High-throughput modular click chemistry synthesis of catechol derivatives as covalent inhibitors of SARS-CoV-2 3CL ^pro

Abstract

The 3C-like protease (3CL^pro) is a crucial target in anti-Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) drug design. Herein, we performed high-throughput synthesis of catechol derivatives from the bioactive catechol-terminal alkyne scaffold A4, by using modular click chemistry. Subsequently, we conducted two rounds of SARS-CoV-2 3CL^pro inhibition screening and selected seven compounds for synthesis and further efficacy validation. Compound P1-E11 had potent inhibitory effects toward SARS-CoV-2 3CL^pro (IC₅₀ = 2.54 ± 0.46 μM); exhibited good selectivity toward the human cysteine proteases cathepsins B and L; and demonstrated superior anti-SARS-CoV-2 potency (EC₅₀ = 4.66 ± 0.58 μM) with low cytotoxicity (CC₅₀ > 100 μM) in A549-hACE2-TMPRSS2 cells. The irreversible covalent mechanism of P1-E11 was confirmed through time-dependent experiments, enzyme kinetic studies, and dilution and dialysis assays. The binding affinity between P1-E11 and SARS-CoV-2 3CL^pro with a K_D value of 0.57 μM was validated through surface plasmon resonance (SPR) experiments. Molecular docking provided insights into the binding mode of P1-E11 to the target protein. This study demonstrated the feasibility and efficacy of modular click reactions in natural-product-based structural modifications and presents a novel approach for leveraging this strategy in antiviral drug discovery.

Main article text

1. INTRODUCTION

SARS-CoV-2, the viral agent responsible for the global coronavirus disease 2019 (COVID-19) pandemic, has resulted in approximately 770 million infections and 6.96 million deaths worldwide to date, thus posing a serious threat to human life and health [1]. Although several vaccines have been developed that effectively curb the spread of COVID-19 and decrease the number of severe cases, the emergence of high-frequency mutations in the virus has led to potential drug resistance and declining vaccine efficacy [2, 3]. Given the limitations in vaccine distribution and the need for improved vaccines targeting these mutations, the discovery of novel and potent antiviral treatments remains crucial to fight against both the current COVID-19 pandemic and future viral outbreaks.

The 3C-like protease (3CL^pro) substantially contributes to the life cycle of SARS-CoV-2 by cleaving 11 sites on polyprotein precursors (pp1a and pp1ab), thereby producing multifunctional non-structural proteins and facilitating viral replication [4, 5]. Importantly, the 3CL protease is relatively evolutionarily conserved in pathogenic β-coronaviruses, thus decreasing its susceptibility to mutation. Consequently, 3CL protease inhibitors may achieve broad-spectrum anti-coronavirus activity, even against evolving strains. In addition, the low sequence similarity of 3CL protease to human proteases with similar functions minimizes the risk of off-target effects [6, 7]. Therefore, 3CL^pro has emerged as a critical target for developing broad-spectrum anti-coronaviral drugs. In pursuit of this goal, scientists have used various strategies, including drug repurposing, high-throughput physical screening, virtual screening, and AI-assisted drug design, to develop 3CL protease inhibitors, including substrate-derived peptides, and covalent and non-covalent small-molecule inhibitors. These inhibitors are considered necessary additions to the COVID-19 therapeutic armamentarium [8–16]. However, the traditional process of developing new small-molecule drugs is resource-intensive and time-consuming: 3−5 years is typically required for candidates to advance to clinical trials [17, 18]. Given the urgent public health crisis posed by rapidly spreading and highly infectious pathogens such as SARS-CoV-2, a pressing need exists for strategies to expedite the development of antiviral drugs.

In recent years, rapid assembly in microtiter plates by using click chemistry and in situ high-throughput screening has emerged as an efficient approach for establishing structure-activity relationships and discovering bioactive molecules [18–20]. The concept of click chemistry, initially proposed by Sharpless and coworkers, was aimed at rapidly identifying compounds with desirable properties, novel structures, and specific functions through modular synthesis [21]. Over more than two decades of development, click chemistry has not only made great contributions to organic synthesis but also become a critical concept across fields including drug discovery, biomedical applications, and materials synthesis [22–25]. Currently, copper (I)-catalyzed azide-alkyne cycloaddition (CuAAC) and sulfur (VI) fluoride exchange (SuFEX) are prominent methods for assembling functional molecules efficiently and reliably [26–28]. Furthermore, in 2019, Dong et al. introduced an efficient diazo-transfer reaction using fluorosulfuryl azide (FSO₂N₃) to generate diverse organic azides from various primary amines. These azides can then undergo CuAAC reactions with specific alkyne compounds directly in microtiter plates without requiring purification or isolation. The conversion rates of the synthesized triazole derivatives exceed 70%, thus rendering them suitable for direct screening of biological activity. This highly predictable high-throughput synthesis method offers a rapid and gentle way for generating biologically active lead molecules [29].

Natural products serve as crucial reservoirs for new chemical-entity drugs in drug discovery, and their structural modification with click chemistry strategies has been extensively documented for various pharmacological activities [30, 31]. In a previous study, we performed a high-throughput fluorescence resonance energy transfer (FRET) based activity screen against a natural product library and consequently identified hydroxytyrosol’s promising inhibitory effects toward SARS-CoV-2 3CL^pro (IC₅₀ = 2.42 ± 0.03 μM). Subsequent investigations indicated that the catechol group is critical for this activity, through covalent binding to SARS-CoV-2 3CL^pro [32]. In this study, we performed modular click chemistry synthesis of catechol derivatives to efficiently obtain 1152 new compounds within a single day. These catechol derivatives were directly subjected to high-throughput screening for SARS-CoV-2 3CL^pro inhibitory efficacy in microtiter plates. The molecules with promising activity were then resynthesized as pure products, and their inhibitory activity was confirmed at the enzyme level. Subsequently, the efficacy of these pure products was validated through in vitro antiviral experiments in an A549-hACE2-TMPRSS2 cell model. The compound P1-E11 exhibited the strongest anti-SARS-CoV-2 effects, with an EC₅₀ value of 4.66 ± 0.58 μM. Mechanistic studies were subsequently performed on P1-E11, thus leading to rapid identification of a novel and potent anti-coronaviral candidate.

2. MATERIALS AND METHODS

2.1 Chemistry

The azide library was prepared as previously described, with a single azide solution at a concentration of approximately 50 mM [29]. The reactions were monitored with a Waters ACQUITY UPLC H-Class system or TLC (HSG F254, Jiangyou). All compounds listed herein were purified via column chromatography on silica gel (300–400 mesh) and visualized under a UV light monitor. ¹H and ¹⁹F NMR spectra were captured with a Bruker Avance III instrument at 400 MHz and 376 MHz, respectively, ¹³C NMR spectra were obtained with a Quantum-I Plus instrument at 101 MHz. The NMR spectra for ¹H, ¹³C, and ¹⁹F were captured at 297 K in CD₃CN, CD₃OD, or DMSO-d ₆; chemical shifts (δ) are expressed in parts per million (ppm), and coupling constants (J) are expressed in hertz (Hz). High-resolution mass spectra were also determined with a Waters Acquity 2D-UPLC/Acquity UPC2/Xevo G2-XS QTOF instrument.

General procedure 1 : construction of 1152 triazole-containing catechol derivatives

A total of 96 azide solutions from the azide library, each at a 50 mM concentration in DMSO, were added in 10 μL volumes to each 350 μL well of a 96-well microtiter plate. To each well of the newly loaded plate, catechol-terminal alkyne solution (25 mM, 20 μL per well) was added, followed by CuSO₄/THPTA (1:1) aqueous solution (2.5 mM, 10 μL per well) and NaAsc aqueous solution (250 mM, 10 μL per well). Each well contained approximately 50 μL mixture. The plate was sealed tightly and then shaken at 800 rpm at 40°C for 12 h.

