New tools of analysis
Graduate student Nathan C. VerBerkmoes and ORNL researcher Dorothea Thompson at an ion trap mass spectrometer used to identify proteins in Shewanella oneidensis bacteria (in sample bottle) before and after exposure to chromium.
Click here for a high resolution photograph.
In the mid-1990s at Ohio State University, Dorothea Thompson studied a single gene and a single promoter regulating that gene as part of her doctoral thesis research. Today Thompson, a molecular microbiologist in ORNL's Environmental Sciences Division (ESD), acknowledges that today's Ph.D. candidates in genomics no longer focus on, say, determining the DNA base sequence of a single gene or predicting the structure of one protein formed according to one gene's instructions. Instead, like Thompson, these graduate students are taking a systems-level approach to describing biological organisms.
They are engaged in systems biology, invented by physical scientists who apply systems analysis tools to biological problems.
Consider Nathan VerBerkmoes, a third-year Ph.D. candidate in the University of Tennessee-ORNL Graduate School of Genome Science and Technology (GST), who is working with Bob Hettich, a mass-spectrometry expert in the Organic and Biological Mass Spectrometry Group in ORNL's Chemical Sciences Division (CSD). By developing and demonstrating a mass-spectrometry-based technology platform for systems biology studies, VerBerkmoes has been able to contribute as a first author or coauthor on at least 10 scientific journal papers that have been published, are in press, or are under review.
"Many researchers have spent their entire careers investigating a particular protein--its interactions, its regulation, and its pathways," Hettich says. "Systems biology is the opposite of this conventional biological approach because it takes a global view down rather than a reductionist view up. You don't target any particular gene or protein but rather take a snapshot of the whole organism and all of its parts working together."
In the case of a bacterial cell, systems biology attempts to integrate all the DNA information (the genome), the RNA information (the transcriptome), the protein information (the proteome), and the metabolite information (metabolome). "The integration of this global information should provide a composite description of the whole function of the organism," Hettich says.
In a couple of systems biology projects, VerBerkmoes and Hettich collaborate with Thompson, who has obtained microarray data on specific activated genes of the Shewanella oneidensis bacterium in the presence of radionuclides and toxic metals, such as strontium and chromium. Using mass spectrometers to analyze S. oneidensis as it makes a metal less soluble and more likely to stay put in sediments or soil, VerBerkmoes cranks out lists of proteins and their relative concentrations.
Thompson's microarray data show that the expression level of some genes has risen, and VerBerkmoes' mass spec data indicate an increase in the abundance of corresponding proteins encoded by those genes. They send their data to computational biologists for additional interpretation and analysis. From this kind of collaboration emerges a scientific paper.
Genomics, proteomics, and bioinformatics have all come into play in this research. The work is an example of "microbial functional genomics," says Thompson, who coauthored the first textbook on the subject with three other researchers, including her colleague Jizhong Zhou, a pioneer in the environmental applications of DNA microarray technology. But, broadly speaking, understanding the roles of genes, proteins, protein complexes, regulatory sequence elements, and complex regulatory networks within cells and how they operate together to enable cells to function and survive is systems biology at work. Changes in system components and their interactions can best be studied when the cells are either grown in different ways or stressed by exposing them to toxic or radioactive metals.
From Genomics to Proteomics
Hettich explains that genomics embraces not only the order of the DNA bases but also the location of all the genes in a particular genome. The genome is translated into the transcriptome--the RNA level indicating which genes are turned on and which ones have a high expression level.
The next level is the proteome--the proteins produced by the cell in response to instructions from the expressed genes.
Shewanella oneidensis bacteria.
Click here for a high resolution photograph.
The proteins interact with each other and form protein complexes, which carry out much of the work of the cell. Identifying and analyzing protein complexes in two bacterial species--Shewanella oneidensis and Rhodopseudomonas palustris--is the goal of the Department of Energy's Genomics: GTL Center for Molecular and Cellular Systems, of which Michelle Buchanan is scientific director.
Buchanan, CSD director, calls R. palustris "a good bug for DOE's missions because it might be useful for hydrogen production, carbon sequestration, and waste remediation. That's partly why we chose it. Depending on how R. palustris is grown, it will turn on different parts of its apparatus to take different pathways to ensure survival. That's also why we want to understand from a systems approach how we can control those metabolic pathways to get the microbe to do all the things we want it to do simultaneously."
Frank Larimer, leader of the Genome Analysis and Systems Modeling Group in ORNL's Computer Science and Mathematics Division, says that computational biologists collaborate with experimenters in an iterative process. "Experimenters help us fine-tune a computer model of a bacterium, such as R. palustris," he says, "to get a comprehensive view of this biological system and to predict how best to re-engineer it to maximize its ability to achieve a desired function, such as producing hydrogen."
"Systems biology looks at a microbe as a cell--as a microbial system with all the genes, RNA, ribosomes, proteins, and regulatory sequences that are active all the time as a total system," says Brian Davison, director of ORNL's Life Sciences Division. "Then it examines groups of microorganisms, communities, and ecosystems of microorganisms all working together."
In the past, ecologists have tried to understand the functioning of ecosystems--from forests to wetlands to deserts--in terms of different species, including bacteria, plants, and animals. The approach has not worked. Some ecologists now argue that it may be possible to understand ecosystems by starting with DNA molecules. The concept often elicits laughter, but according to ESD's Steve DiFazio, there are reasons to believe this approach could work.
"Multicellular organisms share the same genetic code and are related evolutionarily. For example, most organisms respire in about the same way. We can exploit this shared ancestry and conservation of function to design genomic tools that can be used to elucidate the underlying rules that govern the organization and functioning of ecosystems.
"In traditional ecological research, we isolate an individual plant in the laboratory and study how it responds when grown under changing conditions," he continues. "But when we grow the same plant in a natural ecosystem in the field, we find that typically it will respond differently in the field than in the lab. The organism will interact with other plants, as well as fungi and bacteria in the soil.
"Often, the responses from individual lab experiments don't allow ecologists to predict accurately the results in the field because of interactions among unknown organisms in natural ecosystems. Thus, ecologists must ultimately study and manage organisms in the context of ecosystems. In much the same way, individual genes cannot be studied in isolation but must be functionally characterized in the context of living organisms. This is the signature purpose of systems biology. Just as parts of cells exchange protein subunits, organisms in an ecosystem can exchange metabolites and signals that help each other survive," DiFazio says, explaining that metabolites are compounds containing carbon, nitrogen, and other elements that provide cells with energy and building blocks of proteins.
"For example, certain fungi (mycorrhizae) can survive only by growing on tree roots. Without the fungi, the tree will not grow as well or survive severe drought. Genomics provides tools and information that will allow us to understand the mechanistic basis of these complex interrelationships."
"We are sequencing the fungi that affect the poplar tree's mineral uptake and drought tolerance," says ESD's Jerry Tuskan. "We are beginning to build bioinformatics resources that will allow us to look at how enzymatic systems within organisms interact, how organisms interact among themselves, and how those interactions are shaped by other organisms and physical components in the environment."
To understand ecosystems, researchers may need to get down to the molecular level, to the level of genes and proteins, to understand why components that are similar yet different--like ORNL staff members and UT graduate students--are so dependent on each other.