By Dean A. Haycock

Special To BioWorld Today

The arms race that helped define the cold war is over. The arms race between the immune system and viruses rages on. The strategies and counter-strategies in the latter struggle are played out on the molecular level in a way more sophisticated and ingenious than any developed by cold war defense contractors.

"Viruses have been studying immune systems for millions of years. It stands to reason they have learned tricks we don't know about yet," said Hidde Ploegh, professor of biology at the Massachusetts Institute of Technology, in Cambridge.

Herpes viruses such as cytomegalovirus (CMV) hide in cells and establish chronic infections. Uninfected cells normally display molecular signals on their surfaces called major histocompatibility complex class I (MHC) molecules. The MHC molecules include bits of molecules found inside of cells. They label the cell as healthy when the bits are all recognized as "self" by circulating T cells. When MHC molecules incorporate bits of foreign, virus peptide into their structures, it serves as a signal for the T cells to attack and kill the infected cell, thus destroying the virus and limiting the infection.

To succeed in their undercover role, CMVs can prevent infected cells from displaying any MHC signals on their surfaces. With no MHC displayed, there is nothing for the T cells to recognize as "non-self" and the virus avoids destruction.

It has been known for some time that many viruses employ this trick to avoid destruction by T cells. Four genes in the human CMV work together to prevent cells they infect from presenting MHC molecules that would give away their presence.

The immune system, however, has other resources to call on. One is the natural killer (NK) cell. NK cells terminate cells that fail to present any MHC "all clear" signals.

The latest dispatches from the front appear in the current issue of Nature. Both articles report a new strategy employed by CMV to outwit the immune cells that actively hunt them.

In "The class I MHC homologue of human cytomegalovirus (UL18) inhibits attack by natural killer cells," post-doctoral fellow Hugh Reyburn, Higgins Professor of Biochemistry Jack L. Strominger and their co-authors from Harvard University report that the CMV can outsmart NK cells by producing a protein that mimics MHC molecules on the surface of infected cells.

The "decoy" MHC molecule wards off roaming NK cells, thus allowing the virus to survive in infected cells. The decoy also fools T cells into acting as if infected cells are healthy when they are not. Reyburn et al. studied the CMV gene product UL18 in cultured cells.

Australian researcher Helen Farrell and her colleagues from the University of Western Australia describe the phenomenon in vivo in mice, in a paper titled "Inhibition of natural killer cells by a cytomegalovirus MHC class I homologue in vivo."

These papers "show yet another example of a trick used by a virus to elude the immune system," Ploegh said.

The experiments indicate that homologous genes in both human and mouse cytomegalovirus produce the decoy MHC.

The human CMV class I gene homologue is named UL18. When the Harvard scientists cloned and expressed it in a line of human cells that lacked MHC class I on their surfaces, they transformed the cells from being NK-susceptible to NK-resistant.

UL18 interacts with a receptor called CD94 found on NK and T cells. The interaction appears to be based on carbohydrates, which make up part of the structure of UL18.

"The most important thing we can do is to identify the carbohydrate on the UL18," Strominger told BioWorld Today.

Strominger next would like to show that the CD94 and the UL18 gene product interact directly.

"One could imagine that if one knew a great deal about the interaction, we could figure out ways to block it so infected cells would get killed more readily. That is a long way down the road," Strominger said. In collaboration with researchers from Cambridge, U.K., the researchers from Cambridge, Mass., will be studying CMVs with and without the MHC mimicking gene.

"We can try and have a look at the function of this gene in context of the virus-infected cell as opposed to the gene in isolation," Reyburn said. The research described in the two papers is significant because of what it tells us about the importance of NK cells, which are more often considered in the context of tumor immunology than of viral immunology.

"If the virus bothered to have a strategy to avoid attack by natural killer cells, then that would suggest that natural killer cells are an important part of the antiviral immune response," Reyburn said. Ploegh agreed.

"It shows that viruses care not only about T-lymphocytes and antibodies, but they also care about cells from the innate immune system (which includes NK cells)," Ploegh said.

Ploegh added a note of caution.

"At the same time one should not be uncritically enthusiastic, either. One should acknowledge that while these experiments on UL18 demonstrate that in vitro this molecule has the properties ascribed to it, it is much less clear what the relevance of this particular gene product is to the virus at any point in the infectious cycle. That is true for almost all 'evasive' examples of this kind. It is one thing to show that in the test tube these proteins do this, that or the other thing, but to prove that the virus actually evolved this product for the specific purpose that the scientists ascribe to it is a different story altogether."

Ploegh said the paper by the Australian group reinforces the picture presented by Reyburn and Strominger and begins to address the role of the decoy in determining the virulent properties of the virus.

"As such, it is an important step forward because the Australian group appears to be close to establishing the in vivo relevance," Ploegh said. *