastic change in national character. Perhaps a significant contributing factor to some of our institutional ills is the food we eat, the corresponding imbalance in gut flora which arises from eating such food, and the subtle yet powerful-in-aggregate dysfunctional personality changes that e.g., biologically-active metabolites of such non-symbiotic gut flora might cause.

I sense some reader eyebrows being raised at this point– that bacteria could influence, let alone commonly influence, personality may seem quite a stretch. But consider the case of toxoplasma gondii (among other examples): it’s a protozoan parasite that spends part of its lifecycle in cats and part in other mammals, and shows clear signs of manipulating host behavior for its own ends. Infected rats, for instance, actually seek out the scent of cat urine, since when the rat gets eaten the toxoplasma can complete its lifecycle in the cat. What’s mind-boggling, though, is that humans who show tell-tale immunological signs of a past toxoplasma infection score statistically different on personality tests than do those who have not been infected (infected men tend to score higher in paranoia, whereas for women toxoplasmosis seems to lead to higher levels of social trust and sexual promiscuity). It’s unknown how or why toxoplasma causes such subtle personality changes in humans– though likely it’s a result of many generations of toxoplasma getting progressive fitness benefits from honing its initially accidental effects on rat fear/motivation, and since rats and humans are both mammals, some of those same psychological buttons toxoplasma has evolved to exploit in rats are hooked up to things in our brains, too. The fundamental point I would take from this is that there’s ample evidence that pathological microorganisms can and do subtly affect personality[6]. Personally? I believe these external pathogenic influences on personality – real as they are – will pale in comparison to that mediated by our gut floras.

Let me be perfectly clear: whether diet-gut flora interaction could be a commonly significant or significant-in-aggregate factor in personality is mostly just an intuition, built on the connections between gut flora and health, and health and personality, how tightly coupled bacteria and their metabolites are to our bodies and the scope of functional possibilities where such metabolites might- intentionally or unintentionally- act as e.g., hormone mimics, how many nerves (100 million+) are in our guts and how connected they are to our brain, the many possible feedback mechanisms between bacteria and neurotransmitters[7], how ‘clever‘ and environmentally manipulative bacteria can be[8], and how connected I think physiology and personality are. A sore tooth can influence personality; surely something that can not only affect our nerves, but many aspects of our biochemistry and nutrition as well, may do the same. I think it’s premature to really push any specific hypothesis about this gut flora-personality connection, since a lot of the basic science isn’t there to build on. But it’s a hypothesis we should be open to, and if I were in charge of the NIH or a well-endowed charitable foundation, I’d heavily prioritize flora research in general, particularly in the context of twin studies: we know so very little about gut flora (other than that it’s important) that the expected return is extremely high[9].

Update 10/13/07:

Since posting, there have been at least two major developments on this topic:
1. A theory that the appendix functions as a “safehouse” for good bacteria, especially with respect to repopulating the body’s GI tract after diarrhea, has been gaining traction. See

Bollinger RR, Barbas AS, Bush EL, Lin SS, Parker W. 2007. Biofilms in the large bowel suggest an apparent function of the human vermiform appendix. J Theor Biol (in press) doi:10.1016/j.jtbi.2007.08.032

2. A study has linked preference for chocolate with gut flora metabolism. Specifically, people who identified themselves as “chocolate desiring” had significantly different metabolic profiles and significantly different gut flora activity profiles than those who self-identified as “chocolate indifferent”. It’s both clear that this is an important result, and extremely hard to tease apart the causality involved at this point.

Results to be published in the Nov. 2 issue of American Chemical Society’s Journal of Proteome Research (via PhysOrg).

Update 8/1/10:

A new frontier in gut flora research is exploring the viral symbionts of peoples’ gut flora. Recent research points to a viral counterpart to our gut flora which
1. We know very little about (a recent study matched only 20% of the viral biome to existing databases);
2. Varies significantly between individuals (with identical twins having no more similarity with each other vs unrelated people, contra gut flora);
3. Is surprisingly stable (5% change over the course of a year, 1% for the most common viromes– indicating a benign, possibly functional symbiosis rather than an arms race).

The functional significance of the viral counterpart to our gut flora remains unknown. Presumably, though, it’s important.


