With primetime attention thanks to outbreaks of anthrax and SARS in the past several years, research into infectious disease has been catapulted out of the ignorance afforded to most so-called neglected diseases and into the spotlight.
It made sense that genomics and its brethren technologies would be major players in this scene: after all, the first living organism to have its genome sequenced was Haemophilus influenza. But the community has come a long way since then. Scientists have employed the whole arsenal: microarrays, sequencing, genotyping, bioinformatics, proteomics, you name it — they've all had a hand in helping researchers better understand these pathogens, the mechanisms they use to attack humans, and new ways to stop them in their tracks.
In fact, these technologies have been so successful that leading researchers in the field say the major challenges facing the field are no longer about getting better tools. "We're in a period of rapid technology expansion," says Joe DeRisi at the University of California, San Francisco, pointing to low-cost array tools and next-gen sequencing as examples. "The major hurdle in all of this infectious disease work is getting the perfect samples, the most pedigreed samples. … Sometimes the perfect sample just doesn't exist for the disease you'd like to look at."
Beyond that, much of the challenge in actually making a real-world impact is in stirring up commercial interest in the field, says Phil Hanna at the University of Michigan. While there's tremendous interest in public-sector institutions for basic research, and there's just as much enthusiasm in places like the US Food and Drug Administration to see therapeutics or vaccines to fight infectious disease, Hanna contends that there's a lack of interest in the key area between those two endpoints. "It's really a matter of a commercial venture," he says. "There doesn't seem to be enough interest at that level to put the investment into it."
But in the meantime, scientists are pushing ahead in their investigations. While studies of pathogen biology have dominated much of the last several years, attention is now turning toward the host-microbe interaction and to figuring out the immune response to disease.
In the pages that follow, GT profiles some of the pioneers and leaders of infectious disease research and the 'omics technologies they're using.
Fraser-Liggett: From Pathogen HQ TIGR to Human-based Disease Studies
You can't possibly have a thorough conversation about genomics and infectious disease without talking about Claire Fraser-Liggett, who helped shape the Institute for Genomic Research into a powerhouse of pathogen science. Under her stewardship, TIGR sequenced the genomes of some 30 or 40 pathogens, and more than 200 strains of those organisms, Fraser-Liggett estimates.
She says the institute got its start with the H. influenza shotgun sequencing project, and TIGR's first grant in the area was from the National Institute of Allergy and Infectious Diseases for sequencing the pathogen responsible for syphilis. From that point on, the priority list for sequencing these organisms exploded — and looking back over all those sequences, "it's safe to say that in all cases, these projects revealed a lot of new information about these organisms," says Fraser-Liggett, who is now at the University of Maryland. "Some themes … emerged, particularly related to evasion of host immune response — information that was critical for going forward to start functional genomic studies."
Among the lessons learned early on was the unexpected and tremendous diversity of these species. Researchers found in a comparison of several E. coli strains that the bacteria shared just 40 percent of their gene complement across all strains, says Fraser-Liggett, adding that such species may have to be thought of as having a "pan-genome" that includes all the possible sequences instead of one specific genome. "For many pathogens it is probably impossible to think about defining the extent of diversity within a single genome," she says. A significant amount of resources are going into trying to determine how pathogen genomes got so diverse to begin with — mechanisms like lateral gene transfer and gene acquisition help answer some questions — and all of that "has implications for vaccine development," Fraser-Liggett says.
So far, tremendous effort has gone into understanding the biology of these pathogens — but recently, much attention has turned toward understanding the human half of these relationships. Research is going into taking the measure of immune response, and still more is looking into the "effect of pathogens on human hosts, [where pathogens] can actually affect the host signaling pathways," Fraser-Liggett says.
As for her own research, Fraser-Liggett has become very interested in better understanding the metagenomics of microbial communities within humans, and how shifts in those communities may play a role in the susceptibility to, onset of, or other relationship with infectious disease. "Even things we don't think of as having infectious etiology like obesity," she says, "may somehow be correlated with a shift in the overall composition of these microbial communities. … There may be instances where there are diseases that have an infectious etiology, but it's not due to the presence or absence of a single pathogen."
Figuring out what "normal" is in those microbial communities, as well as whether and what kind of role they play in disease state, "will require additional sequencing as well as transcriptome and proteome analysis," Fraser-Liggett adds. "It's really a brand new field, and at this point the list of important questions that we would all like to see answered is extremely long."
