December 4, 2006

Volume 84, Number 49

pp. 47-54

Anthrax Sleuthing

Science aids a nettlesome FBI criminal probe

Lois R. Ember

It was a tense, unsettling time. A mere week after the Sept. 11, 2001, terror attacks, anthrax-laced letters began coursing through the mails on their way to several news organs and two U.S. senators, delivering death to five and mayhem to a nation.

This first major act of bioterrorism on U.S. soil triggered one of the largest, most complex, and costliest investigations ever undertaken by the Federal Bureau of Investigation, and still the person who mailed the letters remains at large.

This September, Joseph Persichini Jr., acting assistant director of the FBI's Washington field office, acknowledged the major, if unheralded, role science is playing in the probe. Yet the FBI has said little about what science has revealed, citing the criminal nature of the case as its reason. What scientific tidbits the public has been fed come from media reports, and most of these have been incorrect or incomplete.

Since finding an unopened anthrax letter addressed to Sen. Patrick J. Leahy (D-Vt.) in late 2001 and the letter's dramatic handover to scientists at Fort Detrick in Maryland, the FBI has clamped down on information on the probe. The embargo has been so tight that a former top military scientist who now works for a government contractor tells C&EN that he was consulted before the Leahy letter, but afterward, he could get no updates on progress being made even from friends in the FBI.

Though massive resources have been devoted to solving the case, many FBI critics attribute FBI's silence to the fact that the probe initially was misdirected and is now stalled.

Inexplicably, that silence was broken this August. Then, Douglas J. Beecher, a microbiologist in the FBI's hazardous materials response unit, published a paper in Applied & Environmental Microbiology, a well-respected but not well-known journal. It took the media a month to publish accounts of Beecher's article, which they generally interpreted as indicating that the FBI initially had misunderstood the nature of the anthrax used in the attacks.

After reading those news accounts, Rep. Rush Holt (D-N.J.), a member of the House Permanent Select Committee on Intelligence, wrote the FBI, requesting that it brief the committee on the status of the investigation. Assistant FBI Director Eleni P. Kalisch summarily rejected Holt's request.

Kalisch said that briefing the intelligence committee on a criminal investigation would be inappropriate. She also said the FBI and the Justice Department had decided long ago to stop briefing members of Congress after sensitive, classified information found its way into media accounts citing congressional sources. A Holt spokesman told C&EN the intelligence committee received "three limited briefings in 2002 and 2003, and no committee member has ever been implicated in leaks."

Angered by the FBI's refusal to brief Congress, Sen. Charles E. Grassley (R-Iowa), in late October, blasted the FBI's investigation for its "dead-ends" and "lack of progress." In a letter to Attorney General Alberto R. Gonzales, Grassley listed a litany of questions he wanted the department and the FBI to answer. He is still awaiting answers.

Beecher's peer-reviewed paper set off heated discussions not only in Congress but also in the arms control community and among government and academic scientists. The seven-page article chronicles the methodology the FBI used to uncover the Leahy letter, which, because it was unopened, contained the most unadulterated powder recovered from any letter.

What sparked debate was one paragraph in the discussion section that a military analyst, who asked not to be named because he still works with the FBI, says "clearly had nothing to do with the content of the article."

The first anthrax-laced letter destined for the Senate reached the office of former Sen. Thomas A. Daschle (D-S.D.) and was opened by one of his aides on Oct. 15, 2001. That simple act unleashed a fluffy light tan powder that wafted through the office and traveled the air ducts to contaminate the entire Hart Senate Office Building. Offices in the Hart building were evacuated, and eventually, other Senate and House offices were shuttered as well. The work of Congress came nearly to a standstill.

Five years later—after the FBI had conducted more than 9,100 interviews and 67 searches and had issued 6,000 grand-jury subpoenas—the case remains unsolved. FBI Director Robert S. Mueller III expects the case will eventually be solved. But the FBI's critics agree with Daschle, who contends that "the investigation's trail has gone cold."

In an Oct. 16 Washington Post OpEd, Daschle alludes to the Beecher article and writes that questions still "remain in the scientific community about the composition of the anthrax and the level of technological expertise required to manufacture it."

Given how easily the powder in the Daschle letter aerosolized, government officials, military scientists, and academic anthrax experts were quoted in the media as claiming the anthrax spores in the letter had to have been "weaponized." That is, the spores had to have been specially treated or processed—milled and coated with an additive such as silica—to make them float in the air. But in his article, Beecher, almost as an aside, dismisses this possibility.

In the paragraph that set the scientific and arms control communities abuzz, Beecher writes: A "widely circulated misconception is that the spores were produced using additives and sophisticated engineering supposedly akin to military weapon production."

