The Scientific Activist

The New York Times reported yesterday that "scientists find new receptor for HIV," referring to a paper published online in Nature Immunology on Sunday by Arthos et al. This is basically correct, although it would be more accurate to call the new receptor a co-receptor, since the infection of a cell with HIV still depends on the primary receptor, CD4, in combination with either CCR5 or CXCR4. The newly-identified co-receptor, just like the other HIV receptors, is a protein located on the surface of white blood cells (T-cells, specifically). HIV, like any other virus, can only replicate when it's inside of a living cell. To get inside, however, the virus must first recognize and attach to the target cell, using these receptors.

The reason that I'm blogging about this, though, is that the new co-receptor is the alpha-4 beta-7 integrin--a member of the family of proteins that I study. The integrins are a family of large cell-surface proteins. They're heterodimeric, meaning that each integrin contains two distinct molecules--an alpha subunit and a beta subunit. In mammals, there are 18 different alpha subunits and 8 different beta subunits, and alpha-4 beta-7 is one of 24 unique alpha-beta pairs that exist in the body. The main job of the integrins is to anchor cells to the extracellular matrix (although they also play important roles in cell signaling, as I've written about previously). This is only true, however, in cells that are normally adherent to the matrix. In non-adherent cells (various types of blood cells), integrins have different roles. In platelets, for example, the alpha-IIb beta-3 integrin is normally inactive. When the body sustains a wound, alpha-IIb beta-3 is activated, allowing it to bind to fibrinogen and causing a blood clot to form. In white blood cells, on the other hand, integrins (with the beta-2 and beta-7 subunits) are used to let the cell roll along the blood vessel wall and then infiltrate it and migrate to the site of an infection.

HIV replicates primarily in the part of the immune system associated with the digestive system (the gut-associated lymphoid tissue or GALT), where it rapidly depletes T cells, which are positive for CD4. Arthos et al. set out to find out why this is. Their first clue was that gp120--the protein on the surface of HIV that binds to CD4--disrupts natural killer cells, which don't express CD4. Some other protein, then, must have been binding to gp120 in these cells, and they identified the binding partner as the alpha-4 beta-7 integrin. They then went on to demonstrate that gp120 binds to alpha-4 beta-7 integrins in T cells.

This finding alone basically answered the question of why HIV thrives in the GALT. Integrins only bind other extracellular proteins when they are activated. As is the case with non-adherent cells in general, the T cell integrins are by default inactive. However, due to the presence of retinoic acid in the GALT, alpha-4 beta-7 integrins there exist primarily in an active state. This is necessary for their biological function, but it also means that these T cells with active alpha-4 beta-7 are more susceptible to HIV.

The really interesting thing about this story is how binding to alpha-4 beta-7 assists HIV infection. HIV cannot use alpha-4 beta-7 directly, but instead when it binds to alpha-4 beta-7 this initiates a signaling cascade inside the target cell that causes the activation of a completely different integrin, alpha-L beta-2. The active alpha-L beta-2 integrin then binds to another protein on the surface of an infected cell. This establishes a virological synapse, allowing much more effective cell-to-cell transmission of the virus.

This is also where my only major gripe about the paper comes in (in addition to the fact that reading a paper where 90% of the data is from flow cytometry is extremely tedious). The paper (and the press coverage) exclusively refers to alpha-L beta-2 as "LFA-1". The integrin scientist in me cringes at this, because "LFA-1" is an outdated terminology that's still unfortunately used in immunology. This detracts from one of the coolest parts of the story (and something that The New York Times missed completely): here you have HIV binding to one integrin and--through outside-in signaling--initiating a signaling pathway that activates a completely different integrin--through inside-out signaling--which aids in transmission of the virus from an infected cell to an uninfected cell. That's interesting, damn it!

The authors do some additional work to identify the binding site on gp120, and it turns out to be the same amino acid sequence that alpha-4 beta-7 uses to bind to fibronectin: leucine-aspartate-valine. As expected, mutating this sequence (aspartate to alanine) greatly decreased binding to alpha-4 beta-7 (but not to CD4) and inhibited viral replication. Interestingly, though, after a few weeks in cell culture, a highly infective strain emerged. This strain had spontaneously mutated back to the wild type sequence (another indication of why HIV is so difficult to combat due to its ability to overcome challenges through spontaneous mutation).

Beyond this, though, the details of the picture remain quite vague. But, this should be a promising area of future research--both in elucidating the chronology of the signaling pathways involved and the structural aspects of the various protein-protein interactions taking place here. Most significantly, this paper identifies alpha-4 beta-7 as a potential target for anti-HIV therapy. The integrins are currently a hot area in drug design, and a couple of drugs have already become available, with mixed success. Many others are in the works, including some targeting alpha-4 beta-7 to treat inflammatory and autoimmune disorders. Based on the work of Arthos et al., it looks like these could have the added benefit of being potential anti-HIV drugs as well.

Arthos, J., Cicala, C., Martinelli, E., Macleod, K., Van Ryk, D., Wei, D., Xiao, Z., Veenstra, T.D., Conrad, T.P., Lempicki, R.A., McLaughlin, S., Pascuccio, M., Gopaul, R., McNally, J., Cruz, C.C., Censoplano, N., Chung, E., Reitano, K.N., Kottilil, S., Goode, D.J., Fauci, A.S. (2008). HIV-1 envelope protein binds to and signals through integrin alpha-4 beta-7, the gut mucosal homing receptor for peripheral T cells. Nature Immunology DOI: 10.1038/ni1566