Ohio State Project: Google Earth for Viral Spread
Health officials and researchers the world over have a new way to keep tabs on the spread of the avian flu virus (H5N1) and other deadly viruses, thanks to a group of researchers led by Dan Janies, assistant professor in the department of biomedical informatics at Ohio State University. Janies and his team recently unveiled an online global map of H5N1 in the form of phylogenetic trees projected onto a map of the Earth using the same programming language and features that power Google Earth, the interactive online global map application.
The new "supermap" provides not just geographic information on virus outbreaks, but contains data on the virus's genetic mutations and evolutionary information that could help predict where the next outbreak is most likely to occur. The phylogenetic tree appears as a scaffold starting with the possible origin of the virus high up in the sky, flows down through branches of its inferred ancestors as well as data on hosts, and finally, terminates with the tree's tip as the current version of the virus reported at that particular location on Earth.
Different viral hosts, including birds, insects, humans, and other animals, are delineated on the map by intersecting multicolored lines and symbols. Every node in each tree representing either a particular strain of a virus or its host can be clicked on to reveal a popup window displaying data on the mutations associated with that particular strain. Links in the window connect back to GenBank, the map's primary source of genomic data. The tree can also be animated using TimeSpan, another feature native to Google Earth, to show the spread and mutations of the virus over the last decade.
"I can go to my map and see where a particular mutation is occurring in other strains that may have been sequenced, but that have never been the subject of laboratory studies," says Janies. "I can also look at a certain mutation that's spreading westward, for example, and know that that's one we should be watching, so it's predictive in a transitive sense."
In the beginning, Janies and the other researchers had to pass huge files of data back and forth to build and update the map. But now, the process of keeping the map up to date has been streamlined. "We're building a much more robust workflow system," he says. "We have ways of downloading data from GenBank in an intelligent way, update when there's a critical mass of new data, pipeline it to a tree-building process, output that to a KML process, [and] get a new visualization."
The biggest stumbling block in developing such a sophisticated interactive tool is not technological, he says, but rather, the social and political hang-ups countries have when it comes to releasing data on avian flu outbreaks. "What we envision is a future where some of the sociological problems get overcome and we have a rapid DNA sequencer on the ground," Janies says, "and we get the data up-linked to a central location, do the calculations overnight, and have a new map on the president's desk in the morning."
At Hanna Lab, the Spotlight Is on Anthrax
You can't accuse Philip Hanna of being a fair-weather fan. A scientist at the University of Michigan, Hanna's research has been focused on anthrax since long before the agent burst into public attention several years ago.
Starting on anthrax in 1990 as a postdoc, Hanna has worked since then on spore-forming bacteria that cause disease — a particularly insidious group that includes tetanus, botulism, and, of course, anthrax.
That research got a serious kick in the pants when anthrax came on the scene as a potential bioterror threat. Nick Bergman, a member of Hanna's lab, designed an array containing all the anthrax genes (it's now sold through Affymetrix) that enabled scientists to perform global screens against the bacterium. It also allows for the comparison of samples from different infections, which Hanna hopes will help researchers home in on the specific genes contributing to pathogenesis of the organism.
They certainly weren't the only ones tackling the new threat, and Hanna says getting the B. anthracis genome sequence "was a huge advance" that opened the door to all the other 'omic technologies, including array work and shotgun proteomics.
Hanna and Bergman say that the organism's relatively slow life and reproductive cycle have been a boon to researchers. Because anthrax traditionally forms spores and lies dormant in soil for years and years, it "doesn't go through that many generations relative to other bugs," says Bergman, noting that this has contributed to more monomorphic strains than is characteristic of many bacteria. But while that helps in the general biological study of anthracis, it hinders efforts in forensics or strain typing since there are so few differences to be found. "It's very difficult to trace B. anthracis back to its source," Bergman says. "We're having to develop other methods relying more on epigenetics than genetics."
Beyond strain typing, major efforts in the lab are targeted at understanding the mechanisms and biology of anthracis. "We still don't know how this organism ends up killing animals or people," Hanna says. In his lab, researchers are working to figure out the pathophysiology of the anthrax infection. They're also focused on "finding new choke points in the anthrax infectious cycle" that could presumably aid in the design of antibiotics or at least diagnostics for the biomedical community
The Hanna lab and its collaborators have unleashed a number of technologies on anthracis. With John Yates at Scripps, they're using high-throughput shotgun proteomics and various mass spec approaches to help understand the actual content of the bacterial spore. Bergman says next-gen sequencing technology is also being put to task for strain comparison studies and sequencing of an mRNA pool to generate expression profiles in work with Timothy Read at the Naval Research Center. "As scientists, we're always kind of junkies for new technology," Hanna says.
