Studies towards the Design and Synthesis of Novel 1,5-Diaryl-1H-imidazole-4-carboxylic Acids and 1,5-Diaryl-1H-imidazole-4-carbohydrazides as Host LEDGF/p75 and HIV-1 Integrase Interaction Inhibitors

Two targeted sets of novel 1,5-diaryl-1H-imidazole-4-carboxylic acids 10 and carbohydrazides 11 were designed and synthesized from their corresponding ester intermediates 17, which were prepared via cycloaddition of ethyl isocyanoacetate 16 and diarylimidoyl chlorides 15. Evaluation of these new target scaffolds in the AlphaScreenTM HIV-1 IN-LEDGF/p75 inhibition assay identified seventeen compounds exceeding the pre-defined 50% inhibitory threshold at 100 µM concentration. Further evaluation of these compounds in the HIV-1 IN strand transfer assay at 100 μM showed that none of the compounds (with the exception of 10a, 10l, and 11k, with marginal inhibitory percentages) were actively bound to the active site, indicating that they are selectively binding to the LEDGF/p75-binding pocket. In a cell-based HIV-1 antiviral assay, compounds 11a, 11b, 11g, and 11h exhibited moderate antiviral percentage inhibition of 33–45% with cytotoxicity (CC50) values of >200 µM, 158.4 µM, >200 µM, and 50.4 µM, respectively. The antiviral inhibitory activity displayed by 11h was attributed to its toxicity. Upon further validation of their ability to induce multimerization in a Western blot gel assay, compounds 11a, 11b, and 11h appeared to increase higher-order forms of IN.


Introduction
The human immunodeficiency virus type 1 (HIV-1), the causative agent of acquired immune deficiency syndrome (AIDS), is one of the world's greatest health and development challenges. Despite chemotherapeutic interventions for the virus being developed through active research, the emergence of highly drug-resistant strains against the current approved Food and Drug Administration (FDA) antiretroviral drugs takes a significant toll on the global fight against HIV-1 [1][2][3][4][5][6][7]. Thus, there has been an ongoing search for new treatment options targeting novel steps in the HIV-1 replication cycle. Targeting the active host protein and HIV-1 protein interactions has become a viable therapeutic approach. A typical example is a lens epithelium-derived growth factor (LEDGF/p75) and HIV-1 integrase (IN) protein-protein interaction [8][9][10]. IN is one of three multifaceted essential enzymes required for HIV-1 replication. This 32 kDa protein consists of 288 amino acids and is composed of the catalytic core, N-terminal and C-terminal functional domains [11][12][13][14]. The main protein function is to allow reliable integration of viral DNA genomes into host cells [11,15]. There are currently only four IN inhibitors on the market that have been clinically tested and approved by the FDA, including elvitegravir, dolutegravir, raltegravir, and bictegravir [16][17][18][19]. These inhibitors are referred to as strand transfer inhibitors (INSTIs) and share the identical mechanistic mode of action of blocking the strand transfer step through binding to the active site of viral IN [14][15][16][17]. Despite the success of highly active antiretroviral therapy (HAART), a new generation of drugs is needed to overcome multi-drug resistance, enhance adherence, and lower toxicities [1][2][3][4][5][6][7].
A survey of the literature reveals that a few synthetic methods for constructing 1,5disubstituted-imidazole-4-carboxylates using α-isocyanoacetate synthons have been reported. One of the first syntheses was described in 1978 by Hunt and co-workers [38], where the condensation of the potassium salt of ethyl isocyanoacetate with isothioureas in hexamethylphosphoric triamide (HMPT) at room temperature catalyzed by copper (I) chloride (CuCl) afforded a mixture of disubstituted imidazole-4-carboxylate esters in various yields as presented in Scheme 1a. Another approach was reported in 1994 by Nunami and co-workers [39] in which key intermediate 3-bromo-2-isocyanoacrylates (BICA) were easily derived from isocyanoacetate esters, followed by reaction with a variety of primary amines in the presence of triethylamine (Et 3 N) in N,N-dimethylformamide (DMF) as solvent (Scheme 1b). Furthermore, in 2011, Meng and co-workers [33] demonstrated a strategy to introduce a range of 1,5-disubstituted-1H-imidazole-4-carboxylate ester derivatives as intermediates towards the synthesis of chromeno [3,4-d]imidazol-4-one by using readily available ethyl isocyanoacetate, primary aliphatic amines, and an aldehyde in the presence of copper iodide (CuI) and pyridine in methanol (MeOH) heated at 50 • C (Scheme 1c). Aguilar and co-workers [40] reported a one-pot two-step strategy for the preparation of a disubstituted imidazole-4-carboxylate ester. This procedure involves first coupling 4-chloroaniline with 4-iodobenzoyl chloride using 1-ethyl-(3-dimethylaminopropyl)carbodiimide (EDCI) to produce the amide. The amide is then treated with chlorophosphate to generate unstable iminophosphates in situ, followed by condensation under basic conditions with ethyl isocyanoacetate to produce the desired compound (Scheme 1d). This procedure only reported one example and the approach required strictly anhydrous conditions. Additionally, Cai and co-workers [41] in 2019 reported an annulation between α-isocyanoacetates and ketenimines in the presence of a base, potassium tert-butoxide (t-BuOK), in DMF at 40 • C as an efficient and straightforward synthesis of disubstituted imidazole-4-carboxylate ester derivatives in good yields (Scheme 1e). Additionally, Hu and co-workers [42] in 2020 reported a new method for preparing 1,5-disubstituted imidazole-4-carboxylate esters by t-BuOK-mediated annulation of ester isocyanoacetates with trifluoroacetimidoyl chlorides (Scheme 1f). The methodology includes mild reaction conditions and uses readily available reagents. However, only two examples were reported utilizing ester isocyanoacetate synthons in this approach. More recently, Lv and co-workers [43] in 2021 described the construction of these compounds in good yields by reacting α-isocyanoacetates with nitrones catalyzed by silver trifluoromethanesulfonate (AgOTf) in the presence of potassium trifluo-romethanesulfonate (KOTf) in toluene (PhMe) as a solvent under a nitrogen atmosphere (Scheme 1g). Scheme 1. Synthesis of 1,5-disubstituted-imidazole-4-carboxylate esters using α-isocyanoacetates.
As part of our ongoing interest in the design of possible inhibitors of the HIV-1 IN and LEDGF/p75 protein-protein interaction [30], we directed our attention to the synthesis of 1,5-disubstitued-1H-imidazole-4-carboxylate esters. Herein we describe a novel and efficient approach for the synthesis of diaryl-imidazole-4-carboxylate esters via cycloaddition of an α-isocyanoacetate with stable N-aryl-benzimidoyl chloride in-termediates (Scheme 1h). None of the previously reported synthetic methods to these compounds [33,[40][41][42][43] make use of diarylimidoyl chlorides and isocyanoacetate esters. The carboxylate esters prepared using this new methodology were subsequently converted into the corresponding 1,5-diaryl-1H-imidazole-4-carboxylic acids and 1,5-diaryl-1H-imidazole-4-carbohydrazides, which were then evaluated for their ability to bind to the novel HIV-1 IN dimer interface pocket.

Testing the Hypothesis In Silico
In order to test our hypothesis and provide some proof of concept for preparation of the proposed analogues (Figure 2), target candidates 10a and 11a were docked into the LEDGF/p75-binding pocket of HIV-1 integrase to gain a deeper understanding of the possible binding interactions. The X-ray crystal structure of HIV-1 IN dimer (PDB code: 4NYF) in complex with the small molecule, BI 224436 (2), was used for a better understanding of binding interactions of the tested ligands [27]. The model developed suggested a clear difference in the positioning of the two ligands within the binding site. The carboxylic acid moiety of compound 10a was shown to form hydrogen bonds with critical histidine (His) 171, glutamic acid (Glu) 170, and threonine (Thr) 174 amino acid residues of HIV-1 IN, while the imidazole motif is predicted to exhibit hydrogen bonding with the glutamic acid 170 residue as presented in Figure 3a,b. Moreover, the 4-fluorophenyl moiety is shown to be positioned towards hydrophobic alanine (Ala) 128 and alanine 129 residues while the 4-bromophenyl moiety is positioned towards the hydrophobic pockets occupied by threonine 125. On the other hand, the carbohydrazide motif of 11a is predicted to interact with threonine 125 and glycine (Gln) 95 amino acid residues. In addition, the 4-fluorophenyl moiety is projected towards the Glycine 168 residue while the 4-bromophenyl moiety projects towards histidine 171 and glutamic acid 170 as shown in Figure 3c,d. The modelling data suggest that these scaffolds interact with critical amino acid residues in the LEDGF/p75 binding site of HIV-1 integrase.

Synthesis of Key 1,5-Diaryl-1H-imidazole-4-carboxylate Esters
Possible synthetic pathways were considered for the synthesis of the desired target 1,5diaryl-1H-imidazole-4-carboxylic acids 10 and 1,5-diaryl-1H-imidazole-4-carbohydrazides 11. The approach adopted was to start with reaction of an acyl chloride 12 with an aniline derivative 13 to give the corresponding amide 14 that could be readily converted into the corresponding imidoyl chloride 15 (Scheme 2). The key step in the synthesis was the cycloaddition reaction between ethyl isocyanoacetate 16 and a range of imidoyl chlorides to construct a range of key 1,5-diaryl-1H-imidazole-4-carboxylate ester intermediates 17. Facile conversion of the ester intermediates into their corresponding carboxylic acid and carbohydrazide derivatives ultimately gave rise to the desired targets 10 and 11, respectively. Judicial choice of differently substituted starting materials using this approach enabled a detailed and thorough structure-activity relationship (SAR) analysis of the desired final compounds to be undertaken. The route was initially tested starting from commercially available 4fluorobenzoyl chloride 12a and 4-bromoaniline 13a. A mixture of 4-fluorobenzoyl chloride 12a and 4-bromoaniline 13a in ethyl acetate in the presence of Et 3 N was reacted at room temperature for 16 h to afford N-(4-bromophenyl)-4-fluorobenzamide 14a in excellent yield, which was utilized directly in the next step without purification [51][52][53]. The amide 14a was then subjected to a neat reaction by heating under reflux with excess SOCl 2 for 8 h to give N-(4-bromophenyl)-4-fluorobenzimidoyl chloride 15a. This imidoyl chloride was isolated in quantitative yield and was subsequently used without purification [51,[54][55][56]. Compound 15a was then subjected to reaction with ethyl isocyanoacetate 16. Specifically, a mixture of stoichiometric amounts of ethyl isocyanoacetate 16, 1,8-diazabicyclo [5.4.0]undec-7-ene (DBU) in dry THF under argon was cooled to −78 • C, followed by subsequent addition of 15a. The reaction was allowed to warm to room temperature and stirred for 48 h, affording the desired novel ethyl 1-(4-bromophenyl)-5-(4-fluorophenyl)-1H-imidazole-4-carboxylate 17a in 64% yield, after column chromatographic purification.

Synthesis of Key 1,5-Diaryl-1H-imidazole-4-carboxylate Esters
Possible synthetic pathways were considered for the synthesis of the desired target 1,5-diaryl-1H-imidazole-4-carboxylic acids 10 and 1,5-diaryl-1H-imidazole-4-carbohydrazides 11. The approach adopted was to start with reaction of an acyl chloride 12 with an aniline derivative 13 to give the corresponding amide 14 that could be readily converted into the corresponding imidoyl chloride 15 (Scheme 2). The key step in the synthesis was the cycloaddition reaction between ethyl isocyanoacetate 16 and a range of imidoyl chlorides to construct a range of key 1,5-diaryl-1H-imidazole-4-carboxylate ester intermediates 17. Facile conversion of the ester intermediates into their corresponding carboxylic acid and carbohydrazide derivatives ultimately gave rise to the desired targets 10 and 11, respectively. Judicial choice of differently substituted starting materials using this approach enabled a detailed and thorough structure-activity relationship (SAR) analysis of the desired final compounds to be undertaken. The route was initially tested starting from commercially available 4-fluorobenzoyl chloride 12a and 4-bromoaniline 13a. A mixture of 4-fluorobenzoyl chloride 12a and 4-bromoaniline 13a in ethyl acetate in the presence of Et3N was reacted at room temperature for 16 h to afford N-(4-bromophenyl)-4-fluoroben- The optimized conditions were then applied to various commercially available benzoyl chlorides 12b-f and aniline derivatives 13a-h to establish the scope and limitations of the reaction. First, we studied the transformation of the amides 14b-p into the essential N-aryl benzimidoyl chloride intermediates 15. It was found that intermediates 15b-n were successfully synthesized in quantitative yields after removal of excess SOCl 2 and compounds were used in the subsequent reaction without purification. However, when the aniline ring had two chlorine substituents (at the paraand ortho-positions, i.e., 14o and 14p), no product was obtained and TLC analysis indicated the presence of unreacted starting materials. The series of N-aryl benzimidoyl chloride intermediates 15b-n was then subsequently subjected to cyclization with the isocyanide 16. After column chromatographic purification, only the ethyl 1,5-diaryl-1H-imidazole-4-carboxylate derivatives derived from aniline motifs substituted with the para-electron-withdrawing groups (i.e., 17a-l in Table 1) were isolated in adequate purity. It was noted that scaffolds containing electron-donating substituents on the aryl ring directly attached to the 5-position of the imidazole motif were obtained in far poorer yield than those containing electron-withdrawing groups. For instance, the reasonable yields of 64% for 17a, 58% for 17b, 72% for 17g, and 67% for 17h were achieved when the electron-withdrawing fluorine atom occupied the paraor the metapositions on the 5-aryl ring. On the other hand, the presence of an electron-donating group (i.e., methyl and methoxy substituents) attached to the paraor the metapositions of the 5-aryl group led to poor reactivity and gave 17c-e and 17i-k in poor yields of between 18% and 41%. It was further observed that the absence of substituents on the 5-aryl ring also negatively affected the yields of the products (17e was obtained in 22% and 17l in 37% yield). Unfortunately, the presence of electron-donating methyl groups at the paraand orthopositions on the aryl ring derived from the aniline motif (i.e., substrates 15m and 15n in Scheme 2) was shown to hinder product formation entirely. It is not clear if the difficulty in forming the imidazole ring under the reaction conditions described can be attributed primarily to the steric hindrance of these groups or their electron-donating nature. However, electron-donating groups are expected to strongly disfavor the reaction by decreasing the electrophilicity of the imidoyl chloride carbon atom that undergoes nucleophilic attack by the α-carbon of the isocyanide. The structures of compounds 17a-l are in accordance with their 1 H and 13 C NMR spectra (see Supplementary Materials, Figures S1-S24) and MS-ESI. The 1 H NMR spectra of compounds 17a-l are characterized by the presence of the methine proton of the imidazole ring at ca. 7.6 ppm, the methylene protons of the ethyl group at ca. 4.2 ppm, as well as the methyl protons which appear at ca. 1.2 ppm as a triplet (Figures S1-S24). The 13 C NMR spectra of the title compounds displayed the methine-, methylene-, and methyl carbons at ca. 137, 60, and 14 ppm, respectively. To the best of our knowledge, there are no previous reports on the cycloaddition reactions of ethyl isocyanoacetate and N-aryl benzimidoyl chloride intermediates for the preparation of these novel ethyl 1,5-diaryl-1H-imidazole-4-carboxylate derivatives 17a-l as described herein.  A mechanistic pathway for the formation of 1,5-diaryl-1H-imidazole-4-carboxylate esters 17 is proposed in Scheme 3. The initial step is thought to involve activation of ethyl isocyanoacetate 16 under basic conditions to generate anion I [42]. The α-carbon of the intermediate I then undertakes a nucleophilic attack on the C=N double bond of 15 to afford intermediate II, with the elimination of a chloride ion as the DBU hydrochloride salt. Intermediate II tautomerizes to produce III, which then undergoes in situ cyclization and proton transfer to give the desired product 17. Scheme 3. Plausible mechanism for the formation of product 17.

