Photo by Thomas Splettstoesser, courtesy of Oak Ridge National Laboratory
Most of the mercury absorbed by humans through the food chain is in the form of methylmercury manufactured by microbes.
Most of us are at least vaguely aware of the risks of mercury exposure associated with eating fish—it's pretty widely known, for example, that a constant daily diet of albacore tuna is not to be recommended. What is less widely understood is that consumption of fish and shellfish is actually the main source of human exposure to mercury across the globe. And mercury exposure—which at high levels can lead to serious neurological and other disorders and even death—is something that most of us would presumably prefer to avoid. Indeed, government regulators recommend that pregnant and nursing mothers and young children abstain from or limit consumption of certain kinds of fish (see link below). But the interesting story here is that the ultimate culprit behind all this turns out not so much to be the fish or shellfish themselves, but rather microbes.
Certain microbes living in rivers, lakes, and oceans take a form of mercury—ultimately derived from burning coal from power plants, erupting volcanoes, and other environmental sources, and absorbed into the water—and convert it into a highly toxic substance called methylmercury (CH3Hg+). It is this methylmercury that is consumed by fish and shellfish all over the world. Not only is this methylmercury a particularly nasty toxin; it is also "bioaccumulative," meaning that it accumulates in the body and works its way up the food chain, as fish ingest other fish that have ingested methylmercury in turn. (For that reason, the oldest and biggest fish—such as albacore tuna and shark, for example—also tend to have the highest methylmercury levels.)
All that has been known for quite a while. But only lately have we begun to understand the mechanism by which this microbial production occurs. A research team led by the U.S. Department of Energy's Oak Ridge National Laboratory has recently achieved major new insights into the mechanism of microbial "mercury methylation" by identifying two key genes in microbes that are responsible for this process. The discovery, which cracks a mystery that scientists have pursued for more than 40 years, could open the way to potential advances in both detection and remediation of environmental mercury contamination. The work also illustrates how the new tools and methods of contemporary "systems biology" are enabling researchers to penetrate biological mysteries that have eluded previous generations of investigators. The research, which also included collaborators from the University of Tennessee and the University of Missouri, was recently published in the journal Science.
Oak Ridge National Laboratory has a special interest in mercury, dating from the Cold War years. In the 1950s and 1960s, the laboratory's neighboring Y-12 nuclear plant, another DOE facility, used large quantities of mercury for the enrichment of lithium in the production of hydrogen bombs. Tons of that mercury were lost or unaccounted for, and mercury contamination at the Oak Ridge site consequently remains an issue. For example, the East Fork Poplar Creek, which originates within the Y-12 Site, has been declared a mercury hazard since 1982.
Partly for this reason, DOE's Office of Science (SC) has invested substantially in research at Oak Ridge aimed specifically at basic science designed to advance the understanding of mercury's properties and behavior in contaminated environments.
The work has been part of a larger effort by SC's Office of Biological and Environmental Research (BER) to support genomics-based research probing the microbial world for solutions to a range of energy and environmental challenges. The effort began to take root during BER's participation in the Human Genome Project in the 1990s and became the theme of a major report by BER's formal federal advisory committee in 2000. The consensus at the time was that bringing the new tools of genomics-based biology to bear on the microbial world could help find solutions in a wide range of DOE mission-related areas, from subsurface contamination at DOE sites to the potential development of new energy sources and bio-based products.
What the researchers needed was a way to narrow their search. Using chemical principles and structural bioinformatics they focused on identifying proteins that might carry out a specific kind of methyl transfer. They then hunted for genes encoding such protein targets among known methylating bacteria in the genome database, and it was this strategy that ultimately led them to the discovery of the genes.
Not surprisingly, therefore, the first step toward unlocking the mercury methylation mystery—and indeed typically the first step in almost any systems biology research effort today—was the genomic sequencing of some of the microbes involved. In 2003, BER supported the sequencing of the genome of one major mercury methylator, the bacterium Geobacter sulfurreducens PCA. (At the time, the sequencing was motivated not primarily by an interest in mercury methylation, but rather by what was thought to be the microbe's potential to neutralize radioactive metals in the environment—another legacy issue at DOE weapons sites.)
Sequencing is a prerequisite for understanding an organism, but by itself it doesn't necessarily tell you a lot. The genome of G.sulfurreducens PCA, for example, has more than 3.8 million "base pairs" (base pairs being the units of DNA) and is thought to encode more than 3,500 genes. One of the oldest techniques for determining the function of a gene is to "knock it out," or delete it from the genome, and see what happens to the organism as a result. Some researchers did try this approach. But with thousands of genes to explore, the knock-out method was the equivalent of searching for a needle in a haystack.
