For a number of years, I’d been interested in studying how human immunodeficiency virus enters cells. In particular, what elements of the envelope glycoproteins of the virus mediate the attachment of the virus to the target cell, and then mediate the fusion of the viral membrane with the target cell membrane to allow the entry of the interior of the virus (the capsid) into the cytoplasm of a healthy cell? It had been known that the CD4 glycoprotein acted as the initial receptor for HIV-1. That’s the first receptor, but it was also shown that expression of CD4 was not sufficient to render a cell susceptible to HIV-1 infection. That and other pieces of evidence suggested there might be a second receptor for HIV.

“What we’ve been trying to do over the years is to understand how HIV works—how HIV operates to make more of itself and to cause disease, and once we can understand those operating principles in detail, we can find very precise ways to interrupt the process.”

And so the work in that Cell paper identifies one of the major co-receptors for HIV, CCR5, which is used by most of the clinically relevant isolates of HIV-1 to allow the completion of the entry process. First, the virus binds its envelope protein to CD4 and then that binding of CD4 to the envelope protein allows the envelope protein to bind to the second receptor, CCR5. The second receptor binding leads ultimately to the fusion process, whereby the viral membrane and the target cell membrane fuse. So there are two gateways for HIV-1 to enter cells: CD4 and one of the chemokine receptors, which in this case is CCR5. Earlier in that same year, Ed Berger had identified another chemokine receptor, CXCR4, which could also act as a second receptor for certain strains of HIV-1.

It’s clear that certain isolates of HIV-1 can use CCR3, and there may be certain cell types in the body where CCR3 is actually utilized. We do think that CCR5 is the major receptor and probably the most important co-receptor for clinical isolates, although CCR3 can also be used. Subsequently people have identified other chemokine receptor relatives that can be used at least in tissue culture by virus to enter cells. For the most part, it seems that those are less clinically important. CCR5 is probably the major receptor, and after that CXCR4 is used by some strains of HIV-1.

I think that certainly in this case it was primarily identifying CCR5 among the potential candidates. The chemokine receptors are part of a family of receptors known as seven transmembrane segment G-protein coupled receptors. There are many of those that potentially could have been a candidate for this second HIV-1 receptor. We did have some clues, though: some work suggesting that some of the beta chemokines that were known to interact with chemokine receptors could inhibit HIV infection. That suggested to us that the chemokine receptors, which are a subset of the bigger family of G-protein coupled receptors, were probably involved in the entry process. We also had access to collaborators cloning cDNA encoding chemokine receptors. So at that point we were able to narrow the field enough to try a few different chemokine receptors, and demonstrate that CCR5 was the most potent in facilitating HIV entry. Some others, like CCR3, showed more modest effects.

We were aware of the importance of the work when we were doing it. Certainly the field at that point was ready to make these discoveries and there were many competing groups working on this. In fact, there were several papers on the chemokine receptors published in 1996 by other groups, as well. The problem had been lingering for several years as to what the second receptor might be.

At the beginning of 1996, nobody had much of a clue. By the end of that year, there were very many papers on the subject and it was clear it was quite important. Subsequently, chemokine receptors have been targeted by several pharmaceutical companies as potential ways to inhibit HIV infection in the clinical setting. So there are now drugs targeting CCR5 that are in clinical trials. We were pretty aware all along of the general importance of what we were doing, and it’s been satisfying to see some of those discoveries have had practical applications.

We continued to try to understand the molecular details of the HIV entry process, with the idea that as we understand those details we can define targets, either on the virus or on the target cell, for intervention. In particular, we’ve been interested in defining the structure of the HIV envelope glycoprotein, how those envelope glycoproteins mediate virus entry and how that entry process can be inhibited by, for example, neutralizing antibodies generated by the immune system or by small-molecule drugs. Some of the things we have been working on since 1996 involve the determination of structures of the HIV envelope glycoprotein itself. That work has been done in collaboration with Wayne Hendrickson at Columbia and Peter Kwong at the NIH.

