Discovery of 2-(furan-2-ylmethylene)hydrazine-1-carbothioamide derivatives as novel inhibitors of SARS-CoV-2 main protease

Abstract

The COVID-19 posed a serious threat to human life and health, and SARS-CoV-2 M^pro has been considered as an attractive drug target for the treatment of COVID-19. Herein, we report 2-(furan-2-ylmethylene)hydrazine-1-carbothioamide derivatives as novel inhibitors of SARS-CoV-2 M^pro developed by in-house library screening and biological evaluation. Similarity search led to the identification of compound F8–S43 with the enzymatic IC₅₀ value of 10.76 μM. Further structure-based drug design and synthetic optimization uncovered compounds F8–B6 and F8–B22 as novel non-peptidomimetic inhibitors of M^pro with IC₅₀ values of 1.57 μM and 1.55 μM, respectively. Moreover, enzymatic kinetic assay and mass spectrometry demonstrated that F8–B6 was a reversible covalent inhibitor of M^pro. Besides, F8–B6 showed low cytotoxicity with CC₅₀ values of more than 100 μM in Vero and MDCK cells. Overall, these novel SARS-CoV-2 M^pro non-peptidomimetic inhibitors provide a useful starting point for further structural optimization.

Graphical abstract

1. Introduction

The outbreak of coronavirus infectious disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has been recognized as a serious threat to human life and health [[1], [2], [3]]. As of March 27, 2021, the disease has caused over 480 million people infections with more than 6 million deaths globally [4]. SARS-CoV-2 is a positive-strand RNA enveloped beta-coronavirus, and similar to severe acute respiratory syndrome coronavirus (SARS-CoV) and Middle East respiratory syndrome coronavirus (MERS-CoV), containing genome encodes non-structural proteins including main protease (M^pro, also known as 3-chymotrypsin-like protease, 3CL^pro), papain-like protease (PL^pro), helicase, and RNA-dependent RNA polymerase (RdRp) [5,6]. Among them, M^pro is a key enzyme in the viral life cycle, which is involved in the virus’ replication process, and results in the maturation of at least 12 non-structural proteins [[7], [8], [9]]. Furthermore, the highly conserved M^pro in coronavirus and the absence of closely related homologs in humans make M^pro an attractive target for the discovery of broad-spectrum antiviral drugs [10,11].

To date, there is only one M^pro inhibitor on market, although a series of SARS-CoV-2 M^pro inhibitors have been identified (Fig. 1 ) [10,12]. Among them, the organoselenium derivative 1 (Ebselen), was first disclosed as a covalent inhibitor of SARS-CoV-2 M^pro by screening the approved drugs and drug candidates [13]. It is reported that Ebselen could inhibit SARS-CoV-2 M^pro activity with an IC₅₀ value of 0.67 μM, and SARS-CoV-2 in Vero cells with an EC₅₀ value of 4.67 μM¹³. A variety of significant peptidomimetic inhibitors of SARS-CoV-2 M^pro have been reported, including compound 2 (N3), 3 (13b), 4 (11a), 5 (Boceprevir), 6 (GC-376), 7 (MI-09), and 8 (PF-07321332), which exhibited high SARS-CoV-2 M^pro inhibitory activity and SARS-CoV-2 inhibition at micromolar to sub-micromolar levels [[13], [14], [15], [16], [17], [18], [19], [20], [21]]. Furthermore, PF-07321332 was approved for mild and moderate symptoms caused by SARS-CoV-2 infection by US Food and Drug Administration (FDA), and PF-07321332 was combined with protease inhibitor and cytochrome P450 3A4 (CYP3A4) inactivator ritonavir to reduce its metabolism by CYP3A4 [21].

Fig. 1.

Chemical structures of represented inhibitors of SARS-CoV-2 M^pro.

Although the peptidomimetic inhibitors are highly effective against SARS-CoV-2 M^pro, the stability of hydrolase limited the application of these inhibitors [[21], [22], [23], [24]]. In contrast, the non-peptidomimetic inhibitors are less developed [25,26]. Su et al. and Liu et al. simultaneously reported that compound 9 (Baicalein) was a non-peptidomimetic inhibitor of SARS-CoV-2 M^pro with the enzymatic inhibitory ability and antiviral activity both at micromolar levels [27,28]. Meanwhile, compound 10, an N-substituted isatin derivative, was identified as a potent non-peptidomimetic inhibitor of SARS-CoV-2 M^pro, with an IC₅₀ value of 0.045 μM²⁹. Furthermore, compound 11 (Masitinib), an orally bioavailable tyrosine kinase inhibitor, was revealed as a potent inhibitor of SARS-CoV-2 M^pro, with an IC₅₀ value of 2.50 μM, through screening a library of 1900 clinically safe drugs [30]. Masitinib could block the replication of coronavirus in the lungs and nose of SARS-CoV-2 infected mice [30]. Moreover, Kneller and coworkers reported that compound 12 (HL-3-68) exhibited SARS-CoV-2 M^pro inhibitory activity, with an IC₅₀ value of 0.29 μM, through X-ray/neutron crystallography guided drug design [31]. Additionally, compound 13 (Jun8-76-3A) was discovered as a non-peptidomimetic inhibitor of SARS-CoV-2 M^pro by the modification of known M^pro inhibitor ML188 [32]. Overall, the chemical diversity of the identified non-peptidomimetic inhibitors is still highly limited. Therefore, it is urgent to develop novel non-peptidomimetic inhibitors against SARS-CoV-2 M^pro for the broad-spectrum antiviral drug candidate discovery, especially the new scaffold inhibitors that were not reported in previous coronavirus studies.

Herein, we report the identification of 2-(furan-2-ylmethylene)hydrazine-1-carbothioamide derivatives as novel non-peptidomimetic inhibitors of SARS-CoV-2 M^pro by screening our in-house library and subsequent similarity search. Interestingly, compound F8 was disclosed with the dissociation constant (K_D) value of 27.7 μM, by surface plasmon resonance (SPR) assay, which also exhibited an IC₅₀ value of 21.28 μM by enzymatic assay. Further two-dimensional similarity search based on the structure of compound F8 led to the identification of a series of 2-(furan-2-ylmethylene)hydrazine-1-carbothioamide derivatives as novel SARS-CoV-2 M^pro inhibitors. Among them, the most potent compound, F8–S43, exhibited the IC₅₀ value of 10.76 μM against SARS-CoV-2 M^pro. Then, three rounds of optimization based on the structure-based drug design and synthetic modification discovered compounds F8–B6 and F8–B22 as non-peptidomimetic inhibitors of M^pro with IC₅₀ values of 1.57 μM and 1.55 μM, respectively. Moreover, enzymatic kinetic and mass spectrometry studies demonstrated that F8–B6 was a reversible covalent inhibitor of M^pro. Besides, F8–B6 exhibited no obvious cytotoxicity in Vero and MDCK cells with CC₅₀ values over 100 μM. Furthermore, the structure-activity relationship (SAR) of the newly identified scaffold was discussed, which provided useful guidance for further chemical optimization.

2. Results and discussion

2.1. Identification of novel M^pro inhibitors by screening in-house library

To discover novel SARS-CoV-2 M^pro inhibitors, especially the new chemical structures that were not previously reported in coronavirus study, an in-house library was screened by SPR assay (Fig. 2 A). As drawn in Fig. 2B, compound F8 showed the high possibility as a new inhibitor of SARS-CoV-2 M^pro, with the K_D value of 27.7 μM. The further enzymatic assay demonstrated that compound F8 inhibited SARS-CoV-2 M^pro activity with an IC₅₀ value of 21.28 μM (Fig. 2C). Similarity analysis of compound F8 showed that it is not similar to any of the previously reported SARS-CoV-2 M^pro inhibitors with the Tanimoto coefficient below 0.14 (Fig. 2D). All of these indicated that the identified non-peptidomimetic inhibitor F8 expanded the chemical space of SARS-CoV-2 M^pro inhibitors and provided a useful starting point for further structural optimization.

Fig. 2.

The discovery of SARS-CoV-2 M^pro inhibitors. (A) Screen the in-house library through SPR assay. (B) The chemical structure and K_D value of compound F8. (C) The dose-dependent curve of F8 against SARS-CoV-2 M^pro. (D) The similarity analysis of compound F8 with known SARS-CoV-2 M^pro inhibitors.

2.2. Similarity search and preliminary SAR study

To investigate SAR and find more potent inhibitors of this new scaffold, a two-dimensional similarity search was performed based on the chemical structure of compound F8. A total of 70 compounds were captured and purchased from the ChemDiv and SPECS commercial databases. As exhibited in Table 1 , the replacement of the 3-carboxyl phenyl group of F8 with a 3-trifluoromethyl phenyl group (F8–S1), a 2-carboxyl phenyl group (F8–S2), a 2-nitro-4-methoxy phenyl group (F8–S3), a 4-nitrophenyl group (F8–S4), a 3-methyl-4-nitrophenyl group (F8–S5), or a 2,3-dimethyl-4-nitrophenyl (F8–S6), led to a decrease of potency, indicating that the 3-carboxyl phenyl group of F8 was essential for maintaining SARS-CoV-2 M^pro inhibitory activity. Furthermore, the replacement of the R¹ group of F8 with a 10-methylacridin-9(10H)-one moiety (F8–S7 to F8–S16), or a poly-substituted pyridine ring (F8–S17 to F8–S20), the inhibitory ability of M^pro was decreased. Meanwhile, the replacement of the R¹ group with a 1,2,5-oxadiazol-3-amine group (F8–S22), or a 1-methyl-1H-tetrazol-5-amine group (F8–S24), produced no significant effect on inhibitory potency. Besides, the introduction of the large hydrophobic substituents group (F8–S25, F8–S26, F8–S28 to F8–S39), was not conducive to the maintenance of inhibitory ability, which indicated that the R¹ group of F8 was more suitable for small substituents. Notably, the replacement of a urea linker with a thiourea linker was likely to improve the inhibitory potency (F8–S40 to F8–S43). Among them, the most potent compound, F8–S43, displayed an IC₅₀ value of 10.76 μM against SARS-CoV-2 M^pro (Fig. S1).