A second important sequencing project focused on the genome of another microbe known to methylate mercury, the bacterium Desulfovibrio desulfuricans ND132. This project, completed in 2011, was actually undertaken by the Oak Ridge mercury researchers and was specifically aimed at understanding mercury methylation.
But the genome of D.desulfuricans ND132 posed a similar puzzle. It too was 3.8 million base pairs long, also with a predicted 3,500 or so genes.
What the researchers needed was some means of narrowing their search. To do this they turned to another major technique of today's systems biology—comparative sifting of biological databases. Once upon a time, biology research tended to involve a single researcher or small research team working at a lab bench and performing a series of "wet" experiments to test a hypothesis. That kind of biology has by no means disappeared. But in the post-Human Genome Project world of systems biology, research is often as likely to take the form of scouring large and growing databases containing information on genomes and their function (i.e., individual genes and the proteins they encode) for hundreds or even thousands of organisms, in search of illuminating comparisons.
In this case, that approach enabled the researchers to identify nine genes possibly linked to mercury methylation. But while this finding put the team on the right track, they could not clinch the case simply by this method. The researchers were only able to pinpoint the genes they were searching for after carefully considering the actual biochemistry involved in mercury methylation.
The Oak Ridge researchers spent some time thinking about the biochemical reactions needed to bind a methyl group (CH3) to mercury (Hg), and which proteins from various biochemical pathways in microbes might be involved in such a reaction. With the help of that analysis, they identified a protein from a different microbe that is known to transfer a methyl group (not to mercury). They reasoned that the mercury methylators might have a similar protein with a similar-looking gene. This proved to be an excellent bet.
Using an algorithm called BLAST (for "Basic Local Alignment Search Tool"), the researchers compared a section of the gene for their target protein to the genome of D.desulfuricans ND132. One of the nice features of BLAST is that it develops statistics on similarity, which means that one isn't necessarily relying on an exact match. Using BLAST, the researchers found a close enough candidate gene in D.desulfuricans ND132.
Photo courtesy of Oak Ridge National Laboratory
Oak Ridge National Laboratory researcher Stephen J. Tomanicek performed experiments in the study.
The team then compared their candidate gene to genes in five other known mercury methylators for which complete genomic sequences existed. They found similar genes (encoding similar proteins) in all five microbes. They also found that this candidate gene was not present in the genomes of closely related microbes that do not methylate mercury.
Finally, by examining the genomes of the six known mercury methylators for which complete genomic sequences existed, they noticed that their candidate gene seemed to be paired with a second gene that all six had in common.
This pair of genes therefore seemed to hold the key to mercury methylation (though researchers acknowledged that other, additional genes must also be involved). Importantly, neither of the two genes was present in eight closely related microbes that were also sequenced and known not to methylate mercury.
So the researchers hypothesized that the two genes—and their corresponding proteins—were essential to mercury methylation. To test their hypothesis, they employed the tried-and-true method of gene knock-out. They deleted the two genes, singly and in pairs, from both microbes. It turned out that if either gene was missing, the microbes could no longer methylate mercury. Conversely, when researchers restored both genes, mercury methylation resumed. (Similar genes were found in more than 50 other microbes for which sequences exist, suggesting that the ability to produce methylmercury may be more widespread in the microbial world than previously recognized.)
For the first time, therefore, researchers have access via the two genes to two key proteins that are required for mercury methylation. This paves the way for the first insights into the actual mechanism of mercury methylation and should lead, down the road, to improved detection, and perhaps ultimately remediation, of mercury contamination and its potential impact on the environment and human health.
--Patrick Glynn, DOE Office of Science, Patrick.Glynn@science.doe.gov
Jerry M. Parks, Alexander Johs, Mircea Podar, Romain Bridou, Richard A. Hurt, Steven D. Smith, Stephen J. Tomanicek, Yun Qian, Steven D. Brown,, Craig C. Brandt, Anthony V. Palumbo, Jeremy C. Smith,, Judy D. Wall, Dwayne A. Elias, and Liyuan Liang, "The Genetic Basis for Bacterial Mercury Methylation," Science 339, 1332 (2013).
B. A. Methé et al., "Genome of Geobacter sulfurreducens: Metal Reduction in Subsurface Environments," Science 302, 1967 (2003).
Steven D. Brown et al., "Genome Sequence of the Mercury-Methylating Strain Desulfovibrio desulfuricans ND132, Journal of Bacteriology 193, 2078 (2011).
DOE Office of Science, Office of Biological and Environmental Research
Oak Ridge National Laboratory, Geochemical and Molecular Mechanisms Controlling Mercury Transformation in the Environment
DOE Office of Science, Office of Biological and Environmental Research, Climate and Environmental Sciences Division, Subsurface Biogeochemical Research
What You Need to Know About Mercury in Fish and Shellfish: EPA and FDA Advice For Women Who Might Become Pregnant, Women Who are Pregnant, Nursing Mothers