Dr. Gary Gibbons, Director, NHLBI:  During the course of this, what was the most surprising finding, or intriguing finding that – I know you've been studying this for, for some period of time, following this line of research. What was the most surprising to you?
Dr. Stan Hazen, Cleveland Clinic:  The most surprising finding was having such a dramatic difference in the metabolism of a common food substance, carnitine, in omnivores as compared to, I'll just say vegans. And it turns out, so carnitine is something that we don’t derive calories from. We, humans, Homo sapiens, have no catabolic enzymes, no way to cleave or degrade carnitine. We either absorb it whole or it just passes right on through and is eliminated in our feces. And so the individuals who had a chronic low carnitine exposure, and that's a vegan or a vegetarian, since carnitine is almost exclusively found in flesh. The root word for carnitine is carnis. It's the same root word as carnivore. And it's found predominantly in flesh, especially red meats. And if you're on a low flesh diet, you’re a vegetarian or a vegan, the types of microbes that can actually chew up and use carnitine as a fuel, they kind of die away. And so those individuals, they ate the one time bullous of carnitine in the form of, whether it's a steak, or a capsule, and they simply did not make the metabolite. But on the other hand, the omnivore, who's chronically exposed to the carnitine, they generated quite a bit of the TMAO. And it was highly variable from person to person, consistent with the variability in the intestinal microbiota of each of us.
I would like to say one thing about this. People are probably wondering well what is this thing TMAO and how does it promote atherosclerosis. I didn't mention this in the summary. What we actually found, and this was the other, there's two things I thought were most surprising. One was how different metabolism could be between an omnivore and a vegan. And we're measuring something in blood and it would be like either almost all there or not, even though you're ingesting the same thing. 
But the second thing was, is mechanistically how the TMAO was linked to heart disease, I found, was very surprising, in how widespread an effect it seemed to have on cholesterol metabolism in multiple different compartments in the body. So TMAO was changing cholesterol metabolism, or facets of it, in cells of the artery wall, in cells in the liver, and in the intestine. And the net affect is enhanced rate of cholesterol deposition because of up regulation of scavenger receptors and macrophages and cells in the artery wall. And decreased reverse cholesterol transport, decreased elimination of cholesterol from, for example, a peripheral cell that we then injected and then followed and traced the cholesterol as it left the peripheral macrophage into the blood, into the liver, out in the feces. There's about a 30% reduction in reverse cholesterol transport that was caused by this, and multiple candidate genes in the various reverse cholesterol transport pathway were affected.
So, this metabolite that gets made by gut flora is influencing cholesterol and sterol metabolism and this, we think, is the mechanism of how, or a mechanism of how it is contributing to the association of enhanced atherosclerosis whenever you have enhanced levels of t-mano, or TMAO.
Dr. Gibbons:  So that raised a couple of questions. One was, your paper was obviously very provocative and, and you’ve done some large-scale studies. But, obviously, this was a smaller subset and, indeed, that one brave soul for the beneficence of science, who ate the steak. How generalizable do you think this is given that variation?
Dr. Hazen:  Well, we looked at five different individuals with the isotope labeled ingestion that were vegetarian or vegan and through the full antibiotic course, before and after and before again, and in those subjects, we saw, in all five, very little TMAO production. We then did, also, a much larger comparison of amongst over 25 vegetarians or vegans versus 30 to 40 omnivores, only at baseline, instead of doing before versus after antibiotics. And we saw similar results. But there was clearly a lot of variability, especially in the omnivore group. And that just has to do partly with, I believe, differences in our diet and clearly differences in the synthetic capacity to make this metabolite. 
So, I think it's fairly generalizable. Certainly, what we see in the human situation is also being paralleled by what we're seeing in the mouse scenario or the mouse models. And I think whether or not bacteria grow in a flask, or a culture, or a petri dish, or in our intestines, if you give them the food that they like, they start to have a selective advantage and will replicate faster. If you start to starve them of a nutrient that they prefer, then they will have a selective disadvantage. And others will have an advantage because it’s zero some gain, we're all, it's a terrarium inside –
Dr. Gibbons:  Right.
Dr. Hazen:  – and there's only so much space and nutrients.
Dr. Gibbons:  Very good.
Dr. Gibbons and Dr. Hazen Transcript
Segment 2