Table 1.

The chemical structures and enzymatic activities of the 43 analogs of F8.

Compounds	R [1]	X	R [2]	Inhibition% (50 μM) a	IC50 ± SD (μM) a
Tideglusib	–	–	–	98.6	0.30 ± 0.02
F8		O		65.0	21.28 ± 0.89
F8–S1		O		25.5	N.T. b
F8–S2		O		41.7	N.T.
F8–S3		O		<20.0	N.T.
F8–S4		O		30.1	N.T.
F8–S5		O		42.9	N.T.
F8–S6		O		<20.0	N.T.
F8–S7		O		21.3	N.T.
F8–S8		O		26.8	N.T.
F8–S9		O		54.5	N.T.
F8–S10		O		55.4	N.T.
F8–S11		O		22.5	N.T.
F8–S12		O		<20.0	N.T.
F8–S13		O		<20	N.T.
F8–S14		O		30.0	N.T.
F8–S15		O		24.1	N.T.
F8–S16		O		26.0	N.T.
F8–S17		O		37.3	N.T.
F8–S18		O		36.2	N.T.
F8–S19		O		32.9	N.T.
F8–S20		O		38.3	N.T.
F8–S21		O		31.2	N.T.
F8–S22		O		70.9	25.02 ± 1.11
F8–S23		O		34.5	N.T.
F8–S24		O		86.0	17.50 ± 1.65
F8–S25		O		45.3	N.T.
F8–S26		O		22.4	N.T.
F8–S27		O		29.4	N.T.
F8–S28		O		44.6	N.T.
F8–S29		O		33.7	N.T.
F8–S30		O		35.1	N.T.
F8–S31		O		35.8	N.T.
F8–S32		O		24.2	N.T.
F8–S33		O		<20.0	N.T.
F8–S34		O		<20.0	N.T.
F8–S35		O		62.8	N.T.
F8–S36		O		23.3	N.T.
F8–S37		O		<20.0	N.T.
F8–S38		O		23.0	N.T.
F8–S39		O		38.3	N.T.
F8–S40		S		85.1	10.88 ± 0.16
F8–S41		S		51.4	N.T.
F8–S42		S		55.9	N.T.
F8–S43		S		95.0	10.76 ± 0.48

^aData are presented as geometric mean values of at least two independent runs.

^bNot tested.

As shown in Table S1, the replacement of the thiourea linker of F8–S43 with a hydrazine linker (F8–S44 to F8–S48), an imine linker (F8–S49 to F8–S52), or a double bond linker (F8–S53 to F8–S65) led to the decrease of inhibitory potency, suggesting that the thiourea linker might be a good skeleton for these newly identified inhibitors. As for the right moiety of F8, the removal of the benzene ring resulted in the loss of potency (F8–S66 to F8–S70), indicating that the benzene ring of F8 was likely to play a hydrophobic role in the binding pocket of SARS-CoV-2 M^pro. As displayed in Table 2 , the replacement of the 4-nitrophenyl group of F8–S43 with a 4-bromobenzenyl group (F8–S43–S1), led to the remarkable loss of inhibitory activity, and the reduction of the volume of the R¹ group was beneficial to the preservation of inhibitory activity (F8–S43–S2, IC₅₀ = 8.08 μM; F8–S43–S3, IC₅₀ = 9.69 μM). Notably, the replacement of the thiourea linker to urea or guanidine linker decreased the inhibitory potency (F8–S43–S4 to F8–S43–S6), indicating that the thiourea linker was very important for the maintenance and improvement of the potency. To study the SAR of the right moiety, the hydrophobic substituent of the left moiety and the thiourea linker were fixed according to the structure of F8–S43. As drawn in Table S2, the decrease of the hydrophobic volume of the right moiety was not conducive to the maintenance of activity (F8–S43–S11 to F8–S43–S32). All in all, these results indicate that the newly identified scaffold expanded the chemical diversity of SARS-CoV-2 M^pro inhibitors, which could serve as a starting point for subsequent structural optimization.

Table 2.

The chemical structures and enzymatic activities of the 10 analogs of F8–S43.

Compounds	R [1]	X	R [2]	Inhibition% (50 μM) a	IC50 ± SD (μM) a
Tideglusib	–	–	–	98.6	0.30 ± 0.02
F8–S43		S		95.0	10.76 ± 0.48
F8–S43–S1		S		27.7	N.T. b
F8–S43–S2	NH2	S		81.6	8.08 ± 0.38
F8–S43–S3	NH2	S		81.1	9.69 ± 0.45
F8–S43–S4	NH2	O		<20.0	N.T.
F8–S43–S5		NH		20.1	N.T.
F8–S43–S6		NH		28.7	N.T.
F8–S43–S7		S		28.4	N.T.
F8–S43–S8		S		34.2	N.T.
F8–S43–S9		S		30.2	N.T.
F8–S43–S10		S		<20.0	N.T.

^aData are presented as geometric mean values of at least two independent runs.

^bNot tested.

2.3. Rational design and synthetic optimization

2.3.1. Structure-based design of newly identified scaffold

To guide the structural optimization of the newly identified scaffold, the representative compound F8–S43 was docked into the catalytic site of the SARS-CoV-2 M^pro (PDB ID, 7JU7) [30]. As shown in Fig. 3 A, compound F8–S43 was embedded into the catalytic site of SARS-CoV-2 M^pro and occupied the S1 and S2 sites, while the S1′, S3, and S4 sites were not filled with F8–S43. Furthermore, the 4-nitrophenyl moiety of F8–S43 was located at the S1 site and formed a hydrogen bond with His163, and the furan ring was deeply buried in the S2 site and had a π-π stacking interaction with the imidazole of His41 (Fig. 3B and Fig. S2). Meanwhile, the thiourea linker of F8–S43 formed two hydrogen bonds to the backbone carbonyl of Cys44, so the removal or replacement of the thiourea linker led to the loss of inhibitory ability. Besides, the right moiety of F8–S43 was located at the solvent-exposed area and formed a hydrogen bond with Ser46. As drawn in Fig. 3C, the removal of the dihydro-2H-pyrazol-2′-one moiety of F8–S43 had no impact on enzymatic inhibition in the aforementioned similarity search study, so the solvent-exposed region of this scaffold was structurally modified to improve the inhibitory potency of SARS-CoV-2 M^pro, followed by S1 and S2 pockets occupied moieties.

Fig. 3.

Predicted binding mode of compound F8–S43. (A) The binding pattern of compound F8–S43 with SARS-CoV-2 M^pro in surface and 2D diagram (B). Hydrogen bonds are represented by yellow lines. Images depicting the proposed binding modes were generated using PyMOL software. (C) The design of newly identified SARS-CoV-2 M^pro inhibitors.

2.3.2. Synthesis of target compounds

The target compounds F8-A1 to F8-A9 were synthesized as illustrated in Scheme 1 . Briefly, we synthesized the intermediates (15) from appropriate aryl boronic acid (14) with 5-bromofuran-2-carbaldehyde through Pd(PPh₃)₄ catalyzed Suzuki-Miyaura cross-coupling reaction [33]. Then, corresponding intermediates (15) reacted with substituted thiosemicarbazide to produce imine linkage (F8-A1 to F8-A9) through modification of published procedures or known methods [34,35].

Scheme 1.

Synthetic Route of Compounds F8-A1 to F8-A9. Reagents and conditions: (a) 5-bromofuran-2-carbaldehyde, Pd(PPh₃)₄, K₂CO₃, PhMe/EtOH/H₂O, 90 °C, overnight; (b) R¹NH₂CSNHNH₂, MeOH, 50 °C, 4 h.

The synthetic routes of the target compounds F8–B1 to F8–B13 were described in Scheme 2 . The building blocks (17 and 19) were prepared following previously described procedures [33]. Shortly, the starting material 5-iodofuran-2-carbaldehyde (16) or (5-formylfuran-2-yl)boronic acid (18) reacted with (4-carbamoylphenyl)boronic acid or substituted bromobenzene to form corresponding intermediates (17 and 19) through Suzuki-Miyaura cross-coupling reaction. Then, the aldehyde group of intermediates (17 and 19) condensed with the amino group of thiosemicarbazide to form the imine linkage giving the target compounds (F8–B1 to F8–B13).

Scheme 2.

Synthetic Route of Compounds F8–B1 to F8–B13. Reagents and conditions: (a) (4-carbamoylphenyl)boronic acid, Pd(PPh₃)₄, K₂CO₃, PhMe/EtOH/H₂O, 90 °C, overnight; (b) Thiosemicarbazide, MeOH, 50 °C, 4 h; (c) Substituted bromobenzene, Pd(PPh₃)₂Cl₂, Na₂CO₃, MeCN/H₂O, 90 °C, overnight; or substituted bromobenzene, Pd(PPh₃)₄, K₂CO₃, PhMe/EtOH/H₂O, 90 °C, overnight; or Substituted bromobenzene, Pd(PPh₃)₂Cl₂, 2 M Na₂CO₃, DME/EtOH, 60 °C, overnight.

