A Patent Review on SARS Coronavirus Main Protease (3CL^pro) Inhibitors

Abstract

The severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) pandemic is an unprecedented global health emergency causing more than 4.2 million fatalities as of 30 July 2021. Only three antiviral therapies have been approved or granted emergency use authorization by the FDA. The SARS‐CoV‐2 3CL protease (3CL^pro) is deemed an attractive drug target as it plays an essential role in viral polyprotein processing and pathogenesis, although no inhibitors have been approved. This patent review discusses SARS coronavirus 3CL^pro inhibitors that have been filed up to 30 July 2021, giving an overview on the types of inhibitors that have generated commercial interest, especially amongst drug companies. Insights into the common structural motifs required for active site binding is also discussed.

Necessity is the mother of invention: The current SARS‐CoV‐2 pandemic is causing an unprecedented global health crisis. The coronavirus main protease 3CL^pro is deemed an attractive drug target due to its critical role in viral polyprotein processing. In this patent review, 24 coronavirus 3CL^pro inhibitor patents filed up to 30 July 2021 are described. Important structural features required for protease binding are discussed to facilitate the design of more potent inhibitors.

1. Introduction

In late 2019, a number of severe pneumonia clusters were reported in Wuhan, Hubei province, China.[ ¹ , ² ] Genome sequencing revealed the responsible pathogen was a coronavirus with 80 % nucleotide sequence identity to the severe acute respiratory syndrome coronavirus (SARS‐CoV), responsible for the 2002–2004 outbreak originating from Foshan, Guangdong province, China.[ ³ , ⁴ , ⁵ , ⁶ ] This new coronavirus was named ‘severe acute respiratory syndrome coronavirus 2’ (SARS‐CoV‐2) by the International Committee on Taxonomy of Viruses ^[7] and the disease was called ‘coronavirus disease 2019’ (COVID‐19) by the World Health Organization (WHO). The virus has since spread globally, infecting more than 196 million people and causing more than 4.2 million deaths as of 30 July 2021. ^[8] Although a number of vaccines have been granted authorization, only three antiviral therapies (Remdesivir, Casirivimab+Imdevimab and Sotrovimab)[ ⁹ , ¹⁰ , ¹¹ ] have so far been approved or granted emergency use authorization by the United States Food and Drug Administration (FDA), signifying an urgent need for more antiviral drugs.

The coronavirus 3‐chymotrypsin‐like protease (3CL^pro; also known as ‘main protease’ or ‘M^pro’) is deemed an attractive drug target due to its essential role in coronavirus polyprotein processing. It cleaves the polyprotein at more than 11 sites to yield essential proteins required for virus replication and pathogenesis.[ ¹² , ¹³ , ¹⁴ , ¹⁵ , ¹⁶ ] As 3CL^pro has no reported human homologues, a 3CL^pro inhibitor should not inhibit any human proteases and hence, reduce the risk of side‐effects.[ ¹⁵ , ¹⁶ ] 3CL^pro is a cysteine protease with a unique substrate preference for a P1 glutamine, hydrolysing the peptide bond at the C‐terminus of glutamine.[ ¹² , ¹³ , ¹⁴ , ¹⁵ , ¹⁶ ] Particularly noteworthy is that the 3CL^pro of SARS‐CoV and SARS‐CoV‐2 share 96 % amino acid sequence identity so inhibitors designed for the former should work equally well on the latter.[ ¹² , ¹⁷ ] Indeed, the first coronavirus 3CL^pro inhibitor to enter clinical trials is Pfizer's PF‐07304814 (ClinicalTrials.gov identifier NCT04535167) in September 2020. First described in a patent filed in 2005 as a SARS‐CoV 3CL^pro inhibitor (WO 2005/113580), its synthesis, structure‐activity relationship study and preclinical development details were published only much later in 2020. ^[18] .

2. Patent Review

Although many SARS‐CoV and SARS‐CoV‐2 3CL^pro inhibitors have been reported, collated and reviewed in academic journals since the 2002–2004 SARS‐CoV pandemic, none have so far been approved as antiviral drugs.[ ¹² , ¹³ , ¹⁴ ] A plausible explanation could be that commercial interest in coronavirus drug development waned rapidly after the SARS‐CoV pandemic ended. However, since the start of the SARS‐CoV‐2 pandemic in late 2019, academic and commercial interest in coronavirus 3CL^pro inhibitors returned with a vengeance. This review collates SARS‐CoV and SARS‐CoV‐2 3CL^pro inhibitors reported in patent applications filed since the 2002 SARS‐CoV pandemic up to 30 July 2021. SciFinder® software (CAS) ^[19] was employed and search terms: ‘coronavirus 3CL^pro inhibitor’, ‘coronavirus 3CL protease inhibitor’, ‘coronavirus Mpro inhibitor’, ‘coronavirus main protease inhibitor’ and ‘coronavirus protease inhibitor’ were used. The patents are summarized in Table 1 in chronological order with detailed analysis included vide infra.

