The Novel Cyclophilin Inhibitor CPI-431-32 Concurrently Blocks HCV and HIV-1 Infections via a Similar Mechanism of Action

Abstract

HCV-related liver disease is the main cause of morbidity and mortality of HCV/HIV-1 co-infected patients. Despite the recent advent of anti-HCV direct acting antivirals (DAAs), the treatment of HCV/HIV-1 co-infected patients remains a challenge, as these patients are refractory to most therapies and develop liver fibrosis, cirrhosis and liver cancer more often than HCV mono-infected patients. Until the present study, there was no suitable in vitro assay to test the inhibitory activity of drugs on HCV/HIV-1 co-infection. Here we developed a novel in vitro "co-infection" model where HCV and HIV-1 concurrently replicate in their respective main host target cells--human hepatocytes and CD4+ T-lymphocytes. Using this co-culture model, we demonstrate that cyclophilin inhibitors (CypI), including a novel cyclosporin A (CsA) analog, CPI-431-32, simultaneously inhibits replication of both HCV and HIV-1 when added pre- and post-infection. In contrast, the HIV-1 protease inhibitor nelfinavir or the HCV NS5A inhibitor daclatasvir only blocks the replication of a single virus in the "co-infection" system. CPI-431-32 efficiently inhibits HCV and HIV-1 variants, which are normally resistant to DAAs. CPI-431-32 is slightly, but consistently more efficacious than the most advanced clinically tested CypI--alisporivir (ALV)--at interrupting an established HCV/HIV-1 co-infection. The superior antiviral efficacy of CPI-431-32 over ALV correlates with its higher potency inhibition of cyclophilin A (CypA) isomerase activity and at preventing HCV NS5A-CypA and HIV-1 capsid-CypA interactions known to be vital for replication of the respective viruses. Moreover, we obtained evidence that CPI-431-32 prevents the cloaking of both the HIV-1 and HCV genomes from cellular sensors. Based on these results, CPI-431-32 has the potential, as a single agent or in combination with DAAs, to inhibit both HCV and HIV-1 infections.

Fig 1. CPI-431-32 blocks both HIV-1 and HCV mono-infection.

CD4+ T-lymphocytes (0.5 x 10⁶ cells) (triplicates) were exposed to HIV-1 JR-CSF (1 ng of p24) together with DMSO, the HIV-1 protease inhibitor nelfinavir, the HCV NS5A inhibitor daclatasvir and the two CypI ALV and CPI-431-32. Virus and drugs were washed away after 3 h and HIV-1 replication monitored for a period of 2 weeks every 3 days by quantifying amounts of virus in the cell culture supernatant by p24 ELISA (A). Same as A, except Huh7.5.1 hepatoma cells (0.5 x 10⁶ cells) (triplicates) were exposed to HCV JFH-1 (1 ng of core) and replication monitored by ELISA for HCV core levels in supernatants (B), by RT-qPCR for HCV RNA levels in supernatants (C) or by colony forming assay for HCV infectious units in supernatants (D). Results are representative of two independent experiments.

Fig 2. CPI-431-32 simultaneously blocks HIV-1 and HCV infections.

CD4+ T-lymphocytes (A) and Huh7.5.1 cells (B) (0.5 x 10⁶ cells) (triplicates) were exposed to HIV-1 JR-CSF and HCV JFH-1. Drugs (2 μM) were added 3 h pre-viral exposure (top panels) or 3 days post-viral exposure (bottom panels). Viral replications were monitored by HIV-1 p24 and HCV core ELISAs. C. CD4+ T-lymphocytes (0.5 x 10⁶ cells) (triplicates) were exposed to HIV-1 JR-CSF. Drugs were initially added 3 days post-viral exposure (2 μM) and then daily (0.5 μM). Viral replications were monitored by HIV-1 p24 and HCV core ELISAs. Results are representative of two independent experiments. D. Model for the HIV-1/HCV co-culture system.

Fig 3. CPI-431-32 inhibits the isomerase activity of CypA.

CypA inhibition was assessed with the chymotrypsin-coupled isomerase inhibition assay. Representative data from one experiment is shown. Each symbol is the mean (± SE) of 4 replicate enzyme reactions. Mean IC₅₀ values were 1.8 ± 0.6 nM (mean ± SD), 2.8 ± 0.4 nM, and 16.8 ± 2.3 nM for CPI-431-32, ALV and CsA, respectively.

Fig 4. CPI-431-32 blocks HIV-1 reverse transcription and nuclear import.

TZM cells were infected with DNase-treated NL4-3 HIV-1 together with DMSO or CPI-431-32 (2 μM). Target cell DNA was isolated at the indicated times and was used to detect early (A) and late HIV-1 reverse transcripts (B). TZM cells were infected with NL4.3 in the presence or absence of CPI-431-32 (2 μM). Six hours post-infection, cells were fractionated and cytoplasmic and nuclear extracts analyzed by Western blotting (C). Results are representative of two independent experiments.

Fig 5. CPI-431-32 blocks HCV-mediated double-membrane vesicle (DMV) formation.

T7 polymerase expressing Huh7.5.1 cells were transfected in triplicates with JFH-1 NS3-NS5B plasmid in the presence of DMSO or increasing concentrations of CPI-431-32 and analyzed by EM as well as by Image J and ITEM software for DMVs quantification as described previously [34]. Illustrating EM images of DMVs are presented with an amplification of 25,000x. Data are representative of two independent experiments.