As a practicing gastroenterologist at Mount Sinai Hospital in New York City, Louis Cohen, M.D., was driven to understand how commensal bacteria interacted with host cells in inflammatory bowel disease, Crohn's disease and ulcerative colitis. As a researcher in the laboratory of Sean Brady, Ph.D. at The Rockefeller University, NYC, he found answers.
Several years ago, the Brady research group was using metagenomics—a method of extracting microbial DNA from environmental samples directly, without culturing the bacteria—to characterize bioactive natural products, primarily from soil samples. But Cohen says his interest in patient-oriented research "allowed us to start thinking about another environment—namely, the human microbiome as opposed to the soil microbiome. After that, the idea of using metagenomics as a drug-discovery platform just clicked."
"Metagenomics is the answer to tried-and-true, culture-based small molecule discovery approaches," Cohen, an SLAS2016 podium presenter, explains. Culturing bacteria from a sample and looking at the molecules they produce has "obvious limitations," he says. First, molecules of interest can be mined only in bacteria that can be cultured—and many can't be. "Secondly, we can only investigate molecules we can actually produce in the laboratory, yet many complicated environmental signals are necessary for the production of many molecules, and we can't easily reproduce them."
For example, researchers may know that certain factors are part of a bacterial genome, but nutrients such as iron may also be needed to trigger the production of a particular protein or enzyme. "It takes time and a good deal of trial and error to figure out the right mix of cofactors and signaling factors and nutrients required to make a molecule that might impact host cellular functions," Cohen emphasizes. And that's where metagenomics comes in. As the National Research Council of the National Academy of Sciences explains in The New Science of Metagenomics: Revealing the Secrets of Our Microbial Planet:
"All metagenomics studies take the same first step: DNA is extracted directly from all the microbes living in a particular environment. The mixed sample of DNA can then be analyzed directly, or cloned into a form maintainable in laboratory bacteria, creating a library that contains the genomes of all the microbes found in that environment. The library can then be studied in several ways, based primarily either on analyzing the nucleotide sequence of the cloned DNA or on determining what the cloned genes can do when they are expressed as proteins..."
Sequence-based metagenomics relies largely on next-generation sequencing and robust computational tools to amass and make sense of the huge amounts of data gathered in the study of interactions within microbial communities. Indeed, the fact that next generation sequencing has become much more affordable "has allowed metagenomics to come to the forefront," Cohen says. "Now individual labs have their own sequencers (ours has two), and therefore the ability to identify biosynthetic clusters in their samples."
Functional metagenomics, which is Cohen's area of expertise, enables researchers to investigate the proteins produced by bacteria without first having to understand the microbial community's underlying gene sequence or the structure of the protein of interest. "We can study any bacteria from which we can extract high molecular weight DNA—and for us, so far, that's been all of them," Cohen says. "We've also been able to eliminate the need to go through trial and error with nutrients and various transcription factors by using strong promoters to force genes of interest to switch on."
By providing a systematic way of surveying commensal DNA for genes that encode effector functions, functional genomics also has the potential to "expand how scientists approach drug discovery," Cohen continues. "Because commensal bacteria live inside of us, they are themselves an endogenous drug delivery system. Therefore, the way we normally conceptualize a 'drug'—based on having certain chemical properties that make it amenable to injection, oral administration and so forth—becomes less important, since anything endogenous bacteria produce that affects a host pathway can be considered a 'drug.' And because we can manipulate these bacteria, again, we can think about drug discovery methods in a much broader sense because we know we have this wonderful natural delivery system we can potentially tap into."
Cohen's most recent work offers insights into the mechanisms by which commensal bacteria interact with their human hosts. Using functional metagenomics, he and his colleagues identified a specific commensal metabolite involved in host-microbial interactions. Reporting in a recent issue of Proceedings of the National Academies of Sciences (PNAS), the team describes how they screened commensal DNA cloned from three phenotypically distinct patients for genes that activate NF-kappa B, a transcription factor known to play a key role in mediating cellular responses to environmental stimuli and in regulating the immune response to infection. They identified 26 unique commensal bacteria effector genes (Cbegs), one of which—Cbeg12—was recovered from all three patient libraries. Further analysis revealed that Cbeg12 encodes for the production of N-acyl-3-hydroxypalmitoyl-glycine, or commendamide.
"Commendamide resembles human long-chain N-acyl-amides that function as signaling molecules by activating G-protein-coupled receptors—specifically, GPCR 132/G2A, which is implicated in both autoimmune disease and atherosclerosis," Cohen explains. "Taken together, the findings show the utility of functional genomics for both identifying potential mechanisms used by commensal bacteria for host interactions and outlining a methodical approach to the identification of diverse Cbegs that impact host cellular functions."