Table 1.

Summary of SARS coronavirus 3CL^pro inhibitor patents in chronological order up to 30 July 2021.

Patent number	Applicant/Assignee	Filing date (dd.mm.yyyy)	Inhibitor type	Fig.	Ref.
WO 2004/093860	Pfizer	13.04.2004	peptidomimetic	1	[20]
WO 2004/101742	Cytovia	06.05.2004	peptidomimetic	2	[21]
US 2006/0014821	Agouron Pharmaceuticals	13.08.2004	peptidomimetic	3	[22]
WO 2005/041904	Fulcrum Pharmaceuticals	01.11.2004	small molecule	4	[23]
WO 2005/066123	Taigen Biotechnology	28.12.2004	peptidomimetic	5	[24]
WO 2005/113580	Pfizer	09.05.2005	peptidomimetic	6	[25]
US 2006/0019967	National Health Research Institutes, Taiwan	20.07.2005	small molecule	7	[26]
WO 2006/042478	Tsinghua University, Shanghai Institute of Organic Chemistry	24.10.2005	peptidomimetic	8	[27]
CN 1965833 A	Peking University	18.11.2005	small molecule	9	[28]
WO 2006/061714	Pfizer	06.12.2005	peptidomimetic	10	[29]
WO 2006/095624	Tokyo Medical and Dental University, RIKEN	01.03.2006	small molecule	11	[30]
WO 2007/075145	Singapore Polytechnic, Shanghai Institute of Materia Medica	15.12.2006	small molecule	12	[31]
CN 103159665B	Tianjin International Joint Academy of Biotechnology and Medicine	09.12.2011	small molecule	13	[32]
WO 2013/049382	Kansas State University, Ohio State University, Wichita State University	27.09.2012	peptidomimetic	14	[33]
KR 1020130002975	Chonnam National University	20.12.2012	small molecule	15	[34]
WO 2013/166319	Kansas State University, Wichita State University	02.05.2013	peptidomimetic	16	[35]
CN 106176728 A	Institute of Microbiology, Chinese Academy of Sciences	07.07.2016	small molecule	17	[36]
WO 2017/114509	Shanghai Institute of Materia Medica, University of Lübeck	30.12.2016	peptidomimetic	18	[37]
US 2017/0313685	Purdue Research Foundation	28.04.2017	small molecule	19	[38]
WO 2017/222935	Kansas State University, Wichita State University	16.07.2017	peptidomimetic	20	[39]
CN 108785293 A	Tianjin International Joint Academy of Biomedicine	28.04.2017	small molecule	21	[40]
WO 2018/042343	GSK	30.08.2017	peptidomimetic	22	[41]
WO 2020/030143	Shanghai Institute of Materia Medica, Fudan University	09.08.2019	peptidomimetic	23	[42]
WO 2021/176369	Pfizer	03.03.2021	peptidomimetic	24	[43]

SciFinder® revealed 18 patents on CoV 3CL^pro inhibitors and together with a 2020 patent review on coronavirus therapeutic agents, ^[44] a total 24 patents are summarised in Table 1. Unsurprisingly, majority of the patents (23 out of 24) described inhibitors designed to inhibit SARS‐CoV 3CL^pro as the SARS‐CoV‐2 pandemic started only in late 2019. However, since both viral proteases share 96 % sequence identity,[ ¹² , ¹⁷ ] inhibitors designed for the former should also inhibit the latter protease. Indeed, Pfizer's PF‐00835231 (6 b), first described in WO 2005/113580, was originally designed to inhibit SARS‐CoV 3CL^pro (IC₅₀ 4 nM) ^[18] and was later shown to be a potent inhibitor of SARS‐CoV‐2 3CL^pro (IC₅₀ 8 nM). ^[45]

To our best knowledge, the earliest patent application describing SARS‐CoV 3CL^pro inhibitors was filed by Pfizer in 2004 (WO 2004/093860; Table 1) ^[20] involving 3‐residue peptidomimetics bearing a P1 glutamine or lactam glutamine mimic with a C‐terminal electrophilic α,β‐unsaturated ester warhead (Figure 1). Some inhibitors had their P3 amino acid residue substituted by a 2‐pyridone moiety (1 b), presumably to improve their pharmacokinetic properties. All the described compounds were originally designed to inhibit the rhinovirus 3C protease which shares some structural similarity to coronavirus 3CL^pro. ^[20] Unfortunately, no coronavirus 3CL^pro inhibition data were reported in the patent. We believe this is the first patent describing peptidomimetics bearing a P1 lactam glutamine mimic (1 a) used for inhibiting SARS‐CoV 3CL^pro and were originally designed as rhinovirus 3C protease inhibitors in 1999 by Agouron Pharmaceuticals (WO 99/57135). ^[46] We are currently unable to recommend these peptidomimetics for further drug development until their SARS‐CoV‐2 3CL^pro inhibitory potencies are revealed.

Figure 1.

