The American Society for Microbiology (ASM), which has come under recent attack for publishing information in scientific manuscripts that some see as potentially helpful to terrorists, released new guidelines last month in an effort to address the problem. These guidelines seek to establish a clear strategy for handling manuscripts that deal with microbiological agents that could have a bioterrorist application; however, their creation highlights the lack of a consensus on this area in the scientific community, setting the stage for a debate that strikes at the heart of scientific freedom.

Researchers and security experts have been asking whether certain scientific data should be suppressed since February of 2001, when Australian virologists described a strain of mousepox virus that could kill mice that were immune to the wild-type virus. The work appeared in the ASM-affiliated Journal of Virology (75, 1205; 2001).

But the debate remained academic until last September, when postal anthrax attacks in the US killed five and focused intense public attention on bioterrorism. Last month's suspension of Louisiana State University (LSU) scientist, Steven Hatfill, whose home has been searched by the Federal Bureau of Investigation in relation to the anthrax attacks, has increased attention on the role of biomedical research and its possible exploitation by terrorists.

Hatfill, currently the Associate Director of the National Center for Biomedical Research and Training at LSU, is a former government bioweapons researcher (although his security clearance was cancelled in September of 1999). Because of the nature of his research, Hatfill has been engaged with the US Army Medical Research Institute for Infectious Diseases (USAMRIID) in Frederick, Maryland, and genetic analysis of one of the anthrax-laced letters indicates a strong correlation with a strain developed at USAMRIID. Hatfill made a public statement last month announcing that he has never worked with anthrax. His latest research paper focuses on the mouse adaptation of the Ebola virus as a model to study the pathogenesis of the infection (J. Comp. Pathol. 125, 243; 2002).

The anthrax attacks spurred drastic increases in government funding for biodefense research and inspired new legislation aimed at curbing access to deadly pathogens. Yet scientists have expressed fears that new regulations could seriously harm academic openness, (Nature Med, 8, 197; 2002). In response, the Massachusetts Institute of Technology (MIT) convened a faculty panel to establish new policies on classified biological research. MIT—where radar was invented during World War II—addressed similar concerns in its engineering departments decades ago by establishing a separate campus exclusively for classified engineering research.

The new panel's report, issued 12 June, advocates a similar approach for the biological sciences; classified research or studies involving restricted pathogens would have to occur at a separate, off-campus facility provided by the university as a means of preserving academic openness on the school's main campus.

Hatfill denies anthrax involvement Credit: AP Photo/Rick Bowmer

The ASM's move to draw up specific guidelines relating to terrorist misuse of biological data demonstrates how much of an issue this is becoming for scientific journals. A paper published in Science this July (Science 297, 1016; 2002) describing the synthesis of ainfectious poliovirus from oligonucleotides in a cell-free system raised public concern that a similar process could yield a deadly virus that might be used in a bioterrorist attack. Yet, the manuscript was criticised in a letter to the journal last month by biowarfare expert Steven Block of Stanford University. Block says that Science's decision to publish the article amounted to “little more than a stunt.” He has called for a “national dialog among biologists, health-care experts, politicians and the general public about the future of biological work with biological weapons implicatons.”

Ronald Atlas, president of ASM, says that since October, he has received five requests from researchers to withhold certain data from papers for security reasons, although he refused to specify who wanted to withhold information or from what papers. The requests prompted Atlas to send a letter on 22 July to the National Academy of Sciences (NAS), asking that they convene a conference for journal publishers to discuss the issue.

ASM publishes 11 journals and has around 40,000 members. Among other things, the new manuscript guidelines ask that all reviewers advise the editor, in confidence, “if, in their opinion, the manuscript under review describes misuses of microbiology or of information derived from microbiology.” If it does, appropriate steps will be taken in consultation with the Publications Board. See http://journals.asm.org/misc/ Pathogens_and_Toxins.shtml.

Atlas divides the problem into two separate questions: “One is whether you should publish something or not.” Even if the government considers the research unclassified, ASM's new policy allows journal editors to act as a “secondary screen,” to block the publication of a paper that could aid terrorists. “The second question,” says Atlas, “which is really the one that drove me to send the letter to the National Academy ... is if you are going to publish something can you leave details out to prevent the information from being misused”.

The ASM's answer to this is “No”. According to the society's policy, all papers published in its journals must contain sufficient methodological details to reproduce the work. Atlas is hopeful that the NAS meeting, now planned for this autumn, will help other journal publishers adopt uniform standards.

As for our own policy, Editor-in-Chief of Nature Medicine and Editor of Nature, Philip Campbell says, “We keep in touch with people thinking about these issues to ensure that we are sensitive about them. Several years ago we rejected a paper because it constituted dangerous information—in nuclear technology rather than biology. Referees are typically quick to point to potential abuse of published information. But I've heard nothing so far to make me seriously question our policy that research published in Nature and its related journals should allow authors unrestricted access to key data.”