General procedure 2 : preparation of target compounds

To a solution of amine substrate (0.5–1 mmol) and DMF (1 mmol/10 mL), KHCO₃ solution (3 M aqueous solution, 4 equiv.) and subsequently FSO₂N₃ solution (MTBE solution, 1 equiv.) were added. The resulting mixture was agitated at ambient temperature and monitored with TLC or LC/MS. Subsequently, NaAsc solution (1 equiv.), alkyne substrate (0.8 equiv.) and CuSO₄/THPTA (1:1) solution (0.04 equiv.) were sequentially added to the crude azide solution. The resulting mixture was agitated at ambient temperature and monitored by LC/MS. After completion of the reaction, the mixture was diluted with 30 mL ethyl acetate, then rinsed with water (50 mL × 3) and saline (50 mL × 2). The organic phase was dehydrated with anhydrous sodium sulfate and condensed. The crude product was refined through column chromatography (10:1 CH₂Cl₂:CH₃OH) to yield the intended product.

2.1.1 4-(2-((1-(4-(trifluoromethoxy)phenyl)-1H-1,2,3-triazol-4-yl)methoxy)ethyl) benzene-1,2-diol (P1-E11)

General procedure 2 was followed with 4-(trifluoromethoxy)aniline (88.6 mg, 0.5 mmol), thus yielding a brown oil (150.8 mg, 80% yield). ¹H NMR (400 MHz, CD₃CN) δ 8.12 (s, 1H), 7.90–7.84 (m, 2H), 7.52–7.45 (m, 2H), 6.72–6.69 (m, 2H), 6.67 (s, 1H), 6.63–6.56 (m, 2H), 4.63 (s, 2H), 3.69 (t, J = 6.0 Hz, 2H), 2.74 (t, J = 6.0 Hz, 2H). ¹³C NMR (101 MHz, CD₃CN) δ 150.0, 147.4, 145.8, 144.3, 137.2, 132.7, 123.8, 123.6, 123.2, 122.0, 121.9 (q, J = 257 Hz), 118.8, 117.4, 116.6, 72.8, 64.9, 36.5. ¹⁹F NMR (376 MHz, CD₃CN) δ −57.51. HRMS (ESI): calcd for C₁₈H₁₇N₃O₄F₃ ⁺ ([M+H]⁺) 396.1171, found 396.1169.

2.1.2 4-(2-((1-(4-chloro-2-hydroxyphenyl)-1H-1,2,3-triazol-4-yl)methoxy)ethyl) benzene-1,2-diol (P5-H11)

General procedure 2 was followed with 2-amino-5-chlorophenol (71.8 mg, 0.5 mmol), thus yielding a brown solid (63.9 mg, 35% yield). ¹H NMR (400 MHz, CD₃OD) δ 8.30 (s, 1H), 7.64 (d, J = 8.0 Hz, 1H), 7.08 (d, J = 4.0 Hz, 1H), 7.02 (dd, d, J₁ = 8.0 Hz, 1H, d, J₂ = 4.0 Hz, 1H), 6.68–6.63 (m, 1H), 6.55–6.49 (m, 1H), 4.66 (s, 2H), 3.70 (t, J = 8.0 Hz, 1H), 2.74 (t, J = 8.0 Hz, 2H). ¹³C NMR (101 MHz, CD₃OD) δ 151.8, 146.1, 144.6, 136.4, 131.7, 127.1, 126.6, 124.9, 121.2, 121.1, 118.0, 117.0, 116.2, 73.0, 64.6, 36.5. HRMS (ESI): calcd for C₁₇H₁₇N₃O₄ ³⁵Cl⁺ ([M+H]⁺) 362.0908, found 362.0903.

2.1.3 4-(2-((1-((1H-indol-3-yl)methyl)-1H-1,2,3-triazol-4-yl)methoxy)ethyl)benzene-1,2-diol (P7-H10)

General procedure 2 was followed with (1H-indol-3-yl)methanamine (73.1 mg, 0.5 mmol), thus yielding a yellow oil (67.2 mg, 37% yield). ¹H NMR (400 MHz, CD₃CN) δ 9.44 (br, 1H), 7.55 (s, 1H), 7.51–7.39 (m, 3H), 7.19–7.12 (m, 1H), 7.08–7.01 (m, 1H), 6.77 (br, 1H), 6.71–6.62 (m, 3H), 6.52–6.47 (m, 1H), 5.68 (s, 2H), 3.55 (t, J = 8.0 Hz, 2H), 2.64 (t, J = 8.0 Hz, 2H). ¹³C NMR (101 MHz, CD₃CN) δ 145.7, 145.3, 143.8, 137.5, 132.2, 127.2, 126.3, 123.9, 123.2, 121.4, 120.7, 119.2, 118.4, 116.8, 116.1, 112.7, 110.2, 72.1, 64.4, 46.5, 35.9. HRMS (ESI): calcd for C₂₀H₂₁N₄O₃ ⁺ ([M+H]⁺) 365.1614, found 364.1609.

2.1.4 4-(2-((1-(2-(piperidin-1-yl)ethyl)-1H-1,2,3-triazol-4-yl)methoxy)ethyl)benzene-1,2-diol (P9-C9)

General procedure 2 was followed with N-(2-aminoethyl)piperidine dihydrochloride (201.1 mg, 1 mmol), thus yielding a brown oil (81.1 mg, 29% yield). ¹H NMR (400 MHz, DMSO-d ₆) δ 8.78 (br, 1H), 7.99 (s, 1H), 6.64–6.56 (m, 2H), 6.46–6.40 (s, 1H), 4.50 (s, 2H), 4.43 (t, J = 8.0 Hz, 2H), 3.53 (t, J = 8.0 Hz, 2H), 2.68 (t, J = 8.0 Hz, 2H), 2.62 (t, J = 8.0 Hz, 2H), 2.40–2.32 (m, 4H), 1.48–1.40 (m, 4H), 1.39–1.31 (m, 2H). ¹³C NMR (101 MHz, CD₃CN) δ 145.6, 145.3, 144.0, 132.0, 124.8, 121.3, 117.0, 116.1, 72.1, 64.4, 58.7, 55.0, 48.1, 36.0, 26.4, 24.8. HRMS (ESI): calcd for C₁₈H₂₇N₄O₃ ⁺ ([M+H]⁺) 347.2083, found 347.2076.

2.1.5 (S)-4-(2-((1-((tetrahydrofuran-2-yl)methyl)-1H-1,2,3-triazol-4-yl)methoxy)ethyl) benzene-1,2-diol (P10-G11)

General procedure 2 was followed with (S)-(tetrahydrofuran-2-yl)methanamine (50.5 mg, 0.5 mmol), thus yielding a light-yellow oil (59.1 mg, 37% yield). ¹H NMR (400 MHz, CD₃CN) δ 7.87 (br, 2H), 7.75–7.69 (m, 1H), 6.73–6.62 (m, 2H), 6.65–6.47 (m, 1H), 4.57–4.50 (m, 2H), 4.47–4.38 (m, 1H), 4.37–4.27 (m, 1H), 4.24–4.14 (m, 1H), 3.82–3.57 (m, 4H), 2.06–1.95 (m, 1H), 1.90–1.69 (m, 2H), 1.65–1.51 (m, 1H). ¹³C NMR (101 MHz, DMSO-d ₆) δ 145.0, 144.0, 143.6, 129.6, 124.5, 119.5, 116.3, 115.5, 76.8, 71.0, 67.5, 63.3, 53.1, 35.0, 28.2, 25.1. HRMS (ESI): calcd for C₁₈H₂₇N₄O₃ ⁺ ([M+H]⁺) 347.2083, found 347.2076.

