If you have an ACS member number, please enter it here so we can link this account to your membership. (optional)

ACS values your privacy. By submitting your information, you are gaining access to C&EN and subscribing to our weekly newsletter. We use the information you provide to make your reading experience better, and we will never sell your data to third party members.


Analytical Chemistry

NIH Metabolomics Centers Ramp Up

Six NIH centers have been created to provide researchers with analytical tools and services to study biochemical pathways

by Jyllian Kemsley
November 4, 2013 | A version of this story appeared in Volume 91, Issue 44

This image shows how metabolites are interrelated.
Credit: Oliver Fiehn/UC Davis
The UC Davis metabolomics center has developed metabolite maps to depict the relationships of structurally identified compound classes. Arrangements are determined by chemical similarity (gray lines) and known enzymatic transformations (red lines).

The task of the cells that make up our bodies is metabolism—the process of chemically transforming sugars, amino acids, nucleotides, lipids, steroids, and other molecules in ways that ultimately allow us to grow and reproduce. The systematic study of the collection of small-molecule products of cells, a field called metabolomics, offers a unique window into the effects of nutrition, disease, drugs, or environmental exposures.

If something in a cell is amiss, with metabolomics “you will see immediately any dysregulation,” says Oliver Fiehn, a professor of molecular and cellular biology and director of the West Coast Metabolomics Center at the University of California, Davis.

Metabolomics studies, however, are analytically challenging. Cells act as central processing units for a sea of thousands of molecules, some varying widely in structure and some differing only by placement of a simple methyl group. Some of the chemicals are early risers and others night owls, differing in concentration and reactivity depending on the time of day or when an organism last ate. The vagaries of the molecules can swamp a change in a key biomarker for monitoring health or detecting a disease. Metabolomics studies, therefore, require sophisticated analytical instruments and data analysis tools—not to mention stringent experiment control.

To advance the field of metabolomics, the National Institutes of Health announced last year that it would invest as much as $51.4 million over five years in a metabolomics program. Much of that effort has involved establishing six regional metabolomics centers. The dual mission of the centers is to provide experimental and data analysis services to the research community and to develop new tools for metabolomics research and get them into the hands of scientists.

The first three centers were established last year at UC Davis, the University of Michigan, and North Carolina’s Research Triangle Institute (RTI). NIH added the other three this fall, at the University of Florida, the University of Kentucky, and Mayo Clinic, in Rochester, Minn. Each center’s core analytical services focus on mass spectrometry and nuclear magnetic resonance spectroscopy. Some centers also do imaging using a variety of approaches, including positron emission tomography and mass spectrometry. Additionally, NIH established a data repository and coordinating center at UC San Diego.

Although much of the work of the centers is targeted toward biomedical applications, the centers may range further afield. For example, other areas where metabolomics may be useful include agriculture or alternative energy, says the director of the University of Florida center, chemistry professor Richard A. Yost.

No matter the facility’s focus, the center directors note that their work with clients on experimental design is just as important as the experiments themselves. Strict control of experimental subject behavior is necessary, says Mayo Clinic’s K. Sreekumaran Nair, who is also a professor of endocrinology. For example, if a team’s study involves collecting a blood or urine sample from a research subject in the morning, they should consider that a participant’s metabolite profile will depend on whether the subject ate or exercised before sample collection. Mayo metabolomics experts might recommend putting participants on a standard diet for three days and hospitalizing them overnight to better control sampling conditions.

Researchers must also control sample processing and storage, adds RTI metabolomics center director Susan J. Sumner, who also leads her institute’s efforts to transition advances in the lab to clinical use. For blood samples, several different anticoagulants can be used as part of the processing, but researchers need to know which ones and be consistent throughout the study for all samples. “You don’t want the biggest distinguishing factor between groups to be the type of anticoagulant,” Sumner says.

Another challenge is the sheer variety of metabolites and their concentrations. A single sample preparation may not yield all the desired information—different preps may be necessary for running different analyses on different instruments.

Then there’s some researchers’ expectation that all peaks in a chromatogram or spectrum will be identified. “We find lots and lots of features on a chromatogram, but for the most part we don’t know what the vast majority of them are,” says Michigan’s Charles F. Burant, a professor of metabolism. That doesn’t necessarily diminish a peak’s utility—even if its chemical identity is unknown, it can potentially still be used as a biomarker to pinpoint early stages of a disease. “But some people don’t like not knowing what it is,” Burant says.

Metabolomics in the past has generally provided only a snapshot of a particular portion of a metabolome at a particular point in time. The new centers expect their work to be broader and more informative. In some cases, that means targeting new classes of molecules.

At UC Davis, those metabolites include glycans, steroids, complex lipids, and lipid metabolism mediators known as oxylipins, Fiehn says. Lipids and oxylipins, for example, are of interest in studying ovarian cancer because the cancer cells nest in fat tissue. “Ovarian cancer really eats up the fats,” Fiehn says. “It must have a very different type of metabolism, and there are indications for a big role for oxylipins.”

Other metabolomics studies are moving away from targeted groups and instead are looking for everything that samples may have to offer. “Our expectation is that we are going to be setting up and delivering global, untargeted metabolomics for a variety of samples, including blood, tissues, and plants,” says Florida’s Yost.

Researchers are also starting to look at chemical profiles in flux, not just at one particular moment, to observe how the compounds change over time. Understanding chemical flux through particular metabolic pathways is where Michigan’s Burant believes metabolomics will reveal its utility in explaining an organism’s response to different stimuli or the progression of a disease.

Augmenting mass spec and NMR studies by labeling molecules with isotopes will help. These studies will be a particular focus of the Florida and Kentucky centers. For example, 13C can be added to a nutrient, such as glucose. By using 13C-labeled glucose, researchers can simultaneously track individual metabolic pathways and illuminate networks of pathways, says Kentucky center director and professor of toxicology Richard M. Higashi.

Isotope-labeling experiments have helped cancer researchers start to understand how cancer cells can reprogram metabolic pathways. They’ve also made researchers realize that they know less about metabolism than they had thought. Published metabolic pathway maps still can’t completely account for where 13C atoms turn up when labeled glucose is fed to cells, Burant says.

In addition to efforts to expand their analytical targets, the centers are working to improve instrument sensitivity and to make methods more reproducible. NIH is funding a metabolite synthesis program to prepare compounds as standards to help with validating methods or identifying unknowns.

The centers are also working to improve bioinformatics, a computational approach to help with data analysis, interpretation, and visualization. Data management is an enormous challenge when one sample can yield 1 gigabyte of information, Florida’s Yost says. The Mayo and RTI centers are also involved in translational efforts to connect metabolomics with clinical science, especially in areas such as nutrition, obesity, and diabetes.

Scientists at the centers are excited about where the projects will take them. “As a group, we chemists tend to collaborate with people in a cottage industry model, just you and someone else on a particular problem,” Yost observes. “The centers are an opportunity not just to innovate in chemistry but to have a broader impact.”


This article has been sent to the following recipient:

Chemistry matters. Join us to get the news you need.