As drawn in Scheme 3 , the synthetic methods of F8–B14 to F8–B21 were similar to the aforementioned synthesis of compounds F8–B1 to F8–B13. Simply, 5-bromofuran-2-carbaldehyde (20) reacted with corresponding aryl boronic acid to form intermediates (21), then the newly prepared intermediates condensed with thiosemicarbazide to give target compounds F8–B14 to F8–B21. Moreover, the intermediate (23), 5-benzylfuran-2-carbaldehyde, was synthesized by the starting material 5-(hydroxymethyl)furan-2-carbaldehyde (22) reacted with benzene through substitution reaction following previously described procedures [36]. Then, the intermediate (23) was condensed with thiosemicarbazide to give the target compound F8–B22.

Scheme 3.

Synthetic Route of Compounds F8–B14 to F8–B22. Reagents and conditions: (a) Corresponding aryl boronic acid, Pd(PPh₃)₄, K₂CO₃, PhMe/EtOH/H₂O, 90 °C, overnight; (b) Thiosemicarbazide, MeOH, 50 °C, 4 h; (c) benzene, trifluoromethanesulfonic acid, r.t.

As shown in Scheme 4 , the synthetic methods of F8–C1 to F8–C4 were similar to the aforesaid synthesis of compounds F8–B1 to F8–B13. Briefly, the appropriate aryl bromides (24) reacted with (4-carbamoylphenyl)boronic acid to form intermediates (25) through Suzuki-Miyaura cross-coupling reaction. Then, the corresponding intermediates condensed with thiosemicarbazide to give target compounds F8–C1 to F8–C4. All of the synthesized target compounds were characterized by ¹H NMR, ¹³C NMR, and high-resolution mass spectrometry (HRMS) experiments as single substances.

Scheme 4.

Synthetic Route of Compounds F8–C1 to F8–C4. Reagents and conditions: (a) (4-carbamoylphenyl)boronic acid, Pd(PPh₃)₄, K₂CO₃, PhMe/EtOH/H₂O, 90 °C, overnight; (b) Thiosemicarbazide, MeOH, 50 °C, 4 h.

2.3.3. Biological evaluation and SAR study

The synthesized target compounds were evaluated through aforesaid enzymatic assay, and the results were summarized in Table 3 . To investigate the SAR of the solvent-exposed region of the newly identified scaffold, the other regions were fixed with the same substituents. Notably, with the replacement of the 3-carboxyl phenyl group of F8-A1 with a 4-carboxyl phenyl group (F8-A2), the inhibitory potency was maintained. Furthermore, the introduction of hydrophobic substituents, including methyl, 4-chlorophenyl, 3-trifluoromethyl phenyl, and cyclohexyl amino groups (F8-A3 to F8-A9), led to decreased inhibitory ability. Thus, the R¹ group of this scaffold was fused with a hydrogen atom, and the R² group was further structural modified to enhance the inhibitory ability and explore explicit SAR.

Table 3.

The chemical structures and enzymatic activities of F8-A1 to F8-A9.

Compounds	R [1]	R [2]	Inhibition% (50 μM) a	IC50 ± SD (μM) a
F8-A1 (F8–S40)	H	3-COOH	85.1	10.88 ± 0.16
F8-A2	H	4-COOH	89.4	17.53 ± 3.98
F8-A3		3-COOH	22.2	N.T. b
F8-A4		4-COOH	39.5	N.T.
F8-A5		3-COOH	43.8	N.T.
F8-A6		4-COOH	27.5	N.T.
F8-A7		3-COOH	43.0	N.T.
F8-A8		4-COOH	38.6	N.T.
F8-A9		4-COOH	37.3	N.T.

^aData are presented as geometric mean values of at least two independent runs.

^bNot tested.

As for the phenyl moiety of the newly identified scaffold (Table 4 ), the replacement of 4-carboxyl phenyl group of F8-A2 with a 4-amide phenyl group (F8–B1), a 4-sulfonyl phenyl group (F8–B2), or a phenyl group (F8–B3), resulted in moderate improve in inhibitory activity. When the benzene ring had multiple substituents (F8–B4 to F8–B14), the polar substituents, including carboxyl group, hydroxyl group, and methoxy group, were favorable to maintain or promote the inhibitory ability. Among them, the 3-hydroxy-4-carboxyl phenyl group substituted compound F8–B6 exhibited good activity, with the IC₅₀ value of 1.57 μM (Fig. S1). Whereas the replacement of the 4-carboxyl phenyl group of F8-A2 with a 3, 4-dichlorobenzyl group (F8–B15), led to the loss of inhibitory potency, which meant that the phenyl moiety of the newly identified scaffold was not suitable to be replaced by a completely hydrophobic substituent. Furthermore, this moiety was compatible with pyridine or substituted pyridine rings (F8–B16 to F8–B17). Intriguingly, the naphthalene ring substitution was not conducive to the preservation of inhibitory activity (F8–B18 to F8–B19), while the quinoline ring did not affect the potency (F8–B20 to F8–B21). Therefore, the aforementioned moiety might be more suitable for nitrogen heterocyclic replacement. Notably, the replacement of the 4-carboxyl phenyl group of F8-A2 with a benzyl group (F8–B22), led to the enhancement of potency, and the compound F8–B22 displayed an IC₅₀ value of 1.55 μM (Fig. S1), which potency was 13-fold more than the initial compound F8.

Table 4.

The chemical structures and enzymatic activities of F8–B1 to F8–B22.

Compounds	R [1]	Inhibition% (50 μM) a	IC50 ± SD (μM) a
F8-A2		89.4	17.53 ± 3.98
F8–B1		94.2	4.00 ± 0.25
F8–B2		98.5	5.39 ± 0.14
F8–B3		100.0	4.05 ± 0.26
F8–B4		100.6	8.14 ± 1.14
F8–B5		98.5	4.69 ± 0.34
F8–B6		102.2	1.57 ± 0.08
F8–B7		89.0	32.53 ± 2.33
F8–B8		84.4	40.59 ± 1.06
F8–B9		93.6	25.39 ± 2.05
F8–B10		97.2	10.0 ± 0.50
F8–B11		79.8	N.T. b
F8–B12		114.8	9.09 ± 1.05
F8–B13		106.6	4.14 ± 0.16
F8–B14		102.5	5.81 ± 0.07
F8–B15		49.8	N.T.
F8–B16		95.4	7.09 ± 0.51
F8–B17		97.4	8.76 ± 0.75
F8–B18		73.2	N.T.
F8–B19		54.1	N.T.
F8–B20		92.3	10.40 ± 2.20
F8–B21		88.9	10.30 ± 0.70
F8–B22		100.1	1.55 ± 0.08

^aData are presented as geometric mean values of at least two independent runs.

^bNot tested.

To investigate the furan ring moiety of the newly identified scaffold, the benzene ring moiety was designed to fix with the benzyl group, as the compound F8–B22 displayed an IC₅₀ value of 1.55 μM. However, the designed compounds were difficult to synthesize, so the furan ring moiety was optimized with the 4-carboxyl phenyl group replacement. As drawn in Table 5 , the replacement of 2,5-position substituted furan ring of F8-A2 with a 2,4-position substituted furan ring (F8–C1), a 2,5-position substituted thiophene ring (F8–C2), or a 1,3-position substituted benzene ring (F8–C3), let to maintain or improve inhibitory potency. Whereas, the 1,4-position substituted benzene ring (F8–C4) was not conducive to the preservation of inhibitory activity, indicating that the furan ring moiety acted as an important connecting linker and prefer to three atomic lengths. It is noteworthy that the introduction of a substituent to the furan, thiophene, or benzene ring might be beneficial to the improvement of activity, however the synthesis was too difficult to achieve. Thus, we are trying to perform scaffold hopping based on this scaffold to improve the synthetic accessibility and inhibitory potency in the future.

Table 5.

The chemical structures and enzymatic activities of F8–C1 to F8–C4.

Compounds	Ar	Inhibition% (50 μM) a	IC50 ± SD (μM) a
F8-A2		89.4	17.53 ± 3.98
F8–C1		90.9	24.56 ± 0.70
F8–C2		95.1	5.78 ± 0.30
F8–C3		99.2	5.44 ± 0.16
F8–C4		64.0	N.T. b

^aData are presented as geometric mean values of at least two independent runs.

^bNot tested.

Meanwhile, the mechanism of action of these newly identified inhibitors was investigated through enzymatic kinetic and mass spectrometry assays. As exhibited in Fig. 4 A and B, the Lineweaver-Burk plot with different F8–B6 concentrations yielded an intercept at the X-axis, indicating that F8–B6 is a non-competitive inhibitor of SARS-CoV-2 M^pro. Moreover, prolonged incubation of SARS-CoV-2 M^pro with F8–B6 exhibited a time-dependent increase of inhibition activity (Figs. S3A–3C). The addition of dithiothreitol (DTT) could reverse the inhibitory effect of F8–B6 (Fig. S3D). Further liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis demonstrated that F8–B6 covalently bonded to SARS-CoV-2 M^pro (Fig. 4C), which was similar to the previously reported covalent inhibitor Ebselen [13], with more than one F8–B6 can be covalently bonded to the dimer of M^pro. Notably, thiacetazone moiety was identified as a covalent warhead to cystine in previous work [37,38], and this was consistent with the above SAR that the introduction of thiacetazone moiety led to the increase of inhibitory activity. However, the specific covalent mechanism still needed to be further studied. In addition, reversibility assay of F8–B6 to SARS-CoV-2 M^pro revealed that the ultrafiltration of inhibitor could recover enzymatic activity to a certain extent (Fig. 4D), which indicated that F8–B6 is a reversible inhibitor. Overall, the enzymatic kinetic and mass spectrometry studies demonstrated that F8–B6 is a reversible covalent inhibitor of SARS-CoV-2 M^pro. Interestingly, during the review process of this manuscript, Xu and coworkers reported that thiosemicarbazone is a promising scaffold for the inhibition of SARS-CoV-2 M^pro activity [39]. Besides, thiosemicarbazones are also well-known inhibitors of human Cathepsins. Pandey and coworkers reported that thiosemicarbazones derivatives are inhibitors of Cathepsin B, H, and L through a multi-target approach [40]. Notably, Cathepsin L is a key host cysteine protease utilized by coronaviruses for cell entry and is a promising drug target for novel antivirals against [41,42]. Thus, the selectivity of compounds F8–B6 and F8–B22 against Cathepsin L was investigated through previously reported method [43], and the results demonstrated that compounds F8–B6 and F8–B22 exhibited moderate inhibition of Cathepsin L, with the IC₅₀ values of 16.33 μM and 8.09 μM, respectively (Fig. S4).

Fig. 4.

Michaelis-Menten kinetics analysis (A) and Lineweaver-Burk plot (B) of SARS-CoV-2 M^pro in presence or absence of F8–B6. (C) The liquid chromatograph-mass spectrometer of SARS-CoV-2 M^pro with F8–B6. (D) Reversibility assay of the inhibition ability of F8–B6 against SARS-CoV-2 M^pro.

Meanwhile, two compounds F8–B6 and F8–B22, with good SARS-CoV-2 M^pro inhibitory abilities, were further evaluated for their cytotoxic activities in Vero and MDCK cells by MTT assay. As drawn in Fig. 5 A and B, both compounds F8–B6 and F8–B22 exhibited low cytotoxicity, and F8–B6 showed the CC₅₀ values more than 100 μM in Vero and MDCK cells, while F8–B22 displayed the CC₅₀ values around 100 μM in Vero cells, and above 100 μM in MDCK cells.

Fig. 5.

The cytotoxicity of compounds F8–B6 and F8–B22 in Vero (A) and MDCK cells (B).

3. Conclusion

In the current work, a series of 2-(furan-2-ylmethylene)hydrazine-1-carbothioamide derivatives were identified as non-peptidomimetic inhibitors of SARS-CoV-2 M^pro through screening an in-house library by SPR and enzymatic assays. Further similarity search led to the identification of compound F8–S43, which exhibited an IC₅₀ value of 10.76 μM against SARS-CoV-2 M^pro. Then, three rounds of optimization based on the structure-based drug design and synthetic modification discovered compounds F8–B6 and F8–B22 as non-peptidomimetic inhibitors of M^pro with IC₅₀ values of 1.57 μM and 1.55 μM, respectively. Moreover, enzymatic kinetic and mass spectrometry analysis demonstrated that F8–B6 was a reversible covalent inhibitor of M^pro. Besides, F8–B6 displayed no obvious cytotoxicity in Vero and MDCK cells. Taken together, this chemical series may serve as a good starting point for the further optimization of SARS-CoV-2 M^pro non-peptidomimetic inhibitors.

4. Experimental section

4.1. Chemistry

General method. Synthesis reagents and solvents were obtained from commercial suppliers and used without further purification. Characterizations of compounds are provided in the Supporting Information. ¹H and ¹³C NMR spectra were recorded on Bruker (400 MHz) instruments, using dimethyl sulfoxide (DMSO‑d ₆) as solvents. Chemical shifts are given in parts per million (ppm) downfield from tetramethylsilane (δ) as the internal standard in deuterated solvent and coupling constants (J) are in Hertz (Hz). Data are reported as follows: chemical shift, integration, multiplicity (s = singlet, d = doublet, t = triplet, q = quartet, dd = doublet of doublet, ddd = doublet of doublet of doublet, dt = doublet of triplet, m = multiplet, bs = broad signal), and coupling constants. High-resolution mass spectra were recorded on a Bruker Apex IV FTMS mass spectrometer using electrospray ionization (ESI). All compounds tested in biological assays were >95% pure.

4.1.1. General procedure A

A schlenk tube was charged with 5-bromofuran-2-carbaldehyde (1 mmol), corresponding boronic acid (1.2 mmol), tetrakis(triphenylphosphine)palladium (0.05 mmol), and potassium carbonate (3 mmol). The vessel was evacuated and backfilled with argon. A mixed solvent of Toluene/EtOH/H₂O (5/5/2 mL) was added. Then the schlenk tube was heated to 90 °C overnight. After cooling to room temperature, the reaction mixture was concentrated and diluted with water, filtered and the filtrate was adjusted to pH 2 with 2 N HCl. The precipitate was filtered and dried in vacuo to give the crude products without further purification.

4.1.2. General procedure B

A schlenk tube was charged with (5-formylfuran-2-yl)boronic acid (1.5 mmol), substituted bromobenzene (1 mmol), Pd(PPh₃)₂Cl₂ (0.05 mmol), and sodium carbonate (2 mmol). The vessel was evacuated and backfilled with argon. A mixed solvent of MeCN/H₂O (4/1.3 mL) was added. Then the schlenk tube was heated to 60 °C overnight. After cooling to room temperature, the reaction mixture was concentrated and diluted with water, filtered and the filtrate was adjusted to pH 2 with 2 N HCl. The precipitate was filtered and dried in vacuo to give the crude products without further purification.

4.1.3. General procedure C

A mixture of (5-formylfuran-2-yl)boronic acid (1.5 mmol), substituted bromobenzene (1 mmol), Pd(PPh₃)₂Cl₂ (0.05 mmol) in DME/EtOH (3/3 mL), and 2 M aqueous Na₂CO₃ (3 mL, 6 mmol of Na₂CO₃) was flushed with nitrogen for 3 min and heated at 60 °C overnight under nitrogen atmosphere. The solvents were removed under reduced pressure, the residue was dissolved in water, the mixture obtained was filtered through Celite, and the filtrate was adjusted to pH 2 with 2 N HCl. The precipitate was filtered and dried in vacuo to give the crude products without further purification.

4.1.4. General procedure D

To a solution of the corresponding aldehyde (1.0 mmol) in MeOH (5.0 mL) was added thiosemicarbazide or substituted thiosemicarbazide (1.05 mmol). The mixture was heated to 50 °C and stirred for 4 h. The residue was recrystallized from ethanol, and the precipitate was filtered and dried in vacuo to give target compounds.

4.1.5. (E)-N-(1,5-dimethyl-3-oxo-2-phenyl-2,3-dihydro-1H-pyrazol-4-yl)-2-((5-(4-nitrophenyl)furan-2-yl)methylene)hydrazine-1-carbothioamide (F8–S43)

Compound F8–S43 was purchased from commercial database. Reddish brown solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 12.05 (s, 1H), 9.21 (s, 1H), 8.27 (d, J = 8.9 Hz, 2H), 8.09 (s, 2H), 8.07 (s, 1H), 7.52 (t, J = 7.8 Hz, 2H), 7.47 (d, J = 3.6 Hz, 1H), 7.38 (dt, J = 8.4, 1.1 Hz, 2H), 7.36–7.30 (m, 1H), 7.24 (d, J = 3.7 Hz, 1H), 3.12 (s, 3H), 2.19 (s, 3H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.70, 162.73, 154.95, 152.84, 151.43, 146.73, 135.70, 135.67, 132.34, 129.57, 126.67, 125.12, 124.85, 123.94, 116.30, 113.06, 109.85, 36.36, 11.66. HRMS (ESI) [M + H] + calcd for C₂₃H₂₁N₆O₄S: 477.1345; found: 477.1341.

4.1.6. (E)-3-(5-((2-carbamothioylhydrazono)methyl)furan-2-yl)benzoic acid (F8-A1)

Following the general procedures A and D, compound F8-A1 was obtained in 42% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 13.18 (s, 1H), 11.53 (s, 1H), 8.29 (s, 2H), 8.07 (d, J = 7.6 Hz, 1H), 8.00 (s, 1H), 7.89 (d, J = 7.6 Hz, 1H), 7.79 (s, 1H), 7.57 (t, J = 7.6 Hz, 1H), 7.24 (d, J = 3.6 Hz, 1H), 7.11 (d, J = 3.6 Hz, 1H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.18, 167.43, 153.97, 149.98, 132.40, 132.11, 130.33, 129.84, 129.22, 128.54, 124.82, 115.54, 109.70. HRMS (ESI) [M − H]^- calcd for C₁₃H₁₀N₃O₃S⁻: 288.0448; found: 288.0442.

4.1.7. (E)-4-(5-((2-carbamothioylhydrazono)methyl)furan-2-yl)benzoic acid (F8-A2)

Following the general procedures A and D, compound F8-A2 was obtained in 59% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 13.00 (s, 1H), 11.56 (s, 1H), 8.33 (s, 1H), 8.02–7.90 (m, 5H), 7.83 (s, 1H), 7.27 (d, J = 3.6 Hz, 1H), 7.10 (d, J = 3.6 Hz, 1H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.25, 167.35, 153.85, 150.44, 133.67, 132.14, 130.44, 130.22, 124.29, 115.78, 110.93. HRMS (ESI) [M − H]^- calcd for C₁₃H₁₀N₃O₃S⁻: 288.0448; found: 288.0442.

4.1.8. (E)-3-(5-((2-(methylcarbamothioyl)hydrazono)methyl)furan-2-yl)benzoic acid (F8-A3)

Following the general procedures A and D, compound F8-A3 was obtained in 62% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 13.19 (s, 1H), 11.56 (s, 1H), 8.40–8.32 (m, 1H), 8.30 (s, 1H), 8.07 (d, J = 7.6 Hz, 1H), 8.01 (s, 1H), 7.89 (d, J = 7.6 Hz, 1H), 7.59 (t, J = 7.8 Hz, 1H), 7.25 (d, J = 3.6 Hz, 1H), 7.08 (d, J = 3.6 Hz, 1H), 3.03 (d, J = 4.4 Hz, 3H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 177.93, 167.42, 153.94, 150.10, 132.08, 132.05, 130.35, 129.83, 129.22, 128.54, 124.84, 115.41, 109.77, 31.37. HRMS (ESI) [M − H]^- calcd for C₁₄H₁₂N₃O₃S⁻: 302.0605; found: 302.0601.

4.1.9. (E)-4-(5-((2-(methylcarbamothioyl)hydrazono)methyl)furan-2-yl)benzoic acid (F8-A4)

Following the general procedure, compound F8-A4 was obtained in 51% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 13.01 (s, 1H), 11.61 (s, 1H), 8.34 (q, J = 4.6 Hz, 1H), 8.03–7.90 (m, 5H), 7.28 (d, J = 3.6 Hz, 1H), 7.09 (d, J = 3.6 Hz, 1H), 3.05 (d, J = 4.4 Hz, 3H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 177.97, 167.33, 153.79, 150.56, 133.66, 131.75, 130.43, 130.22, 124.24, 115.54, 110.97, 31.37. HRMS (ESI) [M − H]^- calcd for C₁₄H₁₂N₃O₃S⁻: 302.0605; found: 302.0600.

4.1.10. (E)-3-(5-((2-((4-chlorophenyl)carbamothioyl)hydrazono)methyl)furan-2-yl)benzoic acid (F8-A5)

Following the general procedures A and D, compound F8-A5 was obtained in 74% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 12.97 (s, 1H), 12.01 (s, 1H), 10.05 (s, 1H), 8.33 (s, 1H), 8.12 (s, 1H), 8.08 (d, J = 7.8 Hz, 1H), 7.90 (d, J = 7.8 Hz, 1H), 7.65 (d, J = 8.4 Hz, 2H), 7.59 (t, J = 7.6 Hz, 1H), 7.43 (d, J = 8.4 Hz, 2H), 7.28 (d, J = 3.6 Hz, 1H), 7.24 (d, J = 3.6 Hz, 1H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 176.09, 167.40, 154.29, 149.87, 138.48, 133.14, 132.09, 130.28, 129.87, 129.72, 129.31, 128.61, 128.49, 127.54, 124.91, 116.16, 109.87. HRMS (ESI) [M − H]^- calcd for C₁₉H₁₃ClN₃O₃S⁻: 398.0372; found: 398.0366.

4.1.11. (E)-4-(5-((2-((4-chlorophenyl)carbamothioyl)hydrazono)methyl)furan-2-yl)benzoic acid (F8-A6)

Following the general procedures A and D, compound F8-A6 was obtained in 51% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 13.02 (s, 1H), 12.03 (s, 1H), 10.03 (s, 1H), 8.12 (s, 1H), 8.00 (d, J = 8.4 Hz, 2H), 7.95 (d, J = 8.4 Hz, 2H), 7.64 (d, J = 8.8 Hz, 2H), 7.43 (d, J = 8.8 Hz, 2H), 7.33 (d, J = 3.6 Hz, 1H), 7.25 (d, J = 3.6 Hz, 1H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 176.22, 167.32, 154.19, 150.36, 138.47, 133.60, 132.95, 130.47, 130.32, 129.86, 128.49, 127.86, 124.35, 116.32, 111.12. HRMS (ESI) [M − H]^- calcd for C₁₉H₁₃ClN₃O₃S⁻: 398.0372; found: 398.0361.

4.1.12. (E)-3-(5-((2-((3-(trifluoromethyl)phenyl)carbamothioyl)hydrazono)methyl)furan-2-yl)benzoic acid (F8-A7)

Following the general procedures A and D, compound F8-A7 was obtained in 35% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 12.11 (s, 1H), 10.23 (s, 1H), 8.33 (s, 1H), 8.15 (s, 1H), 8.12–8.03 (m, 2H), 7.98–7.87 (m, 2H), 7.65–7.53 (m, 3H), 7.31–7.21 (m, 2H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 176.15, 167.40, 154.41, 149.86, 140.33, 133.51, 132.12, 130.26, 129.84, 129.71, 129.64, 129.33, 129.29 (d, J = 31.9 Hz), 128.60, 124.95, 124.54 (q, J = 273.3 Hz), 122.11 (d, J = 4.0 Hz), 122.03, 116.24, 109.87. ¹⁹F NMR (376 MHz, DMSO‑d ₆) δ −61.11. HRMS (ESI) [M − H]^- calcd for C₂₀H₁₃F₃N₃O₃S⁻: 432.0635; found: 432.0623.

4.1.13. (E)-4-(5-((2-((3-(trifluoromethyl)phenyl)carbamothioyl)hydrazono)methyl)furan-2-yl)benzoic acid (F8-A8)

Following the general procedures A and D, compound F8-A8 was obtained in 51% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 13.01 (s, 1H), 12.13 (s, 1H), 10.20 (s, 1H), 8.18–7.89 (m, 7H), 7.68–7.52 (m, 2H), 7.34 (d, J = 3.6 Hz, 1H), 7.27 (d, J = 3.6 Hz, 1H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 176.21, 167.35, 154.31, 150.32, 140.26, 133.54, 133.29, 130.44, 130.37, 129.79, 129.56, 129.21, 124.50 (q, J = 273.4 Hz), 124.29, 122.27 (d, J = 4.0 Hz), 122.04, 116.31, 110.98. ¹⁹F NMR (376 MHz, DMSO‑d ₆) δ −61.15. HRMS (ESI) [M − H]^- calcd for C₂₀H₁₃F₃N₃O₃S⁻: 432.0635; found: 432.0627.

4.1.14. (E)-4-(5-((2-(cyclohexylcarbamothioyl)hydrazono)methyl)furan-2-yl)benzoic acid (F8-A9)

Following the general procedures A and D, compound F8-A9 was obtained in 74% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 13.01 (s, 1H), 11.58 (s, 1H), 8.13–7.78 (m, 6H), 7.30 (d, J = 3.6 Hz, 1H), 7.16 (d, J = 3.6 Hz, 1H), 4.28–4.08 (m, 1H), 2.01–1.05 (m, 10H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 175.93, 167.30, 153.89, 150.50, 133.61, 132.04, 130.48, 130.24, 124.22, 115.65, 111.06, 52.97, 32.24, 25.54, 25.24. HRMS (ESI) [M − H]^- calcd for C₁₉H₂₀N₃O₃S⁻: 370.1231; found: 370.1225.

4.1.15. (E)-4-(5-((2-carbamothioylhydrazono)methyl)furan-2-yl)benzamide (F8–B1)

Following the general procedures B and D, compound F8–B1 was obtained in 44% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 11.54 (s, 1H), 8.31 (s, 1H), 8.05 (s, 1H), 8.02–7.81 (m, 6H), 7.43 (s, 1H), 7.24 (d, J = 3.6 Hz, 1H), 7.09 (d, J = 3.6 Hz, 1H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.18, 167.78, 154.16, 150.10, 133.78, 132.32, 132.22, 128.65, 124.08, 115.86, 110.30. HRMS (ESI) [M − H]^- calcd for C₁₃H₁₁N₄O₂S⁻: 287.0608; found: 287.0604.

4.1.16. (E)-2-((5-(4-sulfamoylphenyl)furan-2-yl)methylene)hydrazine-1-carbothioamide (F8–B2)

Following the general procedures B and D, compound F8–B2 was obtained in 63% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 11.57 (s, 1H), 8.34 (s, 1H), 8.09–7.76 (m, 6H), 7.42 (s, 2H), 7.28 (d, J = 3.6 Hz, 1H), 7.11 (d, J = 3.6 Hz, 1H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.23, 153.41, 150.46, 143.45, 132.77, 132.04, 126.82, 124.59, 115.77, 110.99. HRMS (ESI) [M − H]^- calcd for C₁₂H₁₁N₄O₃S₂ ⁻: 323.0278; found: 323.0275.

4.1.17. (E)-2-((5-phenylfuran-2-yl)methylene)hydrazine-1-carbothioamide (F8–B3) [44,45]

Following the general procedures C and D, compound F8–B3 was obtained in 24% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 11.50 (s, 1H), 8.33–8.25 (m, 1H), 7.97 (s, 1H), 7.87–7.81 (m, 2H) 7.78 (s, 1H), 7.47–7.41 (m, 2H), 7.38–7.30 (m, 1H), 7.12 (d, J = 3.6 Hz, 1H), 7.07 (d, J = 3.6 Hz, 1H).

4.1.18. (E)-4-(5-((2-carbamothioylhydrazono)methyl)furan-2-yl)-3-fluorobenzoic acid (F8–B4)

Following the general procedures A and D, compound F8–B4 was obtained in 51% yield. Orange red solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 13.33 (s, 1H), 11.60 (s, 1H), 8.36 (s, 1H), 8.11 (t, J = 8.0 Hz, 1H), 8.01 (s, 1H), 7.93–7.71 (m, 3H), 7.20–7.09 (m, 3H).¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.35, 166.32 (d, J = 2.5 Hz), 158.00 (d, J = 252.1 Hz), 150.58, 147.97 (d, J = 3.4 Hz), 131.95 (d, J = 7.8 Hz), 131.79, 126.75 (d, J = 1.9 Hz), 126.26 (d, J = 3.1 Hz), 121.76 (d, J = 11.9 Hz), 117.16 (d, J = 22.5 Hz), 115.59, 114.91 (d, J = 12.3 Hz). ¹⁹F NMR (376 MHz, DMSO‑d ₆) δ −113.32. HRMS (ESI) [M − H]^- calcd for C₁₃H₉FN₃O₃S⁻: 306.0354; found: 306.0347.

4.1.19. (E)-4-(5-((2-carbamothioylhydrazono)methyl)furan-2-yl)-3-methylbenzoic acid (F8–B5)

Following the general procedures B and D, compound F8–B5 was obtained in 56% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 12.97 (s, 1H), 11.56 (s, 1H), 8.32 (s, 1H), 8.01 (s, 1H), 7.95 (d, J = 8.2 Hz, 1H), 7.89–7.75 (m, 3H), 7.14 (d, J = 3.6 Hz, 1H), 7.04 (d, J = 3.6 Hz, 1H), 2.53 (s, 3H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.23, 167.43, 153.33, 149.90, 134.80, 132.91, 132.76, 132.26, 130.00, 127.58, 127.07, 115.43, 113.80, 22.29. HRMS (ESI) [M − H]^- calcd for C₁₄H₁₂N₃O₃S⁻: 302.0605; found: 302.0597.

4.1.20. (E)-4-(5-((2-carbamothioylhydrazono)methyl)furan-2-yl)-2-hydroxybenzoic acid (F8–B6)

Following the general procedures A and D, compound F8–B6 was obtained in 73% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 11.56 (s, 1H), 8.31 (s, 1H), 7.98 (s, 1H), 7.87 (s, 1H), 7.81 (d, J = 8.0 Hz, 1H), 7.45–7.34 (m, 2H), 7.28 (d, J = 3.6 Hz, 1H), 7.09 (d, J = 3.6 Hz, 1H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.22, 172.05, 161.99, 153.45, 150.60, 136.16, 132.09, 131.42, 115.75, 115.32, 112.49, 112.02, 111.65. HRMS (ESI) [M − H]^- calcd for C₁₃H₁₀N₃O₄S⁻: 304.0398; found: 304.0394.

4.1.21. (E)-4-(5-((2-carbamothioylhydrazono)methyl)furan-2-yl)-2-chlorobenzoic acid (F8–B7)

Following the general procedures A and D, compound F8–B7 was obtained in 83% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 13.36 (s, 1H), 11.57 (s, 1H), 8.31 (s, 1H), 8.06–7.77 (m, 5H), 7.31 (d, J = 3.6 Hz, 1H), 7.08 (d, J = 3.6 Hz, 1H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.27, 166.61, 152.36, 150.78, 133.68, 133.44, 132.32, 132.05, 129.98, 125.85, 122.65, 115.79, 111.76. HRMS (ESI) [M − H]^- calcd for C₁₃H₉ClN₃O₅S⁻: 322.0059; found: 32.0053.

4.1.22. (E)-3-(5-((2-carbamothioylhydrazono)methyl)furan-2-yl)-2-chlorobenzoic acid (F8–B8)

Following the general procedures C and D, compound F8–B8 was obtained in 53% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 13.54 (br, 1H), 11.56 (s, 1H), 8.32 (s, 1H), 8.12 (d, J = 8.0 Hz, 1H), 8.00 (s, 1H), 7.82 (s, 1H), 7.63 (dd, J = 8.0, 1.8 Hz, 1H), 7.52 (t, J = 7.6 Hz, 1H), 7.33 (d, J = 3.6 Hz, 1H), 7.14 (d, J = 3.6 Hz, 1H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.30, 167.77, 150.56, 149.94, 135.47, 132.05, 130.65, 129.45, 129.26, 127.96, 127.01, 115.12, 114.65. HRMS (ESI) [M − H]^- calcd for C₁₃H₉ClN₃O₃S⁻: 322.0059; found: 322.0059.

4.1.23. (E)-5-(5-((2-carbamothioylhydrazono)methyl)furan-2-yl)-2-chlorobenzoic acid (F8–B9)

Following the general procedures A and D, compound F8–B9 was obtained in 50% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 13.61 (s, 1H), 11.54 (s, 1H), 8.30 (s, 1H), 8.13 (s, 1H), 8.03–7.91 (m, 2H), 7.83 (s, 1H), 7.60 (d, J = 8.4 Hz, 1H), 7.27 (d, J = 3.6 Hz, 1H), 7.09 (d, J = 3.6 Hz, 1H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.22, 166.96, 152.92, 150.19, 132.98, 132.15, 131.65, 130.95, 128.96, 127.81, 125.91, 115.72, 110.31. HRMS (ESI) [M − H]^- calcd for C₁₃H₉ClN₃O₃S⁻: 322.0059; found: 322.0059.

4.1.24. (E)-3-(5-((2-carbamothioylhydrazono)methyl)furan-2-yl)-4-chlorobenzoic acid (F8–B10)

Following the general procedures C and D, compound F8–B10 was obtained in 73% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 13.34 (s, 1H), 11.60 (s, 1H), 8.38 (d, J = 2.1 Hz, 1H), 8.33 (s, 1H), 8.04 (s, 1H), 7.85 (dd, J = 8.4, 2.1 Hz, 1H), 7.78 (s, 1H), 7.68 (d, J = 8.4 Hz, 1H), 7.34 (d, J = 3.6 Hz, 1H), 7.20 (d, J = 3.6 Hz, 1H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.32, 166.64, 150.43, 150.05, 133.93, 132.30, 131.82, 130.68, 129.98, 129.03, 128.37, 114.58, 114.26. HRMS (ESI) [M − H]^- calcd for C₁₃H₉N₃O₃S⁻: 322.0059; found: 322.0057.

4.1.25. (E)-3-(5-((2-carbamothioylhydrazono)methyl)furan-2-yl)-2-methoxybenzoic acid (F8–B11)

Following the general procedures C and D, compound F8–B11 was obtained in 59% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 13.14 (s, 1H), 11.54 (s, 1H), 8.31 (s, 1H), 8.12 (d, J = 7.6 Hz, 1H), 8.00 (s, 1H), 7.81 (s, 1H), 7.66 (d, J = 7.6 Hz, 1H), 7.29 (t, J = 7.6 Hz, 1H), 7.11 (s, 2H), 3.79 (s, 3H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.17, 167.60, 155.87, 150.57, 149.05, 132.21, 130.84, 130.10, 127.46, 124.60, 124.41, 116.02, 113.14, 61.64. HRMS (ESI) [M − H]^- calcd for C₁₄H₁₂N₃O₄S⁻: 318.0554; found: 318.0549.

4.1.26. (E)-5-(5-((2-carbamothioylhydrazono)methyl)furan-2-yl)-2-hydroxybenzoic acid (F8–B12)

Following the general procedures A and D, compound F8–B12 was obtained in 67% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 11.48 (s, 1H), 8.25 (s, 1H), 8.14 (d, J = 2.4 Hz, 1H), 8.00–7.92 (m, 2H), 7.74 (s, 1H), 7.07–6.98 (m, 3H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.04, 171.94, 161.37, 154.31, 149.03, 132.64, 131.79, 125.98, 121.69, 118.40, 115.84, 113.94, 107.60. HRMS (ESI) [M − H]^- calcd for C₁₃H₁₀N₃O₄S⁻: 304.0398; found: 304.0394.

4.1.27. (E)-2-((5-(3-cyano-4-hydroxyphenyl)furan-2-yl)methylene)hydrazine-1-carbothioamide (F8–B13)

Following the general procedures A and D, compound F8–B13 was obtained in 52% yield. Pale brown solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 11.49 (s, 1H), 11.43 (s, 1H), 8.27 (s, 1H), 8.10 (d, J = 2.4 Hz, 1H), 7.99–7.90 (m, 2H), 7.81 (s, 1H), 7.13–6.97 (m, 3H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.07, 160.43, 153.52, 149.07, 132.16, 130.84, 129.19, 122.12, 117.25, 116.99, 116.13, 107.94, 100.05. HRMS (ESI) [M − H]^- calcd for C₁₃H₉N₄O₂S⁻: 285.0452; found: 285.0448.

4.1.28. (E)-2-((5-(3,4-dimethoxyphenyl)furan-2-yl)methylene)hydrazine-1-carbothioamide (F8–B14)

Following the general procedures A and D, compound F8–B14 was obtained in 71% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 11.47 (s, 1H), 8.24 (s, 1H), 7.97 (s, 1H), 7.75 (s, 1H), 7.38 (d, J = 9.2 Hz, 1H), 7.34 (s, 1H), 7.08–6.97 (m, 3H), 3.84 (s, 3H), 3.79 (s, 3H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.01, 155.40, 149.62, 149.52, 148.77, 132.57, 122.98, 117.44, 115.97, 112.47, 108.26, 107.47, 56.17, 56.02. HRMS (ESI) [M + H]⁺ calcd for C₁₄H₁₆N₃O₃S⁺: 306.0907; found: 306.0906.

4.1.29. (E)-2-((5-(3,4-dichlorophenyl)furan-2-yl)methylene)hydrazine-1-carbothioamide (F8–B15)

Following the general procedures A and D, compound F8–B15 was obtained in 63% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 11.55 (s, 1H), 8.31 (s, 1H), 8.08 (s, 1H), 7.96 (s, 1H), 7.87 (s, 1H), 7.81 (d, J = 8.4 Hz, 1H), 7.68 (d, J = 8.4 Hz, 1H), 7.28 (d, J = 3.6 Hz, 1H), 7.08 (d, J = 3.6 Hz, 1H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.25, 152.37, 150.29, 132.40, 131.94, 131.60, 130.71, 130.50, 125.90, 124.44, 115.86, 110.73. HRMS (ESI) [M − H]^- calcd for C₁₂H₈Cl₂N₃OS⁻: 311.9771; found: 311.9767.

4.1.30. (E)-2-((5-(pyridin-4-yl)furan-2-yl)methylene)hydrazine-1-carbothioamide (F8–B16)

Following the general procedures A and D, compound F8–B16 was obtained in 72% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 11.60 (s, 1H), 8.60 (d, J = 5.2 Hz, 2H), 8.36 (s, 1H), 7.99 (s, 1H), 7.87 (s, 1H), 7.76 (d, J = 5.2 Hz, 2H), 7.41 (d, J = 3.6 Hz, 1H), 7.19 (d, J = 3.6 Hz, 1H) ¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.36, 152.14, 151.11, 150.72, 136.44, 131.84, 118.24, 115.41, 112.49. HRMS (ESI) [M − H]^- calcd for C₁₁H₉N₄OS⁻: 245.0497; found: 245.0502.

4.1.31. (E)-2-((5-(6-fluoropyridin-3-yl)furan-2-yl)methylene)hydrazine-1-carbothioamide (F8–B17)

Following the general procedures A and D, compound F8–B17 was obtained in 61% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 11.54 (s, 1H), 8.73 (d, J = 2.4 Hz, 1H), 8.42–8.27 (m, 2H), 7.97 (s, 1H), 7.92–7.80 (m, 1H), 7.26 (dd, J = 8.8, 2.4 Hz, 1H), 7.20 (d, J = 3.6 Hz, 1H), 7.06 (d, J = 3.6 Hz, 1H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.24, 162.76 (d, J = 234.7 Hz), 151.26, 150.12, 143.68 (d, J = 15.5 Hz), 137.89 (d, J = 7.9 Hz), 131.90, 124.85 (d, J = 4.4 Hz), 115.68, 110.45 (d, J = 37.9 Hz), 109.96. ¹⁹F NMR (376 MHz, DMSO‑d ₆) δ −69.11. HRMS (ESI) [M − H]^- calcd for C₁₁H₈FN₄OS⁻: 263.0408; found: 263.0403.

4.1.32. (E)-2-((5-(naphthalen-1-yl)furan-2-yl)methylene)hydrazine-1-carbothioamide (F8–B18)

Following the general procedures A and D, compound F8–B18 was obtained in 79% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 11.56 (s, 1H), 8.40 (d, J = 8.0 Hz, 1H), 8.30 (s, 1H), 8.08 (s, 1H), 8.02–7.99 (m, 2H), 7.89 (dd, J = 7.2, 1.2 Hz, 1H), 7.74 (s, 1H), 7.64–7.56 (m, 3H), 7.21 (d, J = 3.2 Hz, 1H), 7.11 (d, J = 3.2 Hz, 1H) ¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.20, 154.42, 149.96, 134.04, 132.70, 129.70, 129.64, 129.19, 127.73, 127.34, 126.87, 126.72, 125.99, 125.30, 115.12, 112.67. HRMS (ESI) [M − H]^- calcd for C₁₆H₁₃N₃OS⁻: 294.0701; found: 294.0703.

4.1.33. (E)-2-((5-(naphthalen-2-yl)furan-2-yl)methylene)hydrazine-1-carbothioamide (F8–B19)

Following the general procedures A and D, compound F8–B19 was obtained in 45% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 11.57 (s, 1H), 8.37 (s, 1H), 8.34 (s, 1H), 8.06–7.79 (m, 6H), 7.60–7.48 (m, 2H), 7.25 (d, J = 3.6 Hz, 1H), 7.13 (d, J = 3.6 Hz, 1H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.16, 155.05, 149.79, 133.53, 133.00, 132.38, 129.03, 128.59, 128.19, 127.38, 127.29, 126.91, 122.77, 115.91, 109.55. HRMS (ESI) [M − H]^- calcd for C₁₆H₁₂N₃OS⁻: 294.0707; found: 294.0705.

4.1.34. (E)-2-((5-(quinolin-7-yl)furan-2-yl)methylene)hydrazine-1-carbothioamide (F8–B20)

Following the general procedures A and D, compound F8–B20 was obtained in 56% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 11.62 (s, 1H), 8.86 (dd, J = 4.0, 2.0 Hz, 1H), 8.47–8.34 (m, 3H), 8.17 (dd, J = 8.8, 2.0 Hz, 1H), 8.03 (d, J = 9.8 Hz, 2H), 7.92–7.83 (m, 1H), 7.53 (dd, J = 8.4, 4.0 Hz, 1H), 7.28 (d, J = 3.6 Hz, 1H), 7.13 (d, J = 3.6 Hz, 1H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.21, 154.33, 151.12, 150.16, 147.77, 136.61, 132.29, 130.11, 128.60, 127.82, 126.33, 122.75, 122.66, 115.83, 110.22. HRMS (ESI) [M − H]^- calcd for C₁₅H₁₁N₄OS⁻: 295.0659; found: 295.0654.

4.1.35. (E)-2-((5-(quinolin-3-yl)furan-2-yl)methylene)hydrazine-1-carbothioamide (F8–B21)

Following the general procedures A and D, compound F8–B21 was obtained in 71% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 11.60 (s, 1H), 9.39 (s, 1H), 8.72 (s, 1H), 8.37 (s, 1H), 8.11–7.97 (m, 3H), 7.89 (s, 1H), 7.75 (t, J = 7.6 Hz, 1H), 7.64 (t, J = 7.6 Hz, 1H), 7.42 (d, J = 3.6 Hz, 1H), 7.17 (d, J = 3.6 Hz, 1H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.25, 152.57, 150.48, 147.60, 147.20, 132.02, 130.20, 129.70, 129.33, 128.81, 127.92, 127.90, 123.38, 115.74, 110.60. HRMS (ESI) [M − H]^- calcd for C₁₅H₁₁N₄OS⁻: 295.0659; found: 295.0654.

4.1.36. (E)-2-((5-benzylfuran-2-yl)methylene)hydrazine-1-carbothioamide (F8–B22)

5-hydroxymethyl-2-furfuraldehyde (0.31 g, 2.45 mmol) was added to the mixture of TfOH (4 mL) and benzene (0.28 mL) in an ice bath. The reaction mixture was stirred at room temperature for 2 h. The mixture was poured into water (30 mL), and extracted with chloroform (3 × 30 mL). The combined extracts were washed with water, the saturated aqueous solution of NaHCO₃, water again, and dried over Na₂SO₄. The solvent was removed under reduced pressure, and the residue was subjected to chromatographic separation on silica gel to give intermediate 5-(phenylmethyl)furan-2-carbaldehyde 0.1 g, yielding 21%. Then following the general procedure D, compound F8–B22 was obtained in 42% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 11.38 (s, 1H), 8.17 (s, 1H), 7.89 (s, 1H), 7.55 (s, 1H), 7.37–7.19 (m, 5H), 6.88 (d, J = 3.6 Hz, 1H), 6.26 (d, J = 3.6 Hz, 1H), 4.02 (s, 2H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 177.98, 157.32, 148.87, 138.02, 133.07, 129.08, 129.01, 127.03, 114.52, 109.71, 34.18. HRMS (ESI) [M − H]^- calcd for C₁₃H₁₂N₃OS⁻: 258.0707; found: 258.0699.

4.1.37. (E)-4-(5-((2-carbamothioylhydrazono)methyl)furan-3-yl)benzoic acid (F8–C1)

Following the general procedures A and D, compound F8–C1 was obtained in 25% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 12.96 (s, 1H), 11.53 (s, 1H), 8.45 (s, 1H), 8.30 (s, 1H), 8.03–7.92 (m, 3H), 7.81–7.65 (m, 3H), 7.53 (s, 1H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.35, 167.45, 151.23, 142.77, 135.97, 132.41, 130.44, 129.88, 127.55, 125.95, 111.02. HRMS (ESI) [M − H]^- calcd for C₁₃H₁₀N₃O₃S⁻: 288.0448; found: 288.0447.

4.1.38. (E)-4-(5-((2-carbamothioylhydrazono)methyl)thiophen-2-yl)benzoic acid (F8–C2)

Following the general procedures A and D, compound F8–C2 was obtained in 51% yield. Yellowish solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 13.02 (s, 1H), 11.56 (s, 1H), 8.27 (s, 1H), 8.22 (s, 1H), 7.98 (d, J = 8.0 Hz, 2H), 7.81 (d, J = 8.0 Hz, 2H), 7.67 (d, J = 4.0 Hz, 1H), 7.64 (s, 1H), 7.50 (d, J = 4.0 Hz, 1H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.09, 167.25, 144.26, 139.89, 137.66, 137.51, 132.33, 130.70, 130.41, 126.43, 125.82. HRMS (ESI) [M − H]^- calcd for C₁₃H₁₀N₃O₂S⁻: 304.0220; found: 304.0215.

