Background: The CC-chemokine receptor CCR5 is a promising target for HIV-1 entry inhibitors. We examined the therapeutic potential of the anti-CCR5 monoclonal antibody PRO140 in the hu-PBL-SCID mouse model of HIV-1 infection.

Methods: SCID-mice were reconstituted with PBMCs from a CCR5 wild-type donor and infected with the primary HIV-1 R5 isolate JR-CSF. When viral steady state was reached, the animals were treated with PRO140 and monitored for viral burden using quantitative RT-PCR. Potential escape mutants were characterized.

Results: Repeated administration of murine PRO140 led to prolonged control of viral replication without viral escape. A humanized version of PRO140 was further tested in the model and was also effective in controlling viral replication upon repeat dosing. In contrast to multidose therapy, a single 1-mg injection of murine PRO140 reduced plasma HIV-1 RNA to undetectable levels in each of 12 mice treated. In three cases, viral replication rebounded due to the emergence of escape mutants. One variant had switched to CXCR4 usage. Sequencing of the envelope gene revealed two point mutations and one 4 amino acid deletion in the V3 loop. Furthermore, sequencing of gp160 from the viral swarm at different time points after treatment revealed the sequential selection of V3 loop variants. Site directed mutagenesis was used to study the role of each mutation in co-receptor usage. The analysis demonstrated that the escape variant had switched to X4 via intermediate dual tropic R5X4 viruses. Importantly, the final X4 variant displayed a remarkably high sensitivity to antibodies, patient sera and viral entry inhibitors in vitro, which was similarly mapped to the V3 loop mutations. Using FACS analysis, we demonstrated that the increase of sensitivity to neutralization was due to an increased affinity of Ab for the escape variant gp120/gp41 trimer.

Conclusions: Our study shows that the anti-CCR5 Ab PRO140 is effective at controlling HIV-1 infection in the hu-PBL-SCID mouse model. However, following single injection, escape may occur, notably by switch to CXCR4. The greater sensitivity of the escape variant to neutralization suggests that the occurrence of such X4 variants may be limited in infected individuals. The findings underscore the potential of anti-CCR5 therapy but also suggest that coreceptor usage should be closely monitored.

Keywords: Antibody anti-CCR5; Hu-PBL-SCID mouse model; HIV-1 escape mutant