I’m not normally a fan of research that claims to prove that weight gain is anything other than a result of eating too many, and burning too few, calories; it seems too much like wishful thinking. But I have to say that the story around gut microbiota being involved in obesity has been getting interesting. The story just took another intriguing turn with a paper that identifies links between gut microbiota and the endocannabinoid system (Muccioli et al. 2010 The endocannabinoid system links gut microbiota to adipogenesis. Mol Syst Biol. 6 392. PMID: 20664638.)
The endocannabinoid system is, as you might guess, the system of receptors that respond to the active principle in cannabis, and the endogenous ligands that activate them. Endocannabinoid signalling is involved in neurotransmission, memory retrieval, and the control of hunger. [Yes, this is why smoking pot gives you the munchies, and affects your short-term memory; and no, the title isn't meant to be sarcastic.]
Muccioli et al noticed two things that made them suspect that there might be a link between the endocannabinoid system and gut microbiota: endocannabinoid signaling seems to be increased in obesity, and lipopolysaccharides — components of the bacterial membrane — have been shown to affect the synthesis of endocannabinoids. Could lipopolysaccharides from gut microbiota affect the production of endocannabinoids, and thus affect obesity? This is a long-ish chain of evidence with some confusing bits, so let’s take it step by step.
First question: do treatments that affect the composition of gut microbiota have any effect on the endocannabinoid system? Muccioli et al. tested five different mouse models in which gut microbiota are either manipulated, or known to be different from normal: mice treated with prebiotics, with or without a high-fat diet, treated with antibiotics, or raised germ-free; plus a mutant mouse that has altered bacteria-host interactions. In each case, they saw significant decreases in the mRNA level for one of the endocannabinoid receptors (CB1) in the colon, but not higher up in the intestine where there are fewer bacteria. The other receptor, CB2, showed no such change. Also, the concentration of the ligand for CB1 goes down in the colons of treated mice, as measured by HPLC-MS.
The authors have previously shown that gut permeability is increased in obesity, causing the escape of lipopolysaccharides (LPSs) from the gut microbiota into your blood and causing inflammation. Now they use CB1 agonists and antagonists to ask, is this connected to endocannabinoids too? Blocking CB1 activity in obese mice, using the antagonist, reduces gut permeability, measured by localization of tight junction proteins, and reduces plasma LPS levels. The agonist, in contrast, increases plasma LPS levels in non-obese mice. So it looks as if overactive CB1 signaling increases gut permeability.
To check that this is a direct effect, Muccioli et al. turned to an in vitro model of the intestinal barrier. In this model, which is essentially a monolayer of gut epithelial cells, both LPS and the CB1 agonist cause decreases in the expression of tight junction protein markers — and, by implication, increase the permeability of the monolayer — while the CB1 antagonist (but not the CB2 antagonist) blocks the LPS effect.
So the story so far is that in obese mice gut microbiota are causing excessive signaling through CB1; this increases the permeability of the gut (which is controlled by CB1), increasing the amount of LPS that gets out into the blood, causing the inflammation that’s associated with obesity. LPS also increases gut permeability in a positive feedback loop. If you make “beneficial” changes to the microbiota, you reduce signaling, blocking LPS release. All very interesting, but does it really matter to the fat cells?
One consequence of the inflammation seen in obesity is increased death of adipocytes and remodeling of adipose tissue. Recent results suggest that one of the problems in obesity is that the generation of new adipocytes is impaired, so lipids accumulate elsewhere, causing insulin resistance. Muccioli et al. treated mice with their gut microbiota-altering regimes and asked what happens to adipocytes as a result. They see (1) reduction of body weight; (2) reduction in CB1 mRNA levels; (3) reduction in the levels of the natural CB1 ligand; and (4) increases in markers of adipocyte differentiation such as PPAR-gamma. The same markers go up if you treat the obese mice with CB1 antagonists. So blocking the excessive CB1 signaling and the resulting inflammation may allow adipocyte development to go back to normal. If you activate CB1 signaling in non-obese mice, however, you also get an increase in adipogenesis markers. The difference here may be that obese mice not only have increased signaling through CB1, they also have increased LPS levels in the blood. In cultured adipose tissue explants, where we don’t have to worry about increased gut permeability, Muccioli et al. show that LPS blocks the CB1-induced increase in adipogenesis markers.
Not a simple picture, but I find myself tentatively persuaded that LPS and the endocannabinoid system may be contributing to mis-regulation of adipogenesis in obesity. And at least in mice, gut microbiota manipulations such as feeding prebiotics appear to cause changes in LPS levels and CB1 activity. None of this constitutes proof that gut microbiota directly contribute to obesity in humans (by the way, check out Jonathan Eisen’s Overselling the Microbiome Award posts), but it’s definitely turning into an interesting story. And now, if you’ll excuse me, it’s time to go grocery shopping. First on the list, brown rice…
Muccioli GG, Naslain D, Bäckhed F, Reigstad CS, Lambert DM, Delzenne NM, & Cani PD (2010). The endocannabinoid system links gut microbiota to adipogenesis. Molecular Systems Biology, 6 PMID: 20664638