2.1.6 2-(4-((3,4-dihydroxyphenethoxy)methyl)-1H-1,2,3-triazol-1-yl)-1-(4-hydroxyphenyl)ethan-1-one (P10-B12)

General procedure 2 was followed with 2-amino-1-(4-hydroxyphenyl)ethan-1-one (93.8 mg, 0.5 mmol), thus yielding a yellow oil (67.5 mg, 36% yield). ¹H NMR (400 MHz, CD₃CN) δ 9.94 (br, 1H), 7.96–7.88 (m, 2H), 7.82 (br, 1H), 7.75 (s, 1H), 7.71 (br, 1H), 6.98–6.95 (m, 2H), 6.72–6.66 (m, 2H), 6.57–6.50 (m, 1H), 5.85 (s, 2H), 4.59 (s, 2H), 3.66 (t, J = 8.0 Hz, 2H), 2.72 (t, J = 8.0 Hz, 2H). ¹³C NMR (101 MHz, DMSO-d ₆) δ 190.1, 163.0, 145.0, 143.9, 143.6, 130.9, 130.6, 129.6, 125.7, 125.7, 119.5, 116.3, 115.6, 115.6, 71.1, 63.3, 55.3, 48.7, 35.0. HRMS (ESI): calcd for C₁₉H₂₀N₃O₅ ⁺ ([M+H]⁺) 370.1043, found 370.1339.

2.1.7 4-(2-((1-((4-(trifluoromethyl)pyridin-2-yl)methyl)-1H-1,2,3-triazol-4-yl)methoxy) ethyl)benzene-1,2-diol (P10-G12)

General procedure 2 was followed with (4-(trifluoromethyl)pyridin-2-yl)methanamine (88.0 mg, 0.5 mmol), thus yielding a brown oil (156.8 mg, 79% yield). ¹H NMR (400 MHz, CD₃CN) δ 8.76 (d, J = 4.0 Hz, 1H), 7.78 (s, 1H), 7.61–7.52 (m, 2H), 6.71–6.67 (m, 2H), 6.55 (dd, J ₁ = 8.0 Hz, J ₂ = 2.0 Hz, 1H), 5.73 (s, 2H), 4.55 (s, 2H), 3.63 (t, J = 8.0 Hz, 2H), 2.70 (t, J ₁ = 8.0 Hz, 2H). ¹³C NMR (101 MHz, CD₃CN) δ 157.6, 152.0, 146.0, 145.4, 143.8, 139.5 (q, J = 34 Hz), 132.2, 125.4, 123.8 (q, J = 273 Hz), 121.4, 120.0 (q, J = 4 Hz), 119.0 (q, J = 4 Hz), 116.8, 116.1, 72.2, 64.4, 55.5, 36.0. ¹⁹F NMR (376 MHz, CD₃CN) δ −64.22. HRMS (ESI): calcd for C₁₈H₁₇N₄O₃F₃ ⁺ ([M+H]⁺) 395.1331, found 395.1327.

2.2 Expression and purification of SARS-CoV-2 3CL^pro and SARS-CoV 3CL^pro

The Smt3-SARS-CoV-2 3CL^pro fusion was inserted into the pET29a(+) vector, which was used to transform E. coli BL21 (DE3) cells for protein expression. The cells were cultured in a shaker (37°C, 220 rpm). When the OD600 reached 0.6–0.8, the temperature was lowered to 16°C. IPTG (0.4 M) was then added to a final concentration of 0.4 mM (at a ratio of 1:1000) to induce overexpression, and the cells were incubated at 16°C with shaking at 220 rpm for 16 h. The cells were then centrifuged at 4000 rpm at 4°C for 15 min to collect the precipitated bacteriophages, and frozen at −80°C for long-term storage. Before protein purification, the bacteria were lysed on ice and resuspended in buffer A (20 mM Tris, pH 8.0, and 300 mM NaCl) to ensure complete mixing. The bacteria were then disrupted under high pressure in a homogenizer until the lysate was no longer viscous. The lysate was collected and centrifuged at 18000 rpm for 30 min with an ultra-fast centrifuge, and the supernatant was subsequently purified. First, Ni-NTA affinity chromatography was used; the lysate supernatant was passed through the packed column at a flow rate of 1 mL/min, to allow the target protein to tightly bind the matrix. Subsequently, buffer A containing various concentrations of imidazole (0 mM, 50 mM, or 100 mM imidazole) was used for elution, to remove unbound heterogeneous proteins and non-specific proteins that were not tightly bound. The target protein was obtained by using buffer A containing 300 mM imidazole. The target protein was then concentrated to a volume of 1–2 mL by centrifugation at 4000 rpm through a 10 KDa ultrafiltration tube. The target proteins were purified with a HiLoad 16/600 Superdex 200 pg (Cytiva) column and buffer B (50 mM Tris pH 7.5, 25 mM NaCl). Finally, the proteins were concentrated by centrifugation to 5–25 mg/mL.

The procedure for expression of SARS-CoV 3CL^pro was as described above. The procedure for purification of SARS-CoV 3CL^pro was as follows. Before protein purification, the bacteria were first suspended in phosphate buffer (50 mM sodium phosphate, pH 7.4, 500 mM NaCl, 10% glycerol) and ultrasonically disrupted in an ice bath (200 V, 5 s + 5 s, 30 min). The bacteria were then disrupted under high pressure in a homogenizer until the lysate was no longer viscous. The lysate was collected and centrifuged at 18000 rpm for 30 min with an ultra-fast centrifuge, and the supernatant was subsequently purified. First, Ni-NTA affinity chromatography was used; the lysate supernatant was passed through the packed column at a flow rate of 1 mL/min, to allow the target protein to tightly bind the matrix. Subsequently, phosphate buffer containing various concentrations of imidazole (0 mM, 20 mM, or 50 mM imidazole) was used for elution, to remove unbound heterogeneous proteins and non-specific proteins that were not tightly bound. The target protein was obtained by using phosphate buffer containing 150 mM imidazole. Subsequently, the target protein was concentrated to a volume of 1–2 mL by centrifugation at 4000 rpm through a 10 KDa ultrafiltration tube. The target proteins were then purified with a HiLoad 16/600 Superdex 200 pg (Cytiva) column and SEC buffer (50 mM Tris, pH 7.30, 1 mM EDTA). Finally, the proteins were concentrated by centrifugation to 5–25 mg/mL.

2.3 Investigation of enzymatic inhibition of SARS-CoV-2 3CL^pro and SARS-CoV 3CL^pro with FRET

The relative enzymatic activity of SARS-CoV-2 3CL^pro and SARS-CoV 3CL^pro in vitro was evaluated through FRET with the fluorescent substrate Dabcyl-KNSTLQSGLRKE-Edan in a reaction system of 100 μL. Various concentrations of compounds were diluted with enzyme buffer (50 mM Tris-HCl, pH 7.3, and 1 mM EDTA) and incubated at 37°C for 60 min. The specific procedures were as follows: 10 μL SARS-CoV-2 3CL^pro or SARS-CoV 3CL^pro with a final concentration of 200 nM, 70 μL enzyme buffer, and 10 μL compound solution with different concentrations were sequentially added to black 96-well plates. After the reaction was completed, 10 μL fluorescent substrate (final concentration of 25 μM) was added to each well, and continuous determination (λex = 340 nm, λem = 490 nm) was performed for 20 min on a Tecan Spark 10 M multimode microplate reader. The fluorescence intensity of the hydrolysate was determined, and the assay was repeated twice for each compound. The collected data were analyzed in GraphPad Prism 8.0 software. The formula for calculation of inhibitory enzymatic activity was as follows:

$$\begin{array}{l}\text{Inhibitory\hspace{0.17em}enzymatic\hspace{0.17em}activity\hspace{0.17em}}(\text{\%})\\ {\text{=\hspace{0.17em}(RFU}}_{\text{2}}-{\text{RFU}}_{\text{1}}{\text{)/(RFU}}_{\text{2}}-{\text{RFU}}_{\text{0}}\text{)\hspace{0.17em}×\hspace{0.17em}100}\end{array}$$

where RFU₂ is the fluorescence intensity without the test compound (maximum enzymatic activity response value), RFU₁ is the fluorescence intensity of the test compound (sample response value), and RFU₀ is the fluorescence intensity of a blank well.