“Probiotics (pills containing bacteria) have resulted in complete elimination of eczema in 80 percent of the people we’ve treated,” says Dr. Joseph E. Pizzorno Jr., a practicing physician and former member of the White House Commission on Complementary and Alternative Medicine Policy. Pizzorno says he’s used probiotics to treat irritable bowel disease, acne and even premenstrual syndrome. “It’s unusual for me to see a patient with a chronic disease that doesn’t respond to probiotics.”

[2] One of my reasons for being so interested in this topic is that I suffer from celiac disease, or at least have some probably-autoimmune-mediated reactions to gluten and casein, and some think celiac disease may be initially, and perhaps chronically, caused by the presence of ‘bad’ bacteria in the gut.

“We found that oral administration of specific Lactobacillus strains induced the expression of mu-opioid and cannabinoid receptors in intestinal epithelial cells, and mediated analgesic functions in the gut–similar to the effects of morphine. These results suggest that the microbiology of the intestinal tract influences our visceral perception.”

[4] – with interesting commentary by John Hawks:
“This is really important stuff — our nutrition is very dependent on these microbes, and there is every reason to think that their ecology affects our overall health status as well. And we know very little about them — heck, these guys are using the same metagenomic techniques to fine organisms in our bodies that are used to find new unidentified ocean life!” e.g., we’re forced to use metabolism and byproduct analysis because most of what lives in our guts can’t be cultured in vitro. Edit, 7/13/07: A recent paper came out in PNAS outlining a technique that may at least partially solve this problem.

[5] Hooper LV, Wong MH, Thelin A, Hansson L, Falk PG, Gordon JI. Molecular analysis of commensal host-microbial relationships in the intestine. Science 2001 Feb 2;291(5505):881-4.

[6] Gregory Cochran has argued from an angle of evolutionary fitness load and allele frequency that many things we think of as being caused by genes or behavior (e.g., rheumatoid arthritis) are probably primarily mediated by pathogens.

[7] See – it’s also a good survey of some other complexities in this topic.

[8] This just scratches the surface of the manipulative potential of bacteria, but one such subtle strategy used by bacteria is (briefly) explained here:
“The genes responsible for toxin production only seem to be expressed during periods of nutrient deprivation. This is consistent with the view that most disease-causing bacteria express their pathogenicity when they are hungry,” says Abraham Sonenshein, professor at the Sackler School of Graduate Biomedical Sciences at Tufts University and at Tufts University School of Medicine, at the 107th General Meeting of the American Society for Microbiology (ASM) on May 24, 2007.

[9] Another related high-return research area which I think is a little better represented in science (but should also get more funding) is examining potential pathogenic influences on personality, e.g., along the aforementioned lines of toxoplasma gondii and
this previous post.

– I got to thinking about this when musing about America’s social/political/institutional ills. Perhaps this is a longshot for trying to help explain our various institutional dysfunctions. But who knows? It could be a contributing factor. Just because the effects of gut flora are incredibly complex doesn’t mean they’re neutral when aggregated into larger contexts.
– At this point I don’t have solid predictions on what sorts of diets tend to strengthen/weaken what sorts of personalities. Perhaps the safest thing to say is that the significant variation in human personality may have gut flora as a significant contributing factor, and also that the physiological stress of coping with a non-symbiotic flora may e.g. lead to personalities more prone to addiction.
– Though there is plenty of literature documenting correlations between gut flora and various diseases, and specific diseases and personality changes, I haven’t found any literature that covers what I would like to study, which is modeling gut flora’s influence on personality without the mediating frame of a specific disease.

If you enjoyed this content, please consider donating.

13 thoughts on “Scientific Research (1/5: Gut Flora)

  1. Thanks. Yeah, gut flora is just linked to *everything* it seems… and the obesity angle certainly makes sense.

  2. I don’t know quite what to say. It’s like you’ve tuned into some of my hitherto private thoughts and amplified them, adding substance and citations.

    But check it.

    One thing I can say is that since many neurotransmitters are metabolites of coding amino acids, we might expect to find these same molecules–as well as metabolites of these molecules which also act on the same receptor sites–in many contexts throughout the ecosystem.

