June 2003
Dental researchers often say studies of the mouth may have important implications in other parts of the body. If ever there was a case in point, it’s research on a common oral pathogen with the tongue-twisting name of Actinobacillus actinomycetemcomitans.
About three years ago, a team of NIDCR grantees discovered a unique cluster of genes that long ago incorporated into the DNA of this organism, and which followup studies indicate originated in distantly related bacteria. According to the scientists, this case of “horizontal gene transfer” - passing genes from one organism to another - allowed Actinobacillus actinomycetemcomitans to stick tightly to teeth and other surfaces, an essential first step in the process of becoming a human pathogen.
The scientists say these findings could have major implications for treating oral conditions associated with the bacterium. They also noted that these discoveries have provided new insights into bacterial evolution and suggest a highly targeted strategy to treat infections throughout the body that involve related bacteria.
“Whenever you investigate a biological problem in depth - be it in the mouth, kidneys, or liver - fundamental themes will emerge that are applicable to other environments,” said Daniel Fine, D.M.D., one of the NIDCR grantees and a scientist at the University of Medicine and Dentistry of New Jersey in Newark. “This is a case in point.”
Actinobacillus actinomycetemcomitans, or Aa, has always been considered a unique bacterium. While other organisms in its genus readily infect livestock, Aa is the one species of Actinobacillus with the ability to inhabit people. Scientists have isolated it from human tonsils, heart, brain, gastrointestinal tract, and genital tract. They also have associated it with a wide variety of infections, including endocarditis, meningitis, and pneumonia.
But Aa is perhaps best recognized in the biomedical literature as a frequent inhabitant of the oral biofilm, the complex microbial community that forms on the surface of our teeth. Scientists estimate Aa is present in the mouths of about 20 percent of healthy adults and children, where it coexists with other species of bacteria and generally seems not to contribute to disease.
The major exception is in some teenagers who develop localized aggressive periodontitis, or LAP, the predominant form of periodontal disease in children. Studies show Aa is present in the mouths of about 90 percent of children with the condition, and the bacterium tends to colonize the gingival pockets surrounding incisors and first molars, the teeth that are affected by LAP.
What has intrigued some microbiologists is that, in culture, strains of the bacterium isolated from the mouth at first retain their ability to grow sticky, finger-like appendages, called pili, which allow them to adhere tenaciously to hard surfaces. But, after being subcultured over several generations, mutants begin to appear that stop adhering altogether and take over the culture. This suggests that Aa contains genes that are critical for adherence, but which are not essential for its actual biological survival. It also suggests that scientists may be able to selectively control this oral pathogen by inhibiting these genes and, specifically for teenagers with LAP, stop it from damaging their gums and destroying affected teeth.
About three years ago, the laboratories of Fine and NIDCR grantee David Figurski, Ph.D., a microbiologist at Columbia University’s College of Physicians & Surgeons, teamed to locate the genes for adherence. They identified seven genes clustered together in a row. Knock out any of these seven so-called tight adherence, or tad, genes, and Aa lost its ability to attach to surfaces. The bacteria could neither grow the bundled filaments of the pili, nor aggregate with other bacteria, a critical activity in forming a potentially infectious colony.
Then their research turned really intriguing. Led by Scott Kachlany, Ph.D., and Paul Planet, Ph.D., of the Figurski lab, in collaboration with Rob DeSalle, Ph.D., of the American Museum of Natural History in New York, the group found that the tad genes were completely novel, meaning they didn’t match any genes previously banked in GenBank. But, when they searched a database of microbial DNA genes of undetermined function, the researchers found partial matches in a variety of other microorganisms for each of the tad genes, an indication that these unknown genes were fairly common in the microbial world.
Remarkably, they noticed strong similarities between the predicted proteins encoded by the tad genes and the predicted proteins of genes in Yersinia pestis, the source of bubonic plague, and other human pathogens, such as Bordetella pertussis, the cause of the whooping cough, and Mycobacterium tuberculosis, the cause of tuberculosis. When the researchers located the tad-like genes in these organisms, they were aligned in exactly the same order, one through seven, as the tad genes in Aa.
“Here we are working on an organism involved in periodontal disease, and it turns out to be related to Yersinia pestis, the cause of the plague,” said Fine. “Who knew?”
But the scientists still didn’t know what the Tad proteins actually did to help Aa adhere to surfaces, information they would need to begin piecing together the precise molecular details of adherence.
That’s when, using special computer programs that predict where a protein might reside in a cell, the group determined that part of the amino-acid structure of the tad proteins resembled proteins involved in forming membranes. The finding was particularly interesting because located near the tad cluster was a gene called flp1. Two years earlier, a Japanese group reported that flp1, encodes a major structural component of the actual filaments that form pili. Given these clues, the scientists predicted that the tad genes are probably involved in forming a secretion machine that transports proteins across the membrane. Moreover, they predicted that the machine most likely plays a role in assembling and secreting the Flp1 protein as part of the sticky filaments of the pili.
