From in silico hit to long-acting late-stage preclinical candidate to combat HIV-1 infection

Abstract

The HIV-1 pandemic affecting over 37 million people worldwide continues, with nearly one-half of the infected population on highly active antiretroviral therapy (HAART). Major therapeutic challenges remain because of the emergence of drug-resistant HIV-1 strains, limitations because of safety and toxicity with current HIV-1 drugs, and patient compliance for lifelong, daily treatment regimens. Nonnucleoside reverse transcriptase inhibitors (NNRTIs) that target the viral polymerase have been a key component of the current HIV-1 combination drug regimens; however, these issues hamper them. Thus, the development of novel more effective NNRTIs as anti-HIV-1 agents with fewer long-term liabilities, efficacy on new drug-resistant HIV-1 strains, and less frequent dosing is crucial. Using a computational and structure-based design strategy to guide lead optimization, a 5 µM virtual screening hit was transformed to a series of very potent nanomolar to picomolar catechol diethers. One representative, compound I, was shown to have nanomolar activity in HIV-1-infected T cells, potency on clinically relevant HIV-1 drug-resistant strains, lack of cytotoxicity and off-target effects, and excellent in vivo pharmacokinetic behavior. In this report, we show the feasibility of compound I as a late-stage preclinical candidate by establishing synergistic antiviral activity with existing HIV-1 drugs and clinical candidates and efficacy in HIV-1-infected humanized [human peripheral blood lymphocyte (Hu-PBL)] mice by completely suppressing viral loads and preventing human CD4⁺ T-cell loss. Moreover, a long-acting nanoformulation of compound I [compound I nanoparticle (compound I-NP)] in poly(lactide-coglycolide) (PLGA) was developed that shows sustained maintenance of plasma drug concentrations and drug efficacy for almost 3 weeks after a single dose.

Keywords: HIV-1; NNRTI; drug synergy; humanized mice; nanoparticle.

Fig. 1.

FEP-guided optimization of a docking hit. The representative steps for improving the diphenylmethane substructure (A) through functionalization to (B) further elaboration to a diphenyl ether scaffold (C) followed by additional substituent changes to the potent difluoro-substituted diphenyl ether (D) and finally, the naphthyl catechol phenyl ether compound I are presented here. EC₅₀ values mentioned here were calculated by the MT-2 T-cell assay using the HIV-1 IIIB virus strain as described in Materials and Methods. The results are from three experiments involving triplicate determination.

Fig. 2.

Synergistic inhibition of HIV-1 replication by combinations of compound I with (A) TDF, (B) FTC, (C) EVG, (D) Ed4T, and (E) EFdA analyzed by the isobologram analysis. The red dotted lines from the upper left corners to the lower right corners indicate that the two drugs are additive, and the curves below the lines indicate that drugs are synergistic. The results are from three experiments involving triplicate determination.

Fig. 3.

Synergistic inhibition of HIV-1 replication by combinations of compound I with (A) TDF, (B) FTC, (C) EVG, (D) Ed4T, and (E) EFdA analyzed by MacSynergy II 3D plots. The peaks above the horizontal plane indicate that the drugs are synergistic. The results are from three experiments involving triplicate determination.

Fig. 4.

Pharmacokinetics of free compound I and compound I-NP in BALB/c mice. (A) Comparison of serum levels of compound I after i.p. administration of 20 mg/kg free (black circles) and NP (red circles) compound I. The Inset depicts the enlarged view of the low serum concentrations observed for compound I-NP (red circle) and compound I (black circle). (B) Comparison of serum levels of compound I after i.p. administration of 100 mg/kg free (black circles) and 190 mg/kg NP (blue circles) compound I. The Inset depicts the enlarged view of the low serum concentrations observed for compound I-NP (blue circle) and compound I (black circle). The blood samples were collected at different intervals and analyzed as described in Materials and Methods. (C) HIV-1 inhibition by serum containing free compound I in TZM-bl cells. The values are mean ± SD from three different experiments involving triplicate measurements. (D) SEM image of compound I-NP. (Scale bar, 1 µM.) (E) HIV-1 inhibition by serum containing free compound I (black) and compound I-NP (blue) in TZM-bl cells. Acc. V, accelerating voltage; Det, detector; Exp, exposure number; LOD, limit of detection of HPLC used to quantify serum concentrations of compound I/compound I-NP; Magn, magnification; Spot, technical term that describes the width of the SEM beam, usually defined in reference to the center point of the beam (reference point and value may change based on specific SEM unit); WD, working distance.

Fig. 5.

Experimental protocols used in efficacy studies to determine the antiviral activity of free compound I and compound I-NP in HIV-1–infected NSG-Hu-PBL mice. (A) Treatment paradigm to determine preinfection antiviral activity of 100 mg/kg free compound I and 190 mg/kg compound I-NP administered i.p. (B) Serum drug levels in HIV-1–infected NSG-Hu-PBL mice treated with 100 mg/kg of free compound I daily and 190 mg/kg of compound I-NP on day −1. Mice receiving free compound I were further divided into two groups. Mice in group I (green circles) received 100 mg/kg of free compound I daily until day 32, whereas mice in group II (red circles) stopped receiving compound I on day 19. For both the groups, the serum drug levels were determined on days 8, 16, 25, and 32. After i.p. administration of 190 mg/kg of compound I-NP (blue circles) on day −1, serum levels of compound I were determined on days 8, 16, 25, and 32. Data points represent ±SD.

Fig. 6.

Antiretroviral activities of free compound I and compound I-NP in NSG Hu PBL mice infected intraperitoneally with HIV-1 (JR-CSF) on day 0, 24 h after drug administration. One hundred milligrams per kilogram of free compound I was administered daily (days -1 to 32) by i.p. route, whereas 190 mg/kg of compound I-NP was administered by i.p. route on day -1. (A) FACS analyses compared the ratio of CD4⁺/CD8⁺ cells in untreated (Top), compound I-NP–treated (Middle), and free compound I–treated (Bottom) mice. (B) Quantification of FACS analysis shows the percentage of human CD4⁺ among total CD3⁺ T cells of peripheral blood. (C) Total PVL determined by RT-PCR in HIV-1_JR-CSF–infected NSG mice. Data points represent ±SD. LOD, limit of detection of assay used to quantify PVLs.