This is the FBI's first public statement on the investigation since it began analyzing the material in the Leahy letter and the first time the bureau has described the anthrax powder. Beecher, however, provides no citation for the statement or any information in the article to back it up, and FBI spokeswomen have declined requests to interview him.

"The statement should have had a reference," says L. Nicholas Ornston, editor-in-chief of the microbiology journal. "An unsupported sentence being cited as fact is uncomfortable to me. Any statement in a scientific article should be supported by a reference or by documentation," he says.

Early news reports, replete with unnamed sources, implied that the universe of potential suspects was fairly narrow. The perpetrator of the attacks, the reports said, was likely to have special technical skills and likely had access to highly contained defense labs such as those operated by the U.S. Army Medical Research Institute of Infectious Diseases (USAMRIID) in Maryland and the Army's Dugway Proving Ground in Utah.

Because the anthrax powders proved to be so deadly, the thinking was that the perpetrator had to have used equipment, additives, and procedures that the Army had used to weaponize biological agents in its offensive bioweapons program before President Richard Nixon shut it down in 1969.

Several former government officials and scientists, who asked for anonymity, say the early media accounts that Beecher says mischaracterized the anthrax powders can be traced to the government's struggle to deal quickly with an unsettling and unfamiliar threat.

At an Oct. 29, 2001, White House press briefing, Maj. Gen. John S. Parker, then-commanding general of the Army's Medical Research & Materiel Command at Fort Detrick, said silica had been found in the Daschle letter. Tom Ridge, then-director of the White House Office of Homeland Security, at a briefing a few days earlier said a binding agent had been used to make the anthrax powders.

As one of the former government officials tells C&EN, "Those judgments were premature and frankly wrong." At the height of the attacks, top government officials with no scientific background received briefings from people who also were not scientists, and "the nuances got lost," he explains.

Sometimes scientists misspoke as well, as was the case with the Armed Forces Institute of Pathology. AFIP studied the anthrax powder from the Daschle letter using energy dispersive X-ray spectrometry, and a top AFIP scientist, Florabell G. Mullick, reported the presence of silica in an AFIP newsletter. Yet, the spectrum AFIP released shows a peak for the element silicon, not silicon dioxide (silica).

Harvard University molecular biologist Matthew S. Meselson, who has consulted for the FBI on the anthrax probe, dismisses these early statements as misunderstandings or misinterpretations of the scientific studies conducted on the Daschle powder. "I don't know of anybody with spore expertise who actually worked on the stuff who said the spores were coated," he says. The FBI has never publicly claimed the spores were coated with silica and, in fact, told members of Congress at classified briefings that the spores were not coated, he says.

Meselson alerted the FBI to a 1980 microbiology paper that reports finding silicon in the spore coat of Bacillus cereus, a cousin to Bacillus anthracis, the bacterium that causes anthrax. The silicon AFIP detected might be a natural element of the anthrax spore coat.

Although the FBI has released no information on studies probing for the presence of silicon in the coat of anthrax spores, and no studies have been published, Peter Setlow suspects that such studies have been done. About two years ago, Setlow, a molecular biologist at the University of Connecticut Health Center, was invited to an FBI-organized meeting of spore specialists.

The explanation for mischaracterizing the attack material is really quite simple, one of the former government officials says. When the attacks occurred, "there was no systematic methodology in place to evaluate a biological powder forensically." Initially, he says, the studies were "done on the fly." And quite frankly, he says, "a lot of people didn't know what they were looking for.

"The pace of the forensic investigation ground to a halt," this official says, "because there was not a lot of available expertise in the scientific toolbox."

Much of the material from the Daschle letter was consumed by destructive tests that produced little useful information, the official says. The government was understandably reluctant to proceed with tests on the Leahy powder until a validated testing protocol was developed, he explains.

So in December 2001, the FBI met with experts selected by the National Academies for advice on how to deal with the Leahy letter, a participant at that meeting says. Six NAS-vetted scientists attended that one-day meeting at the FBI's Washington field office and produced a flow chart, a scientific playbook on how to analyze the powder to garner the most information. Whether that flow chart was ever used is unknown.

The December meeting was among the first of eight the FBI would eventually convene with scientists "to develop a comprehensive analytical scheme for evaluating and analyzing the anthrax evidence," the FBI's Persichini says. In fact, the "FBI has held two outreach sessions in the past 18 months, and Beecher was present at the first one," says Milton Leitenberg, an arms controls expert at the University of Maryland.

Also in his paper, Beecher writes: "Individuals familiar with the compositions of the powders in the letters have indicated that they were comprised simply of spores purified to different extents." His citation for this statement is a 2003 article that investigative journalist Gary Matsumoto published in the news section of Science (302, 1492).

