News & Views | Published:

Cardiovascular disease

The diet–microbe morbid union

Nature volume 472, pages 4041 (07 April 2011) | Download Citation

A common dietary component that some people even take as a supplement is converted by the gut microbiota to harmful metabolites linked to heart disease. This finding has cautionary implications. See Article p.57

Everyone knows that a 'bad diet' can lead to heart disease. But which dietary components are the most harmful? Some lay the blame on saturated fatty acids, others point a finger at excess carbohydrates, which also lead to obesity and insulin resistance. On page 57 of this issue, Wang et al.1 outline a remarkable chain of events that links diet, intestinal bacteria and liver metabolism to the generation of a chemical that promotes the build-up of arterial plaque and cardiovascular disease.

Intestinal bacteria currently hold centre stage for their role in maintaining digestive health2. Although the main focus has been on detailed molecular characterization of the gut microbiome3, there is increasing interest in the impact of these seemingly innocuous gut microorganisms on metabolic disease in humans. Indeed, recent evidence4,5 has implicated gut microbiota in insulin resistance and non-alcoholic fatty-liver disease.

A burgeoning area of research is metabolomics — an unbiased approach to identifying and measuring the small-molecule metabolites in a system — and determining the relationship of the metabolome to disease. Nonetheless, the scope of the blood metabolome that arises from the gut microbiome has not been fully defined; this knowledge could lead to insights connecting diet, the gut microbiota and disease.

Wang et al.1 tell a compelling story — which starts with a metabolomics approach — of their search for circulating small molecules associated with coronary heart disease. They screened blood from patients who had experienced a heart attack or stroke and compared the results with those from blood of people who had not.

The authors found major differences in choline, betaine and trimethylamine N-oxide (TMAO) — three metabolites of the ubiquitous dietary lipid phosphatidylcholine (also called lecithin). Choline is an essential nutrient6, and lack of dietary choline can lead to non-alcoholic fatty-liver disease and muscle damage. This knowledge has sparked the use of choline as a dietary supplement to prevent liver damage and to increase muscle performance7. Furthermore, lecithin is marketed as a dietary supplement to reduce the risk of heart disease, despite the absence of data to support this claim.

After choline is released from phosphatidylcholine by phospholipase enzymes, gut microbiota metabolize much of it into trimethylamine (TMA) — a gas that smells like rotten fish. When TMA reaches the liver, oxidizing flavin monooxygenase enzymes convert it into TMAO (Fig. 1). Wang and colleagues gave mice isotopically labelled phosphatidylcholine orally and later found the label in TMAO in the animals' plasma, confirming the metabolic link between dietary intake of phosphatidylcholine and the production of TMAO. They also report that, in mice prone to atherosclerosis, increased dietary choline leads not only to increased plasma levels of TMAO, but also to greater plaque development in the animals' arteries.

Figure 1: From diet to disease.
Figure 1

High-fat foods are rich in the lipid phosphatidylcholine (PC) and its metabolite choline (C). Intestinal bacteria convert C to TMA. In the liver, the enzyme FMO3 processes TMA to TMAO — a metabolite that makes its way into the blood. Wang et al.1 show that circulating TMAO may contribute to greater plaque development in the arteries, and so to heart disease.

To demonstrate the role of gut microbiota in this process, the researchers treated mice with broad-spectrum antibiotics, effectively abolishing the animals' intestinal flora. In this setting, phosphatidylcholine administration did not result in TMAO in the blood and, more strikingly, a high-choline diet did not increase the severity of atherosclerosis. A lingering question is whether increased TMAO production contributes to cardiovascular disease or is simply a marker of disease risk.

This paper1 raises the possibility of several new approaches to prevent or treat atherosclerosis. The most obvious is to limit dietary choline intake. Although phosphatidylcholine is found in a wide range of foods, it tends to be particularly high in foods with greater fat content6. Indeed, people with trimethyluria, who cannot convert TMA to TMAO, are prescribed a low-fat, low-choline diet to reduce TMA production8. What's more, our diet comparisons show that a very low carbohydrate (Atkins) diet contains roughly 2.5 times more choline than a typical very low fat (Ornish) diet. It is thus tempting to speculate that a very low fat diet may reduce the risk of heart disease in part because of its low choline content (Fig. 1). These results also call into question the safety of using choline and lecithin as dietary supplements.

Another approach is to reduce the load of gut bacteria that generate TMA from dietary choline. Intriguingly, low-dose antibiotics have been used8 to reduce TMA production in people with trimethyluria. It is of note that antibiotic trials in humans9 set up to test the hypothesis that certain microorganisms, such as chlamydia, may directly infect the arterial wall have not shown cardiovascular benefit. If bacterial species responsible for metabolizing choline to TMA are identified, their selective elimination would be ideal because it would be therapeutically sufficient, and less disruptive to the intestinal microbiota than the broad-spectrum antibiotics Wang et al. used in mice.

A third approach is to use probiotics — live microorganisms that both inhibit and promote various species in the gut microbiome. In a mouse model carrying a 'humanized' microbiome10, administration of a certain probiotic reduced TMAO production, whereas another probiotic increased it. Clinical studies of the effect of probiotics on plasma TMAO levels and on cardiovascular disease in humans would be of interest.

Because TMAO is produced in the liver by the action of the flavin monooxygenase FMO3, inhibition of this enzyme in the liver might be another strategy by which to reduce TMAO production and cut the risk of heart disease. Although complete absence of FMO3 — for instance, in the disease trimethylaminuria — is undesirable, its reduced activity might be beneficial. Whether variations in the gene encoding FMO3 that reduce its activity are associated with reduced plasma TMAO levels and, more importantly, with reduced incidence of cardiovascular disease, should be tested.

Although Wang and colleagues' work1 suggests that excess dietary choline might lead to cardiovascular disease, choline is an essential nutrient for several cellular metabolic pathways. So any attempt to reduce the levels of choline or its metabolites for therapeutic purposes requires caution. Nonetheless, this study has added phosphatidylcholine and other sources of dietary choline — such as the widely used food supplements — to the list of dietary culprits with the potential to increase the risk of heart disease. What's more, it implicates the gut microbiome in promoting heart disease in the setting of a high-choline diet. The implications for prevention of cardiovascular disease are tangible, and the subsequent chapters in this story will make fascinating reading.


  1. 1.

    et al. Nature 472, 57–63 (2011).

  2. 2.

    , , , & Adv. Immunol. 107, 243–274 (2010).

  3. 3.

    et al. Science 312, 1355–1359 (2006).

  4. 4.

    , , , & Science 307, 1915–1920 (2005).

  5. 5.

    et al. Proc. Natl Acad. Sci. USA 103, 12511–12516 (2006).

  6. 6.

    , , & J. Nutr. 133, 1302–1307 (2003).

  7. 7.

    et al. Nutr. J. 6, 30 (2007).

  8. 8.

    et al. J. Am. Diet. Assoc. 104, 1836–1845 (2004).

  9. 9.

    , & J. Am. Med. Assoc. 293, 2641–2647 (2005).

  10. 10.

    et al. Mol. Syst. Biol. 4, 157 (2008).

Download references

Author information


  1. Kimberly Rak and Daniel J. Rader are at the Institute for Translational Medicine and Therapeutics, and the Cardiovascular Institute, University of Pennsylvania Schools of Medicine and Veterinary Medicine, Philadelphia, Pennsylvania 19104-6160, USA.

    • Kimberly Rak
    •  & Daniel J. Rader


  1. Search for Kimberly Rak in:

  2. Search for Daniel J. Rader in:

Corresponding author

Correspondence to Daniel J. Rader.

About this article

Publication history



Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing