The primary cause of acquired immunodeficiency syndrome (AIDS) is the human immunodeficiency virus (HIV), which infects target white blood cells by connecting (fusion) the membranes of the host cell and HIV, causing the viral nucleocapsid to deposit in the cytoplasm. HIV's Gp41, a single-pass integral viral membrane protein with a 150-residue soluble domain outside the virus, is involved in fusion catalysis. This soluble ectodomain (SE) adopts a hairpin structure that contains N-helix-turn-C-helix, and two helices align themselves antiparallelly and there is van der Waals contact between them. The final fusion structure also consists of three hairpins joining together to form a six-helix bundle. There are competing hypotheses about the catalytic role in HIV/cell fusion of the SE hairpin structure. One hypothesis is that the hairpin is a “post-fusion” structure with no role in catalysis, whereas another hypothesis postulates an important role for the hairpin in catalysis which is associated with binding to the membrane. The latter hypothesis is supported by rapid vesicle fusion induced by the hairpin under some conditions. Further, mixtures of WT and V513E proteins of a large HIV gp41 ectodomain construct with final trimer of-hairpins structure exhibit V513E (gp41 V2E in absence of gp120)-dominant reduction of vesicle fusion that is quantitatively similar to that for gp160 (glycoprotein 160, which is a precursor protein that gets cleaved into two smaller proteins: gp120 and gp41) fusion and infection. Vesicle fusion and helicity were measured for FP_HM (fusion peptide (FP)+hairpin (H)+membrane proximal external region (MPER)) using trimers with different fractions (f’s) of WT and V2E proteins. Correlation between low helicity vs. fusion of V2E was also found in this study. Reductions in FP_HM fusion and helicity vs. fV2E were quantitatively-similar to those for gp160-mediated fusion and infection. Global fitting of all V2E data supports 6 WT gp41 (2 trimers) required for fusion. These data are understood by a model in which the ~25 kcal/mol free energy for initial membrane apposition is compensated by the thermostable hairpin between the Fp in target membrane and Mper/transmembrane domain in virus membrane. The data support a structural model for V2E dominance with a membrane-bound Fp with antiparallel β sheet and interleaved strands from the two trimers. This increased distance is the hypothesized reason for reduced fusion and infection. Role of hairpin and correlation between helicity vs. fusion further studied through characterization and fusion of G10V_FPHM and L9R_FPHM, respectively. Results established the less helicity in V2E/L9R than WT vs. less fusion in V2E/L9R than WT and similar helicity in WT/G10V vs. similar fusion in WT/G10V. Position of hairpin in membrane could be better understanding of membrane apposition (viral and host membrane) during fusion because deep SE insertion indicates the close membrane apposition and surface insertion predicts far membrane apposition. Moreover, a hairpin contribution to catalysis of membrane fusion is most straightforwardly understood if there is deep insertion as the locations of the surrounding lipid molecules in the membrane would likely be very different than their energetically stable positions in an unperturbed bilayer. We used either 13C-glucose or 13C-glycerol to express a large soluble ectodomain in a minimum medium, with a dilution of the 13C labeling accomplished by mixing with unlabeled glucose or by selectively labeled glycerol. The proximity of protein with 13C-labeled and lipid with 2H labeled positioned either in middle /periphery/ in between of middle and periphery of the membrane were measured using the rotational-echo double-resonance (REDOR) solid-state NMR technique. The production of 13C proteins used a variety of carbon sources, and as a result, various labeled carbon peak assignments on the side chains of amino acids were observed in NMR. There was a significant REDOR dephasing for samples from different carbon sources, which shows that the soluble ectodomain has mostly peripheral location in the membrane.
Read