Meselson, who reviewed Beecher's article for the FBI, was asked to assess scanning electron micrographs of the anthrax powder. Early in 2002, he spent half a day at the FBI's Washington field office and looked at "a large heap of electron micrographs" of the powder from the Daschle letter.

"I saw no evidence of anything except spores, no evidence of silica nanoparticles," Meselson says. "If silica was present, I would have seen it, but nothing could have been purer than what I saw," he insists. Though purified, the preparation "had not been milled," he adds.

A government official who asked not to be named says the FBI knew early on that the Daschle and Leahy powders had a high concentration of spores. "But knowing the specific attributes of the spores took a longer time," he explains.

A former top military scientist speaking on background because his current employer has government contracts, tells C&EN that he, too, "saw scanning electron micrographs" of the powder from the Daschle letter. "I saw only spores and almost no rubbish from the culture media." If the spores had been coated with silica, they would have looked like doughnuts with large sugar particles on them, he says. Instead, "the Daschle spores were clean doughnut holes with no sugars."

He also says, "I had never seen a preparation that pure—10¹² spores per gram—with no rubbish." Curious about the purity of the spores, he contacted William C. Patrick III, who had made bioweapons for the Army when the U.S. had an offensive program. He says Patrick told him it was possible to get rid of nonspore material by repeatedly washing the spores with water and spinning off the culture debris into the supernatant.

This former military scientist never saw the material from the Leahy letter and "heard nothing from the FBI regarding the Leahy letter." So, even though he saw pure spores in the electron micrographs of the Daschle powder he was shown, "It was never clear to me whether the spores were coated or not, because I heard it both ways."

Media reports had described the material released when the Daschle letter was opened as looking like a cloud of smoke. "I had always thought the spores had to be treated to get them to fluff up as they did," he says.

Meselson, however, has another theory. He believes that "if the spores are pure enough, they will be suspended into air, they will fly." He builds his theory on the scientific scaffold of triboelectricity, which, he notes, "aerosol physicists haven't considered."

Triboelectricity occurs, for example, when combing your hair on a dry winter's day causes sparks to fly as electrons move from hair to bind more tightly to the comb. In Meselson's theory, all the purified spores carry the same electrical charge so they will fly apart. And, he says, "you don't need much to fly into the air" to cause harm.

Both Meselson and the former military scientist agree that making the purified preparations didn't require an expensive laboratory setup. As the military scientist says, "A simple facility" is really all that's needed. "I have concluded that maybe the hardest part is doing it safely so you don't hurt yourself. Some experience is needed, but it's probably more an art than a science," he says.

Arms control expert Jonathan B. Tucker, a Bosch Public Policy Fellow at the American Academy in Berlin, says, "The use of the Ames strain, the purity of the spores, and the extreme volatility of the material suggests that it was made by an individual with a high degree of technical sophistication."

Other experts say Beecher's now famous paragraph broadens the scope of potential suspects to include individuals or small groups lacking the resources of large national programs. Rutgers University microbiologist Richard H. Ebright, however, doesn't believe that it does.

As Ebright points out, the anthrax mailer had to have the "requisite microbiological and powder preparation skills." But equally important, the perpetrator "had to have access to the attack strain," which in all the letters was Ames.

Ebright admits that the pool of persons with the required skills is large and many times "larger than the pool of persons with access to the [Ames] strain." Prior to the Sept. 11, 2001, attacks, the Ames strain "was narrowly distributed," probably to "no more than a dozen, certainly no more than 20 laboratories" worldwide, he says. Labs possessing the strain were part of U.S. and allied biodefense and intelligence programs, and the perpetrator "must have obtained the attack strain" from one of these labs, he argues.

On Oct. 5, NBC Nightly News reported: "Investigators tell NBC News that the water used to make [the anthrax spores] came from a northeastern U.S., not a foreign, source." Ebright says, "This information, if correct, would appear to narrow the field" of labs possessing seed cultures of the Ames strain prior to Sept. 11, 2001.

As Ebright explains, "The intersection between institutions in possession of the Ames strain prior to Sept. 11, 2001, and institutions in the northeastern U.S. would appear" to narrow the likely source of the Ames strain to "two or three institutions: USAMRIID; the University of Scranton; and, if one interprets 'northeastern' broadly, Battelle Memorial Institute" in Columbus, Ohio. Battelle does classified research for the Department of Defense. A University of Scranton scientist was using "nucleic acid sequences to develop taxonomies of bioweapons agents, a subject of interest to the Department of Defense," Ebright says.

"If the NBC report on the identification of the water source is correct, it reflects further development of the analytical approach" reported in articles published in 2003 on the use of stable isotope analysis for microbe forensics, Ebright says. Those methods applied to O and H can provide information about the water used for the culture media, Ebright says.

