BioWorld International Correspondent

LONDON - The human immunodeficiency virus successfully infects human cells because one of its proteins overcomes what is otherwise a highly effective antiviral defense mechanism, a new study suggests. The scientists who carried out the work say it may one day be possible to develop antiviral drugs to treat HIV infection by targeting the viral protein that blocks the cellular defenses.

Allan Hance, a research director at INSERM in Paris, told BioWorld International, "It looks like human cells have come up with this antiviral strategy, and the retroviruses, including HIV, have in turn evolved a protein that blocks the defense system."

Hance and his colleagues report their study in the May 16, 2003, Science in a short paper titled "Hypermutation of HIV-1 DNA in the Absence of the Vif Protein."

Scientists have scrutinized and analyzed every single gene and protein of HIV-1 in the quest to understand how these proteins help the virus attack human cells and subvert their genetic machinery. Until now, however, the function of one protein - Vif, which stands for viral infectivity factor - had remained a mystery.

Researchers knew that Vif was essential if HIV were to infect T lymphocytes and monocytes, and stimulate those cells to produce more virions, but no one knew how Vif worked.

There was just one clue. Researchers had observed that some cell lines allowed HIV to infect them and produce further generations of virus even if Vif was not present. Those cell lines were called permissive cell lines. Others, called non-permissive cell lines, could only be infected with production of new viruses, if Vif was present.

In London, a different team of researchers, led by Michael Malim, had carried out a study to find out how the expression of proteins by permissive cell lines differed from that of non-permissive cell lines. They discovered that the key difference was a protein called CEM15, also known as APOBEC3G. It is a member of the family of cytidine deaminases, an evolutionarily ancient enzyme family present in both bacteria and mammals.

Additional studies also had shown that one member of this family, APOBEC1, could edit the messenger RNA coding for the protein apoB in mammalian cells, turning a single cytidine base into uridine, and resulting in the production of a truncated protein. More recently, researchers discovered that another member of the same family of enzymes, a protein called activation-induced deaminase (AID), plays a role in bringing about hypermutation of the DNA encoding the immunoglobulins of B cells, the process that helps to produce an infinite variety of antibodies.

Hance said, "This put the idea into our heads that perhaps APOBEC3G could also edit DNA." Accordingly, he and his colleagues set out to establish whether Vif could prevent editing of single-stranded DNA synthesized by viruses produced in non-permissive cells.

The team took non-permissive cell lines and added either wild-type HIV-1 or a strain of HIV-1 from which the gene encoding Vif had been deleted. After 24 to 48 hours, the viruses produced were recovered and used to infect new target cells.

Next, the researchers isolated the newly produced viral DNA from these cells, which had been manufactured by reverse transcription. They amplified some sections of viral DNA, which were then cloned and sequenced.

They found that there were many places in the DNA synthesized by the HIV-1 that lacked the Vif gene, where guanine bases had changed to adenine, whereas few such changes had appeared in the DNA synthesized by wild-type viruses. The likely explanation for this finding, the authors wrote, is the alteration of cytidine bases to uridine in the single-stranded DNA of viral origin, which has the effect of changing guanine bases to adenine in the opposing DNA strand.

"What seems to happen is that, during the first cycle of infection with viruses that lack Vif, the new viral particles that are formed have the APOBEC3G enzyme contained within them," Hance said. "When these particles try to infect new target cells, the virus enters and starts to synthesize its DNA, but, because Vif is not present, APOBEC3G can start to edit the DNA by changing the cytidines to uridines. This introduces errors into the DNA, so that when manufacture of viral proteins begins, these also have errors."

Hance and his colleagues found that, in the DNA encoding the envelope protein of these virions, several premature stop codons had been introduced, leading to amino acid substitutions that meant this protein would have a truncated and abnormal structure.

He and his team are now embarking on a program to find out how the Vif protein interacts with APOBEC3G. "If we can work this out," Hance said, "and establish how to block this interaction, then we can start thinking about ways of protecting the antiviral activity from Vif, which could presumably block HIV infection."