London, UK - A new kind of association study, assessing metabolic phenotypes rather than genotypes, has turned up some surprising connections between a number of metabolites and blood pressure [1]. This first metabolomewide-association study will be the beginning of a wave of such research, which should yield novel and important information regarding many diseases, say Dr Elaine Holmes (Imperial College London, UK) and colleagues in their paper published online April 20, 2008 in Nature.

One of the senior investigators, Dr Jeremy K Nicholson (Imperial College London), explained to heartwire that the study involved assessing metabolites from the urine of 4630 participants of the INTERMAP epidemiological study, involving 17 population samples aged 40 to 59 in China, Japan, the UK, and the US. They showed significant differences between the populations in terms of urinary metabolite-excretion patterns, with China at one end of the scale and the US at the other. "We found that you can geographically map people according to their metabolism, and that's new."

We found that you can geographically map people according to their metabolism, and that's new.

"There is much more geographic variation in metabolic phenotype than there is in genetic phenotype," he continued. "For example, we found a huge difference in metabolic phenotype between two Asian countries—Japan and China—but there is almost no difference between these two populations in terms of genotype. This is because they have only culturally diverged in the past few thousand years, so they are not genetically different, but their diets are different. The metabolic fingerprint of these people is capturing the environmental differences between them, including diet." The researchers also found that gut microbes appear to vary widely around the world. "Every single population we've looked at is separated at least partly by gut microbial metabolites, and gut microbes have a huge influence on your health overall," Nicholson explains.

In the study, they identified significant associations between around 30 metabolites and increased blood pressure, the most intriguing of which was formate. "Little is known about the physiology of formate," Nicholson says, "but what we've identified from a blind metabolic study just looking for statistical associations is something that fits in with renal physiology and hypertension rather nicely, indicating that if you can modulate formate levels—maybe by changing diet, changing gut microbes or giving probiotics, for example—perhaps you can affect blood-pressure regulation. The number of things this opens up is really quite enormous; this particular study is just scratching the surface."

Nicholson says the whole list of metabolites that are linked to blood pressure that they unearthed is long. "We'd have taken up a whole issue [of Nature] to go through the biochemistry. But there will be more papers in the specialized journals."

The number of things this opens up is really quite enormous; this particular study is just scratching the surface.

He says that systemic metabolic profiling—or metabonomics—is much simpler, cheaper, and quicker than genomewide-association studies: "You can get a metabolic profile out of somebody in just a minute or two, with no sample pretreatment and with a cost of just a few dollars per sample, and theoretically it can be performed on any epidemiological data set. We predict now that this will be possible with nearly every epidemiological study that's ever been done—you can go back and metabolically profile those samples that have been kept well. It's huge added value."

Of course, genomics is still important, he acknowledges. "Genomics is telling you something about the background biology, but whether or not in the short term we will have any outcomes that are really tractable in a clinical sense only time will tell."

By way of illustration, he cites the differences in metabolic phenotype between Chinese and Japanese. "We know they are genetically indistinguishable, but their incidence of stroke, heart disease, and other conditions are totally different. So that tells us that the reasons for these diseases are basically not genetic."

Another example is the enormous rises in diabetes, obesity, and hypertension in the US in the past two decades. "You're getting massive changes occurring in one generation, so they can't possibly be changes due to genetics; they have to be changes due to the environment or the interaction of environment and genes."

So it's not who you are, but where you live and what you do that's important.

He likens genomics to "a telephone directory without the addresses in—you can get some numbers and maybe some names, but you don't know where the people live. What we are doing with metabonomics is giving the complementary information—the physiologic and metabolic addresses that link to those genes." The next step, he says, will be to link genomics with metabonomics, where people will be classified by both kinds of association studies.

"Once again, what this really brings home is how important lifestyle and environment are to your overall health. We found that Japanese Americans—while genetically identical to Japanese in Japan—have all the disease risk factors of Americans; we find them metabolically virtually indistinct from Americans. So it's not who you are, but where you live and what you do that's important."