In mid-to-late 2003, the FBI contracted out some 20-odd studies of the culture media using isotopic analyses to trace to a specific geographic area the water and nutrients used to grow the anthrax. Yet, early in 2002, DNA sequencing of the anthrax taken from the first anthrax victim conducted at the Institute for Genomic Research and other genetic analyses pointed to USAMRIID as the origin of the Ames strain.

The DNA sequencing work was published in Science in 2002 and reported by the media. Also noted in media accounts was the radiocarbon dating analyses by Lawrence Livermore National Lab in June 2002 that found the Ames attack strain was cultured no more than two years before the mailings.

In November 2002, FBI Director Mueller announced that efforts were being made to "reverse engineer" the mailed anthrax. News accounts in spring 2003 reported that the work was being conducted by the Army's biodefense center at Dugway Proving Ground.

These news reports, naming no sources, claimed that Dugway had successfully reproduced the anthrax powder used in the attacks. Dugway, according to the media, concluded that the attack material was made with simple methods and inexpensive equipment and that the spores were not coated with an additive such as silica.

Daniel Martin, a microbiologist in Dugway's Life Sciences Division, tells C&EN that Dugway was asked "to produce materials to see how they compared with the materials the FBI had in its possession." But, Martin says, Dugway did not reverse or back engineer the attack powder. "Back engineering implies that you know exactly what the material is and can replicate the material exactly, step by step." That isn't what Dugway did, he says.

Instead, Martin says, Dugway used the Leahy powder as the culture starter to "produce several different preparations using different media, and different ways of drying and milling the preparation" that the FBI could use for comparison purposes. Dugway, he says, never analyzed the Leahy powder and did no comparative analyses between the preparations made and the Leahy powder.

Indeed, by fall 2003, Michael A. Mason, then-assistant director of FBI's Washington field office, is quoted as saying that the FBI had not been able to re-create the process used to make the anthrax attack material. Still, he said, the FBI had learned enough to believe that the perpetrator had special expertise.

Leitenberg says a well-connected former military scientist told him that Dugway was only able to produce preparations containing "one-fifth the number of spores found in the Leahy powder." This same military source also told Leitenberg that Battelle Memorial Institute was also asked to back engineer the Leahy powder.

Back in 2003, Mason was not certain whether the anthrax case would ever be solved. Even if there was no "successful resolution," Mason said the investigation was "remarkable" because of the scientific and analytical skills employed.

So why, three years after Mason's public remarks and a pretty effective gag order, has the FBI chosen to speak out through Beecher's article? It's possible that the FBI is confident enough in the science "to set the record straight or to deflect ongoing or anticipated criticism," one former government official speculates.

It is also possible that Beecher's famous paragraph may be setting the groundwork for the FBI's defense in the suit brought against it by Steven J. Hatfill, whom former attorney general John D. Ashcroft called "a person of interest," the former official says.

A former FBI laboratory official says the FBI may have realized that the scientific evidence is pointing to a different conclusion than initial speculation that the perpetrator had to be associated with a national program. If so, "then it is very valuable for a number of reasons to have the evidence published in a peer-reviewed scientific journal, which gives it a measure of acceptance and credibility," he says.

To underscore his statement, the FBI lab official points to "the Daubert standard for scientific evidence and associated case law." This standard is a legal precedent by which federal trial judges rule on the admissibility of evidence based on its relevance and reliability (C&EN, Feb. 27, page 36).

Despite Mason's uncertainty three years ago, the FBI now seems confident that the case eventually will be solved. Writing in the Washington Post on Oct. 6, former State Department intelligence analyst Kenneth J. Dillon says there are two possible reasons for that confidence. One is that the FBI actually knows but lacks some confirmatory evidence to nail the perpetrator. The other, he writes, is embarrassment because "the evidence [the FBI might have] points to the clandestine biowarfare program of a close ally as the anthrax source."

If NBC reported the science correctly and the water used to make the anthrax did come from a northeastern U.S. source, Dillon's second supposition falls apart.

Leitenberg says that "scientists in the biodefense programs of several nations allied to the U.S. have frequently expressed the suspicion that the U.S. government is embarrassed to identify segments or individuals of the U.S. biodefense community as responsible for the 2001 anthrax events."

The FBI is not talking about the perpetrator and is saying very little publicly about the science it has called upon in trying to solve the five-year-old case. What the public has been told points to a U.S. biodefense facility as the source of the attack strain of anthrax spores that were not specially treated or engineered but were very pure—and very deadly.

Chemical & Engineering News

ISSN 0009-2347

Copyright © 2011 American Chemical Society