This story is a supplement to the feature "Jeremy Nicholson's Gut Instincts: Researching Intestinal Bacteria" which was printed in the July 2008 issue of Scientific American.

One of the hottest biomedical fields right now is metabolomics—the study of the metabolites and other chemicals that the body and its bacteria produce. The goal is to find out how the compounds can serve as indicators of health and disease. For the Insights story, "Going with His Gut Bacteria," in the July 2008 Scientific American, Melinda Wenner talked with Jeremy Nicholson of Imperial College London. One of the founders of the field, Nicholson thinks that metabolomics may prove that the best medicine actually targets intestinal flora rather than cells of the body. Here is an edited excerpt from the interview.

You were one of the first scientists to study the metabolome, the collection of chemicals produced by human metabolism. Was it hard getting people to take the idea seriously?
Nobody was in the slightest bit interested. I had terrible difficulties getting funding throughout the 1980s in this area. I remember sending a paper to Nature in 1987 that showed how you could use nuclear magnetic resonance and computational pattern recognition to look at urine from animals that had been poisoned with lots of different sorts of drugs. The editor said, "There's no interest [in this] to anybody whatsoever." That would have been 10 years in advance of the first paper that would really call itself metabolomics or metabonomics.

Over the 10 years that followed, I built up a hell of a laboratory, so when our work started to get noticed, we were already one of the best-equipped labs in the world.

Why was no one interested back then?
I don't think it was necessarily willful resistance; there was a lot of other stuff going on. In the '80s molecular biology had just come in. You couldn't get a grant in the U.K. unless you were doing molecular biology, because everybody thought that was going to solve everything. Then, also, in the late '80s you had the idea of genomics coming in.

Why do you think that the metabolome is more likely than the genome to give scientists the answers they want?
Genomics only takes you part of the journey to real biological discovery. The genome is a blueprint for life, but it doesn't tell you how the thing works. If you had a blueprint for a nuclear power station, it would tell you exactly how to build one, but it wouldn't tell you anything about quantum mechanics, physics, the idea of nuclear fission, radioactive decay or anything that made it work. You can look at the genome the same way. It may well have a blueprint for building life, but it doesn't tell you how the parts fit together.

And your work has shown that the environment makes a huge contribution to your health.
People talk about the genes that make you fat, but really, if you sit on your butt eating pork rinds and Big Macs and watching television, you will get fat, no matter what your genes say. What you do to yourself is really important. Metabolism captures environmental signatures as well as genetic. Your environment involves things like drugs you're exposed to, the pollutants you're exposed to, the products of your gut microbes, the metabolic products of your diet—so when we do a broad-screen metabolic profile, we're capturing all of that information, plus information that links to genome variation. For me, metabonomics is the most holistic of the "-omics." In principle, it can capture the signature of everything.

We've found that humans are far more metabolically diverse than genetically diverse. For instance, Chinese and Japanese people are actually metabolically very distinct, despite the fact they're genetically near identical. And they have very different incidences of diseases.

How could scientists use this information to inform medicine?
I have this new concept of metabolome-wide association study. It will allow us to sample the genetic and the environmental things that cause diseases in people. We've found metabolic biomarkers that link to things like blood pressure in humans. Using this approach, we can generate new hypotheses in physiology that can be tested and may ultimately result in new drug discovery.

And you believe many of our metabolic differences have to do with gut bacteria. How did you come to realize that these microbes were so important for our health?
I've always known, ever since we started working on metabolic profiling, that there were metabolites that came from the gut microbes. We never really paid a lot of attention to it until maybe about seven or eight years ago, though. It was not just me—it was also Professor Ian Wilson [a scientist at AstraZeneca in England]. He became intrigued because he looked at colonies of rats—supposedly very, very similar groups of rats—but some produced one set of metabolites and others produced a different set. And yet they were from the same breeder; they were the same genetic strains. The differences were down to different gut microbial populations in rats residing in different parts of the laboratory.

The more we looked into it, the more we realized that microbes were so intimately involved in animal metabolic processes that they might have contributions to disease development in ways that hadn't really been thought of before. We're really just starting to expand this now, thinking about how gut microbes influence all sorts of things. They have influences on liver diseases and gut pathology like Crohn's disease and irritable bowel syndrome; there's even evidence that autistic children have very, very different gut microflora [than other children]. Almost every sort of disease has a gut–bug connection somewhere. It's quite remarkable.

What, ultimately, are you hoping to achieve with metabolomics?
We want to be able to take a set of biological data from a human being, and then, based on what we know about the metabolic makeup of that person, say how long they're going to live, what diseases they're likely to suffer from, how to treat those diseases, and how to manage their lifestyle and drug therapy optimally. We're opening up sets of doors here into the future of health care—the manipulation of biology that would be just unimaginable five years ago.

Any funny or surprising moments you'd like to share from your research?
We did some work about 10 years ago at another person's laboratory on something called magic-angle spinning spectroscopy [a kind of NMR spectroscopy that relies on spinning the sample to achieve higher resolution data]. What I was interested in was whether or not we could get some extra information out of lipoprotein signals by spinning the probe very, very fast. I put the blood plasma sample in and the spectrum that came out was totally nothing like plasma is normally. I thought, absolutely fantastic! We've liberated all this new information! We tried several more samples and the same thing happened, and so I started to chat with one of the guys in this laboratory. I said, "We got an amazing spectrum, it looks nothing like plasma spectra should be." And he said, "Oh, show me!" And I showed him and he said, "Hmm, that looks very familiar." To cut a long story short, what happened was that the previous week the guy had been running samples of blue cheese—a food science company had been conducting experiments. Rather than discovering a new part of the fundamental dynamics of lipoproteins, we discovered how to detect blue cheese in plasma.