Selection with a peptide fusion inhibitor corresponding to the first heptad repeat of HIV-1 gp41 identifies two genetic pathways conferring cross-resistance to peptide fusion inhibitors corresponding to the first and second heptad repeats (HR1 and HR2) of gp41

Abstract

We generated four HIV-1 cultures that are resistant to a peptide fusion inhibitor corresponding to the first heptad repeat of gp41 in order to study mechanisms of resistance and gain insights into envelope glycoprotein-mediated membrane fusion. Two genetic pathways emerged that were defined by acquisition of a specific mutation in either the first or second heptad repeat region of gp41 (HR1 or the HR2, respectively). Each pathway was enriched in mutations that clustered in either HR2 and V3 or in HR1 and residues in or near CD4 contact sites. The gp41 mutations in both pathways not only accounted for resistance to the selecting HR1 peptide but also conferred cross-resistance to HR2 peptide fusion inhibitors and enhanced the stability of the six-helix bundle formed by the self-assembly of HR1 and HR2. The gp120 mutations alone enhanced fusion but did not appear to directly contribute to resistance. The implications of these findings for resistance mechanisms and regulation of envelope-mediated fusion are discussed.

Fig. 1.

Diagram of domains in gp41 and amino acid sequences of corresponding peptides. HR1 and HR2 correspond to the first and second heptad repeat motifs in gp41, respectively. FP, fusion peptide; TM transmembrane domain; C-tail, cytoplasmic tail. Numbers denote amino acids starting at gp160 according to the nomenclature of the HXB2 reference clone. The sequence of gp41 corresponds to the JR-CSF molecular clone used for these studies a and d indicate positions in the heptad repeats.

Fig. 2.

Summary of resistance pathways that emerged after selection with the N44 peptide. The mutations and their locations in specific regions of gp120 and gp41 subunits of the envelope glycoprotein are shown. E560K and E648K are common early mutations that are used to define pathways I and II, respectively. Mutations contributing to pathways I and II are shown in white and shaded circles, respectively. Mutations shown in shaded rectangles contribute to both pathways. Arrows indicate pathway, not the order of emergence of the mutations.

Fig. 3.

Sensitivity of the envelope glycoprotein (Env) to peptide fusion inhibitors. Wild-type Env (WT), Envs containing all mutations in both gp120 and gp41, or Envs containing all mutations in either gp120 or gp41 from viruses in the N44 resistant cultures 1, 2, 3, and 4 were incorporated into HIV pseudoviruses to assess susceptibility to the inhibition by N44 (A), IZN36 (B), T20 (C), and C34 (D) peptides on U87 CD4⁺ CCR5⁺ cells with high levels of receptor and coreceptor and by N44 on RC4 cells with lower levels of receptor and coreceptor (E). Values shown are the ratios of the IC₅₀s of the resistant Envs compared to the IC₅₀ of the WT Env, shown as means ± SDs from at least three independent experiments. *, P < 0.01 compared to WT (t test). Cltr, culture.

Fig. 4.

Effect of resistance mutations on envelope glycoprotein (Env) fusion and receptor use. Wild-type Env (WT) and Envs containing resistance mutations in both gp120 and gp41 or in gp120 or gp41 alone were tested for Env-mediated cell-cell fusion using target cells expressing various levels of CD4 and CCR5: RC4 cells, low levels of CD4 and CCR5 (A); RC49 cells, low level of CD4 and high level of CCR5 (B); JC55 cells, high level of CD4 and low level of CCR5 (C), and JC53 cells, high levels of CD4 and CCR5 (D). Values are the ratios of the GM of fusion (β-galactosidase activity) of Envs with resistance mutations compared to WT Env, shown as means ± SDs from three independent experiments. *, P < 0.01 compared to WT (t test); Cltr, culture.

Fig. 5.

Thermal denaturation studies of equimolar mixtures of the HR1 (N36) and HR2 (C34) peptides containing wild-type residues or mutations, as indicated in Fig. 1. (A) The α-helical conformation of the complex formed by N36-C34 mixtures (10 μM each in phosphate-buffered saline) was analyzed by CD spectroscopy. (B) The fraction of unfolded N36-C34 mixtures was calculated from CD measurements at 222 nm as a function of temperature. Two independent experiments gave identical results. The N36 and C34 mixtures are as indicated.