In recent years, chemicals that inhibit the cell cycle have found a home in cancer treatment. Now some of these same drugs are being tested, with surprising success, as antiviral compounds.

When developing therapeutic compounds for treating infectious agents, parasites, bacteria, or viruses, experts say there are two general approaches that can be taken - attack the invader, or interfere with the host in ways that stop the invader.

Traditionally, researchers have used the former approach, focusing on the infectious agent's own proteins and enzymes. Theoretically, this approach provides specificity for the drug, protecting the host cells from further damage. For example, a drug might inhibit the virus's polymerase enzyme - as many currently available antiviral compounds do - but leave the host cell's polymerases functioning normally.

This specificity for the invader is important because if a drug knocks out an essential function in both the host cells and the infectious agent, it might not be very helpful in the long run. Or, at least, that's the dogma as to what should happen.

But, recently, two groups are challenging that notion. Tom Albrecht's group at the University of Texas Medical Branch at Galveston shows that chemical inhibitors of the host's cell cycle block cytomegalovirus (CMV) replication and production of infectious progeny. And following his lead, Priscilla Schaffer's group, previously at the University of Pennsylvania and now at Beth Israel Deaconess Medical Center in Boston, Massachusetts, demonstrates that herpes simplex virus (HSV) replication is also effectively blocked. Importantly, host cells didn't show significant signs of toxicity in either case.

Albrecht says, "Quite honestly, we were taken aback in two ways: it is a very good block, and it appears to have limited toxicity."

Although Albrecht is surprised at the effectiveness of the block, previous data from his lab and a large number of other labs show that CMV infection induces the host cell to enter the cell cycle. In fact, the virus uses a very clever trick to increase its own replication efficiency. Once inside the cell, the virus stimulates the host cell to enter the cell cycle. Progressing through the G1 phase, the cell prepares for DNA synthesis, making all of the components necessary for rapid DNA replication. "Then," says Albrecht, "this dastardly, devious little virus blocks cellular DNA synthesis, which allows the virus uncompeted access to all of these building blocks or precursors that are now used to make viral DNA."

Given this data, Albrecht reasoned that if he could prevent the host cell from entering the cell cycle, he could block viral replication. And sure enough, by adding inhibitors of cdk2 (a kinase required for cell cycle progression) to tissue culture cells recently infected with CMV, the scientists prevented the production of new viruses and effectively limited the original infection.

Thus far, the researchers have focused on two inhibitors of cdk2 - roscovitine and olomoucine. Both compounds work in the assay, although roscovitine seems to be slightly more effective.

While the results from both groups are promising, it is still early, according to Albrecht and others in the field. "If the process [of developing antiviral compounds] has ten hurdles," says Edward Mocarski of the Department of Microbiology and Immunology at Stanford University School of Medicine in California, "the first hurdle or first two have been cleared, with eight or nine to go for roscovitine."

Toxicity is a major concern with these compounds and is, as yet, largely untested in the antiviral work. Schaffer's group has some very preliminary data suggesting that roscovitine is effective in blocking HSV infection in mice and that the drug has limited toxicity. But, perhaps, more importantly, there is substantial work in the cancer field using related drugs, which demonstrates that cdk2 inhibition is relatively well tolerated in humans. Flavopiridol, a distinct, but chemically related, inhibitor of cdk2 has already passed through Phase I clinical trials and is in Phase II testing for use as a chemotherapy agent. Although each drug will need to be tested independently, the limited toxicity of one cdk2 inhibitor bodes well for another.

But why take this new approach of targeting host proteins when there has been significant success already targeting the viral machinery? The main reason, according to some scientists, is that the risk of developing drug-resistant strains of virus decreases if the drug acts on a host protein as opposed to acting on a viral one. If the drug binds to a viral protein, there will be strong selection pressure favoring mutations that limit affinity for the drug. If the drug interacts with a host protein, this pressure is, at least, indirect, if not reduced. Also, it appears that by targeting one part of the cellular machinery, multiple stages of the viral life cycle are blocked, and this is added insurance against developing drug resistance.

The other reason to look at new approaches to drug development is that none of the drugs available is a panacea. According to David Koelle, a clinician in infectious diseases in the Division of Allergy and Infectious Diseases at the University of Washington at Seattle, the drugs available for treating CMV patients are hard to use and can be quite toxic over time, and he sees a medical need for more effective and easier-to-use drugs.

As Mocarski points out, it is too early to tell if these drugs will make it through the maze of drug development; but, a number of virologists and clinicians interviewed are cautiously enthusiastic about this new approach. "It sounds good to me," says Koelle. "We know viruses need to hijack cellular machinery for many steps of their own replication pathway." But, he adds in a more cautious tone, "It seems like the inevitable dilemma of pharmacology that when you start mucking with some cellular machinery, which the cell probably evolved for some perfectly good reason, that you could potentially lead to toxicity, in addition to inhibiting the virus."