Microbiome and colitis: a new approach

one issue in current microbiome research is that there are often associations between certain microbial species found in the gut microbiome and certain disease or health states, but the direct cause/effect link is not clear. these researchers, looking at a colitis model in mice, derived an improved approach in assessing causality (see doi:10.1038/nature25019​).

Background:

--they studied gnotobiotic mice (ie, mice with known gut microbiomes), some mice with mouse microbiota (MMb) and some with human microbiota (HMb); and also some specific-pathogen free (SPF) mice

--these mice were exposed to DSS (dextran sodium sulfate), known to cause colitis in mice

    ​--in MMb mice, the colitis was clinically associated with severe weight loss and 100% mortality

    ​--in HMb mice, as well as SPF mice, they lose less weight and rarely die; but the degree of microscopic colonic inflammation on day 5 did not differ between the groups, though there were some small difference at day 10 (ie, the toxic DSS evoked equivalent inflammatory responses)

--in assessing the microbiome differences, prior studies identified more than 150 bacterial taxa that were differentially abundant in the HMb vs MMb mice; and 100 differentially abundant species in MMB vs SPF mice

--and, directly evaluating/comparing these 100-150 specific microbiome differences would be a rather daunting task

--focusing on those with the most differences narrowed the list to 26 taxa present in one group and absent in the other

--they chose 4 present in HMb mice, gave them to the MMb mice, then challenged these mice with DSS, finding that none of these organisms significantly augmented disease severity, reflecting the problem to trying to isolate the offending organism by using this type of dichotomy (present vs absent in the microbiome)

--so, they decided to co-house the different mice (MMb and HMb) to generate mice with hybrid-microbiota to see if there would be a shorter list of phenotype-associated microbial taxa (ie, there should be fewer differences between the groups if their baseline microbiomes were more similar) and then, if there were differences in the susceptibility to disease, be able to compare the differences between the original mice and vs the hybrid ones to assess disease-modifying microorganisms, a process they called "triangulation", but with a shorter list of likely suspects

Details:

--they co-housed HMb and MMb mice for 3 weeks, finding bidirectional microbe transfer between the two microbiomes. then they decreased the co-housing time to the shortest interval allowing for bidirectional microbial transfer in order to minimize further microbiome changes, and found that after only 1 day of co-housing, there were significant survival differences after giving the toxic DSS. these hybrids housed microbiomes which were intermediate between the HMb and MMb mice, also found at 3 days and 14 days of co-housing. but there were distinctive microbiota in each of these time-periods of co-housing.

--so, since there were intermediate survival differences at each time, they reasoned that the specific culprit microbial species would be present in each of the time-groups, finding only one such taxon: Lachnospiraceae (a gram-positive, anaerobic, non-spore-forming bacteria) that was associated with survival from DSS-induced colitis.  There were almost none of these species in MMb mice, intermediate in the hybids and SPF mice and high in HMb mice (and, of note, this species had been identified in several human studies, also finding an inverse relationship with inflammatory bowel disease)

--then, to assess causality: they orally administered cultured feces from HMb mice (with enhanced Lachnospiraceae ) to MMb mice, finding the mice were protected from clinically significant colitis

--then, they identified one Lachnospiraceae isolate which was predominantly present in HMb feces, a previously unidentifed species subsequently named Clostridium immunis, gave that to MMb mice and found that they were protected from colitis

Commentary:

--microbiome-wide association studies, looking at a broad array of microbial species, have found lots of associations with different diseases (see link below to many such studies), though often one cannot assert causality. the above approach raises a few issues:

    --it offers a new and easier technique to identify potentially protective (and, i would imagine, harmful) microbial species

    --and, i would also imagine, that similar techniques could be used looking at the human microbiome in patients with and without diseases likely associated with the microbiome (and there are lots of them). initially perhaps through human microbiomes transplanted into germ-free mice

    --and it could lead to the development of more useful probiotic supplements to prevent or potentially modify disease

    ​--though, it might turn out that disease (or health) may depend on a more complex interaction between different microbial species being present simultaneously (ie, not a single species by itself): as is often true for dietary micro- and macro-nutrients: ie no single magic bullet

so, this study does add to reinforcing the importance of a healthy microbiome in preventing disease, that many diseases are caused by distorted microbiomes, and that there may be ways to restore healthy microbiomes by specific probiotic concoctions. i would add that many features of a healthy lifestyle (eating well, maintaining good body weight, exercising, sleeping well, avoiding stress, avoiding antibiotics, etc etc) are paramount in developing a healthy microbiome and preventing disease. but, of course, many people cannot do so (unable to do these healthy things because of lack of access to good foods/safe exercise venues/etc because of their socio-economic-political situations, or they just really do need antibiotics for clearcut indications, etc.)  However, i think it is really important not to descend into a blame-the-victim mentality of "well, you eat at MacDonalds, so you deserve your disease" instead of looking at the overall social environment which reinforces the MacDonalds and many many other unhealthy lifestyle issues, or makes healthy "choices" inaccessible. The reality is that most of the chronic diseases we clinicians manage are a result of our dysfunctional social environment, one which effectively promotes unhealthy eating, smoking/alcohol, overweight, lack of exercise, heightened stress, etc. And, although we should be engaged to change these underlying conditions, we still need to treat patients with meds. so, overall, exploring the utility of corrective probiotics, i think, is still welcome....

for a slew of microbiome articles, see: http://gmodestmedblogs.blogspot.com/2017/11/microbiome-hypertension-and-atopic.html​ which links to many other relevant microbiome blogs