Chloropyridinyl Esters of Nonsteroidal Anti-Inflammatory Agents and Related Derivatives as Potent SARS-CoV-2 3CL Protease Inhibitors

We report the design and synthesis of a series of new 5-chloropyridinyl esters of salicylic acid, ibuprofen, indomethacin, and related aromatic carboxylic acids for evaluation against SARS-CoV-2 3CL protease enzyme. These ester derivatives were synthesized using EDC in the presence of DMAP to provide various esters in good to excellent yields. Compounds are stable and purified by silica gel chromatography and characterized using 1H-NMR, 13C-NMR, and mass spectral analysis. These synthetic derivatives were evaluated in our in vitro SARS-CoV-2 3CLpro inhibition assay using authentic SARS-CoV-2 3CLpro enzyme. Compounds were also evaluated in our in vitro antiviral assay using quantitative VeroE6 cell-based assay with RNAqPCR. A number of compounds exhibited potent SARS-CoV-2 3CLpro inhibitory activity and antiviral activity. Compound 9a was the most potent inhibitor, with an enzyme IC50 value of 160 nM. Compound 13b exhibited an enzyme IC50 value of 4.9 µM. However, it exhibited a potent antiviral EC50 value of 24 µM in VeroE6 cells. Remdesivir, an RdRp inhibitor, exhibited an antiviral EC50 value of 2.4 µM in the same assay. We assessed the mode of inhibition using mass spectral analysis which suggested the formation of a covalent bond with the enzyme. To obtain molecular insight, we have created a model of compound 9a bound to SARS-CoV-2 3CLpro in the active site.


Introduction
Novel Coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is responsible for the ongoing COVID-19 pandemic [1,2]. The first cases of the disease were reported in Wuhan, China and then rapidly spread worldwide, overwhelming health care systems, disrupting economies, and leading  We subsequently designed a series of 5-chloropyridin-3-yl esters and demonstrated both enzyme inhibitory and antiviral activity. Compound 3 exhibited a SARS-CoV 3CL protease inhibitory IC 50 value of 250 nM and an antiviral EC 50 value of 2.8 µM in VeroE6 cells [25]. Based upon our previous results, we recently developed indole chloropyridin-3-yl esterderived SARS-CoV-2 3CL protease inhibitors and demonstrated that prototype compounds, such as inhibitor 3, are potent inhibitors of SARS-CoV-2 3CL protease. It exerts comparable antiviral activity to remdesivir (5), an RNA-dependent RNA-polymerase inhibitor [26,27]. Furthermore, we have shown that compound 3 blocked the infectivity and cytopathic effect of SARS-CoV-2 wk-521 in VeroE6 cells in our immunocytochemistry assay.
Compound 4 has also shown potent enzyme inhibitory and antiviral activity in our immunocytochemistry assays [28]. Our X-ray structural analysis of inhibitor 4 and SARS-CoV-2 3CL protease complex demonstrates that the mode of inhibition involved the formation of a covalent bond with the inhibitor carbonyl group and catalytic Cys145 in the active site as shown in Figure 2. Furthermore, our recent structure-activity relationship (SAR) suggested that the position of the carboxylic acid on the indole scaffold is critical for the enzyme activity. Based upon our X-ray structural studies and mode of inhibition, we have further investigated other aromatic and heteroaromatic scaffolds and their ability to block SARS-CoV-2 3CL protease activity as well as antiviral activity in VeroE6 cells [29]. In particular, we plan to synthesize 5-chloropyridin-3-yl esters of widely used nonsteroidal anti-inflammatory agents (NSAIDs) [30,31] and evaluate their potential as irreversible inhibitors of SARS-CoV-2 3CLpro enzyme. Presumably, such acylated thioesters of these NSAIDs would hydrolyze slowly over time and release parent NSAIDs in the cell [32]. Interestingly, this would lead to inhibition of cyclooxygenase in the cell, leading to analgesic and anti-inflammatory effects. In the present studies, we report a series of 5-chloropyridinyl ester derivatives of salicylic acid, ibuprofen, naproxane, indomethacin, and related interesting derivatives. A number of compounds exhibited potent enzyme inhibitory and antiviral activity. pyridinyl ester derivatives of salicylic acid, ibuprofen, naproxane, indomethacin, and related interesting derivatives. A number of compounds exhibited potent enzyme inhibitory and antiviral activity.

