Forty years ago, Jay Levy co-discovered HIV. He has spent the four decades since studying how the virus works inside human cells, and, when he could, how to stop it. Last month, his original frozen virus samples, saved from patients during the early AIDS epidemic, landed on the lab bench of a different team at a different institution entirely. The Gladstone-UCSF team needed real virus to test something no one had done before: a genome-wide map of every human protein that either helps HIV spread or tries to block it, measured not in immortalized cell lines but in the actual immune cells the virus infects in people.
The result, published this week in the journal Cell00382-X), is the most comprehensive catalog yet of what scientists call host factors: the roughly 25,000 human proteins that HIV either exploits to replicate or runs into when it tries. Among them, the team found two proteins that had never been documented blocking HIV before. They then showed that one of them, a protein called PPID, could be engineered in the lab to be roughly 10 times more potent than its natural version at stopping the virus.
"We can now infect up to 70 percent of the cells in a dish with HIV," Eli Dugan told EurekAlert, a co-first author of the study and a researcher at the Gladstone-UCSF Institute of Genomic Immunology. Under standard lab conditions, HIV infects just one or two percent of primary human CD4+ T cells: the exact immune cells the virus targets in the bloodstream. The improvement in infection efficiency, achieved by concentrating the virus and isolating the cells under conditions that mimic the human body more closely than standard cell lines, is what made a genome-wide screen possible for the first time.
The two newly identified antiviral proteins are called PI16 and PPID. PI16 appears to work at the moment HIV tries to fuse with a T cell, blocking the virus before it even enters. PPID acts after the virus has already slipped inside, limiting its ability to reach the cell nucleus where it would normally integrate its genetic material and begin churning out new virus particles. Both proteins restricted even aggressive HIV strains from the early epidemic when tested in T cells carrying elevated levels of them.
The research was led by Alex Marson, a senior author and director of the Gladstone-UCSF Institute of Genomic Immunology, with Ujjwal Rathore as co-first author and corresponding author. Nevan Krogan, a senior investigator at Gladstone and director of the HIV Accessory and Regulatory Complexes Center at UCSF, was also a senior author. The team used two complementary CRISPR approaches, one that switches genes off and one that switches them on, to test nearly every human gene simultaneously.
PPID is structurally related to a protein called CypA, which HIV actually uses to its advantage; the new work shows PPID binds the same part of the virus that CypA binds but produces the opposite effect. Through structural modeling and evolutionary analysis, the team identified the specific amino acid residues that give PPID its antiviral activity, findings that could eventually inform drug design, though no therapeutic has been developed yet.
The link to Levy's original isolates is not incidental. Many lab strains of HIV have been passaged for decades under conditions that select for viruses adapted to grow in artificial cell environments. Those strains behave differently from the viruses that actually circulated in people during the 1980s and 1990s. By testing their screen against Levy's early epidemic samples, the team gained confidence that the host factors they identified are relevant to the real pathogen, not an artifact of lab adaptation.
"This was the first genome-wide effort to show how human genes affect HIV infection in cells taken directly from human blood samples," Dugan said.
The practical path from this finding to a human therapy is long and uncertain. No study of this kind has yet produced an approved drug for HIV, and the leap from a protein identified in a dish to something that works in a patient involves years of pharmacology, safety testing, and clinical trials that the current work does not attempt to address. The team is making its data publicly available so other researchers can explore the host factors further.
For Levy, who retired from UCSF a decade ago but keeps a laboratory, the collaboration represents a kind of scientific continuity that rarely happens at this scale. His archive of early epidemic isolates has been used for decades to study how HIV has evolved, and now, for the first time, it helped define which human proteins the virus encounters when it enters a real immune cell. The two proteins his samples helped identify are not a cure. But they are the most concrete new therapeutic targets to emerge from host-factor research in years, and they came, in part, from a virus sample he put in a freezer when Ronald Reagan was president.
