Constructing artificial proteins to inhibit HIV

The Berlin Patient, an American living in Germany, was cured of HIV in 2007. The patient had developed leukemia, and his doctors decided to try to kill two birds with one stone. He received a bone marrow transplant from a donor who lacked the gene for CCR5, a receptor required for HIV transmission. With these HIV-resistant donor cells replacing his own blood cells, the Berlin Patient’s T-cell population increased until HIV could no longer be detected in his system.

Bone marrow transplants with donor cells, however, are not a feasible way of routinely knocking out the CCR5 receptor—the risk of infection is high and the donor cells may attack the recipient’s tissues. The only FDA-approved CCR5 inhibitor, a drug called Maraviroc, alters CCR5’s conformation so that HIV doesn’t recognize it. Over time, however, strains of the virus can learn to recognize the altered receptor, resulting in drug resistance.

Researchers led by Daniel DiMaio, M.D., Ph.D., the Waldemar Von Zedtwitz Professor of Genetics and deputy director of the Yale Cancer Center, have taken a different approach—preventing the receptor from reaching the cell surface in the first place. Their findings appeared online in the Journal of Virology on July 18, 2012. Elizabeth Scheideman, Ph.D., a postdoctoral fellow in genetics, is the first author.

“We did not at all think we were studying HIV,” said DiMaio, who has devoted almost 30 years to studying papillomaviruses. The HIV research arose when a “very strange” bovine papillomavirus protein called E5 led him to make a “conceptual leap.” DiMaio noticed that E5, a chain of 44 amino acids, slips into cell membranes and activates a receptor for cell growth factors. DiMaio wondered whether he could build proteins similar to E5 with such different targets as CCR5.

He decided to create artificial proteins modeled after E5 that throw a “monkey wrench” into CCR5 to prevent it from functioning properly. He called these proteins “traptamers” since they traverse cell membranes and are apt to bind to other proteins. Ideally, traptamers will cause the CCR5 receptors to become so misshapen that they degrade before reaching the cell surface.

“The great thing about CCR5,” DiMaio said, “is that we—you and I—don’t need it.” Other chemokine receptors perform the same function, receiving signals that direct the immune system to infection sites. But these other receptors don’t allow HIV infection.

DiMaio’s team made a library of 300,000 randomized genes that coded for proteins similar to E5—hydrophobic and about 40 amino acids in length—then screened the list for traptamers that might work. Once they had promising candidates, they used retroviruses to insert individual genes from the library into mouse cells, and cells that expressed CCR5 were sorted out. They continued weeding out traptamers in the remaining cells until only six were left.

Further testing ensured that the traptamers were working the way scientists hoped—they inhibited 70 to 80 percent of HIV infections. Still, DiMaio said, “we thought we could do better.”

So they “sprinkled a few mutations” into the traptamers to see if the tweaks could enhance activity against CCR5. “It worked great,” DiMaio said. Now some of the traptamers inhibit more than 95 percent of infections.

DiMaio said he hopes this approach could be applied to any transmembrane protein target, including those for diseases other than HIV. And thanks to biological selection, researchers don’t even need to know the target’s sequence or structure. Since many drugs already target transmembrane proteins like CCR5, traptamers would present an entirely new class of therapeutic agents.

“A number of groups around the country are looking at ways to use gene therapy approaches to knock out the CCR5 gene to follow up on the results with the Berlin Patient.” said Ronald Desrosiers, Ph.D., an HIV expert and professor of microbiology and immunobiology at Harvard Medical School who was not involved in the study. “The DiMaio publication provides a totally new approach for knocking out CCR5 function and is crying out for more investigation.”