Microbiome change and weight loss

​​​since i have not done any blogs on the microbiome recently, it seems only fair to have a few. this study found that the microbiome composition might affect weight loss achieved through diet (see microbiome change dec weight intjobes2017 in dropbox, or doi: 10.1038/ijo.2017.22​).  i am happy to send along any of the articles noted below; just let me know

Details:

--62 people with increased waist circumference were randomly assigned to an ad libitum New Nordic Diet (NND) vs an Average Danish Diet (ADD), for 26 weeks

    --NND: a whole-food diet with very high dietary fiber, whole grain, fruit and vegetables

    --ADD: the best description i can find on the internet is: meat balls, pizza, spaghetti with meat sauce, dairy, bread, potatoes, rice, pasta, meat and veges (esp cucumber and tomato)

--median age 42, BMI 31, body fat 40%, fasting glucose 5.3 mmol/L (97 mg/dl), fasting insulin 51 pmol/L

--participants were divided into 2 groups depending on their microbiome (enterotypes): those with pretreatment high ratio of Prevotella spp to Bacteroides spp (P/B ratio) vs low ratio [the ratios were 0.28 in the high group and 0.000007 in the low one]

--NND vs ADD had higher dietary fiber (43.3 vs 28.6g/10MJ), protein (18.1 vs 16.4%, though presumably vegetable-based protein), but lower in fat (30.4 vs 33.8%) [all with p<0.001) but no difference in carbohydrates

Results:

--at 26 weeks, those with high P/B ratio on the NND had significant changes in the followings:

    ​--4.97 kg body fat loss in NND vs 1.82 kg loss in ADD [difference of 3.17 kg, p<0.001] vs no significant difference between the NND vs ADD diets in those with low P/B ratio

    --4.58 kg more weight loss in NND vs 1.09 kg loss in ADD [difference of 3.49 kg, p<0.001] vs no significant difference in diets in those with low P/B ratio

    ​--5.19 cm decreased waist circumference in NND vs 0.44 cm in the ADD group [difference of 4.75 cm, p<0.001] vs no significant difference in diets in those with low P/B ratio​

    --the groups on the NND diet had significantly more weight loss if they had a high vs low P/B ratio: 4.97 kg vs 3.41 kg, p=0.04

--at 1 yr follow-up period, in those changing from ADD to NND:

    --patients with high P/B ratio maintained their weight (-1.23 kg, but nonsignificant)

    --patients with low P/B ratio regained 2.76 kg (1.27-4.24), p<0.001

            --this reflected a 3.99 kg difference between the P/B groups when changing to NND diet

Commentary:

--the above study found that people with high baseline P/B ratio seem to be more susceptible to weight loss, even when eating all they want of a diet rich in fiber, whole grain, fruit and vegetables. these researchers had previously reported that those on ad lib NND overall lost more weight than those on ADD. This current study added to that by finding that the NND did work better, but was limited to those with a baseline high P/B ratio​

--other studies have found that those with more Prevotella (more often found in strict vegetarians)​ in their microbiome had lower LDL levels, despite similar anthropometric profiles

--an observational study of 1632 healthy female twins in the UK found that Bacteroides was associated with weight gain (and those with higher Bacteroides overall had less microbiome diversity), and those who ate more dietary fiber had more bacterial diversity (a reflection of a healthier microbiome)

--and, a recent study found that dietary fiber from barley kernel-based bread (as in NND) led to improvements in postprandial blood sugar and insulin in nondiabetic patients, but only in those with high gut P/B ratios. And germ-free mice given a fecal transplant from those who responded to the high fiber diet (and had high P/B ratios) also showed improved glucose metabolism​

--Bacteroides and Prevotella are both genera from the phylum of Bacteroidetes and tend to be inversely present in the gut (ie those with higher Prevotella have less Bacteroides, as if they both compete for the same niche in the gut).