Synthesis of Novel 5-Diaryl-1H-imidazole-4-carboxylic Acids and 1,5-Diaryl-1H-imidazole-4-carbohydrazides
When compounds 17a-j were hydrolyzed (using conditions d in Scheme 4), the resulting novel 1,5-diaryl-1H-imidazole-4-carboxylic acids 10a-l were obtained in yields ranging from 68 to 83% (Table 2) while the reaction of 17a,b, 17d,e, and 17g-l with hydrazine monohydrate in ethanol heated at reflux temperature (conditions e in Scheme 4) yielded the desired novel 1,5-diaryl-1H-imidazole-4-carbohydrazide derivatives 11a,b, 11d,e, 11g-l in yields ranging from 58 to 76% (Scheme 4) [57]. Unfortunately, there was insufficient material available for the conversion of compounds 11c and 11f to the corresponding hydrazide derivatives. The complete hydrolysis to products 10a-l was evidenced by the disappearance of the signals corresponding to the ethyl group in their 1 H and 13 C NMR spectra ( Figures S25-S48) and the appearance of a broad, downfield signal at ca. 12 ppm corresponding to the acidic proton of the carboxylic acid group (COOH) in the 1 H NMR spectra. Meanwhile, the formation of carbohydrazides 11 was evident from the appearance of signals between ca. 9.2-9.3 and 4.3-4.4 ppm corresponding to NH and NH 2 , respectively ( Figures S49-S68).

HIV-1 IN-LEDGF AlphaScreen TM Assays
The inhibition of LEDGF/p75-HIV IN protein-protein activity by the newly synthesized 1,5-diaryl-1H-imidazole-4-carboxylic acids 10a-l and 1,5-diaryl-1H-imidazole-4-carbohydrazide derivatives 11a,b, 11d,e, and 11g-l was assessed using an Amplified Luminescent Proximity Homogeneous Assay (AlphaScreen™ assay) at 100 µM compound concentration [58]. MUT 101 (3), a compound that inhibits both IN-LEDGF/p75 interaction and IN strand transfer activity while enhancing IN-IN interaction, was used as a positive control [21]. Experimental protocols were conducted in triplicate, and the percentage inhibition results are shown in Table 3. Compounds showing percentage inhibition rates of higher than 50% were considered to be active. The biological evaluation data revealed a total of seventeen potential inhibitors (out of twenty-two tested compounds) that exceeded our pre-defined 50% inhibitory threshold of disrupting the HIV-1 IN-LEDGF/p75 interaction. Carbohydrazide 11e, with the para-methylphenyl group at the 5-position and the para-bromophenyl group at the N-site of the imidazole moiety exhibited the highest inhibition of 89%, which is close to the value obtained for reference compound MUT 101, with 90% inhibition. Carboxylic acid 10k, with the para-methylphenyl ring at the 5-position and the para-chlorophenyl unit at the N-site of the imidazole motif showed the second-highest level of inhibition, at 85%. This was followed closely by carboxylic acid 10d, with the metamethylphenyl and the para-bromophenyl motifs at the 5-and Nsites, respectively, which showed 84% inhibition for disruption of the LEDGF/p75-IN interaction. Interestingly, these top 3 inhibitors all have a methyl substituent attached to the aryl ring at the 5-position of the imidazole motif, which is also present in the majority of known LEDGF/p75 interaction inhibitors. This indicates that the presence of the methyl substituent on the structure could be essential to the potency of the compound. The carboxylic acids, 10a (with the para-fluorophenyl and the para-bromophenyl motifs attached at the 5-and N-positions, respectively) and 10e (with the para-methylphenyl and the para-bromophenyl units at the 5-and N-site) displayed the fourth and fifth best inhibition of 83% and 82%, respectively. Compounds 10b, 10i, 10j, 11d, and 11k exhibited good overall inhibition, with percentage inhibitions ranging from 67% to 78% while chemical entities 10c, 10h, 11a, 11b, 11h, and 11i disrupted the HIV-1 IN and LEDGF/p75 interaction by 50-59%, demonstrating moderate inhibitory activity. Marginal percentage inhibition values of 17, 35, and 33% were observed for compounds 10f, 10l, and 11j, respectively, while compounds 11g and 11i were the only inactive molecules. According to the data, with the notable exception of 11g and 11i, the results did support the initial hypothesis that incorporation of carboxylic acid and carbohydrazide moieties onto the structure of hit 9 would have a significant impact on the binding capabilities of the compounds.