2.4 Inhibitory activity of 1152 catechol derivatives toward SARS-CoV-2 3CL^pro

The 1152 catechol derivatives (in 12 96-well plates) efficiently constructed through the CuAAC reaction were directly transferred for inhibitory activity testing without purification and separation. Concentrations of 50 mM in a total volume of 50 μL were placed in each well. First, the inhibitory activity of the compounds toward SARS-CoV-2 3CL^pro was assayed in the plates at a concentration of 3 μM. The compounds to be measured were placed in transparent 96-well plates at 30 μM dilution. Subsequently, 10 μL diluted solution was transferred directly to black 96-well plates containing 10 μL SARS-CoV-2 3CL^pro at a final concentration of 200 nM in 70 μL enzyme buffer. The reaction was preincubated for 60 min at 37°C. Subsequently, 10 μL fluorescent substrate (final concentration of 25 μM) was added to each well, and continuous determination (λex = 340 nm, λem = 490 nm) was performed for 20 min on a Tecan Spark 10 M multimode microplate reader. The fluorescence intensity of the hydrolysate was determined and analyzed in GraphPad Prism 8.0 software.

2.5 Inhibitory mechanism of P1-E11

The procedure for P1-E11 time-dependent inhibition assays was as described in Section 2.3, except that incubation was performed for 5 min, 15 min, 30 min, or 60 min. The residual enzymatic activity was calculated as follows:

$$\begin{array}{l}\text{Residual\hspace{0.17em}enzymatic\hspace{0.17em}activity\hspace{0.17em}}(\text{\%})\\ \text{=\hspace{0.17em}1}-\text{inhibitory\hspace{0.17em}enzymatic\hspace{0.17em}activity\hspace{0.17em}}(\text{\%})\text{.}\end{array}$$

The procedure for analysis of the inhibition kinetics of SARS-CoV-2 3CL^pro by P1-E11 was as previously reported [32].

2.6 Cysteine protease enzymatic assays

Enzymatic responses toward other cysteine proteases (SARS-CoV-2 PL^pro, cathepsin B, and cathepsin L) were measured as previously described [32]. All enzymatic reactions were conducted in duplicate at room temperature for 30 min in 50 μL mixtures containing 50 mM MES, pH 5.00, 100 mM NaCl, 0.05% CHAPS, 5 mM DTT, cathepsin B substrate at 2 μM final concentration, cathepsin B enzyme at 1 ng/well, and test compound. Fluorescence intensity was measured with a Tecan Infinite M1000 microplate reader with excitation at 400 nm and emission at 505 nm. Protease activity assays were performed in duplicate at each concentration. The fluorescence intensity data were analyzed in GraphPad Prism software version 8.0. Cathepsin L used the corresponding substrate, and the determination method was as described above.

2.7 Dilution assays

The dilution assays on SARS-CoV-2 3CL^pro were conducted as previously reported [32].

2.8 Dialysis experiments

The dialysis experiments on SARS-CoV-2 3CL^pro were conducted as previously reported [32].

2.9 Cytotoxicity assays

A549-hACE2-TMPRSS2 cells were first seeded in 96-well plates in volumes of 100 μL per well. After the cells had attached, various drug concentrations in drug diluent containing 2 μM efflux inhibitor elacridar were sequentially added. After 24 h of incubation at 37 °C, the old culture medium was removed, and 10% Cell Counting Kit-8 (CCK-8) reagent was added to each well. The absorbance at 450 nm was measured after culture at 37°C for 2 h to calculate the cytotoxicity of each drug.

2.10 Immunofluorescence assays

A549-hACE2-TMPRSS2 cells were seeded in 96-well plates, allowed to adhere, and then infected with wild-type SARS-CoV-2 that had been isolated from patients with COVID-19 and stored in the BSL-3 Laboratory of the Guangzhou Customs District Technical Center. The experiment was conducted using a full time dosing treatment, and the drug was diluted in DMEM containing 2% FBS. Before infection, the cells were inoculated with the indicated drug concentration and incubated for 2 h. Subsequently, the cells were infected at a multiplicity of infection of 0.03 for 1 h. The inoculator was subsequently removed, 100 μL drug at the indicated concentration was added to each well, and the cells were cultured for 24 h. In this experiment, elacridar, an efflux inhibitor, was added to all diluents at 2 μM. Subsequent immunofluorescence treatment was conducted as previously reported [33].

2.11 Surface plasmon resonance assays

The binding affinity of P1-E11 toward SARS-CoV-2 3CL^pro was assessed with established methods [34]. Notably, the immobilization level of the binding protein achieved 15822.5 resonance units. At a flow rate of 30 mL/min, the contact duration lasted 120 s, whereas the dissociation period extended to 180 s.

2.12 Determination of metabolic stability in liver microsomes

Microsomes in 0.1 M Tris buffer, pH 7.4 (final concentration 0.33 mg/mL), the co-factor MgCl₂ (final concentration 5 mM), and the tested compound (final concentration 0.1 μM, in 0.01% DMSO and 0.005% bovine serum albumin) were incubated at 37°C for 10 min. The reaction was started by the addition of NADPH (1 mM final concentration). Aliquots were sampled at 0, 5, 15, 30, and 60 min, and methanol (cold in wet ice) was added to terminate the reaction. After centrifugation (4000 rpm,5 min), samples were then analyzed with LC-MS/MS.

2.13 Molecular docking

The covalent docking between P1-E11 and SARS-CoV-2 3CL^pro was performed as previously reported [32].

2.14 System setup for molecular dynamics simulations

Molecular dynamics (MD) simulations between P1-E11 and SARS-CoV-2 3CL^pro were performed as previously reported [32].

2.15 Statistical analysis

Statistical analysis was performed in GraphPad Prism 8.0. Experiments were performed in triplicate or as indicated, and all data are presented as mean ± standard deviation.

3. RESULTS

3.1 Design strategy

In our previous study, we discovered that oleuropein is a potent inhibitor of SARS-CoV-2 3CL^pro (IC₅₀ = 4.18 ± 0.19 μM) [32]. Subsequent structure-activity studies revealed that hydroxytyrosol is important for this activity (IC₅₀ = 2.42 ± 0.03 μM). Notably, the probe A4, which was prepared and used in the covalent mechanism experiments, showed enhanced inhibitory effects toward SARS-CoV-2 3CL^pro (IC₅₀ = 1.27 ± 0.07 μM) ( Figure 1A and 1B ). Antiviral studies indicated that compound A4 exhibited the most potent anti-SARS-CoV-2 activity (EC₅₀ = 25.12 ± 0.19 μM) ( Figure 1C ). Accordingly, a class of catechol derivatives with diverse structures, based on the active molecule A4, were efficiently synthesized through a modular click chemistry approach. High-throughput activity screening was directly conducted without purification or isolation, and was followed by preparation of the active molecule and confirmation of its activity ( Figure 2 ).

Figure 1 |

(A) Structures of three compounds. (B) Inhibitory effects of three compounds toward SARS-CoV-2 3CL^pro. (C) Efficacy of hydroxytyrosol and A4 against SARS-CoV-2 in A549-hACE2-TMPRSS2 cells. Data represent two separate experiments.

Figure 2 |

Schematic flow diagram for the efficient discovery of lead compounds based on compound A4, through a modular click chemistry strategy.

3.2 Chemistry

3.2.1 Design and synthesis of 1152 triazole-containing catechol derivatives

The designed catechol derivatives were constructed through CuAAC reactions between terminal alkyne A4 and various azides obtained through diazo-transfer in 96-well microplates. Specifically, as shown in Figure 3 , 1152 azides, derived from primary amines, were added to each well containing alkyne A4. Subsequently, an aqueous solution of CuSO₄/THPTA (a commercially available ligand used in the CuAAC reaction) and aqueous NaAsc solution were introduced. After sealing and reaction under conditions of 800 rpm at 40°C, 1152 triazole-containing catechol derivatives were easily prepared. Meanwhile, five controls were set up to investigate the effects of the reaction system: a 10 mM alkyne DMSO solution, 0.5 mM CuSO₄ aqueous solution, 0.5 mM THPTA aqueous solution, 0.5 mM CuSO₄/THPTA (1:1) aqueous solution, a mixture of alkyne (10 mM), CuSO₄/THPTA (1:1, 0.5 mM), and NaAsc (50 mM).

Figure 3 |

Modular click chemistry for the preparation of 1152 triazole-containing catechol derivatives.

3.2.2 Preparation of hit compounds

The desired compounds were synthesized via a two-step protocol of diazo-transfer-CuAAC reactions, as shown in Scheme 1 . Initially, primary amines served as starting materials for the diazo-transfer reaction with KHCO₃ solution and FSO₂N₃ solution (MTBE solution) in DMF solvent. The resulting azides, requiring no purification, were used directly in their solution form with the alkyne derivative A4 in the subsequent CuAAC reaction, to yield the target compounds.

Scheme 1 |

Synthesis of hit compounds.

3.3 Biological assays

3.3.1 Screening of 1152 triazole-containing catechol derivatives

The 1152 obtained catechol derivatives were used directly for high-throughput activity screening of SARS-CoV-2 3CL^pro without purification. First, the control experiment demonstrated that the reaction ligand did not influence enzymatic activity ( Figure 4A ). Subsequently, we tested 1152 compounds across 12 plates for their inhibitory activity toward SARS-CoV-2 3CL^pro at 3 μM. As depicted in Figure 4B , we identified 47 compounds with >90% inhibition of SARS-CoV-2 3CL^pro at this concentration. We then subjected these 47 compounds to concentration-dependent inhibition testing and discovered 20 compounds with IC₅₀ values below 1.30 μM ( Figure 4C ).

Figure 4 |

(A) Inhibition rates of five controls toward SARS-CoV-2 3CL^pro. (B) Screening of 1152 compounds in 12 plates for inhibitory effects toward SARS-CoV-2 3CL^pro at concentrations of 3 μM. (C) Screening of 47 compounds for concentration-dependent inhibition of SARS-CoV-2 3CL^pro.

3.3.2 Inhibitory activity studies of seven selected catechol derivatives toward SARS-CoV-2 3CL^pro and SARS-CoV 3CL^pro

Through structural analysis of the 20 screened catechol derivatives, we selected seven catechol derivatives representing various structural types, including aryl, heteroaryl, and heterocyclic compounds, for subsequent preparation. The inhibitory activity of these seven catechol derivatives was reassessed by using myricetin, a covalent inhibitor of SARS-CoV-2 3CL^pro with a polyphenolic backbone, as a control ( Table 1 ). It was observed that these pure compounds did not significantly improve the inhibitory activity against SARS-CoV-2 3CL^pro, with IC₅₀ values ranging from 2.54 to 5.37 μM. Additionally, we evaluated the inhibitory potential of these seven catechol derivatives toward SARS-CoV 3CL^pro and found that all these compounds demonstrated robust inhibitory effects, with IC₅₀ values of 1.82–2.92 μM. Myricetin showed strong inhibitory activity toward SARS-CoV-2 3CL^pro and SARS-CoV 3CL^pro, with IC₅₀ values of 0.43 ± 0.02 μM and 0.18 ± 0.03 μM, respectively. Remarkably, the inhibitory activity of these compounds against SARS-CoV 3CL^pro was similar to that of SARS-CoV-2 3CL^pro.

Table 1 |

Inhibitory potential of seven catechol derivatives toward SARS-CoV-2 3CL^pro and SARS-CoV 3CL^pro.

Compound	R	IC50 ± SD (μM)
		SARS-CoV-2 3CLpro (well plate)	SARS-CoV-2 3CLpro (pure compound)	SARS-CoV 3CLpro (pure compound)
P1-E11		1.30 ± 0.03	2.54 ± 0.46	2.24 ± 0.35
P5-H11		0.87 ± 0.13	2.82 ± 0.12	1.72 ± 0.02
P7-H10		1.14 ± 0.07	4.29 ± 0.35	2.72 ± 0.57
P9-C9		1.24 ± 0.03	3.06 ± 0.13	1.82 ± 0.01
P10-G11		1.18 ± 0.04	5.18 ± 0.01	2.57 ± 0.06
P10-B12		1.04 ± 0.04	5.37 ± 0.23	2.92 ± 0.29
P10-G12		0.85 ± 0.03	4.25 ± 0.08	2.65 ± 0.45
A4	/	/	1.27 ± 0.07	1.24 ± 0.04
Myricetin		/	0.43 ± 0.02	0.18 ± 0.03

3.3.3 Profiling of target selectivity toward cathepsin B and cathepsin L

Off-target effects are a substantial challenge associated with covalent inhibitors, and may prompt safety concerns limiting their widespread clinical application. To further assess the target selectivity of seven catechol derivatives, we evaluated their inhibitory activity toward two crucial cysteine proteases, cathepsin B and cathepsin L, and used myricetin as a control. All seven catechol derivatives demonstrated weak inhibitory activity toward cathepsins B and L at a concentration of 5 μM ( Figure 5 ). Similarly, myricetin exhibited weak inhibitory activity at a concentration of 10 μM. These results suggested that the compounds had good selectivity, thus potentially indicating favorable safety profiles.

Figure 5 |

Target selectivity of SARS-CoV-2 3CL^pro inhibitors toward cathepsin B and cathepsin L.

3.3.4 Cell-based antiviral activity studies

The antiviral effectiveness of these seven catechol derivatives toward wild-type SARS-CoV-2 in A549 lung cancer cells expressing human ACE2 and TMPRSS2 was further assessed. Initially, the antiviral efficacy of the seven compounds was determined at a concentration of 25 μM with full time treatments. Most compounds exhibited sustained anti-SARS-CoV-2 activity, as compared with that of compound A4 ( Figure 6A ). In particular, the antiviral activity of compound P1-E11 was significantly enhanced compared to compound A4, while the antiviral activities of P10-B12 and P10-G11 were weaker. Subsequently, we assessed the antiviral activities of these five compounds at various concentrations. Four compounds (P5-H11, P7-H10, P9-C9, and P10-G12) showed antiviral activities comparable to that of compound A4 (EC₅₀ = 25.12 ± 1.85 μM), with EC₅₀ values of 21.67–37.44 μM ( Figure 6B ). Notably, the antiviral efficacy of compound P1-E11 was significantly enhanced, exhibiting an EC₅₀ value of 4.66 ± 0.58 μM, approximately 6-fold higher than that of compound A4. Furthermore, we conducted CCK-8 assays to evaluate the cytotoxicity of these compounds toward A549-hACE2-TMPRSS2 cells. The CC₅₀ values were above 60 μM for all five compounds and were lower than that of compound A4. Notably, compound P1-E11 exhibited the most potent antiviral activity and the least cytotoxicity, with a CC₅₀ value exceeding 100 μM. Given its superior antiviral activity and low toxicity, compound P1-E11 was selected for further mechanistic studies.

Figure 6 |

(A) Antiviral efficacy of these derivatives toward SARS-CoV-2, assessed through immunofluorescence assays at 25 μM with full time treatments. (B) Cell viability effects of five catechol derivatives on A549-hACE2-TMPRSS2 cells and antiviral efficacy evaluations of five catechol derivatives toward SARS-CoV-2 in A549-hACE2-TMPRSS2 cells with full time treatments. Data represent two separate experiments.

