Unlike most people, Dr. Martin Hugh-Jones looks forward to getting bloody swabs in the mail. He collects them to track anthrax. Dr. Hugh-Jones, a professor of epidemiology at Louisiana State University School of Veterinary Medicine, and collaborator Paul Keim, PhD, a professor of microbiology at Northern Arizona University, have built and maintain one of the largest anthrax databases in existence today. Databases such as theirs, and another at Los Alamos National Laboratory in New Mexico, are the ones that investigators are turning to as they attempt to identify anthrax samples used in recent bioterrorist attacks.
Dr. Hugh-Jones is no stranger to bioterrorism. In 1979, when a human anthrax epidemic broke out in Sverdlovsk, a city of the former Soviet Union, the England-educated veterinarian participated in the investigation. He was also in Russia in 1992 when its government finally admitted that the outbreak was caused by an accidental spore emission from a biological warfare facility. Dr. Hugh-Jones, along with other American scientists, befriended Russian pathologists involved in the investigation and managed to obtain tissue samples from people infected in the outbreak. These samples were the start of the Midwestern and Los Alamos databases.
After returning from Russia, Dr. Hugh-Jones sought out Dr. Paul Keim, whom he calls a genius at analyzing genes, and a team effort began. "I am the collector," explained Dr. Hugh-Jones. "I find out where the outbreaks are happening worldwide, contact the people concerned, and get the cultures. I then send them to Paul Keim's laboratory and he does the genetic analysis." Dr. Keim's laboratory is also involved in collecting some samples.
Usually, people send specimens for the Midwestern database to Dr. Hugh-Jones, but he also obtains the specimens himself from bloody swabs or dirt. The epidemiologist keeps track of outbreaks by monitoring the literature and doing good, old-fashioned detective work. He is also moderator for correspondence concerning anthrax for Promed mail, a program of the International Society for Infectious Diseases. This Internet collaboration of international doctors, academics, public health workers, and government officials tracks outbreaks of infectious diseases.
Thus far, the Midwestern database has roughly 1,200 isolates, which include specimens from most parts of the world. The Los Alamos database has had similar success, and the two groups collaborate. Each isolate can be tied to a specific geographic area, using genetic fingerprinting. On the basis of similarities in the bacteria's DNA, anthrax can be organized into categories. These categories are called strains by some researchers, but other scientists prefer to define them as different genotypes.
Investigators rely mainly on two techniques to create genetic fingerprints of Bacillus species in the databases. One technique, known as AFLP (amplified fragment length polymorphism), relies on analyzing short DNA sequences that contain genetic information. Using AFLP, scientists extract DNA from bacteria and cut it into small fragments, using nature's natural scissors—restriction enzymes that recognize and cut specific stretches of DNA. After being amplified to improve the interpretation process, these fragments are analyzed and compiled into a fingerprint that is added to a database.
The second genetic fingerprinting technique, MLVA (multiple locus variable number of tandem repeats analysis), relies on analyzing the pieces of DNA that don't code for proteins. Although believed to be functionally useless, these strands contain repeated sequences of base pairs that are more highly variable from strain to strain than "useful DNA." The repeated sequences are usually surrounded by specific markers, which allow researchers to find and cut them. These fragments are then fingerprinted and added to a database.
On Oct. 25, Office of Homeland Security Director Tom Ridge announced that investigators had made progress in identifying anthrax samples used in the October bioterrorist attacks.
"Department of Defense DNA tests show the anthrax samples from Florida, New York, and Washington are indistinguishable, meaning that they all come from the same strain of anthrax or the same family of anthrax," announced Ridge—"Ames strain."
Using DNA analysis, investigators had created genetic fingerprints of the samples from affected people, sifted through the vast databases, and found a match.
Army records showed that the Ames strain had originally come from a sick cow in Iowa and was sent to the U.S. Army Medical Research Institute for Infectious Diseases in 1980. The Army, determining that the strain was unique, named it Ames, simply because of its origin.
Since then, the Army has supplied it to various laboratories in the United States, and the strain has been disseminated by other routes to facilities around the world. Because of the strain's popularity, the initial identification did not uncover the senders of the contaminated letters.
Many people investigating the bioterrorist attacks have been instructed to refrain from directly commenting on the progress of the investigation. Dr. Kimothy Smith, is a veterinarian and an associate professor of epidemiology in Dr. Keim's laboratory at Northern Arizona University. Dr. Smith says he cannot confirm or deny working on the bioterrorism case, but when asked about his workload in early November, he responded, "I've been working 24/7 in the last month."
Dr. Smith is hopeful that science will provide a break in the case. All genetic markers, the veterinarian said, are not created equal; some allow for the differentiation of very closely related isolates better than others, especially when investigators are using MLVA.
As of early November, work at NAU was showing that, using ultra-sensitive markers, one could trace strains of a different type of anthrax, called Sterne, to a particular laboratory. Sterne has also been disseminated all over the world. "We can go into old strains like Sterne, [and] theoretically Ames, look much closer into those strain types, and differentiate types from different laboratories," Dr. Smith says.
This promising sleuthing ability is made possible by three factors: bioterrorists need to grow large batches of anthrax for it to be useful as a biological weapon, anthrax undergoes genetic mutations every 1,000 generations, and mutations lead to different fingerprints.
For a given sample of anthrax, Dr. Smith says, "We can go back and estimate the number of generations that separate two isolates. We can build a tree of relatedness, saying this strain is so many generations from this laboratory strain and this many generations from this other one, etcetera."
As of Nov. 5, investigators had not traced the anthrax strain used in the October attacks to a specific laboratory. Workers involved with the Midwestern database and the Los Alamos database continued to toil. But, one thing was certain: the databases, started almost 10 years ago, have played an important role in the investigation.