Direct ELISA Assay
The new compounds 10a-l, 11a,b, 11d,e, and 11g-l were also examined for their ability to bind to divalent cations in the catalytic core of integrase using ELISA assays at 100 µM concentration [59]. The assay was carried out in order to determine if the newly synthesized compounds act similarly to MUT 101 (3), which inhibits both IN-LEDGF interaction and IN strand transfer. A control compound, raltegravir, known to bind divalent metals within HIV-1 IN, was also tested alongside the test compounds. The experimental protocols were carried out in duplicate, and their relative percentage inhibitory values are shown in Table 3 and Figure 4. The data showed that most of the compounds showed no significant inhibition of strand transfer. Only three compounds, viz. 10a, 10l, and 11k, showed marginal percentage inhibition of 44%, 39%, and 43%, respectively, at 100 µM compared to a 95% inhibition for raltegravir at 10 µM concentration as shown in Table 3. Thus, these results demonstrate that these two classes of compounds (10 and 11) may be specifically binding to the LEDGF/p75 binding site in HIV-1 integrase (Figure 4).   (10) and carbohydrazide (11) inhibition percentages in AlphaScreen TM (blue) and ELISA (orange) assays at 100 µM.

Cytotoxicity and Antiviral Assays
The newly synthesized compounds 10a-l and 11a,b, 11d,e, and 11g-l were further examined for cytotoxicity and antiviral activity using methods similar to those previously described [30]. The assays were performed in the metallothionein type 4 (MT-4) cells, a human cell line that is highly susceptible to HIV infection. Auranofin, a compound known for its toxicity to mammalian cells, was screened alongside the compounds. Cytotoxicity screening indicates the tolerance of the system to the compounds, as well as indicating the relative toxicity of these compounds to live cells [60]. Cytotoxicity results are represented in Table 4. The data show that thirteen compounds are marginally toxic or non-toxic, having CC 50 values of greater than or equal to 100 µM, while nine compounds exhibited slightly higher toxicity, with CC 50 values of less than 100 µM. However, these scaffolds were not found to be overly toxic in comparison to the control, auranofin, which exhibited a CC 50 value of 1.6 µM. The CC 50 value of control compounds was comparable to the previously described CC 50 values of 4.5 µM against human HepG-2 human liver cells [61]. The results indicate that most of the compounds were tolerated well by the cells. Finally, the compounds were evaluated for their capability to inhibit the viral outgrowth of the HIV-1 subtype C replicating in a cell-based antiviral assay at 100 µM single dose concentration [62]. The antiviral percentage inhibition values of the evaluated compounds were calculated and are recorded in Table 4. The antiviral percentage activities were found to range from low to moderate, with none found to be higher than 50% at 100 µM compound concentration. Although compound 11h was the best performer with an antiviral percentage of 40%, its potency level was possibly induced by its toxicity of 50.4 µM ( Table 4). Taken together, compounds 11a, 11b, and 11g with CC 50 values of >200 µM, 158.4 µM, and >200 µM, and antiviral percentage activities of 35%, 36%, and 33%, respectively, were considered to be promising scaffolds. This indicates that the selectivity of inhibition over toxicity for these compounds was favorable. Furthermore, it was interesting to observe that the compounds with marginal antiviral activities belong to the carbohydrazide family of compounds. Additionally, although, the antiviral percentage activities were minimal in comparison to raltegravir (tested at 10 µM), the results indicate that the inclusion of the carbohydrazide motif positively influenced the binding more than the carboxylic acid moiety.

