Anti-HIV crotocascarin ω from Kenyan Croton dichogamus

Abstract An anti-HIV methanol-soluble fraction of a 1:1 CH2Cl2:CH3OH extract of twigs of a Kenyan Croton dichogamus yielded seven compounds, the new crotocascarin ω (1), the known β-oplopanone (2), dihydroconiferyl acetate (3), 3’(4’’-hydroxyphenyl)-propyl benzoate (4), lupeol, sitosterol and stigmasterol. Crotocascarin ω (90%) inhibited HIV-1 replication with an IC50 value of 5.3 nM, and the compound was cytotoxic towards MT-4 cells presenting an IC50 value of 84 µM. In silico modelling showed that the anti-HIV activity for compound 1 could be through the HIV-1 protease inhibition. Graphical Abstract


Introduction
The anti-HIV potential of several members of the Croton genus has been documented, and compounds with anti-HIV activity have been previously reported from Croton echinocarpus (Ravanelli et al. 2016), Croton megalobotrys (Tietjen et al. 2016(Tietjen et al. , 2018 and Croton tiglium (El-Mekkawy et al. 2000). In a recent study, we also demonstrated that crude extracts from three Kenyan Croton plants, Croton dichogamus, Croton macrostachys and Croton megalocarpus (Terefe et al. 2021) have anti-HIV potential. In addition, we recently reported anti-HIV potential of crotofolanes from C. megalocarpus (Terefe et al., 2022a). In this study, we report anti-HIV activity of a crotofolane diterpenoid, crotocascarin x, from C. dichogamus.
replication at the lowest IC 50 value of 0.002 ± 0.01 mg/mL (5.3 nM) (Figures S10 and S11). A synergistic effect of the two compounds cannot be discounted.

Molecular docking results for crotocascarin x (1)
To propose a mode-of-action of compound 1, in silico inhibition modelling assays were carried out using HIV reverse transcriptase and HIV protease. Docking studies were performed on HIV-1 RT in complex with known inhibitor nevirapine (PDB ID 1JLB) as well as HIV-1 PR in complex with known antiviral atazanavir (PDB ID: 3EL9) using MOE2015 software ( Figure S12). The predicted free energy of binding obtained for compound 1 against HIV-1 RT was higher (DG À1.38 kcal/mol) compared to the known inhibitor nevirapine (DG À7.60 kcal/mol), due to the lack of the crucial p-H interaction shown by Nevirapine. The main interaction contributing to the binding was the hydrogen bond between hydroxy group at C-1 and Cys181 ( Figure S13). For HIV-PR, the predicted free energy of binding exhibited by compound 1 was DG À6.25 kcal/mol, which was also higher compared to the positive control atazanavir (DG À11.49 kcal/mol). The main interaction contributing to the binding was the hydrogen bonding between one of the epoxide oxygens and the backbone N-H of Asp29(B). Asp29(B) also shows a key hydrogen bonding interaction with one of the amide carbonyls of Atazanavir ( Figure S14). Protease inhibitors usually contain a hydroxyethylene core which mimics the transition state of the hydrolysis step by binding with the catalytic aspartic residues Asp29(B) (Ghosh et al. 2016; Figure S14). Therefore, the residues of HIV protease that interact with the inhibitors, such as Gly-27, Asp-29, Asp-30, and Gly-48, are highly conserved (Lv et al., 2015).The formation of hydrogen bonds with these residues at the catalytic region will impair the enzyme's activity. In biological complexes, hydrogen bonds are the most common directed intermolecular interactions, and they contribute significantly to the specificity of molecular recognition. With these results it can be hypothesised that the potential anti-HIV activity of crotocascarin x could be caused by HIV-1 PR inhibition.
As a consequence, the motifs present in crotocascarin x could be used as a starting point to maximise binding interactions with the S2 subsite containing Asp25, but the structure should be modified to interact with the rest of the active site in order to be considered as a potential HIV-PR inhibitor.

Experimental
The twigs of C. dichogamus were collected and extracted as described in our earlier work . The methanol soluble fraction of a 1:1 CH 2 Cl 2 :CH 3 OH extract of twigs of C. dichogamus was subjected to column chromatography at the Department of Pharmacology and Pharmacognosy, United States International University, Kenya. Commercial silica gel (100 À 200, 200 À 300, and 300 À 400 mesh; Qingdao, China) was used for column chromatography (CC). Sephadex LH-20 (Amersham Biosciences) was also used for CC. All solvents used for column chromatography were of analytical grade (Shanghai Chemical Reagents Co., Ltd.). Analytical TLC using precoated aluminum-backed plates (silica gel 60 F 254 , Merck) was used. Spots were detected on TLC under UV light at 254 or 365 nm, followed by spraying with 1% vanillin-sulfuric acid spray reagent and warming. 1 D and 2 D NMR spectra were recorded in CDCl 3 on a 400 MHz Bruker AVANCE NMR instrument at room temperature. Chemical shifts ( The human T-lymphocytic MT-4 cells (ARP-120) were obtained through the National Institute of Health (NIH) HIV Reagent Program, Division of AIDS, National Institute of Allergy and Infectious Diseases (NIAID), NIH: MT-4 Cells, ARP-120, contributed by Dr. Douglas Richman. Human immunodeficiency virus type 1 (HIV-1) IIIB (also referred to as HTLV-IIIB) was obtained through the NIH HIV Reagent Program, Division of AIDS, NIAID, NIH: Human Immunodeficiency Virus-1 IIIB, ARP-398, contributed by Dr. Robert Gallo. The cytotoxicity test was conducted by measuring cell death caused by the test compounds. The assay was conducted using MTT colorimetric assay as described previously by Mosmann and Pauwels (Mosmann 1983;Pauwels et al. 1988). The MTT assay was based on the reduction of the yellow-colored tetrazolium salt MTT 3-(4,5dimethylthiazol-2-yl)-2,5-diphenyltet diphenyltetrazolium bromide) by NAD(P)Hdependent cellular oxidoreductase enzymes to an insoluble dark-blue colored formazan that can be measured spectrophotometrically (Berridge et al. 2005). More details on the protocol followed for this assay is provided as supplementary material.
The docking studies were performed on MOE2015 software package using HIV-1 reverse transcriptase (HIV-1 RT) in complex with known inhibitor nevirapine (PDB ID: 1JLB) and wild-type HIV-1 protease complexed with known antiviral atazanavir (PDB ID: 3EL9). The proteins were prepared by first removing all water molecules, and in the case of HIV-1 PR also the sulfate and formate ions present in the PDB file. Then, hydrogens were added, and the structures were protonated. For the ligand, energy minimisation was performed using molecular mechanics forcefield MMFF94x. To validate the docking protocol used, known inhibitors nevirapine and atazanavir were removed from their corresponding binding pockets and redocked. The Root Mean Square Deviation (RMSD) value from the known co-crystallised conformation was 0.0551 Å and 1.6038 Å respectively. Compound 1 was docked using MOE2015 with triangle matcher, scoring by London dG, 100 poses as placement method and rigid receptor, GBVI/WSA dG 5 poses as refinement method in both targets. The docking procedure was repeated in three independent runs. The lowest scoring affinity pose in each ligand was used to study the ligand interactions (Rotich et al. 2021).

Conclusions
We conclude that crotocascarin x (1) is a potent anti-HIV compound. This compound is from the rare crotofolane diterpenoid class, which possess a fused 5-, 6-and 7-membered rings, biosynthesised from cembranes via casbane, and lathyrane through cross annular cyclisation (Kawakami et al. 2015). Therefore, there is need to subject the over 38 crotofolanes from the Croton genus to anti-viral assays.