DeRisi's Chip Provided First SARS Clues, Continues to ID New Viruses
Half of Joe DeRisi's lab at the University of California, San Francisco, focuses on better understanding the biology of Plasmodium falciparum, one of the parasites that causes malaria — but it was the other half of his lab that brought DeRisi the accolades of the infectious disease research community early in 2003. The PI's other area of interest is in viral diagnostics and discovery, and it was DeRisi who, four years ago, was the first scientist able to definitively identify the mysterious SARS virus as a member of the coronavirus family.
DeRisi's lab took advantage of conserved sequences to design a microarray for the identification of viruses. While many diagnostic leaders focus on technologies that identify known viruses, DeRisi pushed for a chip that would be helpful in identifying unknown viruses as well. "We overrepresent the most conserved nucleic acid sequences among the viruses," he says of the content of the virus microarray composed in his lab. "We try to maximize the possibility that if there is an uncharacterized virus," the hybridization experiment will report enough information to make some basic observations about it.
"This is exactly what happened with the SARS virus," DeRisi says. His lab got samples of the virus, which at the time was completely unknown to the research community. "When we put it on our array, we were able to see" a match with the conserved sequences indicative of coronavirus.
That chip became the cornerstone of the viral diagnostics and discovery branch of DeRisi's lab, which recently published on a novel retrovirus they detected that appears to be associated with prostate cancer. But finding and identifying new viruses is only half the battle. "Just because we find a novel virus doesn't mean it's causal in any way," DeRisi says, using the prostate cancer link as an example. "To follow up on one of these viruses takes years." The discovery work keeps his staff and the lab's partners busy with a growing pipeline to pursue with more epidemiological studies.
Sequence for Mosquito that Plays Host to Yellow and Dengue Fevers
A large-scale sequencing project recently completed the first genome draft of the mosquito species responsible for spreading the yellow fever and dengue fever viruses. The Aedes aegypti mosquito strain is credited with an estimated 200,000 cases of yellow fever and roughly 100 million cases of dengue virus in unvaccinated populations each year, according to the World Health Organization. Both viruses mostly occur in warm climates such as Africa, southeast Asia, and Central America, where Aedes aegypti tend to thrive. The Aedes genome contains about 16,000 genes, roughly five times the number of its distantly related cousin, the malaria-carrying Anopheles gambiae mosquito species. David Severson, a biology professor at the University of Notre Dame, helped lead the worldwide effort, which was comprised of more than 25 institutions, among them the Broad Institute, the Institute for Genomic Research, the University of Iowa, Colorado State University, and Virginia Tech.
Severson cautions that there is still much to be accomplished before the genome data translates into effective methods for combating the deadly mosquito. "We have a pretty good product, but there's a lot of work to be done in terms of enhancing the genome assembly itself and being able to position the assembly pieces onto chromosomes in appropriate linear orders," he says. "We have a lot done, but we still have a long way to go." Even though powerful algorithms can aid in the analysis of the Aedes genome, ultimately it rests upon the research community to eliminate mistakes in the annotation to ensure that they have accurate data going forward, he adds.
Severson works with various Aedes laboratory strains that his team has identified as either being very susceptible to dengue virus or especially poor vectors. So far, they have conducted several different projects involving microarray analysis that looked at the effect on the transcriptome of exposing either resistant or susceptible Aedes strains to infected blood meals.
Currently, he is focused on understanding the molecular basis for what makes Aedes a competent vector to support and transmit these viruses to humans. One possible answer may lie in transposable elements, movable pieces of DNA that control genome size and mutations. About half of the mosquito's genome contains transposable elements and while many of them appear to do nothing more than replicate, there are a significant number of active transposable element copies involved in biological functions. Severson hopes that these active transposable elements can be used eventually to alter the biological mechanisms that actually allow Aedes to spread these viruses.
"We're looking at how we can take advantage of that, and the genome may allow us to leverage some knowledge to interrupt the transfer of the virus from an infected human through the mosquito so it doesn't actually replicate in the mosquito," he says. "We can break the cycle and prevent disease transmission.