    “I got to thinking about this when musing about America’s social/political/institutional ills. Perhaps this is a longshot for trying to help explain our various institutional dysfunctions.”

    Yeah, I think it’s a longshot to say this might explain our shortcomings, but it very well may be a big piece of the puzzle. Of course, most Americans don’t accept evolutionary theory; I imagine they’ll be much less receptive to the idea their mood and behavior may be significantly influenced by microorganisms. The boomer generation is still pretty paranoid about germs in a kill-them-all-let-God-sort-it-out kind of way.

    Any thoughts on candida albicans?

  3. That is interesting! Thanks for the comment, and I’m glad to see this thought represented elsewhere.

    Candida albicans is a tough topic– there are so many claims and so much corresponding skepticism that it’s really impossible to figure out what’s going on. Hopefully respectable scientists can at least get funding to study it, and in a couple years the dust will settle and we can figure out if it’s really as big of an influence on modern health as some people believe. I suspect it’ll be somewhere in the middle: there are tons of interesting and probably largely factual claims about the mechanisms by which it can influence health, but there are just *so many other* similar organisms that probably do similar things that I think it’s unwise to classify it as some sort of manipulative superbug. At this point.

    If I were a scientist studying candida albicans I’d probably focus on the interplay between candida and the immune system. For the claims about candida (which I have some sympathy with) to hold water, candida’s gotta be fairly unique in how it finesses the immune system.

  4. Do either of you have any research concerning the taxonomy of flora comparing different species and their respective organs?
    Could fungus considered plant flora because of how it fixes nitrogen and enriches the soil bedding?
    Don’t know much about c. albicans other than I think that it’s primarily throat and vaginal flora.

  5. I’m sympathetic to them too, though, yeah, there’s a lot of speculation that needs to get sorted out.

    I was surprised to recently see that the wikipedia entry presents the subject (candida is implicated in autoimmune dysfunction) as though there is no controversy nor has there ever been. Nothing on the talk page either. And they have citations. Just six months ago, there was more controversy.

    I brought it up as a way of attempting to address the last sentence in your post: “I haven’t found any literature that covers what I would like to study, which is modeling gut flora’s influence on personality without the mediating frame of a specific disease.” I still can’t quite get at it, and I’m getting too tired to try, so I’ll just give you a link to something which is related to this topic but where I don’t come right out and say what I’m actually thinking.

    I’m going to poke around your site a bit more tomorrow.

    Oh, yeah, Jordan: Don’t know, don’t know, and I think that candida is actually supposed to be found in the lower intestine. The eruptions of candida in the mouth and the vagina are supposedly due to the overuse of antibiotics.

  6. tate- interesting link (I’ve added your blog to my feeds). I’m excited to see this topic get more awareness. I do encourage you to poke around the blog, particularly the 2006 posts.

    Jordan- connecting different bacteria to different organs would be an interesting task, but a very hard and fuzzy one, so much so that I don’t think much of interest can be said about it yet.

  7. Think of what happens in birth: the baby’s head (body) passes down the birth canal/vagina. I’d say that the majority of bacteria inhabiting our guts came from the birth canal, not from breast milk.

  8. Anon-

    That seems to make a lot of intuitive sense. I suppose the question is, how much of a factor would it be? It’d certainly be another data point for twin studies (tracking differences in gut flora where one twin was delivered normally and the other via cesarean…).

  9. Mike,

    Twins are rare enough (1/90 births) so to have a combined vaginal/caesarean birth is much, much more rare (1/2250.)
    Nevertheless, it’s the rare things that get studied most.

    I haven’t found a study that focused on gut flora diffrences in twins from combined births yet, but I’m still looking.

    Evidence that c-sections predispose infants to illness is widespread, however.;=89&fp;=993&view;=short

    But there is also proof that vaginal delivery has its own risks.

    Pretty fascinating if you ask me.

  10. Anon- I have, here and there, though I only got up to 3/5 in this specific series before the ideas turned into larger projects.

    If you work backward from my most recent posts there are a lot of research ideas I'd love to pursue. The neuroscience series, dark energy, and evolution distance specifically.

Comments are closed.