Considering the sometimes glacial pace of biomedical research and drug development, the response to severe acute respiratory syndrome has been extremely nimble.

In December, 25 months after the first SARS cases came to the attention of the international medical community, and less than two years after the SARS coronavirus behind the disease was identified and sequenced, the first experimental vaccine against SARS entered U.S. clinical trials at the National Institutes of Health. Another vaccine entered clinical trials in China (which, along with Hong Kong and Singapore, had the bulk of SARS cases) in May.

While the Chinese vaccine is based on inactivated SARS coronavirus, the NIH vaccine, produced by San Diego biotechnology company Vical Inc., is DNA based. It targets the spike protein, a protein that the virus uses for entering host cells via the angiotensin-converting enzyme receptor.

Unfortunately, the SARS coronavirus, and specifically the S-protein, appear to be no slouch, either. In research published in today's issue of the Proceedings of the National Academy of Sciences but available via early online publication, Gary Nabel and his colleagues from the National Institute of Allergy and Infectious Diseases and the Institute for Research in Biomedicine in Bellinzona, Switzerland, showed that the spike protein has not only mutated several times since the emergence of SARS in November 2002, but also that antibodies that protected against entry of early strains of SARS can enhance the entry of some of the newer strains into cells.

"This virus is evolving; it is responding to antibodies differently than it used to," senior author Gary Nabel told BioWorld Today. Nabel is the director of the NIAID vaccine research center. The vaccine being tested by the NIH was developed in his laboratory.

The scientists tested how the antibodies to spike protein from the S. urbani strain, an early strain of SARS on which the current vaccine is based, affected the ability of spike proteins from others strains of SARS coronavirus to enter host cells. They included strains isolated from patients early and late in 2003 during separate outbreaks. They also looked at strains isolated from palm civets, mongoose-like animals that are thought to be reservoirs for the SARS coronavirus, in their study.

To compare the effects of different spike proteins, the scientists grafted those proteins onto an unchanging backbone to make a so-called pseudovirus. The advantage of that approach is that the effects of changes in spike protein can be investigated independently from changes that might have occurred in other parts of the virus, though it is not entirely clear to what extent findings in the model system apply to whole viruses.

One Strain's Vaccine Is Another Strain's Enhancer

Each pseudovirus was incubated with polyclonal mouse antibodies raised against the spike protein of the S. Urbani strain, and the scientists tested the ability of the different pseudovirus strains to enter 786-O cells. While the antibodies inhibited all pseudoviruses based on spike proteins from early 2003, pseudoviruses based on spike proteins from late 2003 were much less inhibited, and in several cases, antibodies to the S. urbani strain actually enhanced the entry into cells of pseudoviruses based on palm civet-derived spike proteins. Further experiments showed that antibodies that could enhance palm civet spike protein entry could be detected in samples from former SARS patients. The degree of inhibition correlated with the strength of binding to the angiotensin-converting enzyme receptor.

In a final series of experiments, the researchers attempted to build an improved immunogen by modifying the spike protein to prevent entry of the S. urbani strain without enhancing palm civet strain entry. While most antibodies that were effective against S. urbani enhanced the entry of palm civet spike protein, a few did not, suggesting that the development of better imunogens should be possible - though the SARS coronavirus, of course, also is continuing to evolve in its animal reservoirs.

When asked whether his data implied either that a different protein might be a better target for vaccines, or that a SARS vaccine, like the flu vaccine, might have to be manufactured annually to protect against current strains, Nabel replied, "I don't think we're at that point yet." Though he also acknowledged that "we worry about enhancement of SARS because there are times when vaccines can make infections worse."

Nabel thinks his results point to the "need to identify strategies for broadest coverage" against SARS. One possibility is to make sure the cellular arm of the immune system is stimulated, which is the case with the DNA vaccine currently in trials at the NIH. Another is to add the gene for another viral protein, or genes for multiple forms of spike protein, to the current vaccine.

Given that no new human cases of SARS have been reported since December 2003, there is at least a small possibility that all that effort is directed at a virus that the medical community will never see again. Asked whether the reappearance of SARS is a foregone conclusion, Nabel replied, "It's very difficult to predict - as difficult as whether H5N1 will cross over and cause a pandemic."

But he said he believed that it is better to assume SARS will reappear.

"I'd like to think that at least momentarily, we have good measures in place," he said, adding that given the brief time period since the SARS coronavirus was identified and characterized, the international medical community is "remarkably well prepared" for a possible reemergence. "The wise thing now is to be prepared so that if [SARS] should arise at any point in the future, we won't be caught flat-footed like we were."