Despite every scrupulous effort of personal hygiene, fecal excretion deposits millions of contaminating bacteria around the anal aperture. But not to worry; within a couple of hours endogenous microorganisms that colonize the skin wipe out the intruders from their home territory.
Every square centimeter of a human body’s skin is home to some 100,000 bacteria. Add them all up and they outnumber the planet’s human population. But of course not all the bugs that inhabit the human form are commensal and beneficial. Some are openly infectious, and fair game for antibiotics. Others hide their pathogenicity.
Microbiologists have traditionally identified disease-causing organisms by growing them laboriously in a laboratory dish from a sample of infected tissue.
“A lot of diseases are caused by infectious agents we’ve been unable to identify,” observed molecular biologist Matthew Meyerson, at Harvard-affiliated Dana-Farber Cancer Institute in Boston. “For example,” he added, “take multiple sclerosis. MS is much more prevalent in the north of the U.S. than in the south. If people move from the north to the far south before a certain age, they get the risk of the north. The suspicion is the viral link of MS to climate and latitude. I think nobody will really know until we find the microorganism.
“Some people talk about atherosclerosis as an infectious disease,” he went on. “Many are putatively autoimmune or inflammatory, like Type I diabetes or lupus, and it may be that they really are infectious. We don’t know this for sure.”
Meyerson recalled that “doctors used to blame stress and diet for stomach and duodenal ulcers. Only in recent years have ulcers been shown to be caused by infection by a bacterium Helicobacter pylori. Some people even talk about atherosclerosis as an infectious disease. Many are putatively autoimmune or inflammatory, and it may be that they really are infectious. We don’t know this for sure.”
Among these possibly infectious, mysterious, chronic afflictions, he listed rheumatoid arthritis, Type I diabetes, lupus, coronary artery disease, inflammatory bowel disease and several types of cancer including bladder, lung and lymphomas.
Human Genome Pays Off Again
Meyerson is senior author of a brief communication in Nature Genetics for February 2002, but released online Jan. 14. Its title: “Identification of foreign gene sequences by transcript filtering against the human genome.”
“The technique we describe is good for investigating all these chronic diseases of unknown origin,” Meyerson told BioWorld Today. “We report using a sequence-based approach to identify pathogen sequences in human disease. The novelty is that nobody ever thought to my knowledge about doing this before. What makes it possible is the Human Genome Sequence Project, which wasn’t available until now.
“If you have infected tissue,” Meyerson explained, “it can contain microbial sequence and human sequence. Now that we know the human sequence in principle, we can remove it, leaving us with microbial sequence.” He cited one specific clinical example of the approach he and his co-authors developed:
“Our source was the National Center for Biotechnology Information, which provided a database of 16,793 complementary DNA sequences strings of A, G, C and T nucleotides made from messenger RNA. These were from a liver cancer patient diagnosed with hepatocellular carcinoma. We compared them to the human genome sequence, which presumably contains normal DNA, and looked for regions of similarity or identity. We found that most of the regions were the same, but there were some sequences not in the human genome. This library included hepatitis B virus sequences, which we would expect to find in a liver cancer patient and we did.
“Transcript filtering,” Meyerson pointed out, “is pretty much the same thing as computational subtraction. That’s the process we used of comparing sequences to the human genome and removing everything that matched.”
Human Millions Minus Microbial Thousands
The team analyzed 3,287,578 human sequences that make up all the 30,000 or so human genes. Subtraction with seven database filters left a set of 65,839 ESTs expressed sequence tags. This 2 percent remainder included mainly sequences from known pathogenic and commensal organisms, microbial sequences likely to represent experimental contamination, and transcripts from unsequenced regions of the human genome.
Besides the unknown human liver test run, the co-authors also tried their method on a human tissue sample known to contain a papillomavirus, which causes cervical cancer. As expected, the technique highlighted that viral pathogen in the nonhuman leftover DNA. They also found DNA from such organisms as cytomegalovirus, Kaposi’s sarcoma herpesvirus and Epstein-Barr virus.
Since submitting their paper to the journal last August, Meyerson and his group are now “doing some pilot libraries and their performance, trying to identify diseases and tissues to pursue. We’ve not discovered any new organisms, but we’re working on it. We propose to generate sequence and filter cDNA libraries from tissues of diseases such as systemic lupus erythematosus and extraintestinal Crohn’s disease, which are candidates for infectious etiology.”
The university, Meyerson said, has applied to patent, “A method for microbe discovery of viruses, bacteria, fungi and protozoans all microorganisms.” He is now in discussions with pharmaceutical and biotechnology companies, as well as venture capital firms. A potential licensee would “discover sequences of a new organism for use in therapeutics or diagnostics. Assuming we have a licensing agreement,” he went on, “we anticipate after that some practical outcomes might come out pretty quickly within a year or so.
“I think this is going to be a very powerful method for discovering new microbes,” Meyerson concluded, “and will end up changing the way we think about diagnosing and treating a lot of common human diseases.”