4.1.39. (E)-3'-((2-carbamothioylhydrazono)methyl)-[1,1′-biphenyl]-4-carboxylic acid (F8–C3)

Following the general procedures A and D, compound F8–C3 was obtained in 23% yield. White solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 13.00 (br, 1H), 11.52 (s, 1H), 8.27 (s, 1H), 8.22 (s, 1H), 8.18 (s, 1H), 8.13 (s, 1H), 8.03 (d, J = 8.0 Hz, 2H), 7.88 (d, J = 8.0 Hz, 2H), 7.79–7.74 (m, 2H), 7.52 (t, J = 7.4 Hz, 1H) ¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.52, 167.59, 144.13, 142.40, 139.90, 135.50, 130.34, 130.27, 129.88, 128.72, 128.03, 127.48, 125.61. HRMS (ESI) [M − H]^- calcd for C₁₅H₁₃N₃O₂S⁻: 298.0650; found: 298.0656.

4.1.40. (E)-4'-((2-carbamothioylhydrazono)methyl)-[1,1′-biphenyl]-4-carboxylic acid (F8–C4)

Following the general procedures A and D, compound F8–C4 was obtained in 12% yield. White solid. ¹H NMR (400 MHz, DMSO‑d ₆) δ 13.01 (s, 1H), 11.52 (s, 1H), 8.26 (s, 1H), 8.10 (d, J = 5.1 Hz, 2H), 8.03 (d, J = 8.4 Hz, 2H), 7.92 (d, J = 8.4 Hz, 2H), 7.85 (d, J = 8.4 Hz, 2H), 7.77 (d, J = 8.4 Hz, 2H). ¹³C NMR (100 MHz, DMSO‑d ₆) δ 178.47, 167.56, 143.92, 142.08, 140.46, 134.60, 130.45, 130.32, 128.44, 127.62, 127.24. HRMS (ESI) [M − H]^- calcd for C₁₅H₁₃N₃O₂S⁻: 298.0650; found: 298.0653.

4.2. Reagents and compounds

Tideglusib was purchased from Topscience Co. Ltd. SARS-CoV-2 M^pro fluorescent substrate Dabcyl-KTSAVLQSGFRKM-E(Edans)-NH₂ was synthesized by GL (Shanghai) Biochem Ltd. (Shanghai, China). Compounds used for SARS-CoV-2 M^pro inhibitors screening were purchased from ChemDiv (https://www.chemdiv.com/) and SPECS (https://www.specs.net/) commercial databases.

4.3. Cloning, expression, and purification of SARS-CoV-2 M^pro

The full-length gene encoding SARS-CoV-2 M^pro was synthesized for Escherichia coli (E. coli) expression (Hienzyme Biotech). The expression and purification of SARS-CoV-2 M^pro were carried out using the reported protocol [29].

4.4. SPR assay

The SPR assay was used to analyze the interaction between SARS-CoV-2 M^pro with compounds on a Biacore 8 K instrument (GE Healthcare). SARS-CoV-2 M^pro was immobilized on a sensor chip (CM5) via Amine Coupling Kit (GE Healthcare, Buckinghamshire, UK) at levels of approximately 10,000 response units (RU). The first flow channel without immobilized protein was set as a reference, and the compounds were injected at the concentration of 50 μM for screening and 0.2 μM–100 μM for binding study in a period of 60 s. Dissociation was measured for 100–200 s at a flow rate of 30 μL/min using the following assay running buffer: 10 mM phosphate buffer containing 2.7 mM KCl, 137 mM NaCl, and 0.05% surfactant P20 (pH 7.5). All of the data were analyzed through Biacore evaluation software (8 K version 1.0), and the curve was fitted with a 1:1 kinetics binding model.

4.5. Enzymatic assay of SARS-CoV-2 M^pro

A fluorescent substrate Dabcyl-KTSAVLQSGFRKM-E(Edans)-NH₂ (GL Biochemistry Ltd) and assay buffer (40 mM PBS, 100 mM NaCl, 1 mM EDTA, 0.1% Triton 100, pH 7.3) was used for the inhibition assay. For the preliminary screening and IC₅₀ measurements, 0.5 μM protease was incubated with inhibitor at room temperature for 30 min, and then the reaction was initiated by adding 20 μM substrate. The fluorescence signal generated by the cleavage of the substrate was monitored for 20 min at an emission wavelength of 460 nm with excitation at 360 nm using a plate reader (Synergy, Biotek). IC₅₀ values were fitted with the Hill1 function of Origin 2018. For the enzymatic kinetic assay, 0.5 μM SARS-CoV-2 M^pro was pre-incubated with DMSO or F8–B6 for 3 h and different concentrations of the fluorescent substrate were added to initiate the reaction. Data was collected from three replicates and curve-fitted by Origin 2018. For the DTT assay, 0.5 μM SARS-CoV-2 M^pro was premixed with 400 mM DTT or H₂O. Then, the protease solution was co-preincubated with DMSO or inhibitor at various concentrations for 30 min. Enzyme activity was tested and data was collected from three replicates. For time-dependent inhibitory measurement, various concentrations of inhibitors were pre-incubated with SARS-CoV-2 M^pro at a different time at room temperature before the addition of fluorescent substrate.

4.6. Mass spectrometry

2 μM protease was co-incubated with 15 μM F8–B6 on ice for 3 h. Then, the complex solution was analyzed by Quadrupole-TOF LC-MS/MS System (Vion, Waters). Raw data of mass signal was deconvoluted to obtain the total mass of the protein.

4.7. Enzymatic reversibility assay

SARS-CoV-2 M^pro (10 μM) was incubated with 100 μM F8–B6 for 180 min and divided into four parts which were ultra-filtrated for different times. In each time of ultrafiltration, an equal volume of buffer used in the enzyme activity assay was added into the protease solution and ultra-filtrated together for 5 min at 4 °C, 12,000 rpm. Protease left on the upper layer of Millipore was collected and diluted to the final concentration of 0.5 μM and enzyme activity was tested. All data were collected from three replicates to obtain average enzymatic activities and error bars.

4.8. Cathepsin L inhibition assay

The inhibition assay of Cathepsin L was performed as previously reported method [41,43]. Briefly, compounds F8–B6 and F8–B22 were tested using the commercial Cathepsin L Inhibitor Assay Kit (Abcam, Cat# ab197012). And the known Cathepsin L inhibitor FF-FMK was used as the positive compound.

4.9. Cell culture

Vero cells were cultured in (MEM, M&C Gene Technology, Beijing, China) and MDCK cells were cultured in Minimum Essential Medium Dulbecco's modified Eagle's medium (DMEM, M&C Gene Technology, Beijing, China), supplemented with 10% fetal bovine serum (FBS) (PAN Seratech, Aidenbach, Germany) and 100 U/mL penicillin/streptomycin (P/S), and maintained in a humidified atmosphere of 95% air and 5% CO₂.

4.10. Cell viability assay

MTT assay was performed to examine the cell viability of the newly identified SARS-CoV-2 M^pro inhibitors. Vero and MDCK cells were seeded in 96-well plates. After 18 h incubation, SARS-CoV-2 M^pro inhibitors were added to each well. Corresponding vehicles were simultaneously added for blank well. Cell viability was determined 48 h after drug treatment.

4.11. Molecular docking

All of the chemical structures were processed using the LigPrep module in Schrödinger 10.2 software (Schrodinger, LLC, NY, USA) [46]. The OPLS3 force field was adopted to perform energy minimization. Default settings were used for all other parameters. The crystallographic structures of the M^pro in complex with inhibitors were retrieved from the RCSB Protein Data Bank (PDB) and crystal structures (PDB IDs: 7JUJ) were prepared by the Protein Preparation Wizard module in Schrödinger 10.2 software. Default settings were used for all parameters based on the OPLS3 force field. The molecular docking was performed using the Glide module with XP (extra precision) in Schrödinger 10.2 software with default settings for all other parameters [47,48].

4.12. Similarity search

The similarity search was performed using Pipeline Pilot 8.5 of Accelrys. The similarities were calculated through the Tanimoto coefficient based on the fingerprint ECFP_6 of each structure. Then, 70 analogs of compound F8 and 32 analogs of compound F8–S43 were captured and purchased for biological evaluation.

Declaration of competing interest

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

Abbreviations used

SARS-CoV-2

severe acute respiratory syndrome coronavirus 2

COVID-19

coronavirus infectious disease 2019

SARS-CoV

severe acute respiratory syndrome coronavirus

MERS-CoV

Middle East respiratory syndrome coronavirus

M^pro

Main protease

3CLpro

3-chymotrypsin-like protease

PL^pro

papain-like protease

RdRp

RNA-dependent RNA polymerase

FDA

US Food and Drug Administration

CYP3A4

cytochrome P450 3A4

K_D

dissociation constant

SPR

surface plasmon resonance

IC₅₀

half-maximal inhibitory concentration

SAR

structure-activity relationship;

S2_C

S2 channel

HRMS

high-resolution mass spectrometry

PDB

protein data bank

RU

response units

MTT

3-(4,5-dimethylthiazolyl-2)- 2,5-diphenyltetrazolium bromide;

CC₅₀

half-maximal cytotoxic concentration

log P

octanol/water partition coefficient

Appendix A. Supplementary data

The following is the Supplementary data to this article: