The role of the N-terminal segment of CCR5 in HIV-1 Env-mediated membrane fusion and the mechanism of virus adaptation to CCR5 lacking this segment

Abstract

Background: HIV-1 envelope glycoprotein (Env) induces membrane fusion as a result of sequential binding to CD4 and chemokine receptors (CCR5 or CXCR4). The critical determinants of CCR5 coreceptor function are the N-terminal domain (Nt) and the second extracellular loop. However, mutations in gp120 adapt HIV-1 to grow on cells expressing the N-terminally truncated CCR5(Delta 18) (Platt et al., J. Virol. 2005, 79: 4357-68).

Results: We have functionally characterized the adapted Env (designated Env(NYP)) using a quantitative cell-cell fusion assay. The rate of fusion with target cells expressing wild-type CCR5 and the resistance to fusion inhibitors was virtually identical for wild-type Env and Env(NYP), implying that the coreceptor affinity had not increased as a result of adaptation. In contrast, Env(NYP)-induced fusion with cells expressing CCR5(Delta 18) occurred at a slower rate and was extremely sensitive to the CCR5 binding inhibitor, Sch-C. Resistance to Sch-C drastically increased after pre-incubation of Env(NYP)- and CCR5(Delta 18)-expressing cells at a temperature that was not permissive to fusion. This indicates that ternary Env(NYP)-CD4-CCR5(Delta 18) complexes accumulate at sub-threshold temperature and that low-affinity interactions with the truncated coreceptor are sufficient for triggering conformational changes in the gp41 of Env(NYP) but not in wild-type Env. We also demonstrated that the ability of CCR5(Delta 18) to support fusion and infection mediated by wild-type Env can be partially reconstituted in the presence of a synthetic sulfated peptide corresponding to the CCR5 Nt. Pre-incubation of wild-type Env- and CCR5(Delta 18)-expressing cells with the sulfated peptide at sub-threshold temperature markedly increased the efficiency of fusion.

Conclusion: We propose that, upon binding the Nt region of CCR5, wild-type Env acquires the ability to productively engage the extracellular loop(s) of CCR5 - an event that triggers gp41 refolding and membrane merger. The adaptive mutations in Env(NYP) enable it to more readily release its hold on gp41, even when it interacts weakly with a severely damaged coreceptor in the absence of the sulfopeptide.

Figure 1

Kinetics of Env(wt)- and Env(NYP)-induced fusion with HeLa-CD4 cells expressing CCR5(wt) (A) or CCR5(Δ18) (B). (A) Fusion with CCR5(wt)-expressing cells induced by Env(wt) and Env(NYP) is shown by filled circles and filled squares, respectively. Fusion after establishing the 18°-TAS (pre-incubation for 2 hr at 18°C) is shown by open circles (Env(wt)) and open squares (Env(NYP)). (B) Fusion with CCR5(Δ18)-expressing cells induced by Env(wt) and by Env(NYP) is shown by semifilled circles and filled triangles, respectively. The kinetics of fusion between Env(NYP)- and CCR5(Δ18)-expressing cells after establishing the 27°-TAS (pre-incubation for 2 hr at 27°C) is shown by open triangles. The experimental points are means ± SE.

Figure 2

Inhibition of fusion with HeLa-CD4/CCR5(wt)-expressing cells by the CCR5 binding inhibitor, Sch-C. (A) The inhibitory activity of Sch-C was measured upon direct co-culture of effector and target cells at 37°C for 2 hr (filled symbols) and after creating TAS (open symbols). TAS was created by pre-incubating cells at 18°C for 2 hr, and fusion was triggered by additional incubation at 37°C for 1.5 hr. Fusion mediated by Env(wt) and Env(NYP) is shown by squares and circles, respectively. (B) Effector cells expressing Env(wt) (squares) or Env(NYP) (circles) were pre-incubated with HeLa-CD4/CCR5(wt) cells at 18°C for varied times, exposed to 1.35 μM Sch-C for 5 min, and warmed to 37°C for 1.5 hr. (C) Following the creation of TAS (18°C, 2 hr), the cells were incubated for additional 5 or 50 min at 18°C with or without 300 nM Sch-C before raising the temperature to 37°C. The resulting fusogenic activity was normalized to the extent of fusion without the inhibitor. The Env(wt)- and Env(NYP)-induced fusion is shown by squares and circles, respectively.

Figure 3

Inhibition of fusion between Env(NYP)- and CD4/CCR5(Δ18)-expressing cells by Sch-C. The extent of fusion upon direct co-incubation at 37°C for 3 hr (filled circles) and after establishing 27°-TAS (open circles) was measured in the presence of different concentrations of Sch-C, as described in the legend to Figure 2. The TAS was created by a 2 hr pre-incubation at 27°C. This temperature was not permissive to fusion with cells expressing a low (R5d18.2, panel A) and relatively high (R5d18.23, panel B) level of the truncated CCR5. Inhibition of fusion by Sch-C added after creating 18°-TAS (2 hr at 18°C) with cells expressing a higher level of CCR5(Δ18) is shown by gray circles.

Figure 4

Inhibition of cell-cell fusion by the C34 peptide. The extent of fusion between Env(wt)-expressing (open squares) or Env(NYP)-expressing (open triangles) cells and CD4/CCR5(wt) cells was measured following a 2 hr-incubation at 37°C in presence of indicated concentrations of C34. When CCR5(Δ18)-expressing cells were used as targets, the C34 was added either at the beginning of co-culture with Env(NYP)-expressing cells (2 hr at 37°C, open circles) or after capturing cells at TAS (pre-incubation at 27°C for 2 hr, filled circles). When added at TAS, the C34 was allowed to bind for 5 min before warming the cells to 37°C.

Figure 5

Env(NYP) is more readily inactivated by treatment with sCD4 than Env(wt). Cells expressing either Env(wt) or Env(NYP) were collected from the culture dish using a non-enzymatic cell dissociation solution and treated with 25 μg/ml sCD4 for 30 min at 37°C. Cells were washed twice to remove free sCD4 and co-incubated with HeLa-CD4/CCR5(wt) cells for 2 hr at 37°C. The results are plotted as percentage of fusion observed for untreated effector cells.

Figure 6

The effect of S22 peptide on virus infectivity. A. Infections of HeLa-CD4 cells expressing CCR5(Δ18) (R5d18.23 cells, 6.6 × 10⁴ coreceptors/cell) were carried out in the absence and the presence of varying concentrations of the S22 peptide (0, 25, 100, and 200 μM). The replication competent wild-type (open circles) or CCR5(Δ18)-adapted (filled circles) JRCSF isolates were tested. Infections performed in the presence of S22 were normalized to those obtained in cells expressing wild-type CCR5 in the absence of peptide. The graph shows a representative experiment performed in duplicate. Error bars are the range. B. Reconstruction of CCR5(Δ18) function by S22 peptide using a single-cycle infectivity assay. Infectivities of viruses pseudotyped with JRCSF wt (open squares) and S298N mutant (filled squares) were determined in the presence of varied concentrations of the sulfopeptide. Titers were normalized as in panel A. Data points represent averages of two experiments performed in duplicate. Error bars are SE. Note that the overall infectivity of pseudoviruses on these cells was much lower than those obtained for replication-competent viruses (panel B vs. panel A).

Figure 7

Rescue of the CCR5(Δ18) coreceptor function by the S22 peptide. (A) Cells expressing Env(wt) (open circles) or Env(NYP) (filled circles) were co-cultured with HeLa-CD4 cells expressing a relatively high density of CCR5(Δ18) (the R5d18.23 line) for 3 hr at 37°C in the presence or in absence of S22. For comparison, cell-cell fusion induced by JRFL Env in the presence of varied concentrations of the S22 peptide is shown by open triangles. (B) Fusion between Env(NYP)- (filled bars) or Env(wt)-expressing (open bars) cells and CD4/CCR5(Δ18) cells after establishing a 27°-TAS. Two μM of C52L was added (first column) or not added (second column) after creating TAS (27°C, 2 hr) prior to warming cells to 37°C and incubating for additional 2 hr. Third column shows fusion observed when 200 μM S22 was present during the last 30 min of a 2 hr pre-incubation at 27°C required to create TAS. Alternatively, a 27°-TAS was created in the presence of 200 μM S22 peptide, and cells were additionally incubated for 2 hr at 37°C (fourth column).