--diet and the microbiome: a really great review noted that acute changes in diet leads to rapid changes in the microbiome: change to a diet which is strictly animal-based or plant-based alters the microbiota within 24 hours, which then reverts within 48 hours on resumption of the old diet (similar time interval for stress or inflammation-induced microbiome changes). these changes mirrored the microbiomes found naturally in herbivorous vs carnivorous animals. In terms of protein: vegetable-based protein (whey, pea) leads to increases in bifidobacterium and lactobacillus and decreases in bacteroides. Animal-based protein increases bacteroides. This article also reviews the data linking the microbiome to disease: inflammatory bowel disease, psoriasis, atopic dermatitis, autoimmune arthritis, type 2 diabetes, obesity, atherosclerosis. even perhaps Alzheimers (see microbiome and diet JTranslMed2017 in dropbox, , or DOI 10.1186/s12967-017-1175-y)

--given these studies, it is a bit surprising that those taking an NND diet but with a low P/B ratio for 6 months (and then at the 1 year mark) as in the above study did not have a more favorable microbiome change in the process of the study. ??reason.  they comment in the paper that the gut microbiome is overall pretty stable, and that it tends to take more than 6 months of diet change to be effective. [they do note that more "extreme changes" in diets have more rapid effects (eg the 24-hour one noted above).] A Korean twin study found that overall 72% of twins shared their enterotypes (they looked specifically at Prevotella and Bacteroides), and that this was quite stable over time.  and they cite some other studies suggesting overall microbiome stability in the shorter term. But long-term dietary habits were associated with changes: gut Prevotella, as opposed to  Bacteroides, was associated with higher dietary fiber, minerals (potassium and iron), and vitamins (A, C, E, folate, carotene and retinol).] One concern in the above study is that there was no measure of adherence to the different diets.

--one general concern is that we avoid becoming too reductionist here. the gut microbiome has trillions of microorganisms, there are undoubtedly multiple factors leading to changes in the microbiome, these changes involve many different microorganisms, these microorganisms undoubtedly interact with each other, so looking specifically at Prevotella and Bacteroides is undoubtedly an oversimplification. The main point here, i think, should be that a healthy microbiome, one supported by a healthy vegetable-dominant diet, exercise, less stressful existence, and minimizing antibiotic use (especially broad-based ones) is likely to lead to fewer of the above diseases associated with a less healthy microbiome

--an interesting related issue: a prior blog suggested that metformin increases the microbiome content of Akkermansia and that transplanting this bacteria into diabetic mice improved their glucose tolerance, as well as another article finding that eating meats leads to changes in the microbiome leading to increased production of TMAO, a profound atherogen (see http://gmodestmedblogs.blogspot.com/2015/08/early-life-stress-in-mice-changes-in.html )

--there are likely multiple mechanisms as to why the microbiome might influence weight changes, potentially affecting: 

    --the amount of energy extracted from ingested foods: short-chain fatty acids secreted by bacteria in the gut (including Prevotella) can stimulate glucagon-like peptide-1 and peptide YY secretion (GLP-1 and PYY suppress appetite)​

    --the gut/brain axis (see microbiome gut brain axis JClinInvest2015 in dropbox, or doi:10.1172/JCI76304), showing that the gut microbiome can affect emotional behavior, stress- and pain-modulation, and brain neurotransmitters; some of these changes could affect appetite/diet/weight.

so, this study suggests a pretty complex interaction between diet and weight loss: those on a "healthier" NND type diet lose weight only when they have more Prevotella in their gut (and this is on an ad lib, eat all-you-want-to-eat diet). but overall people seem to need to eat such a diet for many months in order to change their microbiomes and achieve these types of weight/waist circumference changes associated with the more favorable P/B ratio (unless they get into the "extreme" diet change approach) . Perhaps this is part of the reason that we clinicians, and our patients, see quite variable outcomes in patients on similar "healthy" diets with lots of veges, fiber, etc.  so, the take-home points of this article to me are:

--losing weight may really vary from one individual to another on the same diet (as we often see in our patients, though the above gives an interesting mechanistic reason)

--but that it seems that being on a healthier diet does lead to a healthier microbiome, at least in the 6-month to 1-year timeframe, and eventually should really affect the desired clinical outcomes of weight loss (esp body fat loss) and decreased waist circumference

--so, our struggle (clinicians and patients) needs to be a long-term one, and that lack of weight loss in the beginning should not deter any of us from continuing with the healthier, vegetable-based diet with lots of fiber.....