3.3.5 Exploration of the covalent binding mechanism of P1-E11

3.3.5.1 Time-dependent inhibition of P1-E11 toward SARS-CoV-2 3CL^pro

A time-dependent inhibition experiment was conducted on SARS-CoV-2 3CL^pro to clarify the mechanism of action of compound P1-E11. SARS-CoV-2 3CL^pro was assayed with P1-E11 for 5 min, 30 min, or 60 min ( Figure 7A ). The efficacy of P1-E11 in inhibiting SARS-CoV-2 3CL^pro increased as the pre-incubation period was prolonged. The IC₅₀ values at these time points were 13.30 ± 2.28 μM, 5.09 ± 0.72 μM, and 2.54 ± 0.46 μM, respectively. Thus, compound P1-E11 significantly inhibited SARS-CoV-2 3CL^pro in a time- and concentration-dependent manner consistent with the modes of action of covalent inhibitors.

Figure 7 |

(A) Time-dependent inhibition of SARS-CoV-2 3CL^pro by P1-E11. (B) Illustration of the covalent reaction. (C) Enzymatic kinetics study of P1-E11 toward SARS-CoV-2 3CL^pro through curve fitting.

3.3.5.2 Inhibition kinetics of SARS-CoV-2 3CL^pro by P1-E11

Covalent inhibitors engage with their targets largely through a dual-step process: initially, the inhibitor effectively positions the electrophilic warhead of the inhibitor near the reactive residue of the target, through a combination of non-covalent interactions and target affinity. The binding affinity of the inhibitor at this stage is expressed by the inhibition constant (K _i). Subsequently, a covalent reaction and formation of a covalent complex occurs, with k _inact representing the rate constant ( Figure 7B ) [35]. The kinetic constant k _inact/K _i for covalent inhibition is generally used to accurately gauge the efficacy of covalent inhibitors. Hence, we conducted an enzymatic kinetic study for P1-E11 to identify the reversible binding constant K _i as well as the irreversible binding parameter k _inact. P1-E11 exhibited superior reversible binding affinity (K _i = 11.38 μM) and a relatively minor irreversible binding capacity (k _inact = 0.066 min⁻¹) ( Figure 7C ). The measured k _inact/K _i for P1-E11 was 97.20 M⁻¹S⁻¹. These findings indicated that P1-E11 was an effective covalent inhibitor.

3.3.5.3 Dilution and dialysis assays

To further validate the reversible or irreversible covalent inhibition properties of compound P1-E11 toward SARS-CoV-2 3CL^pro, we performed both dilution and dialysis assays. The widely recognized reversible covalent inhibitor PF-07321332 (nirmatrelvir) was used as a positive control. To ensure total suppression in the dilution experiments, we treated SARS-CoV-2 3CL^pro with P1-E11 (10 × IC₅₀ concentration). Subsequently, the mixture was diluted 20 fold, thereby resulting in a 10 × IC₅₀ to 0.5 × IC₅₀ concentration. As illustrated in Figure 8A , the activity of SARS-CoV-2 3CL^pro was not significantly restored, thus suggesting that P1-E11 and the target formed an irreversible covalent bond. Nevertheless, we observed a clear revival in the enzymatic activity of the reversible inhibitor PF-07321332. Simultaneously, we conducted dialysis assays for P1-E11 and SARS-CoV-2 3CL^pro. After dialysis of the mixture for 8 h and sample collection at various intervals (0 h, 2 h, 4 h, and 8 h), we measured the enzymatic activity. The enzymatic activity of P1-E11 did not significantly recover after 8 h of dialysis, whereas the enzymatic activity of PF-07321332 did recover ( Figure 8B ). Collectively, these results confirmed that P1-E11 is an irreversible covalent inhibitor of SARS-CoV-2 3CL^pro.

Figure 8 |

Validation of irreversible covalent inhibition: (A) Effects of P1-E11 and PF-07321332 on SARS-CoV-2 3CL^pro, determined through a dilution assay. (B) Effects of P1-E11 and PF-07321332 on SARS-CoV-2 3CL^pro, determined through a dialysis experiment.

3.3.6 Investigation of the binding affinity between P1-E11 and SARS-CoV-2 3CL^pro

To obtain a deeper understanding of the interactions between P1-E11 and SARS-CoV-2 3CL^pro, we performed surface plasmon resonance (SPR) assays with a Biacore T200 optical biosensor. The SPR binding kinetics indicated that P1-E11 considerably increased the resonance units in a dose sensitive manner, with a K_D value of 0.57 μM ( Figure 9 ). This result demonstrated substantial binding affinity between P1-E11 and SARS-CoV-2 3CL^pro.

Figure 9 |

Surface plasmon resonance sensorgrams of P1-E11 with SARS-CoV-2 3CL^pro.

3.3.7 Metabolic stability of P1-E11 in liver microsomes

Subsequently, we assessed the metabolic stability of P1-E11 in liver microsomes from humans, rats, and mice. Samples of the incubation solution were collected at 0, 5, 15, 30, and 60 minutes, then analyzed by LC-MS/MS. As shown in Table 2 , the half-life (T_1/2) of P1-E11 in human, rat, and mouse liver microsomes was 6.4 minutes, 1.4 minutes, and 1.9 minutes, respectively, thus indicating poor metabolic stability of P1-E11 in liver microsomes.

Table 2 |

Metabolic stability of P1-E11 in liver microsomes. The concentration of P1-E11 was 0.1 μM.

Species	P1-E11 (0.1 μM)
	T1/2 (min)	CLint in vitro (mL/min/g protein)
Human	6.4	326
Rat	1.4	1497
Mouse	1.9	1083

CLint, Intrinsic Clearance.

3.4 Molecular docking

SARS-CoV-2 3CL^pro is usually a homodimer, wherein each protomer of SARS-CoV-2 3CL^pro is composed of three domains (I, II, and III), with four highly conserved subsite substrate binding domains (catalytic dyad) at His41 and Cys145 between domains I and II. Domain III plays an important role in the dimerization of 3CL^pro through salt bridge interactions [36]. To elucidate the interaction mode between P1-E11 and SARS-CoV-2 3CL^pro, we performed covalent docking analysis in the homodimeric state. The catechol structure of P1-E11 can be oxidized to o-benzoquinone, which can be subsequently attacked via the sulfhydryl of Cys145, thereby forming a covalent bond. As shown in Figure 10 , trifluoromethoxy phenyl of P1-E11 is surrounded by a pocket formed by two loops, AA 165–168 and AA 189–192, and the oxygen atom forms a hydrogen bond with Gln192 at a distance of 2.6 Å. Another hydrogen bond is formed between the NH of Gly143 and phenol hydroxyl group of P1-E11. In addition, an amide-π interaction forms between Gln189 and triazole ring of the ligand.

Figure 10 |

Predicted binding mode between P1-E11 and SARS-CoV-2 3CL^pro (PDB ID: 7NBY).

3.5 Molecular dynamics simulation

Further MD simulations were performed to investigate whether the docking model might reveal the real binding pattern of P1-E11. Compared with the equilibrium conformation of apo-3CL^pro, the β-strand around Thr25 of the complex is disordered and shifted toward the pocket ( Figure 11C ). The binding pocket of the covalent complex adopts a closed conformation to fit the compound ( Figure 11A and 11B ). By examining the residues within 5 Å, we observed a H-bond network between the hydroxy of catechol and AA 142–145 of the protein, and a π-π interaction between the side chain of His41 and catechol. No other stable polar interaction was found between the remaining groups and the protein scaffold, indicating the significant role of hydrophobic contact ( Figure 11D ). Collectively, these interactions are critical in enhancing the potent inhibitory activity of P1-E11 toward SARS-CoV-2 3CL^pro.

Figure 11 |

Equilibrium conformation from MD simulation of SARS-CoV-2 3CL^pro (PDB ID: 7NBY) in a dimeric state.