Multimerization Gel Assay as a Secondary Screening Assay
Chemical covalent crosslinking of IN has been shown to stabilize the multimer forms when observed on SDS-PAGE [63]. This is particularly useful due to the disruption of these multimers due to SDS and reducing conditions in the gel. The IN cross-linking assay was performed as described previously [63]. This allowed for the capture of multimer forms and provided useful information in determining the effects that the compounds had on IN multimerization . Four compounds 11a, 11b, 11g, and 11h, exhibiting marginal antiviral percentage inhibitory activities were selected for the multimerization studies in a Western blot gel multimerization assay at concentrations of 50, 100, and 200 µM. Compounds 11a, 11b, and 11h appear to cause an increase in higher-order oligomeric forms of IN as shown in Figure 5. There is a clear shift in the oligomeric equilibrium as a result of increasing concentrations of the compounds. These results indicate that the tested compounds cause multimerization, similar to previously described allosteric integrase inhibitors [64,65].  11a, 11b, 11g, and 11h were selected for multimerization studies. Compounds were tested at 50 µM, 100 µM, and 200 µM. MUT101 was included as a control and tested at 50 µM only. Between the no-compound control and MUT101, an increase of higher order IN forms is noticed. An increase in higher order IN oligomers is most pronounced in compounds 11a, 11b, and 11h and the equilibrium shifts to higher order forms as a factor of compound concentration.

Materials and Methods
All commercially available reagents were purchased from Sigma Aldrich (Schnelldorf, Germany) and used without further purification. Dry solvents were used from an LC-Tech SP-1 Solvent Purification System (LC Technology Solutions Inc., Seabrook, New Hampshire, USA) and kept under argon. TLC was performed on pre-coated Merck silica gel F254 plates and viewed under UV light (254 nm) or following exposure to iodine vapor. Column chromatography was performed using Merck Silica gel 60 [particle size 0.040-0.063 mm (230-400 mesh)]. NMR spectra were recorded on a 400 MHz Bruker Avance Spectrometer (Bruker BioSpin, Rheinstetten, Germany) at 298 K equipped with a BBI 5 mm probe and solvent calibrated. High-resolution mass spectra were recorded on an LC-MS system consisting of a Dionex Ultimate 3000 Rapid Separation LC system equipped with a C-18 pre-coated column and coupled to a MicrOTOF QII Bruker mass spectrometer (Bruker Daltonik GmbH, Bremen, Germany) fitted with an electrospray source and operating in positive mode. Sodium formate clusters were used for internal calibration. The chemical shifts (δ) are expressed in parts per million, according to the solvent residual peak resonances and calibrated using solvent signals To a stirred solution containing aniline (14.4 mmol) and Et 3 N (25.1 mmol) in EtOAc (30 mL) was added benzoyl chloride (15.1 mmol) at 0 • C. After addition, the resulting reaction mixture was allowed to warm to room temperature and stirred for 16 h (monitored by TLC). Upon completion, EtOAc (20 mL) was added, followed by washing with 1M HCl, saturated NaHCO 3 and water. The organic layer was dried over anhydrous MgSO 4 , and the solvent evaporated and dried in vacuo to afford pure N-aryl-benzamide derivatives [53].

General Procedure for the Synthesis of N-Aryl-benzimidoyl Chloride Intermediates 15a-n
A mixture of N-aryl-benzamide (1.0 molar equivalent) and thionyl chloride (6.0 molar equivalents) was heated to reflux until reaction completion (8 h). Excess thionyl chloride was removed by co-evaporation with toluene to afford, after drying under high vacuum, diarylimidoyl chloride and were directly used in the next step without further purification [54][55][56].

Modeling Protocols
The X-ray crystal structure of HIV-1 IN dimer (PDB code: 4NYF) in complex with the small molecule, BI 224436 (2) was used for a better understanding of binding interactions of the tested ligands [27]. The HIV-1 IN protein used in the modeling was prepared using the "Protein Preparation Wizard" in Maestro (Schrodinger suite, Schrödinger, Inc, Cambridge, MA, USA) [66]. This protocol involves the addition of hydrogen atoms, bond order assignment, and optimization of bond lengths, bond angles, torsion angles, and non-bonded interactions using the OPLS force field. Furthermore, missing loops and side chains were filled in using Prime with a length of axes set to 30 Å and a value of the scaling factor chosen to be 1.00. Docking grids were produced with the default settings in Glide utilizing the co-crystalized ligand, BI-224436 binding site to express the focus of the grid box. Before docking, structures were minimized to their lowest energy and all possible 3D structure of molecules were created utilizing Ligprep protocol in Maestro [67]. The protocol encompasses the addition of hydrogen, transforming 2D structures to 3D, creating stereoisomers, neutralizing charges, and forecasting the probable ionization state using an OPLS3 force field. The scaffolds were then primarily positioned to the LEDGF/p75 binding site in HIV-1 integrase using the Schrödinger Glide protocol with the standard precision (SP) mode, which resulted in the generation of hundreds for poses of each compound [68].

