Abstract Body

Background:

Second-generation integrase (IN) strand transfer inhibitors (INSTIs) are highly potent antiretroviral compounds that exhibit a high genetic barrier to resistance. Recent clinical studies concluded that some INSTI-treated individuals experience virological failure in the absence of resistance mutations in IN. The aim of the study is to elucidate INSTI resistance mechanisms and pathways.

Methods:

One-year passaging of HIV-1 was conducted using the SupT1 T-cell line and primary PBMCs with an escalating concentration of the INSTI dolutegravir (DTG). We evaluated the impact of the selected mutations on replication kinetics and viral infection through cell-free virion and cell-cell contact and performed an array of biochemical and structural analyses on the selected mutants.

Results:

HIV-1 became resistant to DTG by sequentially acquiring mutations in Env, Gag-nucleocapsid (NC), and, occasionally, IN. By cloning env from the DTG-treated viruses selected in the SupT1 T-cell line or PBMC, we obtained heavily mutated Env clones, 7XEnv and WD-3, respectively. Both Env mutants exhibit faster-than-WT replication in spreading infection. 7XEnv exhibits resistance to multiple classes of antiretrovirals, with the fold resistance being ~2-logs higher for INSTIs than for other classes of drugs. WD-3 confers 5-fold resistance to DTG in PBMC. Viral transmission of 7XEnv through cell-cell contact is more efficient than that of WT. In contrast, WD-3 exhibits more efficient cell-free infection than WT. These results suggest that the selected Env mutations confer resistance to INSTIs by increasing infection capacity through cell-cell transmission or cell-free viral infection. Viral infection over a range of multiplicities of infection (MOI) revealed that INSTIs are more readily overwhelmed by high MOI than other classes of drugs, leading to high-level resistance to INSTIs. The NC mutations selected with DTG conferred modest (3-5-fold) resistance to INSTIs. Significantly, the NC mutations do not affect cell-free infectivity but accelerate the kinetics of early post-entry events, suggesting that they may limit the window of opportunity for INSTIs to bind intasomes and block integration.

Conclusions:

These findings demonstrate that multiple regions in the HIV-1 genome – Env, NC, and IN – collectively contribute to INSTI resistance. The results provide clues to understanding high-level resistance to INSTIs and support the need for genotypic analysis outside of IN in individuals on INSTI-containing regimens.