The HIV-1 Env trimer in HD

Abstract

Understanding the structure of the native HIV-1 envelope spike protein is critical for the development of vaccines and antiviral therapies. In this issue of Structure, Guttman and colleaguesuse Hydrogen/Deuterium(HD) exchangeto provide new insights into the structure of the HIV-1 Envtrimer and enhance our understanding how HIV-1 Env is activated for virus fusion.

The HIV-1 envelope (Env) spike protein mediates entry into cells. Since the Env trimer is the only viral protein on the surface of HIV-1, understanding its structure and activation by the CD4 receptor and coreceptor is vital for the design of vaccines and antiviral therapies. Due to itsheterogeneity, complex glycan shield, and conformational flexibility, the Env trimer has, until recently, eludedstructural characterization at high resolution. However, the last few months have seen an incredible advance in the understanding of the molecular structure of the HIV-1 Env trimer(Julien et al., 2013; Lyumkis et al., 2013). This breakthrough in structural characterization was possible thanks to the generation of soluble, cleaved HIV-1 Envtrimersby a team of researchers from Weill Cornell Medical College and The Scripps Research Institute(Binley et al., 2000; Ringe et al., 2013; Sanders et al., 2013). This is an example of howpersistence over a 15-year periodcan open the door to structural investigations that are of vital interestfor vaccine efforts against HIV-1/AIDS. This effort involved several key milestones: the generation of soluble Envtrimers that lack the hydrophobic transmembrane anchor and the membrane proximal extracellular region (MPER);identification ofthe clade-AHIV-1 isolate BG505 thatnaturally exhibits a more stable trimer; increasing trimer stability by introducing a disulfide bridge (referred to as SOS);optimizing the cleavage efficiency of the precursor into the mature trimer;and stabilizing the native ground state conformation by introducing a I559P mutation in the HR2 region (referred to as IP). This engineeringresulted in the cleaved BG505 SOSIP.664Env trimer. The antigenic features of the BG505 SOSIP.664 trimer indicate that it exhibits a near-native structure (Ringe et al., 2013; Sanders et al., 2013). Other widely usedsoluble trimers, from which the cleavage site has been removed, represent aberrant proteins that are not native trimers (Guttman and Lee, 2013; Ringe et al., 2013). The stability of the engineered BG505 SOSIP.664 Env trimer facilitated determination of its structure at ~5-Å resolution(Julien et al., 2013; Lyumkis et al., 2013), which has providedunprecedented insights into the spatial arrangements of HIV-1 Env, including how the variable V1/V2/V3 loops form the trimer association domain at the tip of the trimer, the gp120/gp41 interface, HR1 and HR2 helices in gp41, and peripheralglycans. Despite this major advance, the resolution is too low to thread the amino acid chain into the gp41 density. Moreover, the molecularunderstandingof how the trimer is activated by receptor CD4, and whether the trimer is conformationally dynamic, remains elusive.

In this issue, MiklosGuttman and colleagues use hydrogen/deuterium exchange (HDX) and oxidative labeling to address these open questions(Guttman et al., 2014). HDX is based on the exchange of hydrogen with the twice-as-heavy deuterium when the protein is switched fromnormalaqueous solvent (H₂O)to deuterated solvent (D₂O). Following the incubation in D₂O, the protein is protealyticallydigested into small peptides, and the molecular mass of each peptideis determined by mass spec. Whereverhydrogen molecules were replaced with deuterium the mass of the peptide increases. HDX assists in characterizing the structure of the HIV-1 trimer by identifying accessible and poorly hydrogen-bonded regions of the timer. Conformational changes in response to ligands such as the receptor CD4 result in altered accessibility. Finally, the method alsoallows insights into conformational dynamics, or the “breathing” of the molecule. By incubating the protein in D₂O for increasing periods of time, deuterium can reach areas of the trimer that are only rarely exposed. Applying HDX to the BG505 SOSIP.664, the authors confirm that the V1/V2/V3 regions have highly ordered segments, which is consistent with their role in trimer association (Julien et al., 2013; Lyumkis et al., 2013). TheN- and C-termini of gp120 that are entirely accessible in the monomeric gp120 are highly protected within the trimer. Thus, they are predicted to play an important role in the gp120/gp41 interaction. The authors then apply HDX to understand the structural changes in response to binding of receptor CD4. Naturally, the CD4-bindingsite becomes more protected upon interaction with CD4 (Figure 1). However, the V1/V2/V3 trimerization domain opens, exposing the V3 loop to allow coreceptor binding. CD4 also alters the gp41 subunit indicating that receptor binding primes the gp41 fusion machine. While for gp120 the CD4 mimetic NBD-556 causes structural changes similar to CD4, it doesnot have the same effect on gp41. Finally, with respect to conformational dynamics, the SOSIP.664 is indeed conformationally dynamic, and not locked in a closed ground-state conformation, as might have been suspected. Thus, this study adds to our emerging knowledge about the structure and dynamics of the HIV-1 Env trimer, and can guide the placement of the amino acid chain into the density of higher resolution structures. Ultimately, the structural aspects of this current study will likely soon be complemented by structures of sufficient resolution to depict gp41, and to characterize the structure of the CD4-bound conformation. However, HDX will continue to provide helpful insightsinto the dynamic nature of HIV-1 Env that other methods cannot easily access.

Figure 1. CD4-induced changes in the BG505 SOSIP.664 as determined by HDX(A).

Ground state structure (Julien et al., 2013; Lyumkis et al., 2013)fitted into the electron density map of (Liu et al., 2008): V1/V2 (yellow) and V3 (green), the bridging sheet (red), HR1 (purple) and gp120 inner domain (black). (B) Hypothetical CD4-bound structureof the trimer with the CD4-bound gp120 structure (Pancera et al., 2010)modeled into the electron density maps of (Liu et al., 2008)with CD4 shown in blue. Figures adapted from Guttman et al., 2014.