Researchers create unusually stable and homogenous version of Env trimer

A research team led by Bing Chen, a structural biologist at Harvard Medical School, has created a lab-made version of the HIV Envelope trimer—from which the spikes on the outer surface of the virus are made—that is similar to the naturally occurring trimer. It is also more stable and more homogenous than previous lab-made versions, perhaps enough so that it could be used as an immunogen in a candidate vaccine for evaluation in human trials (Proc. Natl. Acad. Sci. 109, 12111, 2012). “I think we have made a truly stable and homogeneous Envelope trimer preparation,” Chen says, adding that this is the first time this has been achieved. 

The natural HIV Env trimer is ordinarily cleaved from a precursor protein into two parts, with a gp120 portion ultimately appearing on the outside of the viral membrane, and gp41 traversing the membrane from the inside. These proteins are only held together by weak, non-covalent forces. The artificial trimer constructed by Chen and his team, on the other hand, is made directly from the uncleaved precursor. To make the protein soluble in water, the researchers removed the innermost parts of the intact precursor protein that span the viral membrane and extend inside the virus. The resulting gp140 trimer contained all parts of the natural HIV Envelope trimer that lie outside the membrane of the virus—corresponding to all of gp120 and part of gp41. 

To ensure that the resulting protein had sugar groups and other modifications on its surface that reflect those of the native Env trimer, the researchers grew the protein in a human cell line called 293T. Finally, they purified their protein preparation to make sure it only contained trimers and not dimers or monomers.  

Still, Chen says, many of the trimers made this way fell apart; his team had to make gp140 proteins from about 20 different HIV strains to find two—from a clade A and a clade C strain of HIV-1—that formed trimers that were stable. To Chen, this suggests that it is primarily differences in the protein sequence that somehow make some lab-made Env trimers more stable than others. 

Chen and his team also showed that the properties of their gp140 trimer protein preparation were biologically closer to the native Env trimer than they were to those of gp120 monomers. For example, the new gp140 trimers only bound neutralizing antibodies specific to the CD4 binding site, such as the broadly neutralizing antibody (bNAb) VRC01. But they did not bind non-neutralizing antibodies of similar specificity, such as b6. In contrast, gp120 monomers made from the same sequence bound to both the bNAb VRC01 and the non-neutralizing b6 antibody. “We found our trimers only bind the neutralizing antibodies,” Chen says. “That indicates that the epitope for this non-neutralizing antibody is occluded on the native trimers. The property of our trimer is really consistent with the native, functional trimer.” 

One important question is whether the new gp140 trimer preparation is not only stable enough but also a better immunogen than monomeric gp120, the version of Envelope that has been used in many human clinical trials of AIDS vaccine candidates, including RV144, the first human trial that demonstrated any measure of protection from HIV. Chen and his colleagues found that the answer seems to be yes, at least in guinea pigs: Antibodies induced by six intramuscular immunizations with the new gp140 trimers were up to 15-fold more potent at neutralizing HIV-1 strains that are considered either easy or moderately difficult to neutralize than those induced by six immunizations with monomeric gp120 proteins made from the same gene sequence. 

This is an important advance, Chen says. AIDS vaccine researchers have been using monomeric gp120 proteins in human clinical trials because they didn’t have much evidence that gp140 trimers were more immunogenic than gp120 monomers in small animal experiments. The likely reason for that, he adds, is that previously evaluated gp140 trimer preparations were not very stable, so that they often behaved like three gp120 monomers only held together by a common stem. “All kinds of previous gp140 preparations actually have gp120-like characteristics,” Chen says. 

Researchers are now planning to test the safety and immunogenicity of the new gp140 trimer preparation compared with gp120 monomers in humans. To this end, Chen says, Dan Barouch, one of the coauthors of the study, has already asked the Dutch biopharmaceutical company Crucell to manufacture a version of Bing’s clade C-derived gp140 trimer that is pure enough for use in humans to be able to do small-scale clinical trials. In addition, Nelson Michael, director of the US Military HIV Research Program, says he plans to collaborate with Barouch to test whether immunization with this trimer is safe and whether it improves the immunogenicity of an adenovirus serotype 26 (Ad26)/modified vaccinia Ankara (MVA) vaccine regimen in nonhuman primates and humans. 

Michael, who was not involved in the recent study by Chen and colleagues, says the new gp140 trimer preparation is the purest Envelope trimer preparation and therefore closer to the natural Envelope trimer than any other he is aware of. But he cautions that it remains to be seen whether that translates into better immunogenicity. “It is more physiologic, [but] that does not necessarily portend that it’s going to be a better vaccine,” Michael says. 

The new gp140 trimer preparation also appears to be homogenous and stable enough to be suitable for structural studies, says Chen. Determining the structure of the native Env trimer, or a structure close to it, is a major goal of HIV researchers. But the structural heterogeneity and instability of previous Env trimer preparations have impeded progress in that endeavor. Chen now plans to see if his preparation can overcome those obstacles and if he can create crystals to determine its structure by X-ray crystallography. “It’s not gonna be easy,” he concedes.