We’ve also been interested in how the envelope glycoproteins function in the cell to contribute to the death of the cell. We’ve been interested in the molecular process whereby HIV kills its target cell, what role that plays in the depletion of CD4 T-cells in HIV-infected people. We have gathered evidence that the envelope glycoproteins of HIV are among the more toxic proteins that it makes; that those proteins can dictate how efficiently T-cells get destroyed during natural HIV infection. To that end, we’ve been establishing animal models in monkeys to study the effect in vivo of changes in the envelope glycoproteins of HIV.

We have also been studying, over the last six years or so, some of the processes that happen immediately after entry of the virus into the cell. The virus, once it enters the cells, introduces its capsid, which is basically an electron-dense core that contains its RNA. That capsid gets introduced into the cytoplasm, and we found that HIV and some other retroviruses often encounter blocks at that stage of their replication cycle in the cells of particular species. For example, HIV-1 can enter the cells of Old World monkeys, but the virus is blocked very soon after entry in Old World monkey cells. And there are other examples of other retroviruses that encounter such species-specific blocks. One of the questions that we pursued was, what mediates these species-specific blocks?

It turns out that some mammals make proteins that are called TRIM proteins, for tripartite motif proteins, and the block to HIV is mediated by one of these TRIM proteins, which is called TRIM5alpha. We now think that TRIM5alpha is part of our innate immune system and that it’s evolved in primates and in some other mammals, as well, essentially to deal with retroviral infection. The interesting thing is that humans do make a TRIM5alpha protein, but ours is only partially effective at inhibiting HIV. There is only a single amino acid chain that makes human TRIM5alpha less efficient than Old World monkey TRIM5alpha at blocking HIV. One of the things we’re trying to understand is how to potentiate the human TRIM5alpha. If we could make it a bit more potent, it would be a reasonable inhibitor of HIV infection. It’s an interesting example of innate intracellular immunity, and we’re trying to understand how it works as well as how to potentiate the components of this system in humans.

It’s been gratifying to see that the discoveries have really moved into a practical aspect. Not only do we know what these protein receptors are, but also screens have identified many potent inhibitors that can block HIV infection. I think it’s been pretty satisfying overall to see that in 10 years we now have inhibitors in clinical trials that have really proceeded directly from those discoveries made in 1996. It would always be more satisfying if we knew those inhibitors were totally effective, but we’re not at that stage yet. The clinical trials are still going on. There have also been some side effects identified for some of the inhibitors that have entered clinical trials. So there is still some room for improvement.

I think a lot of discoveries have a serendipitous aspect. One of the key things in this kind of research is that you always needs to be aware of the chance that there might be something unexpected revealed in the data. You have to be careful in analyzing your data to keep an open mind to the unexpected. In this area, the initial discovery of CD4 as a receptor for HIV was a significant advance, but the realization that in certain cell types it wasn’t sufficient to support entry was a very important observation that suggested to people that there might be another receptor out there. And that was somewhat unexpected, because most viruses use only one receptor, if they need a receptor at all.

We think that HIV uses two receptors because it helps to keep its envelope protein components away from immune surveillance mechanisms, like neutralizing antibodies. By using two receptors, HIV can mask some of the critical receptor-binding proteins on its envelope protein. I think that just being open to the possibility of there being more than one receptor was important in ultimately identifying what that receptor was. We were involved in that work as well as several other groups in the HIV field. A number of people were thinking along the same lines at a similar point in time.

That the general theme of our work is that knowledge is power. What we’ve been trying to do over the years is to understand how HIV works—how HIV operates to make more of itself and to cause disease, and once we can understand those operating principles in detail, we can find very precise ways to interrupt the process. That has really been the theme of our work. The more we can understand the molecular details of what the virus is trying to do, the easier it’s going to be to rationally design ways to stop that process.

Joseph G. Sodroski, M.D.
Dana-Farber Cancer Institute
Boston, MA, USA