Synthesis and biological evaluation of novel antiviral agents as protein–protein interaction inhibitors

Abstract

In continuation of our research efforts toward the identification and optimization for novel inhibitors of interaction between human immunodeficiency virus type 1 integrase and cellular cofactor LEDGF/p75, we designed and synthesized a new series of 4-benzylindole derivatives. Most of the title compounds proved to be able to block this protein–protein interaction (PPI), with a percentage ranging from 30% to 90% at 100 µM. The most promising derivative was compound 10b showing IC₅₀ value of 6.41 µM. The main structure–activity relationships (SAR) are discussed and rationalized by docking studies.

• Docking studies
• HIV-1
• indoles
• LEDGF/p75
• synthesis

Introduction

In the last years, the global community has made great progress in responding to the acquired immunodeficiency syndrome (AIDS) epidemic but there are yet about 40 million individuals living with the human immunodeficiency virus type 1 (HIV-1) and every day thousands more people are newly infected with HIV1. United Nations Member States set clear targets for significantly reducing HIV infection and AIDS death and for scaling up treatment by 2015.

The goal is to strengthen and sustain a valid response to AIDS including the prevention of transmission of HIV as well as providing support and care to those already living with the virus, to prevent the HIV/AIDS epidemic from becoming a severe pandemic. Particularly, it is necessary to ensure that 15 million people can be receiving the antiretroviral therapy, thus reducing the transmission of HIV at least among children.

The overall number of people living with HIV has increased as result of new infections and the beneficial effects of the more widely available highly active anti-retroviral therapy (HAART), which changed AIDS from a rapidly lethal disease into a chronic manageable condition; nevertheless, new efforts are needed.

The currently FDA approved anti-HIV drugs belong to several different groups such as nucleoside reverse transcriptase inhibitors (NRTIs), nucleotide reverse transcriptase inhibitors (NtRTIs), nonnucleoside reverse transcriptase inhibitors (NNRTIs), protease inhibitors (PIs), fusion inhibitors (FIs), co-receptor inhibitors (CRIs) and integrase inhibitors (INIs)2.

Despite HAART remarkably decreases viral load and provides a significant improvement in the life expectancy of HIV/AIDS patients, several factors, including the emergence of drug-resistant viral strains, drug toxicity, the poor ability of patients to adhere to the prescribed therapy and costs, require to refine the current therapy and develop new therapeutic paradigms targeting other steps in the viral replication process3,4.

In this context, HIV-1 integrase (IN), the enzyme which catalyzes the integration of viral cDNA into the host genome, represents a validated target for the development of new drugs against HIV-1 infection5–9. Furthermore, discovery of the keto-enol acid class, often referred as diketoacid (DKA) class, and their analogues and bioisosters as IN inhibitors was an additional advance in the validation of this enzyme as a therapeutically viable antiretroviral drug target, representing major leads in the development of anti-HIV-1 IN drugs.

More recently, it was demonstrated that the multifunctional IN interacts with viral DNA and its key cellular cofactor lens epithelium-derived growth factor (LEDGF/p75) to effectively integrate the reverse transcript into a host cell chromosome10,11.

LEDGF/p75 binds HIV-1 IN via a small IN-binding domain (LEDGF_IBD, residues 347–429) within its C-terminal region. IBD is both necessary and sufficient for interaction with HIV-1 IN12.

Taking into account that the association between IN and the cellular cofactor LEDGF/p75 can be considered as one of the most promising IN interaction for the design of protein–protein disrupting therapeutics13, we have directed our research toward the identification of new small molecules that are able to inhibit the protein–protein interaction (PPI) between the enzyme and its cofactor.

In previous research from our laboratory, it was found that N-benzyl-indoles, bearing diketo acid moiety, were able to exert a specific blockade of the IN/LEDGF interaction14^–18 and among them we found CHI-1043, CHIBA-3000 and some methylbenzyl derivatives such as CHIBA-3017 and CHIBA-3053 (Figure 1) as interesting PPI inhibitors14,19.

Figure 1. Chemical structure of CHI-1043, CHIBA-3000, CHIBA-3017 and CHIBA-3053.

Encouraged by these promising results, which could represent an interesting starting point for the development of novel dual drugs, and with the aim of achieving further information about the IN/LEDGF protein interaction, we focused our efforts to synthesize new indole analogues. Hence, several N-benzylsubstituents were inserted and structural modification of diketo acid portion was performed, replacing this critical moiety in a corresponding closed system which was able to maintain the important polar contacts with IBD. Moreover, docking simulations were carried out to investigate the possible binding mode of title compounds.

Material and methods

Computational studies

Docking studies were performed using GOLD software package20 version 4.1.1 (Gold, UK) and GoldScore scoring function as described in our previous paper21.

The crystal structure of the dimeric IN_CCD complexed with the LEDGF/p75_IBD was retrieved from the RCSB Protein Data Bank (entry code 2B4J)22 and used for our docking simulations. The LEDGF structure was removed and hydrogen atoms were added to the IN protein in Discovery Studio 2.5.523. The standard default settings were used in all calculations. A 20.0 Å radius active site was drawn on the original position of the LEDGF_IBD dipeptide Ile365-Asp366 and automated cavity detection was used. Results differing by less than 0.75 Å in ligand-all atom RMSD were clustered together. Two rotatable bonds of the diketo acid moiety of the ligands were kept fixed during docking calculation. The molecules containing the diketo acid moiety were simulated in the mono-ionized form24.

Chemistry

All commercially available reagents and solvents were used without any further purification. The microwave-assisted reactions were carried out in a CEM Focused Microwave Synthesis System, Model Discover (CEM, Matthews, NC), working at the power necessary for refluxing under atmospheric conditions. Melting points were determined on a BUCHI Melting Point B-545 apparatus (Buchi, Milano, Italy) and are uncorrected. Elemental analyses (C, H, N) were carried out on a Carlo Erba Model 1106 Elemental Analyzer (Carlo Erba, Arese (MI), Italy) and the results are within ±0.4% of the theoretical values. Merck silica gel 60 F₂₅₄ (Darmstadt, Germany) plates were used for analytical TLC; column chromatography was performed on Merck silica gel 60 (230–400 mesh) and Flash Chromatography (FC) on a Biotage SP₁ EXP (Biotage, Uppsala, Sweden). ¹H-NMR spectra were recorded in CDCl₃ with TMS as internal standard or [D₆]DMSO on a Varian Gemini-300 spectrometer (Varia, Gemini, USA). Chemical shifts were expressed in δ (ppm) and coupling constants (J) in hertz (Hz). All the exchangeable protons were confirmed by the addition of D₂O.

Synthesis of 3-acetyl-4-methoxy-1H-indole (2a) and 3-acetyl-4-hydroxy-1H-indole (2b) was carried out following the previously reported procedure14,15,25. Phosphoryl chloride (0.92 mL, 10 mmol) was added to ice cold dimethylacetamide (2.79 mL, 30 mmol) under stirring and cooling in ice. 4-Methoxy-1H-indole 1a (147.18 mg, 1 mmol) or 4-hydroxy-1H-indole 1b (133.15 mg, 1 mmol) was added and the reaction mixture was stirred at room temperature for 12 h, and then poured over ice and basified with a 4N aqueous sodium hydroxide solution. The mixture was extracted with ethyl acetate and dried over Na₂SO₄. After the removal of solvent under reduced pressure, the residue was powdered by treating with diethyl ether and recrystallized from dichloromethane.

For the obtained derivatives 2a and 2b, spectral data are in accordance with the literature14,25.

General procedure for the synthesis of 3-acetyl-1-benzyl-1H-indoles (3a, 4–5a,b)

Using the synthetic procedure previously reported by us14,15,25–27, 3-acetyl-4-methoxy-1H-indole (2a) (189 mg, 1 mmol) or 3-acetyl-4-hydroxy-1H-indole (2b) (159 mg, 0.001 mol) was dissolved in DMF (1 mL) at 0 °C and dry sodium hydride (120 mg, 5 mmol) or K₂CO₃ (138 mg, 1 mmol) was added. After stirring for 2 min, the suitable benzyl bromide (1.5 mmol) was added dropwise. The resulting solution was placed in a cylindrical quartz tube (diam. 2 cm), stirred and irradiated in a microwave oven at 100 W and at continuous temperature (50 °C) for 10 min. A saturated NaHCO₃ solution was added. The reaction mixture was extracted with ethyl acetate (10 mL × 3) and dried over Na₂SO₄. After removal of the solvent under reduced pressure, the crude mixture was purified by flash chromatography using a mixture of cyclohexane:ethyl acetate (60:40) as eluent.