Chemistry
The synthesis of various 5-chloropyridinyl esters of common nonsteroidal anti-inflammatory agents [30,31] is shown in Scheme 1. Commercially available aspirin 6 was esterified with 1.2 equivalents of 5-chloro-3-pyridinol using 1.5 equivalents of EDC in the presence of 1 equivalent of DMAP in CH2Cl2 at 23 °C for 12 h. This condition has provided 6a in 12% yield after silica gel chromatography, we have then exposed (S)-and (R)naproxen, racemic ibuprofen and indomethacin under these esterification conditions for the synthesis of other 5-chloropyridinyl esters 8a-11a (46-53% yield). The structures of these ester derivatives are shown in Table 1. We have also prepared chloropyridinyl esters

Chemistry
The synthesis of various 5-chloropyridinyl esters of common nonsteroidal anti-inflamm atory agents [30,31] is shown in Scheme 1. Commercially available aspirin 6 was esterified with 1.2 equivalents of 5-chloro-3-pyridinol using 1.5 equivalents of EDC in the presence of 1 equivalent of DMAP in CH 2 Cl 2 at 23 • C for 12 h. This condition has provided 6a in 12% yield after silica gel chromatography, we have then exposed (S)-and (R)-naproxen, racemic ibuprofen and indomethacin under these esterification conditions for the synthesis of other 5-chloropyridinyl esters 8a-11a (46-53% yield). The structures of these ester derivatives are shown in Table 1. We have also prepared chloropyridinyl esters derived from salicylic acid and its methyl-substituted derivatives as shown in Scheme 2. Initially, we attempted the synthesis of salicylic acid derivatives by using conditions mentioned above. However, the above conditions provided variable results and provided mixture of monomeric and dimeric ester along with small amounts of higher oligomers. The monomeric products were variable. We then carried out the esterification reaction of salicylic acid and its methyl-substituted derivatives by first exposing 0.25 equivalent of acid, 1.2 equivalent of 3-hydroxy-5-chloro-pyridine and 1.5 equivalent of EDC in the presence of 1.0 equivalent DMAP at 23 • C. The mixture was stirred for 2 h and then 0.25 equivalent of acid was added at 23 • C every 2 h interval. The resulting mixture was stirred at 23 • C for 12 h. This condition provided a mixture of monomeric esters (12a-17a) and dimeric esters (12b-16b) respectively, in good yields. In the case of 2-hydroxy-5-methyl benzoic acid 15, in addition to monomer 15a and dimer derivative 15b, we have also obtained triester derivative 15c. 2-Hydroxy-6-methyl benzoic acid 16 provided monoester 16a and diester 16b. The diester 16b was crystalized in CH 2 Cl 2 solution and the identity of the structure was unambiguously determined by X-ray crystallography as depicted in Figure 3 [33,34]. Please see Supporting Information for further details. For our structure-activity relationship studies, we have also prepared 3-acetamido-benzoic acid ester derivatives 18a and 19a in good yield. derived from salicylic acid and its methyl-substituted derivatives as shown in Scheme 2. Initially, we attempted the synthesis of salicylic acid derivatives by using conditions mentioned above. However, the above conditions provided variable results and provided mixture of monomeric and dimeric ester along with small amounts of higher oligomers. The monomeric products were variable. We then carried out the esterification reaction of salicylic acid and its methyl-substituted derivatives by first exposing 0.25 equivalent of acid, 1.2 equivalent of 3-hydroxy-5-chloro-pyridine and 1.5 equivalent of EDC in the presence of 1.0 equivalent DMAP at 23 °C. The mixture was stirred for 2 h and then 0.25 equivalent of acid was added at 23 °C every 2 h interval. The resulting mixture was stirred at 23 °C for 12 h. This condition provided a mixture of monomeric esters (12a-17a) and dimeric esters (12b-16b) respectively, in good yields. In the case of 2-hydroxy-5-methyl benzoic acid 15, in addition to monomer 15a and dimer derivative 15b, we have also obtained triester derivative 15c. 2-Hydroxy-6-methyl benzoic acid 16 provided monoester 16a and diester 16b. The diester 16b was crystalized in CH2Cl2 solution and the identity of the structure was unambiguously determined by X-ray crystallography as depicted in Figure  3 [33,34]. Please see Supporting Information for further details. For our structure-activity relationship studies, we have also prepared 3-acetamido-benzoic acid ester derivatives 18a and 19a in good yield.

Biological Evaluation
We have carried out SARS-CoV-2 3CLpro inhibition assays using the authentic SARS-CoV-2 3CLpro enzyme as described recently [35]. The enzyme inhibitory activity (IC50 values) of synthetic active esters was assessed using a continuous fluorescence assay and the FRET-based substrate UIVT3 (HiLyteFluor488 TM -EATLQSGLRKAK-QXL520-NH2 (HPLC > 90%); Anaspec, Fremont, CA, USA) described by us previously [20,36]. The antiviral activity (EC50 value) of compounds was evaluated using quantitative VeroE6 cellbased assay with RNA-qPCR as described by us recently [28]. The structures and activity

Biological Evaluation
We have carried out SARS-CoV-2 3CLpro inhibition assays using the authentic SARS-CoV-2 3CLpro enzyme as described recently [35]. The enzyme inhibitory activity (IC50 values) of synthetic active esters was assessed using a continuous fluorescence assay and the FRET-based substrate UIVT3 (HiLyteFluor488 TM -EATLQSGLRKAK-QXL520-NH2 (HPLC > 90%); Anaspec, Fremont, CA, USA) described by us previously [20,36]. The antiviral activity (EC50 value) of compounds was evaluated using quantitative VeroE6 cellbased assay with RNA-qPCR as described by us recently [28]. The structures and activity

Biological Evaluation
We have carried out SARS-CoV-2 3CLpro inhibition assays using the authentic SARS-CoV-2 3CLpro enzyme as described recently [35]. The enzyme inhibitory activity (IC50 values) of synthetic active esters was assessed using a continuous fluorescence assay and the FRET-based substrate UIVT3 (HiLyteFluor488 TM -EATLQSGLRKAK-QXL520-NH2 (HPLC > 90%); Anaspec, Fremont, CA, USA) described by us previously [20,36]. The antiviral activity (EC50 value) of compounds was evaluated using quantitative VeroE6 cellbased assay with RNA-qPCR as described by us recently [28]. The structures and activity ture of monomeric and dimeric ester along with small amounts of higher oligomers. The monomeric products were variable. We then carried out the esterification reaction of salicylic acid and its methyl-substituted derivatives by first exposing 0.25 equivalent of acid, 1.2 equivalent of 3-hydroxy-5-chloro-pyridine and 1.5 equivalent of EDC in the presence of 1.0 equivalent DMAP at 23 °C. The mixture was stirred for 2 h and then 0.25 equivalent of acid was added at 23 °C every 2 h interval. The resulting mixture was stirred at 23 °C for 12 h. This condition provided a mixture of monomeric esters (12a-17a) and dimeric esters (12b-16b) respectively, in good yields. In the case of 2-hydroxy-5-methyl benzoic acid 15, in addition to monomer 15a and dimer derivative 15b, we have also obtained triester derivative 15c. 2-Hydroxy-6-methyl benzoic acid 16 provided monoester 16a and diester 16b. The diester 16b was crystalized in CH2Cl2 solution and the identity of the structure was unambiguously determined by X-ray crystallography as depicted in Figure  3 [33,34]. Please see Supporting Information for further details. For our structure-activity relationship studies, we have also prepared 3-acetamido-benzoic acid ester derivatives 18a and 19a in good yield.

Biological Evaluation
We have carried out SARS-CoV-2 3CLpro inhibition assays using the authentic SARS-CoV-2 3CLpro enzyme as described recently [35]. The enzyme inhibitory activity (IC 50 values) of synthetic active esters was assessed using a continuous fluorescence assay and the FRET-based substrate UIVT3 (HiLyteFluor 488 TM -EATLQSGLRKAK-QXL 520 -NH 2 (HPLC > 90%); Anaspec, Fremont, CA, USA) described by us previously [20,36]. The antiviral activity (EC 50 value) of compounds was evaluated using quantitative VeroE 6 cellbased assay with RNA-qPCR as described by us recently [28]. The structures and activity of synthetic ester derivatives are shown in Tables 1 and 2. We first assessed common NSAIDs-derived chloropyridinyl esters shown in Table 1. As can be seen, acetoxysalicylic acid-derived ester 6a exhibited SARS-CoV-2 3CL protease IC 50 value of 360 nM (entry 1). This compound exhibited an antiviral EC 50 value > 100 µM. (S)-Naproxen-derived ester 8a exhibited an enzyme inhibitory activity value of 670 nM (entry 2), while the (R)naproxen derivative 9a exhibited an enzyme IC 50 value of 160 nM (entry 3), an over 4-fold improvement. Racemic ibuprofen-derived ester 10a exhibited an IC 50 value of 810 nM (entry 4). Indomethacin-derived ester 11a exhibited a significant reduction in enzyme activity while structurally related indole derivatives exhibited excellent enzyme inhibitory activity (entry 5) [29]. Interestingly, compound 11a exhibited an antiviral EC 50 value of 30 µM while aspirin, ibuprofen, and naproxen-derived esters did not show appreciable antiviral activity. We presume that the mode of inhibition involves covalent bond formation with catalytic Cys145 as observed in our previous X-ray structural analysis of inhibitorbound SARS-CoV-2 3CL protease as well as mass spectral analysis. [20,29] The irreversible enzyme acylation of the NSAIDs-based inhibitor was examined by using MALDI-TOF. Authentic SARS-CoV-2 3CLpro was incubated with inhibitor 8a and then analyzed with untreated enzyme. As expected, we were able to see a signal for enzyme-bound compound 8a on the LC-MS spectrum that corresponds to acylation of 3CL protease with a mass shift of +212 Daltons.       Based upon the X-ray structure of an irreversible inhibitor (GRL-017-20) bound to SARS-CoV-2 3CL protease (PDB code: 7RBZ), we modeled the complex of inhibitor 9a with SARS-CoV-2 3CL protease [29]. The model of inhibitor 9a bound to the catalytic Cys 145 residue in the active site of the 3CL protease is shown in Figure 4. The sulfur atom of Cys145 forms a covalent bond to the carbonyl and the chloropyridinyl group acts as a leaving group. The ligand sits in the binding pocket formed from Asn142, Met165, Glu166 and Gln 189. The model shows a similar π-π stacking of the aromatic ring with the imidazole ring of the His 41 residue [29]. Unfortunately, we did not observe the anticipated hydrogen bond between the methoxy group oxygen atom of the ligand and the side chain of the Gln 189 residue. The current study can provide a foundation to design new irreversible inhibitors to target SARS-CoV-2 3CL protease. Our laboratory is actively working on the design and synthesis of potent COVID-19 inhibitors. Based upon the X-ray structure of an irreversible inhibitor (GRL-017-20) bound to SARS-CoV-2 3CL protease (PDB code: 7RBZ), we modeled the complex of inhibitor 9a with SARS-CoV-2 3CL protease [29]. The model of inhibitor 9a bound to the catalytic Cys 145 residue in the active site of the 3CL protease is shown in Figure 4. The sulfur atom of Cys145 forms a covalent bond to the carbonyl and the chloropyridinyl group acts as a leaving group. The ligand sits in the binding pocket formed from Asn142, Met165, Glu166 and Gln 189. The model shows a similar π-π stacking of the aromatic ring with the imidazole ring of the His 41 residue [29]. Unfortunately, we did not observe the anticipated hydrogen bond between the methoxy group oxygen atom of the ligand and the side chain of the Gln 189 residue. The current study can provide a foundation to design new irreversible inhibitors to target SARS-CoV-2 3CL protease. Our laboratory is actively working on the design and synthesis of potent COVID-19 inhibitors. Based upon the X-ray structure of an irreversible inhibitor (GRL-017-20) bound to SARS-CoV-2 3CL protease (PDB code: 7RBZ), we modeled the complex of inhibitor 9a with SARS-CoV-2 3CL protease [29]. The model of inhibitor 9a bound to the catalytic Cys 145 residue in the active site of the 3CL protease is shown in Figure 4. The sulfur atom of Cys145 forms a covalent bond to the carbonyl and the chloropyridinyl group acts as a leaving group. The ligand sits in the binding pocket formed from Asn142, Met165, Glu166 and Gln 189. The model shows a similar π-π stacking of the aromatic ring with the imidazole ring of the His 41 residue [29]. Unfortunately, we did not observe the anticipated hydrogen bond between the methoxy group oxygen atom of the ligand and the side chain of the Gln 189 residue. The current study can provide a foundation to design new irreversible inhibitors to target SARS-CoV-2 3CL protease. Our laboratory is actively working on the design and synthesis of potent COVID-19 inhibitors. Based upon the encouraging enzyme inhibitory activity of NSAIDS derivatives, we then prepared a range of salicylic acid derivatives and evaluated their activity. As shown in Table 2, salicylic acid-derived pyridinyl ester 12a exhibited an IC 50 value of 3.47 µM. The corresponding diester derivative 12b exhibited a 5-fold reduction in enzyme activity. However, compound 12b exhibited an antiviral EC 50 value of 64 µM (entry 2). In an effort to modulate activity, we incorporated the methyl group on the aromatic ring. Methyl substitution at C3 resulted in monoester 13a, which exhibited a significant improvement in enzyme activity (IC 50 650 nM) over its unsubstituted derivative 12a. Diester derivative 13b exhibited a reduction in the enzyme IC 50 value but some improvement in antiviral activity (entries 3 and 4). Incorporation of methyl group at C4 provided slight improvement in enzyme activity for both mono-and di-ester derivatives 14a and 14b (entries 5 and 6). Substitution of methyl group at C5 led to the syntheses of mono-ester 15a, diester 15b, and tri-ester 15c. Both mono-ester 15a and diester 15b exhibited improvement in enzyme activity over other substituted derivatives (entries 7-9). Interestingly, substitution of methyl group at C6 resulted in significant loss of enzyme activity (entries 10 and 11). Incorporation of fluorine at C6 also resulted in a further reduction in enzyme inhibitory activity (entry 12). We have also investigated amide derivatives 18a and 19a. Both compounds exhibited enzyme inhibitory activity in low micromolar range (entries 13 and 14). All compounds in Tables 1 and 2 exhibited a cytotoxicity (CC 50 ) value >100 µM. While a number of chloropyridinyl esters exhibited low nanomolar 3CLpro inhibitory activity, the majority of these ester derivatives did not show appreciable antiviral activity, except compounds  11a, 12b, 13b, 15a and 15b, which exhibited antiviral EC 50 values of 24-64 µM. Such high ratios of antiviral EC 50 and enzyme IC 50 values may be due to the expression of the efflux transporter P-glycoprotein in VeroE6 cells. [37,38] We, therefore, examined the antiviral activity of selected compounds (6a, 9a, 11a, 14a, and 15a) in the presence of P-glycoprotein inhibitor, CP-100356 [39]. Interestingly, none of these compounds exhibited any significant antiviral activity in the presence of the P-glycoprotein inhibitor.
Based upon the X-ray structure of an irreversible inhibitor (GRL-017-20) bound to SARS-CoV-2 3CL protease (PDB code: 7RBZ), we modeled the complex of inhibitor 9a with SARS-CoV-2 3CL protease [29]. The model of inhibitor 9a bound to the catalytic Cys 145 residue in the active site of the 3CL protease is shown in Figure 4. The sulfur atom of Cys145 forms a covalent bond to the carbonyl and the chloropyridinyl group acts as a leaving group. The ligand sits in the binding pocket formed from Asn142, Met165, Glu166 and Gln 189. The model shows a similar π-π stacking of the aromatic ring with the imidazole ring of the His 41 residue [29]. Unfortunately, we did not observe the anticipated hydrogen bond between the methoxy group oxygen atom of the ligand and the side chain of the Gln 189 residue. The current study can provide a foundation to design new irreversible inhibitors to target SARS-CoV-2 3CL protease. Our laboratory is actively working on the design and synthesis of potent COVID-19 inhibitors. Based upon the X-ray structure of an irreversible inhibitor (GRL-017-20) bound to SARS-CoV-2 3CL protease (PDB code: 7RBZ), we modeled the complex of inhibitor 9a with SARS-CoV-2 3CL protease [29]. The model of inhibitor 9a bound to the catalytic Cys 145 residue in the active site of the 3CL protease is shown in Figure 4. The sulfur atom of Cys145 forms a covalent bond to the carbonyl and the chloropyridinyl group acts as a leaving group. The ligand sits in the binding pocket formed from Asn142, Met165, Glu166 and Gln 189. The model shows a similar π-π stacking of the aromatic ring with the imidazole ring of the His 41 residue [29]. Unfortunately, we did not observe the anticipated hydrogen bond between the methoxy group oxygen atom of the ligand and the side chain of the Gln 189 residue. The current study can provide a foundation to design new irreversible inhibitors to target SARS-CoV-2 3CL protease. Our laboratory is actively working on the design and synthesis of potent COVID-19 inhibitors.

Chemistry
All compounds were purified by column chromatography. Column chromatography was performed using silica gel 230-400 mesh, with a 60 Å pore diameter. Proton Nuclear Magnetic Resonance NMR ( 1 H NMR) spectra and carbon nuclear magnetic resonance ( 13 C

Chemistry
All compounds were purified by column chromatography. Column chromatography was performed using silica gel 230-400 mesh, with a 60 Å pore diameter. Proton Nuclear Magnetic Resonance NMR ( 1 H NMR) spectra and carbon nuclear magnetic resonance ( 13 C NMR) spectra were recorded on Bruker AV-III-400HD and Bruker AVIII-800 spectrometers. Optical rotations were measured on a Rudolph's AUTOPOL-III automatic digital polarimeter with a sodium lamp and are reported as follows: [α]λ T • C (c = g/100 mL, solvent). High-resolution mass spectrometry (HRMS) spectra were recorded under positive electron spray ionization (ESI+) using a LTQ Orbitrap Mass Spectrometer at the Purdue University Department of Chemistry Mass Spectrometry Center and an Agilent 6550 Q-TOF LC/MS instrument at the Purdue University Analytical Mass Spectrometry Facility.

IC 50 Value Determination
IC 50 values were determined for compounds that covalently inhibit SARS-CoV-2 3CLpro using our recently described assay [28] and data fitting methods that were derived from our previous work on SARS-CoV 3CLpro and inhibition by chloropyridyl esters [22]. The only differences were that pre-incubation of the enzyme with the compounds was 10 min instead of 20 min. In addition, the Morrison Equation was only used to determine the IC 50 values when they were below 1 µM.

Mass Analysis of Enzyme-Inhibitor Complex
Purified SARS-CoV-2 3CLpro was injected onto a Superdex™ 200 Increase 10/300 GL gel filtration column (GE Healthcare) equilibrated in 20 mM HEPES pH 7.5. Fractions containing pure, active protein were pooled for further analysis. Protein was diluted to a final concentration of 2 µM using 20 mM HEPES pH 7.5 and incubated at room temperature with a final concentration of 20 µM compound 8a. The protein and ligand were incubated together for ten minutes before analysis.
Analysis of the proteins was performed on a 6550 iFunnel Q-TOF LC/MS (Agilent Technologies, Santa Clara, CA, USA). A sample (6 ul) was injected on to a Zorbax Extend C18 column (Agilent Technologies) kept at 60 degrees C. The mobile phase consisted of B = acetonitrile and A = 0.1% aqueous formic acid. The flow rate was 0.4 mL/min with a gradient as follows: 0-2 min 3% B; 2-7 min 95% B; 7-9 min 95% B; 9-11 min 3% B. For the first 2 min of the analysis, the column flow was diverted off to waste. TOF MS conditions: