In the Journals

Finding a Chink in the Armor by Investigating the Structure of HIV

A three-dimensional rendering of the structure of trimeric Env bound to 17b is shown. The map was fitted with three copies of the X-ray structures with gp120 shown in purple and 17b in gold. One copy of the gp41 N-terminal helix, shown in cyan was fitted individually into each of the three densities and occupies the central region of the spike, which is essentially a cavity in the unbound state.

The Human Immunodeficiency Virus (HIV) infects and eventually kills CD4-expressing T cells, which are essential for the immune system to function appropriately. Loss of significant numbers of T cells leads to Acquired Immunodeficiency Syndrome (AIDS), a disease that kills over two million people around the world every year. HIV infection depends on two proteins expressed on the virus surface: gp41, which sits in the virus membrane, and gp120, which sits on top of gp41. Three copies, or trimers, of each gp41/gp120 pair make up the envelope glycoprotein, Env. Env coats the virus surface and interacts with its receptor, CD4, and a co-receptor, either CCR5 or CXCR4, on the T cell. Binding to the receptors is thought to cause a structural reorganization of Env, which exposes a fusion peptide that inserts into the T cell membrane and actually forces the virus and host membranes together, initiating an infection. However, the structural details of this process are lacking.

Antibodies directed against different parts of Env may be able to prevent HIV infection, but only if essential structural elements are targeted. Erin Tran, Ph.D., a postdoctoral fellow, and Mario Borgnia, Ph.D., a staff scientist, working with Sriram Subramaniam, Ph.D., in CCR’s Laboratory of Cell Biology, and their colleagues decided to investigate how the structure of Env changes as it binds to its receptor and to HIV-specific antibodies. The conformations of Env in these different situations may suggest new regions for antibody targeting that are likely to block HIV infection.

The researchers began their studies by examining the structure of Env on intact viruses incubated with a soluble version of CD4 to mimic normal receptor binding or with an antibody called 17b, which simulates co-receptor binding. They rapidly froze the samples at -180˚C to preserve the structures and to prevent ice crystals from forming. The frozen samples were then imaged using electron microscopy. Multiple images taken at varying angles were combined to generate the three-dimensional structures. The researchers found that binding to either soluble CD4 or 17b caused the gp120 molecules to rotate outward, forming an opening in the center of the Env structure. This more open conformation is similar to the structure of Env bound to both CD4 and 17b, indicating that 17b binding alone is capable of inducing a conformational change in Env similar to the change induced by the natural CD4 receptor.

The researchers then investigated whether the binding of antibodies that recognize the CD4 receptor site on Env would cause a structural reorganization similar to CD4 binding. Three antibodies, VRC01, VRC02, and VRC03, all target the CD4 site, but instead of promoting the open conformation, all three actually kept Env in a closed state. The researchers suggest that, in addition to physically preventing the CD4 receptor from binding, the broad HIV neutralizing activity of these antibodies may come from their ability to lock Env in this closed conformation. In contrast, a fourth antibody, b12, which also recognizes the CD4 site, was previously shown to require a slight rotation of the gp120 molecules in order to bind. The researchers showed here that the closed Env confirmation caused by binding to the VRC antibodies physically prevents b12 binding. The flexibility required for b12 binding may explain its more limited HIV neutralizing ability; virus strains with gp120 mutations that restrict their movement could not accommodate b12 binding, making them resistant to this antibody.

HIV binding to its co-receptor molecule is known to increase in the presence of CD4. Previous studies showed that binding of soluble CD4 or VRC01 to a soluble gp120 molecule promoted increased binding of the co-receptor mimic, 17b. Because CD4 and VRC01 induced distinct conformations of Env on intact viruses, the researchers wondered whether 17b binding would still be enhanced by VRC01. Incubating the viruses with VRC01 and 17b resulted in two structures: one with only VRC01 bound to Env in the closed conformation and one with both 17b and VRC01 associated with an open Env structure. Since no closed 17b bound structure was detected, these results suggest that if VRC01 binds first, it locks Env in the closed state and prevents 17b binding. When 17b binds first to Env, on the other hand, it induces the open structure, which can still bind VRC01. To test this idea, the researchers incubated viruses with VRC01 before adding 17b. The resulting Env structures were all in the closed conformation with only VRC01 bound. The researchers speculate that, by binding to and retaining the Env closed state, VRC01 is able to neutralize a broad spectrum of HIV strains by blocking receptor and co-receptor binding.

To get an even more detailed picture of Env in the open conformation, the researchers used single particle electron microscopy techniques, which provide higher structural resolution. Incubating soluble Env trimers with 17b again produced the open Env conformation, but this time the researchers were able to detect three helical structures in the center of the Env complex. The researchers attributed these helices to one end of gp41 and suggest that the helices are a part of the HIV fusion peptide, but at a step prior to insertion into a target cell membrane. The researchers also noted that the portion of gp41 that forms each helix is one of the most conserved regions across different HIV strains. This high-resolution structure of Env may provide a new blueprint for building vaccines. Vaccines directed against this structure could stimulate the production of antibodies that will recognize this highly conserved region of Env at a stage where the virus is poised to infect target cells.

Summary Posted: 08/2012

Reference
Structural Mechanism of Trimeric HIV-1 Envelope Glycoprotein Activation. Tran E, Borgnia M, Kuybeda O, Schauder D, Bartesaghi A, Frank G, Sapiro G, Milne J, Subramaniam S. PLoS Pathog. Published online July 12, 2012. Pubmed Link