HIV-1 IN-LEDGF/p75 AlphaScreen TM Assay
An AlphaScreen TM assay was developed and used to determine the percentage inhibition of the newly synthesized 1,5-diaryl-1H-imidazole-4-carboxylic acids 10 and 1,5diaryl-1H-imidazole-4-carbohydrazide derivatives 11. Assay buffer (50 mM Tris pH 7.4, 1 mM MgCl 2 , 1 mM DTT, 0.01% Tween 20, and 2.5% Glycerol) was used to make 10× working stocks of FLAG-IN, His-LEDGF/p75 and test compounds with concentrations of 1 µM, 5 µM, and 1 mM, respectively. Total reaction volumes were 50 µL. First, 35 µL of assay buffer was added to wells of a 1/2 area optiplate. Then, 5 µL of a working stock of FLAG-tagged IN was added to each well containing assay buffer, except for control wells. Then, 5 µL of test compounds were added to each well containing FLAG-IN and assay buffer, except for control wells. Plates were incubated for 30 min at room temperature with gentle shaking. Subsequently, 5 µL of His-tagged LEDGF/p75 was added to each test well, except to control wells, and incubated for 1 h at 30 • C with gentle shaking. A final concentration of 10 µg/mL nickel chelating acceptor beads (Perkin Elmer, MA, USA) and Anti-FLAG donor beads (Perkin Elmer, MA, USA) were added and incubated in the dark at 30 • C with gentle shaking. The plate was subsequently read between 520-620 nm on the EnSpire ® plate reader (Perkin Elmer, MA, USA). The control (3) ran on the plate to ensure accurate reading of the compounds effects were: beads alone, FLAG-IN plus beads alone, His-LEDGF/p75 plus beads alone, FLAG-IN, and His-LEDGF/p75 alone.

Strand Transfer Assay
Reaction buffer (50 mM HEPES pH 7.5, 10 mM NaCl, 5 mM MnCl 2 /MgCl 2 , 15 mM DTT) was added to wells of a round bottom 96 well plate at 10 µL per well. IN and donor DNA were mixed in reaction buffer and 10 µL each was added to the wells containing pre-aliquoted reaction buffer. The final concentrations of IN and donor DNA was 200 nM and 100 nM, respectively, in the context of the total reaction volume. The IN was allowed to complex the DNA for 15 min. Tested compounds were added at 10 µL, to a final of 100 µM, and mixed in wells containing pre-aliquoted reaction buffer. This was incubated at ambient temperature for 15 min. Target DNA was added in a 10 µL volume to a 200 nM final concentration. Total reaction volume was 50 µL. The reaction was immediately incubated at 37 • C for 90 min in a shaking plate incubator. Reactions were stopped by addition of 50 µL ELISA Binding buffer (50 mM Tris-HCl pH 7.5, 4M NaCl, 20 mM EDTA), and immediately added to Streptavidin coated plates (RD Biosystems) and incubated at ambient temperature for 30 min. Non-bound reaction products were removed by washing the wells with TBS-T (25 mM Tris-HCl pH 7.5, 150 mM NaCl, Tween 0.005%) two times with no incubations between washes. Anti FITC-AP linked antibody (Sigma-Aldrich) was added at 200 µL at a 1:5000 dilution in TBS-T and incubated at 37 • C for 120 min. The wells were then washed three times with TBS-T with 5 min incubation per wash. SIGMAFASTTM p-Nitrophenyl phosphate Tablets (Sigma-Aldrich), a phosphatase substrate, was prepared and added to each test well at 100 µL and incubated at 37 • C for 30 min. Absorbance readings were determined at 405 nm on the X-mark spectrophotometer (Bio-Rad, Rosebank, South Africa) [59].

Cellular Toxicity Assay
The day of cytotoxicity testing, MT-4 cells (obtained through the NIH AIDS Reagent Program, Division of AIDS, NIAID, NIH: MT-4 catalogue number 120) were counted and seeded at 5 × 10 4 cells per well. A total of 100 µL of cells were added to each well of a 96-well cell culture plates (TTP) for testing. The volume in the last row of wells of each plate was further increased to 197 µL of media containing MT-4 cells. The plate was placed into the incubator to equilibrate to 37 • C and 5% CO 2 . Test compounds were prepared in plate format and each compound was at 25 mM stock concentration in DMSO. A total of 3.2 µL compound was added to the final row of wells containing 197 µL cells and mixed to ensure the solution was homogeneous with the cells. A 2-fold serial dilution of an 8 times series was performed and the final 100 µL was discarded. The plate was placed at 37 • C and 5% CO 2 for 5 days. On the 5th day the cell viability was assessed using the Cell Titre 96 ® Aqueous One Solution Cell Proliferation assay (Promega) as per manufacturer's instructions. Briefly, 20 µL of MTS solution was added to each well and incubated at 37 • C for 4 h. The absorbance of the plates was then read at 490 nm on a SpectraMax spectrophotometer (Molecular Devices, LLC. San Jose, CA, USA). The data analysis was completed on Origin 8.1 with the log value of the compound molar concentration plotted against the percentage cell viability determined by the absorbance level by sigmoidal fitting. From the curve, a half maximal cytotoxicity CC 50 for compounds was determined. Auranofin was used as a cytotoxic control compound starting at 100 µM across a two-fold serial dilution of an 8 dilution series. DMSO was used as background control and 2-fold serially diluted from 1.5% [60].

