ASSAY ACE: Aaron Diamond's Alexandra Trkola helped develop an assay that characterized the binding mechanisms of the CCR5 coreceptor, which HIV uses, along with CD4, to gain initial entry into cells.
Comments by John P. Moore, staff investigator at the Aaron Diamond AIDS Research Center of Rockefeller University in New York and associate professor at Rockefeller University; and Alexandra Trkola, research scientist at the Aaron Diamond AIDS Research Center.
Before recent HIV X-rays illuminated the virus's instruments of invasion, researchers had to find--and assemble--those pieces in the dark.
"We started out not knowing exactly what we were looking for," Alexandra Trkola, a research scientist at the Aaron Diamond AIDS Research Center of Rockefeller University in New York, recalls of her team's efforts to characterize the earlier-stage, M-tropic virus's binding mechanisms. However, some discoveries prior to their hunt provided piece-meal clues. They knew that the chemokine receptor CXCR4, then called fusin, served as a coreceptor, along with CD4, for the later-stage T-tropic virus (Hot Papers, The Scientist, 12:11, March 30, 1998). They also knew that ß chemokines somehow blocked the entrance of the M-tropic viruses. However, the M-tropic coreceptor and the actual binding mechanisms eluded them.
Initially, so did a viable assay. "I tried a lot of different cell lines," remembers Trkola. Their initial failures proved especially frustrating because the principle behind the experiments seemed so fundamental: since HIV's envelope protein gp120 and the chemokine MIP-1ß probably interacted with the same receptor, they should be able to block MIP-1 ß binding to cells by using gp120. Of course, the reality proved far from simple.
Complicating the search for a workable screen, the CCR5 coreceptor had not yet been identified when the group began its initial experiments, recollects Aaron Diamond's John P. Moore. "I started [the experiments] myself on a cell line called PM1, which I knew must express CCR5, even though we didn't know then exactly what CCR5 was." That didn't work well enough. "The results were frustrating--hints, but nothing good enough to publish." Then five groups independently identified the coreceptor (see above story). "So I asked Alexandra to take over." Trkola then tried an assay based on primary T cells derived from healthy donors. Everything fell into place, Trkola interjects. "Probably because she is better in the lab than I ever was," adds Moore.
The study revealed two fundamental aspects of the virus's entry to cells: that gp120 interacts with the coreceptor to prevent MIP-1 ß binding and that the protein must first bind with CD4 before gp120 links with CCR5. It also showed how neutralizing antibodies can prevent this activity.
Moore guessed that sometimes-collaborator Joseph Sodroski, a pathologist at the Dana-Farber Cancer Institute, was probably following the same course of investigation and had likely reached similar conclusions. Both had done earlier work on the envelope protein and had made some predictions about the coreceptor interaction, based on antibodies that bound to gp120 in other experiments. Moore was right. So the two coordinated independent submissions to the same issue of Nature. Sodroski's group had, indeed, reached the same results, using a different methodology (L.J. Wu et al., Nature, 384:179-83, 1996; also cited in more than 160 papers since publication). "When you have two papers coming from two different labs using somewhat different methods coming to the same conclusion, there's a very good chance it's going to stand up," Moore explains.
Moore adds that the recent crystal structure-related papers (P. D. Kwong et al, Nature, 393:648-59, 1998; R. Wyatt et al, Nature, 393:705-11, June 18, 1998; C.D. Rizzuto et al, Science , 280:1949-53, June 19, 1998; R. Wyatt, J. Sodroski, Science, 280:1884-8, June 19, 1998) provided a 3-D validation of both his team's and Sodroski's group's work (R. Lewis, The Scientist, 12:1, July 6, 1998). "The crystal structure proves that all the models we put together were basically right--which was obviously pleasing to both of us."
The next stage, Moore thinks, is to use the structure to learn more about specific binding sites. That work has already begun (Rizzuto et al.). More detail about the structure and the binding activity will also provide ammunition for developing a small molecule inhibitor to fight HIV. "There are enough compounds kicking around in the labs to know that this is feasible," Moore says. The question is how well they will do in the clinics. Moore predicts that, a year from now, clinical trials will have begun answering that question.