"The biggest challenge we had in our latest study was size, and this is true of most metagenomic studies," says Cohen. "We had to survey 3000 megabases of metagenomic clones to get at the 26 genes we reported on and the single molecule we identified. Because the average bacteria contain about five megabases of DNA, that's a lot of DNA to survey. We know we can't use some of the more robust and reproducible screening technologies—the reagents for luciferase-based screens and other chemical probe-based screenings are very expensive to run at this scale. And because we're using them to generate hypotheses rather than to find a previously identified small molecule of interest, we have to be careful about costs."
That concern led the group to develop a novel and less expensive method to study host-microbial interactions in vitro, using a fluorescent reporter and high-content imaging. The approach works, but has a downside: Because high-content screening is so sensitive, it generates a lot of "noise" (false positives). This meant the group had to undertake several iterations of fine tuning the equipment to ensure the proper experimental conditions, Cohen says.
Cohen will continue using functional metagenomics methods to better understand how commensal bacteria might interact in different patient cohorts. "We know from from sampling DNA from different patients that there's a lot of variability in the types and amounts of bacteria present in individuals. Functional metagenomics will enable us to explore individualized approaches, based on what we find."
He will also be looking at additional ways to facilitate a better understanding of the mechanisms underlying host-microbial interactions. "We're continuing to develop secondary assays to help us get at the biology that's downstream of our recent work," Cohen explains. "We start with a first pass at screening using a low-cost, highly sensitive technology that will get us to numerous bacterial pathways and a slew of molecules that we know have the potential of interacting with host cells. Then we use the secondary assays to focus on specific molecules that are prevalent. And so instead of dealing with 75,000 clones, we end up working with 50, 100 or 1000. Those numbers are much more amenable to some of the traditional chemical-based probes, and allow us to come up with more testable hypotheses about which molecules produced by commensal bacteria are important to specific diseases."
Cohen and his colleagues "are always open to collaboration," he stresses. "Metagenomics, by its nature, spans many disciplines, from microbiology to molecular biology to immunology. We've already started working with immunologists and medicinal chemists. The next round of investigations will require other model systems, so we want to work with researchers who have experience working in different models.
"What excites me more than anything is the possibility of working with a group who has set up a reporter system because they're interested in a particular receptor or pathway but haven't yet been able to identify the factors that modulate it," Cohen continues. "That's a fantastic opportunity to collaborate on building a metagenomic library and using our approach. We've made our methods available in several reports, and it's not very expensive, so there should be no barrier to entry. We want others to use our system and hopefully find it effective, because the amount of pathways and diseases that commensal bacteria interact with are tremendous."
The team has begun talking with other researchers to initiate these kinds of collaborations. "One group has a reporter cell line set up for their pathway and an orphan receptor they haven't been able to find a molecule for," Cohen observes. "They'll be able to use our system and their cell line to very quickly move directly to the process of isolating the gene and then either the molecule or protein or enzyme factor they're seeking." Cohen also is looking forward to meeting potential collaborators at SLAS2016.
Such collaborations are in keeping with the National Research Council's vision for metagenomics:
"Metagenomics will draw on expertise from many disciplines and individuals:
• Those with knowledge of microbiology, including microbial genetics, biochemistry, physiology, pathology, systematics, ecology, and evolution.
• Other biologists, including molecular and cellular biologists and those with knowledge of host organisms, such as humans and other mammals, plants, insects, and microbial hosts with important roles in nature or of economic importance.
• Those with knowledge of the environment, including soil and atmosphere scientists, geologists, oceanographers, hydrologists, and ecosystem scientists.
• Computational scientists, including those with knowledge of statistics, computer science, data mining and visualization, database development, modeling, and applied mathematics.
• Those with expertise in scaling information to large ecosystems, and in evaluating the effects of global change and its interface with policy.
• Engineers, physical scientists, and chemists whose skills and insights are potentially field-transforming in their contribution to new methods, chemistry, devices and applications (within and beyond metagenomics), and the understanding of complexity, networks, and system structure."
Cohen offers insights into his metagenomics research projects at SLAS2016, emphasizing how working with commensal bacteria can broaden the scope of drug discovery to include different types of molecules, as well as the possibility of endogenous delivery systems. His podium presentation, "High-Throughput Screening of Metagenomic DNA Libraries," will be held on Tuesday, January 26, from 3:30-4:00 p.m. at the San Diego Convention Center.
November 23, 2015