This convergence of information led the team to take a closer look at Flp1 and its role in adherence. Kachlany and Planet, along with Jeff Kaplan, Ph.D., of the Fine lab, confirmed that Flp1 was indeed essential for producing filaments and, without a functioning copy of the gene, Aa was incapable of adhering to surfaces. This work also suddenly upped the total number of genes involved in adherence to eight.
More genes were to come. As published in the June 2003 issue of Nature Genetics, Planet, Kachlany, and colleagues found that the original tad cluster actually belonged to a larger 14-gene island, which they defined as a cluster of genes “with a common evolutionary history that is distinct and divergent from the histories of the organisms in which they reside.”
Planet et al. also reported that at least 12 of these 14 genes encode proteins that are critical for filament assembly and thus play a fundamental role in adherence. In fact, they found the Flp1 protein remains trapped inside the cell, unable to be secreted, when any of these 12 genes are mutated, adding support to the assembly and secretion hypothesis.
“This raised the question of whether these genes had been horizontally transferred, that is, passed from another organism to Aa,” said Figurski. “After a rigorous comparison and analysis of a wide variety of available bacterial and archaeal genomes, Paul Planet and Rob DeSalle showed that is most likely to be the case.”
In fact, Planet et al. speculated that the tad genes were transferred either directly or indirectly to Yersinia and Pasteurellacae, the latter being the family to which Aa belongs, from ancestors or close relatives of modern day Rhizobiaceae, a genus of bacteria that live in the soil, often in symbiotic associations with plants. They also noted that though these genes helped transform Aa and other bacteria into human and animal pathogens, the Rhizobiaceae probably used these genes for other, possibly benign purposes, such as attachment to plants.
In performing their analyses, Planet et al. also noted that several of the species had duplicated copies of the adherence genes on plasmids, the small, circular bits of DNA that are also present in some bacteria apart from their chromosomes. The authors say this suggests that plasmids may have been “the driving force” for many of the early transfers of these genes.
“It’s always been supposed that plasmids could serve as vectors that transfer genes among bacteria and catalyze their evolution,” said Figurski, whose specialty, ironically, is plasmid genetics. “But what’s exciting in this case is you actually have some real evidence for the evolutionary events that occurred. It’s not direct evidence at this point, but it’s gratifying to have established a reasonably strong association.”
“What’s more, we now have a solid scientific foundation to build upon in trying to understand exactly what these genes do,” continued Figurski. “The genes have significance in a wide range of bacteria, including some notable pathogens, raising the possibility that their proteins might make excellent drug targets.”
These fundamental discoveries, however, brought the researchers back full circle to the issue of LAP. Was this island of adherence genes also important in the development of LAP? The scientists technically didn’t know for sure, because, though they had established that the genes are essential for adherence in culture, they had yet to demonstrate it directly in the mouth.
To answer the question, the researchers relied on a unique rodent model, developed in the Fine lab, whose oral cavity can be colonized by Aa and later develops symptoms that mimic LAP. Led by Helen Schreiner, Ph.D., of the Fine lab, the scientists fed three groups of rodents a controlled diet containing live Aa for several days: The first group received a strain of Aa with an inactivated flp1 gene, the second received bacteria with an inactivated tadA gene, and the third group consumed diets laced with an intact form of the bacterium. Meanwhile, a fourth group served as an inoculated control group. All of the animals in the study had not been exposed to Aa prior to the study.
As reported in June 10, 2003 issue of the Proceedings of the National Academy of Sciences, Schreiner et al. showed that both the tad - and flp-deficient Aa could not colonize the rodents’ mouths. Conversely, animals that consumed food seeded with intact Aa developed all of the telltale signs of LAP, including bacterial colonization, infection, and bone loss.
Fine said the key now is to combine this growing understanding of adherence with a better understanding of the bacterium’s virulence. “Obviously, if you block attachment, you don’t get disease,” said Fine, noting that eliminating Aa from the mouth altogether could adversely disturb the delicate ecological balance of the oral biofilm, potentially allowing another destructive microbe to dominate and cause disease.
“An alternative and preferred approach would be to knock out genes associated with bone loss, showing that, even though the organism attaches, it will not cause bone loss,” added Fine. “If we can do that, it will lead to a better understanding of LAP.”
According to Fine and Figurski, one of the great satisfactions of performing the work has been the productive give-and-take among the labs over the last 15 years. Fine, a periodontist and microbiologist, noted that the collaboration might not have been consummated had he not spent a year in the Figurski lab during the late 1980s on a senior NIH fellowship to learn how to perform genetic research techniques.
“I knew we would collaborate when Dave one day had to give a talk at his son’s school, and he asked me for a dental x-ray demonstrating the bone loss and periodontal disease,” recalled Fine. “We’ve been collaborating ever since, and, more recently, Rob DeSalle has added his expertise in phylogenetic analysis. I think it shows the real power of multidisciplinary research in attacking a biological problem from all angles. By exploiting each lab’s strengths scientifically, we’ve been able to generate new and quite unexpected data that will likely have far-reaching applications clinically.”