Team Shoots for Portable, Multiplexed TB Detector
An international team of researchers won a $400,000 grant last month from the US Agency for International Development to pursue development of a better tuberculosis diagnostic tool than currently exists.
Led in the US by Paul Luciw and Imran Khan at the University of California, Davis, the project also has a sizable research group in Pakistan, which ranks sixth highest in the world for TB incidence. Khan says one of the goals of the program will be to make sure the diagnostic is on a portable platform so that it can be "taken to the parts of the world where TB really lives."
Luciw and Khan, both in the Center for Comparative Medicine at UC Davis, believe their longstanding work in proteomics will come to the fore in this project. The center's proteomics core has long been based on Luminex technology, and their plan is to build the TB diagnostic around the Luminex platform they already know well. The multiplex capabilities of the platform are critical, says Luciw; current detectors can only look at one antibody at a time. As the group's diagnostic develops, researchers will be able to scan a number of biomarkers at once, including antibodies, cytokines, and chemokines, Luciw adds.
Khan says the beauty of using proteomics instead of the standard PCR test is that results from protein studies can shed light on the patient's history of infection, and perhaps stage of disease, beyond the binary infected/uninfected answer current tests provide. "We're hoping that in humans we'll be able to pick up even latent infections," he says. The other problems with PCR — ease of contamination and inability to pick up trace amounts of bacterial load — could be solved by a protein-based approach.
Novartis Vaccines' Rappuoli: Genomics Turned Vaccine Development Around
Rino Rappuoli has a long list of things to worry about. The global head of vaccine research at Novartis Vaccines and Diagnostics, Rappuoli is responsible for figuring out which vaccines to follow up on, which infectious diseases merit the most resources, and trying to get a handle on preventing, rather than treating, these illnesses. (The ultimate goal, he says, is to "fix people who are healthy so they never get sick.")
As for priorities, Rappuoli says the "big ones" in terms of infectious disease right now include HIV, tuberculosis, malaria, and meningitis. "Those affect millions of people every year," he says. "Then there are ones that are potentially disasters but are not there yet, such as avian influenza."
Rappuoli says genome sequencing has provided a completely new way to approach vaccine development, his area of specialty. "Up to that point vaccines had to be discovered manually," he says, noting that isolating enough of the microorganism and following up on the antigens was a plodding and not spectacularly successful way to come up with vaccines. "When genomics became available … we started designing vaccines not from the lab but from the computer," he says. Rappuoli got into the field in the late ‘90s through a collaboration with the Institute for Genomic Research, where his partners at the time included Craig Venter and Claire Fraser.
Having a pathogen's genome allows his team to run a genome-wide search for potential vaccine candidates, resulting in hundreds of targets instead of the handful of targets more common in the traditional approach. "Now you can collect the best ones," he says. For every single bacterium his team has looked at, "we get new candidate vaccines that have never been available before." Indeed, Rappuoli says, there are promising vaccines in trials right now for diseases that have gone 40 or 50 years without any kind of solution at all.
McGill's Schurr Makes Inroads into Immune Response for TB, Leprosy
Leprosy might not be on your list of top priorities in infectious disease, but Erwin Schurr at Montreal's McGill University sees tremendous potential in it. His team is using leprosy as a model to study host-pathogen interactions in the hopes that not only will it shed light on the pathogenesis of that disease, but also on that of the far more deadly tuberculosis.
Schurr contends that targeting genes will not be the first approach to leprosy or TB, but rather that truly understanding the immune response of humans will enable scientists to come up with new ways to prevent the illnesses from the human side of things. Figuring out the critical pathways may help boost immune response and allow patients to fight off leprosy or TB, he says. "I think genetic diagnosis … has some value eventually," Schurr says. "But the biggest value will clearly come from [targeting] pathogenesis, and then be able to manipulate" the immune response. "The gene itself is not necessarily the target."
In his lab, the goal is to track susceptibility loci. That began years ago with scanning for microsatellites and has continued into the much higher-throughput days of fine mapping for SNPs. For both TB and leprosy, "we look for genetic variants that impact on disease susceptibility," Schurr says. By following up on variants that show significant risk, the team hopes to narrow down the candidate list to "quantitative traits that we believe to be important in the expression of the disease phenotype."