3-Acetyl-1-(4-methyl benzyl)-4-methoxy-1H-indole (3a). Spectral data are in accordance with the literature26.

3-Acetyl-1-(4-trifluoromethyl benzyl)-4-methoxy-1H-indole (4a). m.p. 152–154 °C, yield 85%; ¹H-NMR (δ) 2.70 (s, 3H, CH₃), 3.98 (s, 3H, CH₃), 5.36 (s, 2H, CH₂), 6.71 (d, J = 8.0, 1H, ArH), 6.84 (d, J = 8.0, 1H, ArH), 7.15–7.23 (m, 3H, ArH), 7.57 (d, J = 8.5, 2H, ArH), 7.73 (s, 1H, ArH). Anal. Calcd for C₁₉H₁₆F₃NO₂ – C: 65.70; H: 4.64; N: 4.03. Found – C: 65.87; H: 4.91; N: 4.33.

3-Acetyl-1-(4-trifluoromethyl benzyl)-4-hydroxy-1H-indole (4b). m.p. 190–192 °C, yield 51%; ¹H-NMR (δ) 2.54 (s, 3H, CH₃), 5.38 (s, 2H, CH₂), 6.66–6.75 (m, 2H, ArH), 7.17 (t, J = 7.9, 1H, ArH), 7.25 (d, J = 7.1, 2H, ArH), 7.62 (d, J = 7.4, 2H, ArH), 7.70 (s, 1H, ArH). Anal. Calcd for C₁₈H₁₄F₃NO₂ – C: 64.86; H: 4.23; N: 4.20. Found – C: 64.99; H: 4.54; N: 4.35.

3-Acetyl-1-(4-tert-butyl benzyl)-4-methoxy-1H-indole (5a). m.p. 143–145 °C, yield 73%; ¹H-NMR (δ) 1.29 (s, 9H, CH₃), 2.68 (s, 3H, CH₃), 3.97 (s, 3H, CH₃), 5.25 (s, 2H, CH₂), 6.69 (d, J = 7.9, 1H, ArH), 6.96 (d, J = 7.9, 1H, ArH), 7.09 (d, J = 8.5, 2H, ArH), 7.17 (d, J = 7.9, 1H, ArH), 7.34 (d, J = 8.5, 2H, ArH), 7.72 (s, 1H, ArH). Anal. Calcd for C₂₂H₂₅NO₂–C: 78.77; H: 7.51; N: 4.18. Found – C: 78.95; H: 7.74; N: 4.51.

3-Acetyl-1-(4-tert-butyl benzyl)-4-hydroxy-1H-indole (5b). m.p. 184–186 °C, yield 75%; ¹H-NMR (δ) 1.30 (s, 9H, CH₃), 2.52 (s, 3H, CH₃), 5.27 (s, 2H, CH₂), 6.71–7.67 (m, 8H, ArH), 11.53 (s, 1H, OH). Anal. Calcd for C₂₁H₂₃NO₂ – C: 78.47; H: 7.21; N: 4.36. Found – C: 78.28; H: 7.43; N: 4.21.

General procedure for the synthesis of ethyl 4-[1-benzyl-1H-indol-3-yl]-2-hydroxy-4-oxobut-2-enoates (6a, 7–8a,b).

Adopting the synthetic procedure previously reported by us14,15,19,26,27, a mixture of suitable 3-acetyl-1-benzyl-1H-indole (3a, 4–5a,b) (1 mmol), diethyl oxalate (219 mg, 1.5 mmol) and a catalytic amount of NaOCH₃ was suspended in anhydrous THF (2 mL). The reaction mixture was placed in a cylindrical quartz tube (diam. 2 cm), stirred and irradiated at continuous temperature in a microwave oven for two successive time intervals under the same conditions (250 W, 2 min, 50 °C). The solvent was concentrated under reduced pressure and collected yellow solid was crystallized from ethanol and diethyl ether (1:4).

Ethyl 4-{1-(4-methyl-benzyl)-4-methoxy-1H-indol-3-yl}-2-hydroxy-4-oxobut-2-enoate (6a). Spectral data are in accordance with the literature26.

