New method rapidly detects potential bioterror agent.

A new combination of analytical chemistry and mathematical data analysis techniques allows the rapid identification of the species, strain, and infectious phase of the potential biological terrorism agent Coxiella burnetii.

Research by the Georgia Institute of Technology and CDC has yielded a method that proved to be 95.2 percent accurate in identifying and classifying Coxiella burnetii. The laboratory test delivers results in about five minutes, compared with about two hours for the lab technique currently used.

Facundo Fernandez, an assistant professor of chemistry and biochemistry at Georgia Tech University, presented the findings of the research team at the 230th American Chemical Society National Meeting in Washington, D.C.

Fernandez and his Ph.D. student Carrie Young, a chemist in CDC's Environmental Health Lab, collaborated with CDC researchers in the National Center for Environmental Health and the National Center for Infectious Diseases. They combined mass spectrometry and a mathematical data analysis technique called partial-least-squares analysis. Mass spectrometry allows researchers to look at the profiles of different proteins expressed in a microorganism. Partial-least-squares analysis lets them separate important information from "noise" (biological baseline shifts caused by variations in sample preparation that could corrupt a predictive model).

The combination of these techniques into one method is a novel concept. Fernandez noted that the research also represents the first time that Coxiella burnetii has been detected at the strain level with a rapid detection process. Such classification is a challenging task with bacteria, he added. Researchers believe the technique also will work with other pathogens, which they expect to begin studying this fall.

Coxiella burnetii is a species of concern because it causes the highly infective human disease Q fever, which is transmitted primarily by cattle, sheep, and goats. A human can be infected by as little as one bacterium. Depending on the strain, the disease can manifest as a chronic or acute case. Symptoms can include high fever, severe headache, vomiting, diarrhea, abdominal pain, and chest pain. Q fever can also lead to pneumonia and hepatitis. The chronic form of the disease can cause endocarditis, an infection of a heart valve, and even death.

In addition to being a public health threat, Coxiella burnetii is listed as a Category B bioterrorism agent because of its long-term environmental stability, resistance to heat and drying, extremely low infectious dose, aerosol infectious route, and history of weaponization by various countries, according to CDC.

To date, Georgia Tech and CDC researchers can differentiate between seven Coxiella burnetii strains, which come from Australia, the United States, and Europe. Some strains are more infective than others, and in addition to identifying the strain, the new method can distinguish a Phase I or Phase II strain--that is, the ability to infect.

"The next step is to fine-tune our model and increase the number of strains we can identify," Fernandez said. Separate research involving Fernandez and other Georgia Tech researchers is pursuing development of a field-testing instrument.

The technique is highly sensitive, meaning it can detect Coxiella burnetii strains at very low concentrations, specifically at the attomole level, which is equivalent to 1 X [10.sup.-17] moles. (Moles measure the actual number of atoms or molecules in a sample.)

Until now, the best method for differentiating between strains of Coxiella burnetii was polymerase chain reaction (PCR), a laboratory technique that analyzes the genes of a bacterium and yields results in one to two hours. The new method, which analyzes the proteins of a bacterium, can yield results in five minutes.

"In a bioterrorism event, you want more than one method to determine the strain you are dealing with," Fernandez noted. "So you would use our technique first and then use PCR as a second method to independently confirm your results. Also, our method using mass spectrometry allows you to quickly replicate your analysis--even 10 times if you want to. That gives you an added degree of statistical significance."