(A-B) Surface view of the binding pocket of P1-E11 (A) compared with apo-SARS-CoV-2 3CL^pro (B). (C) Disordered β-strand in complex (green) compared with apo-protein (yellow). (D) Hydrogen bonding and π-π interactions between P1-E11 and SARS-CoV-2 3CL^pro.

As the dimerization state is close to the canonical binding site of 3CL^pro, we conducted additional MD simulations with the protomer P1-E11 complex model. The equilibrium structure was overlaid with chain A of dimerized apo-3CL^pro. As shown in Figure S2 , the dimerization interface in both structures was nearly consistent, thus indicating that the ligand binding scarcely affects the dimerization of SARS-CoV-2 3CL^pro.

4. DISCUSSION

Since the novel coronavirus outbreak, scientists have used a variety of techniques and strategies to accelerate SARS-CoV-2 research and treatment. Among these methods, modular click chemistry is an efficient combinatorial chemistry method that can be used to rapidly discover potential 3CL^pro inhibitors. In this study, a library of catechol derivatives with diverse structures was efficiently constructed for compound A4 through a modular click chemistry strategy. The inhibitory activity of SARS-CoV-2 3CL^pro was directly screened in a high-throughput manner without a need for purification and isolation. This method is more efficient and faster than traditional drug screening. After two rounds of screening, 20 molecules with promising inhibitory activities were identified. Seven compounds representing different structural types were then selected and synthesized. The inhibitory activities of SARS-CoV-2 3CL^pro were tested, and the overall activity was not significantly improved compared to compound A4. In agreement with predictions by the idrug website, the seven compounds showed good lipophilicity, with cLogP values ranging from 1.63 to 3.34, which were more favorable than that of compound A4 (cLogP = 1.29); therefore, the seven compounds had better pharmacokinetic properties. Moreover, the target selectivity of the seven catechol derivatives was verified. These compounds exhibited good selectivity toward human cysteine proteases cathepsins B and L. Subsequently, we evaluated the toxicity in A549-hACE2-TMPRSS2 cells and found that the introduction of fluorine atoms significantly decreased the cytotoxicity, and the CC₅₀ values of P1-E11 and P10-G12 both exceeded 100 μM, with cLogP values of 3.34 and 2.91, respectively. When tested for anti-SARS-CoV-2 activity, compound P1-E11 (trifluoromethoxyphenyl substitution) displayed better antiviral activity, with an EC₅₀ value of 4.66 ± 0.58 μM, than that of compound P10-G12 (4-(trifluoromethyl)pyridine substitution), which showed lower antiviral activity and an EC₅₀ value of 37.44 ± 3.60 μM, possibly as a result of lower targeting affinity. In addition, we examined the covalent mechanism of the compound P1-E11, which had relatively strong antiviral activity and low cytotoxicity. The covalent properties were preliminarily confirmed by time-dependent tests and enzyme kinetic experiments. In view of the reversibility of the covalent reaction process, P1-E11 was verified as an irreversible covalent inhibitor through dilution and dialysis assays. Subsequently, the metabolic stability of P1-E11 was evaluated in human, rat, and mouse liver microsomes, which indicated that P1-E11 showed poor metabolic stability in human microsomes. Finally, the binding characteristics of P1-E11 toward SARS-CoV-2 3CL^pro were validated through SPR experiments and covalent docking.

However, this study has several limitations. First, 100% conversion of the reaction in high-throughput synthesis cannot be ensured, and some active molecules might have been missed; therefore, more suitable reaction conditions for specific terminal alkyne compounds must be identified to ensure the highest possible conversion of the reaction. Second, because the warhead of catechol has poor metabolic stability, it must be replaced with other covalent warheads with favorable druggability at this specific site. Moreover, we used only molecular docking and MD experiments to simulate the interaction mode between P1-E11 and SARS-CoV-2 3CL^pro. However, the actual interaction mode remains unclear, and will be further investigated in the future through protein crystallization.

In the future, we will focus on optimizing the structure of P1-E11 through a bioisosteric strategy to replace the triazole group, aiming to improve the compound’s physicochemical properties. In addition, we will apply target-based drug design according to the results of MD simulations, and will introduce functional groups into the unoccupied pockets to enhance the binding affinity toward the target. On the basis of the click chemistry reaction’s advantages of high efficiency and convenience, the exploration of structure-property relationships and structure-activity relationships will greatly enhance the value of this method in drug discovery. We will continue to use this method for other targets to facilitate the rapid discovery of novel lead structures for a wide range of targets.

Overall, a range of potential inhibitors of SARS-CoV-2 3CL^pro were efficiently and rapidly screened through a modular click chemistry high-throughput synthesis strategy. Among them, the compound P1-E11 exhibited potent anti-SARS-CoV-2 activity (EC₅₀ = 4.66 ± 0.58 μM) and low cytotoxicity (CC₅₀ > 100 μM). This study demonstrated the feasibility and effectiveness of the modular click chemistry in natural-product-based structural modifications, thereby providing a new approach for antiviral drug discovery.

DECLARATION OF COMPETING INTERESTS

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported herein.

APPENDIX A. SUPPLEMENTARY DATA

Supplementary data for this article can be found online at: https://amm-journal.org/wp-content/uploads/2024/09/amm20240028_Suppl.pdf.

REFERENCES

1. WHO Coronavirus (COVID-19) Dashboard. https://covid19.who.intAccessed on March 18, 2024

2. Kushwaha ND, Mohan J, Kushwaha B, Ghazi T, Nwabuife JC, Koorbanally N, et al.. A Comprehensive Review on the Global Efforts on Vaccines and Repurposed Drugs for Combating COVID-19. European Journal of Medicinal Chemistry. 2023. Vol. 260:115719

3. Yang JY, Liu MQ, Liu L, Li X, Xu MX, Lin HF, et al.. The Protective Nasal Boosting of a Triple-RBD Subunit Vaccine Against SARS-CoV-2 Following Inactivated Virus Vaccination. Signal Transduction and Targeted Therapy. 2023. Vol. 8:151

4. Kakavandi S, Zare I, VaezJalali M, Dadashi M, Azarian M, Akbari A, et al.. Structural and Non-Structural Proteins in SARS-CoV-2: Potential Aspects to COVID-19 Treatment or Prevention of Progression of Related Diseases. Cell Communication and Signaling. 2023. Vol. 21:110

5. Li X, Song YC. Structure and Function of SARS-CoV and SARS-CoV-2 Main Proteases and their Inhibition: A Comprehensive Review. European Journal of Medicinal Chemistry. 2023. Vol. 260:115772

6. Jin ZM, Du XY, Xu YC, Deng YQ, Liu MQ, Zhao Y, et al.. Structure of Mpro from SARS-CoV-2 and Discovery of its Inhibitors. Nature. 2020. Vol. 582:289–293

7. Huang C, Shuai HP, Qiao JX, Hou YX, Zeng R, Xia AJ, et al.. A New Generation Mpro Inhibitor with Potent Activity Against SARS-CoV-2 Omicron Variants. Signal Transduction and Targeted Therapy. 2023. Vol. 8:128

8. Harrison C. Coronavirus Puts Drug Repurposing on the Fast Track. Nature Biotechnology. 2020. Vol. 38:379–381

9. Khare P, Sahu U, Pandey SC, Samant M. Current Approaches for Target-Specific Drug Discovery Using Natural Compounds Against SARS-CoV-2 Infection. Virus Research. 2020. Vol. 290:198169

10. Riva L, Yuan S, Yin X, Martin-Sancho L, Matsunaga N, Pache L, et al.. Discovery of SARS-CoV-2 Antiviral Drugs Through Large-Scale Compound Repurposing. Nature. 2020. Vol. 586:113–119