Antiviral Assay
A subtype C virus isolate was obtained through the NIH HIV Reagent Program, Division of AIDS, NIAID, NIH: Human Immunodeficiency Virus Type 1 strain 98/IN/022, ARP-4167, contributed by UNAIDS Network for HIV Isolation and Characterization. MT-4 cells were grown and maintained in RPMI (Gibco) supplemented with 2 mM L-glutamine, non-essential amino acids and 10% heat inactivated fetal bovine serum at 37 • C and 5% CO 2 . Subtype C virus was added to MT4 cells and incubated at 37 • C and 5% CO 2 for two weeks. Supernatants were harvested by centrifugation at 4000× g and stored at −80 • C until use. Harvested supernatant containing HIV-1 was added to fresh MT-4 cells and viral absorption was allowed for 24 h similar to the method by Rosenwirth et al. [69]. After viral absorption, the cells were washed once with fresh media, a sample was taken, and HIV-1 infection was allowed to develop. Seven days after infection, MT-4 supernatant samples were tested for the presence of p24 antigen using the bioelisa HIV-1+2 Ag/Ab kit (biokit) as per manufacturer's instructions. After HIV-1 infection was confirmed, supernatant samples were isolated by centrifugation at 4000× g and immediately used to infect 1 × 10 5 cells/mL of MT4 cells. Virus was allowed to absorb and 24 h later the cells were washed with fresh media and 100 µL of cells, 1 × 10 4 cells, were aliquoted into a round bottom 96 well plate (TPP). Uninfected cells were also prepared. A day 1 sample was taken and stored at −80 • C until final testing. Compounds were diluted in fresh media to 200 µM concentrations and 100 µL was added to each test well, the final concentration of compounds was 100 µM. Raltegravir (NIH, ARP-11680) was used as a control compound at 10 µM, blank wells contained equal volumes of DMSO. Upper and lower asymptotes were determined by untreated MT4 cells either infected with HIV (100%) or not infected with HIV (0%). Four days after infection, p24 presence was determined using the bioelisa HIV-1+2 Ag/Ab kit as per manufacturer's instructions. The data was used to determine if the selected concentration of compounds would reduce the amount of p24 and thereby indicate a reduction in HIV infectivity [62]

Conclusions
A series of novel 1,5-diaryl-1H-imidazole-4-carboxylic acids 10 and 1,5-diaryl-1Himidazole-4-carbohydrazides 11 were designed and synthesized via the formation of 1,5-diaryl-1H-imidazole-4-carboxylate ester intermediates 17 obtained from cycloaddition of ethyl isocyanoacetate 16 and stable diarylimidoyl chlorides 15 under basic conditions. This study was conducted to evaluate the hypothesis that the HIV-1 IN activity of the hit compound 9 could be greatly improved through the addition of carboxylic acid and carbohydrazide motifs. This idea was demonstrated to be productive, since seventeen scaffolds (i. e., 10a-d, 10f, 10h-k, 11a-b, 11d-e, 11h-k) exhibited more than 50% inhibitory activity in disrupting the HIV-1 IN-LEDGF/p75 protein-protein interaction in a direct AlphaScreen TM assay at 100 µM. The specific binding of these compounds to the LEDGF/p75-binding pocket was also established by their further assessment in the HIV-1 IN strand transfer assay at 100 µM, which concluded that none of the compounds (with the exception of 10a, 10l, and 11k with marginal inhibitory percentage activities of 44%, 39%, and 43%, respectively) were actively bound to the active site. In a cell-based assay, the newly targeted compounds exhibited varying degrees of toxicity, with 13 compounds having a CC 50 value of greater than or equal to 100 µM, while 9 compounds were found to exhibit CC 50 values of below 100 µM. Further evaluation of the compounds in an antiviral HIV-1 assay revealed four compounds (11a, 11b, 11g, and 11h) with moderate antiviral percentage activities ranging from 33 to 45% and with CC 50 values ranging from 50.4 µM to >200 µM. The antiviral percentage activity displayed by 11h was credited to its relatively high toxicity of 50.4 µM . Compounds 11a, 11b, 11g, and 11h were further validated for their ability to induce multimerization in a Western blot gel assay, and three compounds, 11a, 11b, and 11h seemed to increase higher-order forms of IN. These data serve as a good starting point for the further optimization of these scaffolds as possible anti-HIV-1 IN drugs. Therefore, efforts are currently underway in our laboratories to improve the potency and enhance the drug-like properties of these compounds.