Ethyl 4-{1-[4-(trifluoromethyl)benzyl]-4-methoxy-1H-indol-3-yl}-2-hydroxy-4-oxobut-2-enoate (7a). m.p. 150 °C dec., yield 89%; ¹H-NMR (δ) 1.25 (t, J = 7.2, 3H, CH₃), 3.86 (s, 3H, CH₃), 4.15 (d, J = 5.5, 2H, CH₂), 5.55 (s, 2H, CH₂), 6.64–7.70 (m, 9H, ArH and CH). Anal. Calcd for C₂₃H₂₀F₃NO₅ – C: 61.75; H: 4.51; N: 3.13. Found – C: 61.99; H: 4.40; N: 3.37.

Ethyl 4-{1-[4-(trifluoromethyl)benzyl]-4-hydroxy-1H-indol-3-yl}-2-hydroxy-4-oxobut-2-enoate (7b). m.p. 263–265 °C, yield 96%; ¹H-NMR (δ) 1.21 (t, J = 7.3, 3H, CH₃), 4.09 (q, J = 7.3, 2H, CH₂), 5.45 (s, 2H, CH₂), 6.25 (s, 1H, OH), 6.64–8.20 (m, 9H, ArH and CH), 13.52 (s, 1H, OH). Anal. Calcd for C₂₂H₁₈F₃NO₅ – C: 60.97; H: 4.19; N: 3.23. Found – C: 61.06; H: 4.33; N: 3.40.

Ethyl 4-{1-[(4-tert-butyl)benzyl]-4-methoxy-1H-indol-3-yl}-2-hydroxy-4-oxobut-2-enoate (8a). m.p. 253 °C dec., yield 96%; ¹H-NMR (δ) 1.20 (s, 9H, CH₃), 1.24 (t, J = 7.3, 3H, CH₃), 3.81 (s, 3H, CH₃), 4.08 (d, J = 7.3, 2H, CH₂), 5.23 (s, 2H, CH₂), 6.54–8.50 (m, 9H, ArH and CH). Anal. Calcd for C₂₅H₂₇NO₅ – C: 71.70; H: 6.71; N: 3.22. Found – C: 71.95; H: 6.89; N: 3.41.

Ethyl 4-{1-[(4-tert-butyl)benzyl]-4-hydroxy-1H-indol-3-yl}-2-hydroxy-4-oxobut-2-enoate (8b). m.p. 219–221 °C, yield 92%; ¹H-NMR (δ) 1.20 (t, J = 7.3, 3H, CH₃), 1.21 (s, 9H, CH₃), 4.09 (q, J = 7.3, 2H, CH₂), 5.27 (s, 2H, CH₂), 5.36–8.13 (m, 10H, ArH, OH and CH), 13.71 (s, 1H, OH). Anal. calcd for C₂₅H₂₇NO₅ – C: 71.24; H: 6.46; N: 3.32. Found – C: 71.33; H: 6.31; N: 3.53.

General procedure for the synthesis of 4-[1-benzyl-1H-indol-3-yl]-2-hydroxy-4-oxobut-2-enoic acids (9–10a,b)

Following the synthetic procedure previously reported by us14,15,19,26,27, the appropriate ethyl 4-[1-benzyl-1H-indol-3-yl]-2-hydroxy-4-oxobut-2-enoate (7–8a,b) (1 mmol) was dissolved in methanol (5 mL) and treated with 2N NaOH (5 mL). The reaction mixture was stirred at room temperature for 1.5 h and then acidified with conc. HCl to give corresponding acids (9–10a,b). The desired products were crystallized from a mixture of ethanol and diethyl ether (1:4).

4-{1-[4-(Trifluoromethyl)benzyl]-4-methoxy-1H-indol-3-yl}-2-hydroxy-4-oxobut-2-enoic acid (9a). m.p. 280 °C dec, yield 72%; ¹H-NMR (δ) 3.90 (s, 3H, CH₃), 5.64 (s, 2H, CH₂), 6.79–7.71 (m, 8H, ArH and CH), 8.57 (s, 1H, ArH). Anal. Calcd for C₂₁H₁₆F₃NO₅ – C: 60.15; H: 3.85; N: 3.34. Found – C: 60.44; H: 4.05; N: 3.56.

4-{1-[4-(Trifluoromethyl)benzyl]-4-hydroxy-1H-indol-3-yl}-2-hydroxy-4-oxobut-2-enoic acid (9b). m.p. 298–300 °C, yield 80%; ¹H-NMR (δ) 5.50 (s, 2H, CH₂), 6.27–8.02 (m, 9H, ArH and CH), 14.56 (s, 1H, OH). Anal. Calcd for C₂₀H₁₄F₃NO₅ – C: 60.44; H: 3.38; N: 3.36. Found – C: 60.61; H: 3.57; N: 3.49.

4-{1-[4-(tert-Butyl)benzyl]-4-methoxy-1H-indol-3-yl}-2-hydroxy-4-oxobut-2-enoic acid (10a). m.p. 215–217 °C, yield 15%; ¹H-NMR (δ) 1.19 (s, 9H, CH₃), 3.75 (s, 3H, CH₃), 5.28 (s, 2H, CH₂), 6.51–9.06 (m, 9H, ArH and CH). Anal. Calcd for C₂₄H₂₅NO₅ – C: 70.74; H: 6.18; N: 3.44. Found – C: 70.98; H: 6.31; N: 3.72.

4-{1-[4-(tert-Butyl)benzyl]-4-hydroxy-1H-indol-3-yl}-2-hydroxy-4-oxobut-2-enoic acid (10b). m.p. 196–198 °C, yield 74%; ¹H-NMR (δ): 1.21 (s, 9H, CH₃), 5.41 (s, 2H, CH₂), 6.55–9.06 (m, 9H, ArH and CH), 11.33 (s, 1H, OH). Anal. Calcd for C₂₃H₂₃NO₅ – C: 70.21; H: 5.89; N: 3.56. Found – C: 70.37; H: 5.61; N: 3.28.

General procedure for the synthesis of 4-[1-benzyl-1H-indol-3-oyl)-3-hydroxyfuran-2(5H)-ones (11a, 12–13a,b)28

A solution of 40% aqueous formaldehyde in water (4 mL) was added to a mixture of appropriate diketoester derivative (6a, 7–8a,b) (1 mmol) in diethyl ether (5 mL). The stirring was then continued until clear layers were formed (usually within 1–2 h). Sometimes, an additional 4 mL of water was added if the reaction was especially thick or when the solid appeared to react slowly. The clear aqueous bottom layer was removed and the organic layer extracted twice with 5 mL of water. The combined aqueous extracts were cooled and acidified with 3 mL of concentrated hydrochloric acid. Then the saturated solution was cooled overnight thus ensuring complete product formation. The resulting solid was collected, dried and recrystallized from ethanol.

4-{1-[4-(Methyl)benzyl]-4-methoxy-1H-indol-3-oyl}-3-hydroxyfuran-2(5H)-one (11a). m.p. 120–122 °C, yield 71%; ¹H-NMR (δ) 2.26 (s, 3H, CH₃), 3.88 (s, 3H, OCH₃), 5.16 (s, 2H, CH₂), 5.44 (s, 2H, CH₂), 6.64–8.21 (m, 8H, ArH). Anal. Calcd for C₂₃H₂₁NO₄ – C, 73.58; H, 5.64; N, 3.73. Found – C, 73.78; H, 5.42; N, 3.91.

4-{1-[4-(Trifluoromethyl)benzyl]-4-methoxy-1H-indol-3-oyl}-3-hydroxyfuran-2(5H)-one (12a). m.p. 170–172 °C, yield 63%; ¹H-NMR (δ): 3.77 (s, 3H, CH₃), 5.05 (s, 2H, CH₂), 5.59 (s, 2H, CH₂), 6.68–7.71 (m, 7H, ArH), 8.21 (s, 1H, ArH). Anal. Calcd for C₂₂H₁₈F₃NO₅ – C: 60.97; H: 4.19; N: 3.23. Found – C: 61.13; H, 4.36; N, 3.45.

4-{1-[4-(Trifluoromethyl)benzyl]-4-hydroxy-1H-indol-3-oyl}-3-hydroxyfuran-2(5H)-one (12b). m.p. 249–251 °C, yield 61%; ¹H-NMR (δ) 5.00 (s, 2H, CH₂), 5.58 (s, 2H, CH₂), 6.44–7.71 (m, 7H, ArH), 9.77 (s, 1H, ArH), 12.79 (s, 1H, OH). Anal. Calcd for C₂₁H₁₄F₃NO₅ – C: 60.44; H: 3.38; N: 3.36. Found – C: 60.63; H, 3.56; N, 3.59.

4-{1-[4-(Tert-butyl)benzyl]-4-methoxy-1H-indol-3-oyl}-3-hydroxyfuran-2(5H)-one (13a). m.p. 179–181 °C, yield 19%; ¹H-NMR (δ) 1.22 (s, 9H, CH₃), 3.76 (s, 3H, CH₃), 5.04 (s, 2H, CH₂), 5.41 (s, 2H, CH₂), 6.68–7.34 (m, 8H, ArH), 8.19 (s, 1H, OH). Anal. Calcd for C₂₄H₂₇NO₅ – C: 71.24; H: 6.46; N: 3.32. Found – C: 71.37; H: 6.58; N: 3.55.

4-{1-[4-(Tert-butyl)benzyl]-4-hydroxy-1H-indol-3-oyl}-3-hydroxyfuran-2(5H)-one (13b). m.p. 212–214 °C, yield 71%; ¹H-NMR (δ) 1.22 (s, 9H, CH₃), 5.12 (s, 2H, CH₂), 5.44 (s, 2H, CH₂), 6.53–7.35 (m, 8H, ArH), 8.81 (s, 1H, OH). Anal. Calcd for C₂₄H₂₃NO₅ – C: 71.10; H: 5.72; N: 3.45. Found – C: 71.32; H: 5.56; N: 3.64.

LEDGF/p75-HIV-1 integrase interaction screening

The AlphaScreen assay was performed as previously described29. Reactions were performed in 25 µL final volume in 384-well Optiwell™ microtiter plates (Perkin-Elmer, Benelux). The reaction buffer contained 25 mM Tris–HCl (pH 7.4), 150 mM NaCl, 1 mM MgCl₂, 0.01% (v/v) Tween-20 and 0.1% (w/v) bovine serum albumin. His6-tagged integrase (300 nM final concentration) was incubated with the compounds at 4 °C for 30 min. The compounds were added in varying concentrations from 1 up to 100 nM. Later, 100 nM of recombinant flag-LEDGF/p75 was added and incubation was extended by another hour at 4 °C. Subsequently, 5 µL of Ni-chelate-coated acceptor beads and 5 µL of anti-flag donor beads were added to a final concentration of 20 µg/mL of both beads. Proteins and beads were incubated at 30 °C for 1 h in order to allow association to occur. Exposure of the reaction to direct light was prevented as much as possible and the emission of light from the acceptor beads was measured in the EnVision plate reader (PerkinElmer, Benelux) and analyzed using the EnVision manager software (Envision, Wichita, KS).

Results and discussion

The most interesting previously reported compounds were benzyl-indole derivatives CHI-1043, CHIBA-3000, CHIBA-3017 and CHIBA-3053 (Figure 1) which proved to be active showing an IC₅₀ value in the low-micromolar range (36.16, 76.44, 67.96 and 3.5 μM, respectively)14,15,19,25.

First, in this study we focused our attention on the benzyl moiety as a relevant lipophilic feature regulating the IN-LEDGF/p75 inhibitory process. Particularly, by means of DRY probe in GRID software, we previously disclosed the hydrophobic region in the LEDGF_IBD binding site, corresponding to IN residue Trp131 (15). This approach is common to other studies and applied also for the analysis of several classical HIV target30,31. Thus, the high inhibitory efficacy of the most efficacious inhibitors could be explained for their ability to map well the DRY area related to the crucial residue Trp131. On this basis, we have explored the effect of bulkier hydrophobic substituents on the benzyl group, such as tert-butyl and trifluoromethyl, that could increase the ability to cover the hydrophobic area on LEDGF_IBD binding site as well as the lipophilicity of our compounds.

In addition, we incorporated the 1,3-diketoacid motif into a closed system characterized by the same chemical functionalities, in order to explore the influence of this chemical manipulation on binding inhibitory effect. Following an earlier reported method, the synthesis of the target compounds was accomplished using the synthetic pathways shown in Scheme 115,27,28.

Scheme 1. Reagents and conditions: (i) POCl₃, CH₃CON(CH₃)₂, RT, 12 h; (ii) benzyl bromide, K₂CO₃ or NaH, DMF, MW: 10 min at continuous temperature (50 °C), 100 W; (iii) diethyl oxalate, dry CH₃ONa, THF, two separate steps under the same conditions MW: 2 min, at continuous temperature (50 °C), 250 W; (iv) 2 N NaOH, MeOH, RT, 1.5 h; (v) CH₂O, Et₂O/H₂O, RT, 2 h.

Suitable 4-substituted indoles (1a,b) were converted into the corresponding 3-acetyl derivatives (2a,b) under Vilsmeier–Haack conditions, in the presence of DMF, acid chloride, such as phosphoryl chloride, and an excess of N.N-dimethylacetamide. The obtained intermediates 2a,b were N-alkylated by reaction with the suitable benzyl bromide. This synthetic step was carried out by the addition of sodium hydride for the methoxy substituted derivatives (a), and potassium carbonate for the hydroxyl substituted derivatives (b). In both cases, the synthesis was performed under microwave (MW) irradiation conditions, thus reducing reaction time, usual thermal reagents degradation and increasing the yields. Reaction of benzyl derivatives (3a and 4–5a,b) with diethyl oxalate under the same conditions: 50 °C for 2 min under 250 W (MW) provided key diketoester intermediates (6a and 7–8a,b), which were converted into the corresponding diketoacids 9–10a,b, by hydrolysis in basic medium, or cyclized in the hydroxyl-furanone derivatives (11a and 12–13a,b), by treatment with formaldehyde aqueous solution. In order to profile inhibitory effects we tested all new synthesized compounds, CHI-1310 and CHI-1164 in AlphaScreen assay (Table 1).

Table 1. Inhibition of IN-LEDGF/p75 interaction of new synthesized compounds, CHI and CHIBA derivatives.

	I					II
	R	R′	%	IC50(μM)		R	R′	%	IC50(μM)
CHI-1043*	OMe	F	75	36.16	CHI-1164	OMe	F	44	–
CHIBA-3000*	OH	F	56	76.44	CHIBA-3004*	OH	F	75	22.28
CHIBA-3017	OMe	CH3	82	–	11a	OMe	CH3	52	–
CHIBA-3053*	OH	CH3	86	67.96	CHIBA-3048*	OH	CH3	81	8.7
9a	OMe	CF3	66	–	12a	OMe	CF3	73	51.3
9b	OH	CF3	30	–	12b	OH	CF3	65	28.28
10a	OMe	C(CH3)3	57	64.00	13a	OMe	C(CH3)3	38	–
10b	OH	C(CH3)3	86	6.41	13b	OH	C(CH3)3	99	23.71

% Inhibition at 100 µM; IC₅₀, concentration required to inhibit the HIV-1 IN-LEDGF/p75 interaction by 50%.

*Data from references14,15,19.

The obtained results showed that derivatives exhibited inhibitory effects at 100 µM concentration with a percentage ranging from 30% to 99%. Analysis of IC₅₀ values pointed out the relevance of the introduction of hydrophobic substituents on benzyl group as previously predicted by our molecular modeling studies15.

In this series of molecules, both for the diketo acids derivatives and 3-hydroxyfuran-2(5H)-ones, generally the presence of hydroxyl group at C4 of benzene fused ring positively influences the inhibitory effects. Overall, the contemporary presence of the alkyl substitution on the benzyl moiety and the hydroxyl group on the indole system provided the most interesting results.

The best activity was displayed when the hydroxyl group of the indole nucleus was associated to the t-butyl substituent of the benzyl moiety (10b IC₅₀ = 6.41 µM). Docking result for the most active compound of the series (10b) into the LEDGF_IBD binding pocket on IN_CCD is reported in Figure 2 and shows nearly the same binding mode of previously reported diketo acid derivatives CHI-1043 and CHIBA-305321. In particular, the diketo acid moiety forms hydrogen bonds with the main chain nitrogens of residues Glu170 and His171 and the side chain of residues Thr174 and Gln95. The fused benzene ring of the indole nucleus projects into the IN hydrophobic pocket, and the N-benzyl substituent presents hydrophobic contacts for the crucial Trp131 residue of chain B. Overall, this new work supports our previous findings on the main structural requirements for IN-LEDGF/p75 inhibition.

Figure 2. Binding mode of compound 10b on IN_CCD. Hydrogen bonds are shown as dotted lines. This figure was produced with pymol³².

Declaration of interest

This work was supported by the European Commission (HEALTH-F3-2008-201032) (THINC project). The authors declare no conflict of interest.