11. Chang YJ, Le UNP, Liu JJ, Li SR, Chao ST, Lai HC, et al.. Combining Virtual Screening with Cis-/Trans-Cleavage Enzymatic Assays Effectively Reveals Broad-Spectrum Inhibitors that Target the Main Proteases of SARS-CoV-2 and MERS-CoV. Antiviral Research. 2023. Vol. 216:105653

12. Zhou YD, Wang F, Tang J, Nussinov R, Cheng F. Artificial Intelligence in COVID-19 Drug Repurposing. The Lancet Digital Health. 2020. Vol. 2:e667–e676

13. La Monica G, Bono A, Lauria A, Martorana A. Targeting SARS-CoV-2 Main Protease for Treatment of COVID-19: Covalent Inhibitors Structure-Activity Relationship Insights and Evolution Perspectives. Journal of Medicinal Chemistry. 2022. Vol. 65:12500–12534

14. Pang XJ, Xu W, Liu Y, Li H, Chen LX. The Research Progress of SARS-CoV-2 Main Protease Inhibitors from 2020 to 2022. European Journal of Medicinal Chemistry. 2023. Vol. 257:115491

15. Gao SH, Huang TG, Song LT, Xu SJ, Cheng YS, Cherukupalli S, et al.. Medicinal Chemistry Strategies Towards the Development of Effective SARS-CoV-2 Inhibitors. Acta Pharmaceutica Sinica. B. 2022. Vol. 12:581–599

16. Song LT, Gao SH, Ye B, Yang ML, Cheng YS, Kang DW, et al.. Medicinal Chemistry Strategies Towards the Development of Non-Covalent SARS-CoV-2 Mpro Inhibitors. Acta Pharmaceutica Sinica. B. 2024. Vol. 14:87–109

17. Ma YS, Xin R, Yang XL, Shi Y, Zhang DD, Wang HM, et al.. Paving the Way for Small-Molecule Drug Discovery. American Journal of Translational Research. 2021. Vol. 13:853–870

18. Blanco MJ, Gardinier KM. New Chemical Modalities and Strategic Thinking in Early Drug Discovery. ACS Medicinal Chemistry Letters. 2020. Vol. 11:228–231

19. Wang XS, Huang BS, Liu XY, Zhan P. Discovery of Bioactive Molecules from CuAAC Click-Chemistry-Based Combinatorial Libraries. Drug Discovery Today. 2016. Vol. 21:118–132

20. Sun YY, Feng D, Zhou ZZ, Zhang T, De Clercq E, Pannecouque C, et al.. In Situ Click Chemistry-Based Discovery of 1,2,3-Triazole-Derived Diarylpyrimidines as Novel HIV-1 NNRTIs by Exploiting the Tolerant Region I in Binding Pocket. Bioorganic & Medicinal Chemistry. 2023. Vol. 96:117484

21. Kolb HC, Finn MG, Sharpless KB. Click Chemistry: Diverse Chemical Function from a Few Good Reactions. Angewandte Chemie (International ed. in English). 2001. Vol. 40:2004–2021

22. Kaur J, Saxena M, Rishi N. An Overview of Recent Advances in Biomedical Applications of Click Chemistry. Bioconjugate Chemistry. 2021. Vol. 32:1455–1471

23. Kondengadan SM, Bansal S, Yang C, Liu D, Fultz Z, Wang B. Click Chemistry and Drug Delivery: A Bird’s-Eye View. Acta Pharmaceutica Sinica. B. 2023. Vol. 13:1990–2016

24. Bauer D, Sarrett SM, Lewis JS, Zeglis BM. Click Chemistry: A Transformative Technology in Nuclear Medicine. Nature Protocols. 2023. Vol. 18:1659–1668

25. Jiang XY, Hao X, Jing LL, Wu GC, Kang DW, Liu XY, et al.. Recent Applications of Click Chemistry in Drug Discovery. Expert Opinion on Drug Discovery. 2019. Vol. 14:779–789

26. Rani A, Singh G, Singh A, Maqbool U, Kaur G, Singh J. CuAAC-Ensembled 1,2,3-Triazole-Linked Isosteres as Pharmacophores in Drug Discovery: Review. RSC Advances. 2020. Vol. 10:5610–5635

27. Dong JJ, Krasnova L, Finn MG, Sharpless KB. Sulfur(VI) Fluoride Exchange (SuFEx): Another Good Reaction for Click Chemistry. Angewandte Chemie (International ed. in English). 2014. Vol. 53:9430–9448

28. Wilson Lucas S, Zijian Qin R, Rakesh KP, Sharath Kumar KS, Qin HL. Chemical and Biology of Sulfur Fluoride Exchange (SuFEx) Click Chemistry for Drug Discovery. Bioorganic Chemistry. 2023. Vol. 130:106227

29. Meng GY, Guo TJ, Ma TC, Zhang J, Shen YC, Sharpless KB, et al.. Modular Click Chemistry Libraries for Functional Screens Using a Diazotizing Reagent. Nature. 2019. Vol. 574:86–89

30. Guo HY, Chen ZA, Shen QK, Quan ZS. Application of Triazoles in the Structural Modification of Natural Products. Journal of Enzyme Inhibition and Medicinal Chemistry. 2021. Vol. 36:1115–1144

31. Jiang XY, Li J, Viayna A, Luque FJ, Woodson M, Jing LL, et al.. Identification of Novel 1,2,3-Triazole Isatin Derivatives as Potent SARS-CoV-2 3CLpro Inhibitors Via Click-Chemistry-Based Rapid Screening. RSC Medicinal Chemistry. 2023. Vol. 14:2068–2078

32. Wang F, Liu DL, Gao DD, Yuan JW, Zhao JX, Yuan S, et al.. Discovery of Natural Catechol Derivatives as Covalent SARS-CoV-2 3CL^pro Inhibitors. International Journal of Biological Macromolecules. 2024. Vol. 264:130377

33. Zhang FM, Huang T, Wang F, Zhang GS, Liu DL, Dai J, et al.. Discovery of Highly Potent Covalent SARS-CoV-2 3CL^pro Inhibitors Bearing 2-Sulfoxyl-1,3,4-Oxadiazole Scaffold for Combating COVID-19. European Journal of Medicinal Chemistry. 2023. Vol. 260:115721

34. Gao DD, Tang S, Cen YX, Yuan L, Lan XJ, Li QH, et al.. Discovery of Novel Drug-Like PHGDH Inhibitors to Disrupt Serine Biosynthesis for Cancer Therapy. Journal of Medicinal Chemistry. 2023. Vol. 66:285–305

35. Lu X, Smaill JB, Patterson AV, Ding K. Discovery of Cysteine-Targeting Covalent Protein Kinase Inhibitors. Journal of Medicinal Chemistry. 2022. Vol. 65:58–83

36. Duan YK, Wang HF, Yuan ZH, Yang HT. Structural Biology of SARS-CoV-2 Mpro and Drug Discovery. Curr Opin Struct Biol. 2023. Vol. 82:102667

Graphical abstract

Highlights

• A series of catechol derivatives were efficiently synthesized using a modular click chemistry approach.

• P1-E11 exhibited superior anti-SARS-CoV-2 activity (EC₅₀ = 4.66 ± 0.58 μM) with low cytotoxicity (CC₅₀ > 100 μM) in A549-hACE2-TMPRSS2 cells.

• P1-E11 has a low KD value of 0.57 μM for SARS-CoV-2 3CL^pro binding.

In brief

A series of potential inhibitors of SARS-CoV-2 3CL^pro were efficiently synthesized and rapidly screened using a modular click chemistry approach. Among them, the compound P1-E11 had potent inhibitory effects against SARS-CoV-2 3CL^pro (IC₅₀ = 2.54 ± 0.46 μM), and exhibited superior anti-SARS-CoV-2 activity (EC₅₀ = 4.66 ± 0.58 μM) with full time treatments and low cytotoxicity (CC₅₀ > 100 μM) in A549-hACE2-TMPRSS2 cells.