General scaffolds from WO 2004/093860.

The second earliest patent application on SARS‐CoV 3CL^pro inhibitors was filed by Cytovia in 2004 (WO 2004/101742; Table 1) ^[21] describing 2‐residue peptidomimetics containing a P1 glutamine residue with a C‐terminal electrophilic monofluoromethylketone warhead (Figure 2). P1 alanine, glycine and N‐dimethylated glutamine mimics were also exemplified although their inhibitory activities were not disclosed. Compound 2 b was revealed to be the most potent inhibitor with an EC₅₀ of 0.02 μg/mL in a SARS‐CoV Vero cell assay. No biochemical 3CL^pro inhibition (IC₅₀) data were reported in the patent. A publication was found for 2 e containing a P1 N‐dimethylated glutamine with an EC₅₀ of 2.5 μM in a SARS‐CoV Vero cell assay. ^[47] The authors revealed that the P1 glutamine residue was incompatible with the C‐terminal electrophilic warhead due to intramolecular cyclisation with the P1 glutamine side‐chain primary amide, resulting in bioactivity loss. ^[47] To solve this problem, Pfizer filed a patent in 2004 describing peptidomimetic inhibitors where the P1 glutamine was modified to either 5‐ or 6‐membered lactams (WO 2004/093860; Figure 1a). ^[20] Notably, this P1 modification strategy was adopted in subsequent patents involving coronavirus 3CL^pro peptidomimetic inhibitors after 2004. In addition, the idea of incorporating a 2‐pyridone moiety at the P3 position, exemplified in compound 2 f, was reported in an earlier Pfizer patent (WO 2004/093860; Figure 1b). ^[20] Although there are currently no follow‐up publications for 2 f, this idea was later adopted for peptidomimetic ketoamide SARS‐CoV‐2 3CL^pro inhibitors in a 2020 Science paper. ^[48]

Figure 2.

Structures from WO 2004/101742. (a) general scaffold; (b–f) exemplified structures.

The third earliest patent on coronavirus 3CL^pro inhibitors was filed in 2004 by Agouron Pharmaceuticals involving 1‐, 2‐ and 3‐residue peptidomimetics bearing a P1 5‐membered lactam glutamine mimic with a C‐terminal electrophilic α,β‐unsaturated ester warhead (US 2006/0014821; Table 1). ^[22] 8 structures were exemplified (Figure 3) but no biochemical 3CL^pro inhibition (IC₅₀) data were reported in the patent. A structure search on all 8 compounds, including compounds 3 c to 3 f, revealed no follow‐up publications. Notably, peptidomimetics with α,β‐unsaturated ester warheads inhibiting SARS‐CoV 3CL^pro were described in an earlier patent (WO 2004/093860; Figure 1). ^[20] We are currently unable to gauge their potential for further antiviral drug development due to the lack of inhibitory potency data. Interestingly, another patent involving peptidomimetic inhibitors with electrophilic α,β‐unsaturated ester warheads was filed in 2004 by Taigen Biotechnology (WO 2005/066123; Figure 5), ^[24] discussed vide infra.

Figure 3.

Structures from US 2006/0014821. (a, b) general scaffolds; (c–f) exemplified structures.

Figure 5.

Structures from WO 2005/066123. (a) general scaffold; (b–f) exemplified structures.

The fourth earliest patent was filed in 2004 by Fulcrum Pharmaceuticals involving small molecules containing one or two boronic acid moieties (WO 2005/041904; Table 1). ^[23] 403 structures were described. The strongest binder was identified to be compound 4 b, also named FL‐166, with a SARS‐CoV 3CL^pro K _i of 22 nM, followed by 4 c to 4 f with K _is ranging between 3.5 and 6.9 μM (Figure 4). Compounds 4 b, 4 c and 4 d were reported in a research article to possess K _is of 0.04, 4.5 and 16 μM against SARS‐CoV 3CL^pro respectively. ^[49] No SARS‐CoV‐2 3CL^pro inhibition data have been reported for these compounds using SciFinder® structure searches. In our view, cell‐based EC₅₀ values should first be obtained for 4 a to gauge its potential for further development as an antiviral drug.

Figure 4.

Structures from WO 2005/041904. (a–e) exemplified structures with the most potent K _is against SARS‐CoV 3CL^pro.

The fifth patent was filed in 2004 by Taigen Biotechnology involving 3‐residue peptidomimetics containing a P1 5‐membered lactam glutamine mimic with a C‐terminal electrophilic α,β‐unsaturated ester warhead (WO 2005/066123; Table 1). ^[24] 145 compounds were reported and 77 were disclosed to be sub‐micromolar SARS‐CoV 3CL^pro inhibitors. Unfortunately, no K _i or IC₅₀ values were assigned to any of the structures. However, a literature search revealed compound 5 b, also named TG‐0203770, to be the strongest binder with a SARS‐CoV 3CL^pro K _i of 58 nM, followed by 5 c and 5 d (K _is 0.66 and 2.26 μM respectively; Figure 5). ^[50] Compound 5 c was later reported to inhibit SARS‐CoV‐2 3CL^pro with an IC₅₀ of 286 nM. ^[45] In contrast, 5 d was reported to be a weak SARS‐CoV 3CL^pro inhibitor (IC₅₀ 85 μM), ^[51] suggesting that a P2 phenylalanine should be avoided in the design of new SARS‐CoV 3CL^pro inhibitors. Compounds 5 e and 5 f, known as TG‐0205486 and TG‐0204998 respectively, were reported with K _is of 99 and 38 nM respectively. ^[52] Notably, two earlier patents involving electrophilic α,β‐unsaturated ester warhead peptidomimetics were filed by Pfizer and Agouron Pharmaceuticals (WO 2004/093860 and US 2006/0014821), discussed vide supra. We opine compounds 5 b, 5 e and 5 f warrant further investigation as SARS‐CoV‐2 drug candidates.

The sixth earliest patent was filed in 2005 by Pfizer involving 2‐residue peptidomimetics containing a P1 5‐membered lactam glutamine mimic with a C‐terminal electrophilic hydroxymethylketone warhead (WO 2005/113580; Table 1). ^[25] 61 compounds were reported with 35 exhibiting sub‐micromolar IC₅₀s against SARS‐CoV 3CL^pro. A literature search revealed PF‐00835231 (6 b; Figure 6) as the lead candidate with a very potent IC₅₀ of 4 nM against SARS‐CoV 3CL^pro and a SARS‐CoV EC₅₀ of 4.8 μM in a SARS‐CoV infection cellular assay while a recent paper reported an IC₅₀ of 8 nM against SARS‐CoV‐2 3CL^pro.[ ¹⁸ , ⁴⁵ ] This strongly suggests that inhibitors designed to inhibit SARS‐CoV 3CL^pro can be repurposed for SARS‐CoV‐2 3CL^pro as both proteases share 96 % sequence identity.[ ¹² , ¹⁷ ] The phosphate prodrug of PF‐0835231 entered phase 1 clinical trials in September 2020 to evaluate safety and pharmacokinetics in hospitalised COVID‐19 patients (ClinicalTrials.gov identifier: NCT04535167). Notably, substituting the P2 leucine with phenylalanine resulted in a 25‐fold activity loss (IC₅₀ 103 vs. 4 nM; 6 e and 6 b respectively), suggesting that the SARS‐CoV 3CL^pro S2 binding pocket preferred smaller sized residues. ^[18] N‐methylating the P2 leucine resulted in a 20‐fold activity loss (IC₅₀ 83 vs. 4 nM; 6 f and 6 b respectively), suggesting that the P2 NH plays an important role in 3CL^pro binding. ^[18] 6‐membered lactam glutamine mimics were also described in the patent but none were exemplified.

Figure 6.

Structures from WO 2005/113580. (a) general scaffold; (b–f) exemplified structures.

The seventh earliest patent was filed in 2005 by the National Health Research Institutes, Taiwan, involving small molecules from a commercially available small molecule library screen (US 2006/0019967; Table 1). ^[26] 21 compounds were reported to possess inhibitory activities with 4 exhibiting ‘very low’ micromolar IC₅₀s against SARS‐CoV 3CL^pro (compounds 7 a to 7 d; Figure 7). Unfortunately, IC₅₀ values were not disclosed in the patent. A literature search revealed compound 7 a to be the most potent with an IC₅₀ of 300 nM against SARS‐CoV 3CL^pro. ^[53] The next most potent is 7 b with an IC₅₀ of 900 nM while 7 c to 7 f exhibited IC₅₀s between 3 to 10 μM. ^[53] Structure searches revealed there are currently no reports on their SARS‐CoV‐2 3CL^pro inhibitory activities. Interestingly, compound 7 c, also known as PNU‐136592, was earlier reported by Pharmacia scientists to inhibit MurB, a bacteria enzyme involved in cell wall peptidoglycan biosynthesis. ^[54] In our view, cell‐based EC₅₀ values should first be obtained to gauge their potential for further antiviral drug development.

Figure 7.

Six most potent SARS‐CoV 3CL^pro inhibitors reported in US 2006/0019967.

The eighth earliest patent was filed in 2005 by Tsinghua University and the Shanghai Institute of Organic Chemistry involving 3‐ and 4‐residue peptidomimetics bearing a P1 5‐membered lactam glutamine mimic with a C‐terminal electrophilic α,β‐unsaturated ester warhead (WO 2006/042478; Table 1). ^[27] 10 compounds were exemplified and no IC₅₀ data was reported (Figure 8). The patent is written in Chinese and an English version could not be found online. A research article published in the same month of patent filing reported inhibitors 8 b and 8 d (named N1 and N3 respectively) with K _is of 10.7 and 9.0 μM against SARS‐CoV 3CL^pro respectively. ^[55] A 2020 Nature paper reported 8 d/N3 to be a very potent SARS‐CoV‐2 3CL^pro inhibitor and that its K _i could not be feasibly measured. ^[15] No SARS‐CoV‐2 3CL^pro inhibition data have been reported so far for compounds 8 c and 8 e using SciFinder® structure searches. In our opinion, 8 d/N3 warrants further investigation as an antiviral drug.

Figure 8.

Structures from WO 2006/061714. (a) general scaffold; (b–e) exemplified structures.

The ninth earliest patent was filed in 2005 by Peking University involving the discovery of the oral H₁ histamine receptor antagonist oxatomide possessing SARS‐CoV 3CL^pro inhibitory activity (CN 1965833 A; Figure 9). ^[28] No IC₅₀ or K _i data was reported. The patent is written in Chinese and an English version could not be found online. A SciFinder® structure search did not reveal any academic journals reporting its SARS‐CoV 3CL^pro inhibitory activity. Due to the lack of inhibitory data, we are currently unable to gauge its potential for further development as an antiviral drug.

Figure 9.

Exemplified structure from CN 1965833 A.

The tenth earliest patent was filed in 2005 by Pfizer involving 2‐ and 3‐residue peptidomimetics containing P1 5‐membered lactam and tertiary amide glutamine mimics with C‐terminal electrophilic aldehyde or ketone warheads (WO 2006/061714; Table 1). ^[29] 15 compounds were exemplified with 4 (10 b to 10 e; Figure 10) exhibiting sub‐micromolar IC₅₀s against SARS‐CoV 3CL^pro. Ketone peptidomimetic 10 f exhibited a moderate IC₅₀ of 2.3 μM, suggesting that aldehyde warheads are much more reactive than ketones. Compound 10 d has been resynthesized and reported to inhibit SARS‐CoV‐2 3CL^pro with an IC₅₀ of 8.5 nM. ^[56] Aldehydes are known to be highly reactive towards endogenous biological nucleophiles, potentially rendering compounds 10 b to 10 e cytotoxic. Aldehydes are also metabolically unstable due to their susceptibility to oxidation and reduction by liver enzymes.[ ⁵⁷ , ⁵⁸ ] Hence, we believe aldehyde peptidomimetics lack potential for further drug development due to their metabolic liabilities.

Figure 10.

Structures from WO 2006/061714. (a) general scaffold; (b–f) exemplified structures.

The eleventh earliest patent was filed in 2006 by Tokyo Medical and Dental University and RIKEN involving small molecules with a 3‐cyanopyridine scaffold (WO 2006/095624; Table 1; Figure 11). ^[30] 102 compounds were described. The patent is published in Japanese and no English version can be found online. To our best knowledge, 3CL^pro IC₅₀ inhibition data was not reported in the patent. Structure searches using SciFinder® did not reveal any follow‐up journal publications. Due to the lack of inhibitory data, we are unable to gauge their potential for further drug development.

Figure 11.

Structures from WO 2006/095624. (a) general scaffold; (b–f) exemplified structures.

The twelfth earliest patent was filed in 2006 by Singapore Polytechnic and Shanghai Institute of Materia Medica involving small molecules with a chromone (1,4‐benzopyrone) scaffold (WO 2007/075145; Table 1). ^[31] 17 structures were exemplified and 4 compounds, 12 b to 12 e (Figure 12), were reported with SARS‐CoV 3CL^pro IC₅₀s ranging from 24 to 49 μM. These compounds were published in a research article in the same year of patent filing. ^[59] Due to their weak inhibitory activities, we opine that these compounds are not potent enough for further drug development.

Figure 12.

Structures from WO 2007/075145. (a) general scaffold; (b–e) exemplified structures.

The thirteenth earliest patent was filed in 2011 by Tianjin International Joint Academy of Biotechnology and Medicine involving 2,3‐dioxindole SARS‐CoV 3CL^pro inhibitors (CN 103159665B; Table 1). ^[32] The patent is written in Chinese and an English version could not be found online. 5 structures (Figure 13) were exemplified with SARS‐CoV 3CL^pro IC₅₀s between 20 to 88 μM. Structure searches on the compounds using SciFinder® did not reveal any follow‐up journal publications. In our opinion, these compounds are not potent enough to be developed as antiviral drugs.

Figure 13.

Structures from CN 103159665B. (a) general scaffold; (b–f) exemplified structures.

The fourteenth patent was filed in 2012 by Kansas State, Ohio State and Wichita State Universities involving 2‐ and 3‐residue peptidomimetics containing a P1 5‐membered lactam glutamine mimic with a C‐terminal electrophilic aldehyde or ketoamide warhead (WO 2013/049382; Table 1). ^[33] A hydroxymethyl sulfonic acid prodrug masking the reactive aldehyde warhead was also described (14 c; Figure 14). Compounds 14 b, 14 c and 14 d exhibited IC₅₀s ranging between 3.5 to 4.7 μM in a SARS‐CoV 3CL^pro biochemical assay. No SARS‐CoV 3CL^pro IC₅₀s were reported for tripeptide aldehyde 14 e and ketoamide 14 f in the patent but were later reported in a research article (0.23 and 0.61 μM respectively). ^[60] Compounds 14 b and 14 c, also named GC373 and GC376 respectively, were reported to inhibit SARS‐CoV 3CLpro with 3.5 and 4.4 μM IC₅₀s in a research paper. ^[61] Interestingly, 14 b/GC373 was shown to be much more active against SARS‐CoV‐2 3CL^pro, exhibiting an IC₅₀ of 0.042 μM in a recent paper. ^[45] Similarly, 14 c/GC376 was also shown to be much more active against SARS‐CoV‐2 3CL^pro, exhibiting IC₅₀s ranging between 0.03 to 0.62 μM.[ ⁴⁵ , ⁶² , ⁶³ ] We opine ketoamide peptidomimetics 14 d and 14 f can potentially be further developed as SARS‐CoV‐2 3CL^pro antivirals because ketoamide peptidomimetic protease inhibitors such as Boceprevir and Telaprevir have been approved for treating hepatitis C virus infections. ^[64]

Figure 14.

Structures from WO 2013/049382. (a) general scaffold; (b–f) exemplified structures.

The fifteenth earliest patent was filed in 2012 by Chonnam National University, South Korea revealing that the plant flavonoid, quercetin (KR 1020130002975; Table 1; Figure 15), was able to inhibit SARS‐CoV 3CL^pro with an IC₅₀ of 73 μM and suggested that it could potentially be used to treat SARS‐CoV infections. The patent is written in Korean and no English version could be found online. In our opinion, we believe that its inhibitory activity is too weak to be considered for development as an antiviral drug.

Figure 15.

Structure of quercetin from KR 1020130002975.

The sixteenth earliest patent was filed in 2013 by Kansas State and Wichita State Universities involving 3‐residue cyclic peptidomimetics and linear piperidine‐2,6‐dione peptidomimetics with C‐terminal aldehyde or ketoamide warheads (WO 2013/166319; Table 1). ^[35] Although SARS‐CoV 3CL^pro IC₅₀s were not disclosed in the patent, a research article by the inventors reported compound 16 c (Figure 16) inhibited SARS‐CoV 3CL^pro with an IC₅₀ of 15.5 μM. ^[65] No SARS‐CoV‐2 3CL^pro inhibitory activities have so far been reported for compounds 16 c to 16 i. In our view, 16 c lacks potency for further drug development.

Figure 16.

Structures from WO 2013/166319. (a, b) general scaffolds; (c–i) exemplified structures.

The seventeenth earliest patent was filed in 2016 by The Institute of Microbiology, Chinese Academy of Sciences involving small molecule disulphide SARS‐CoV 3CL^pro inhibitors (CN 106176728 A; Table 1). ^[36] The patent is written in Chinese and an English version could not be found online. 12 structures were exemplified and the 5 most potent compounds possess IC₅₀s between 1.3 to 2.2 μM (17 b to 17 f; Figure 17). A SciFinder® structure search revealed one follow‐up academic paper reporting their SARS‐CoV 3CL^pro inhibitory activities. ^[66] There are currently no reports on SARS‐CoV‐2 3CL^pro inhibitory activities. We opine that these compounds lack drug development potential due to their disulphide moiety, reported to be unstable in physiological conditions due to the presence of biological reducing agents, free thiols and isomerases. ^[67] Hence, pharmacokinetic studies should be conducted first before these compounds can be evaluated further.

Figure 17.

Structures from CN 106176728 A. (a) general scaffold; (b–f) top five most potent inhibitors.

The eighteenth patent was filed in 2016 by Shanghai Institute of Materia Medica and the University of Lübeck involving 2‐ and 3‐residue peptidomimetics bearing a P1 5‐membered lactam glutamine mimic and a C‐terminal electrophilic aldehyde warhead (WO 2017/114509; Table 1). ^[37] 74 structures were described with SARS‐CoV 3CL^pro IC₅₀s ranging from 30 nM to 7.4 μM. The four most potent structures are shown in Figure 18 (compounds 18 b to 18 e) with SARS‐CoV 3CL^pro IC₅₀s ranging from 30 to 100 nM. The most potent inhibitor 18 b is structurally similar to Pfizer's 10 b (Figure 10), differing at the P2 residue. 18b’s SARS‐CoV 3CL^pro inhibitory activity is 1.5‐fold weaker than Pfizer's 10 b (IC₅₀s 30 and 20 nM respectively), suggesting that SARS‐CoV 3CL^pro prefers a P2 leucine over phenylalanine. This distinction is an important consideration when designing new coronavirus 3CL^pro inhibitors and may be the reason why a P2 leucine was selected for Pfizer's lead compound PF‐00835231 (Figure 6b). ^[18] 18 b was also reported to be the most potent SARS‐CoV‐2 3CL^pro inhibitor (IC₅₀ 34 nM) amongst 13 peptide aldehydes in a recent publication, ^[68] suggesting that inhibitors designed for SARS‐CoV 3CL^pro can be effectively used to inhibit SARS‐CoV‐2 3CL^pro. Interestingly, the addition of a P3 valine to 18 b to yield tripeptide aldehyde 18 d resulted in an approximate 2‐fold activity reduction for SARS‐CoV and SARS‐CoV‐2 3CL^pro (IC₅₀s 30 to 50 nM and 34 to 68 nM respectively). ^[68] Notably, all the inhibitors described in the patent possess highly electrophilic aldehyde warheads which are known to be highly reactive towards endogenous biological nucleophiles, potentially rendering them cytotoxic. Aldehydes are also metabolically unstable due to their susceptibility to oxidation and reduction by liver enzymes.[ ⁵⁷ , ⁵⁸ ] Hence, we believe these aldehyde peptidomimetics lack potential for further drug development.

Figure 18.

Structures from WO 2017/114509. (a) general scaffold; (b–f) most potent exemplified inhibitors against SARS‐CoV 3CL^pro.

The nineteenth earliest patent was filed in 2017 by Purdue Research Foundation involving small molecules with amidophenyl scaffolds (US 2017/0313685; Table 1). ^[38] 77 structures were reported with SARS‐CoV 3CL^pro IC₅₀s between 0.8 to 27.3 μM. The five most potent inhibitors were identified to be compounds 19 b to 19 f (Figure 19) with 0.8 to 2.0 μM IC₅₀s. Interestingly, the structure of compound 19 f was reported in an earlier 2015 research paper with an IC₅₀ of 1.9 μM against HKU4‐CoV 3CL^pro. ^[69] There are currently no reports on their SARS‐CoV‐2 3CL^pro inhibitory activities. We opine that the most potent inhibitor, 19 b, should be further evaluated for drug development.

Figure 19.

Structures from US 2017/0313685. (a) general scaffold; (b–f) most potent exemplified structures.

The twentieth earliest patent was filed in 2017 by Kansas State and Wichita State Universities involving dipeptide peptidomimetics with a P1 5‐membered lactam glutamine analog and a C‐terminal hydroxymethyl sulfonic acid prodrug warhead against the Middle East Respiratory Syndrome (MERS) coronavirus 3CL^pro (WO 2017/222935; Table 1). ^[39] The hydroxymethyl sulfonic acid moiety changes to a reactive electrophilic aldehyde warhead at physiological pH, first described in WO 2013/049382. ^[33] Although no SARS‐CoV 3CL^pro inhibitory activities were disclosed, the MERS‐CoV and SARS‐CoV‐2 3CL^pro share approximately 50 % sequence identity ^[16] so inhibitors designed to inhibit MERS‐CoV 3CL^pro will likely be able to inhibit SARS‐CoV‐2 3CL^pro. 8 compounds were described with IC₅₀s ranging from 0.3 to 3.1 μM with compounds 20 b to 20 f being the most potent against MERS‐CoV 3CL^pro (Figure 20). There are currently no reports on their SARS‐CoV‐2 3CLpro inhibitory activities. As the hydroxymethyl sulfonic acid moiety transforms into a reactive aldehyde warhead in physiological conditions, we opine these compounds lack drug development potential as explained vide supra.

Figure 20.

Structures from WO 2017/222935. (a) general scaffold; (b–f) exemplified structures.

The twenty‐first earliest patent was filed in 2017 by Tianjin International Joint Academy of Biomedicine revealing that the N‐methyl‐D‐aspartate (NMDA) receptor antagonist and gamma aminobutyric acid (GABA) receptor modulator, acamprosate calcium (Figure 21), was able to inhibit SARS‐CoV 3CL^pro (CN 108785293 A; Table 1). ^[40] The patent is written in Chinese and no English version could be found online. Unfortunately, no IC₅₀ or K _i value was reported and there are no follow‐up reports in academic journals. Hence, we are currently unable to comment on its drug development potential for treating COVID‐19.

Figure 21.

Structure of acamprosate calcium from CN 108785293 A.

The twenty‐second patent was filed by GSK in 2017 involving 3‐residue peptidomimetics containing a P1 5‐membered lactam glutamine mimic with a C‐terminal electrophilic ketoamide warhead (WO 2018/042343; Table 1). ^[41] 48 structures were described with SARS‐CoV 3CL^pro IC₅₀s of approximately 7 nM. The five most potent inhibitors against SARS‐CoV 3CL^pro are shown in Figure 22. There are currently no reports on their SARS‐CoV‐2 3CLpro inhibitory activities based on a SciFinder® search. In our opinion, these compounds can potentially be developed as SARS‐CoV‐2 3CL^pro inhibitors to treat COVID‐19 as oral ketoamide peptidomimetic protease inhibitors such as Boceprevir and Telaprevir have been approved for treating hepatitis C virus infections. ^[64]

Figure 22.

Structures from WO 2018/042343. (a) general scaffold; (b–e) exemplified structures.

A special note is also made for WO 2020/030143 (Table 1), filed by Shanghai Institute of Materia Medica and Fudan University in 2019. ^[42] It involves 2‐residue peptidomimetics bearing a P1 5‐ or 6‐membered lactam glutamine mimic with a C‐terminal electrophilic ketoamide warhead designed to inhibit MERS‐CoV 3CL^pro. The patent is written in Chinese and the English version could not be found online. No SARS‐CoV or SARS‐CoV‐2 3CL^pro IC₅₀ data were reported in the patent. However, this patent is included in this review as the MERS‐CoV and SARS‐CoV‐2 3CLpro share approximately 50 % sequence identity ^[16] so inhibitors designed to inhibit MERS‐CoV 3CL^pro will likely be able to inhibit SARS‐CoV‐2 3CL^pro. 253 compounds were described with IC₅₀s ranging from sub‐nanomolar to 786 nM using a MERS pseudovirus neutralization assay. The five most potent MERS‐CoV 3CL^pro inhibitors, 23 b to 23 f, possess sub‐nanomolar to 5 nM IC₅₀s (Figure 23). There are currently no reports on their SARS‐CoV‐2 3CL^pro inhibitory activities based on a SciFinder® search. In our opinion, these compounds can potentially be developed as SARS‐CoV‐2 3CL^pro inhibitors to treat COVID‐19 as oral ketoamide peptidomimetic protease inhibitors such as Boceprevir and Telaprevir have been approved for treating hepatitis C virus infections. ^[64]

Figure 23.

Structures from WO 2020/030143. (a) general scaffold; (b–f) most potent exemplified MERS‐CoV 3CL^pro inhibitors.

The final patent in this review is an application patent involving 13 peptidomimetic protease inhibitors (WO 2021/176369; Table 1). ^[43] Interestingly, 6 peptidomimetics with various C‐terminal electrophilic warheads have been described in an earlier patent (WO 2005/113580), ^[25] all bearing a P1 5‐membered lactam glutamine mimic. No 3CL^pro inhibitory data were disclosed in the patent but SARS‐CoV 3CL^pro IC₅₀s for compounds 24 b to 24 e (Figure 24) were reported in a recent research paper. ^[18] Surprisingly, the patent also described the peptidomimetic rhinovirus 3C protease inhibitor, Rupintrivir (24 f), as a potential COVID‐19 therapeutic despite its lack of SARS‐CoV‐2 3CL^pro inhibitory activity. ^[62] Three hepatitis C and human immunodeficiency virus protease inhibitors were also described; Filibuvir (24 g), Nelfinavir (24 h) and AG‐1859 (24 i). No IC₅₀s were reported in the patent but a literature search revealed Nelfinavir to possess weak SARS‐CoV‐2 3CL^pro inhibitory activity (IC₅₀ 41.5 μM). ^[70] In our view, the peptidomimetic hydroxymethylketones 24 b and 24 c possess the highest potential for further drug development if converted to a phosphate prodrug as observed for clinical candidate PF‐00835231 (6 b). ^[18] In contrast, we opine that Rupintrivir (24 f) and Nelfinavir (24 h) are not potent enough for further development. We are unable to comment on the drug development potential for Filibuvir (24 g) and AG‐1859 (24 i) due to the lack of SARS‐CoV‐2 3CL^pro inhibitory data.

Figure 24.

Structures from WO 2021/176369. (a) peptidomimetic general scaffold; (b–i) exemplified inhibitors.

Lastly, a special mention is made for Pfizer's oral SARS‐CoV‐2 3CL^pro peptidomimetic inhibitor PF‐07321332 (Figure 25) which entered phase 3 clinical trials in July 2021 for treating non‐hospitalised SARS‐CoV‐2 infected adults (ClinicalTrials.gov identifier NCT04960202). ^[71] Like previous Pfizer 3CL^pro inhibitor patents, it bears a P1 5‐membered lactam glutamine mimic and interestingly, a nitrile warhead not described in any patents found in this review. This is particularly intriguing as nitrile warhead peptidomimetics have been reported to be relatively weak SARS‐CoV 3CL^pro inhibitors with IC₅₀s ranging between 4.6 to 49 μM. ^[72] To our best knowledge, there are currently no published reports on its SARS‐CoV‐2 3CL^pro IC₅₀ and a SciFinder® structure search did not reveal any patents for PF‐07321332.

Figure 25.

Structure of PF‐07321332.

3. Summary and Outlook

24 patents describing coronavirus 3CL^pro inhibitors have so far been filed up to 30 July 2021 (Table 1). 9 of the patents were filed by pharmaceutical companies while the remaining 15 from academia. 10 of the 24 patents described inhibitors that have not been reported in academic journals. 14 of the 24 patents involved peptidomimetics with electrophilic warheads, signifying that this modality has generated more commercial interest compared to small molecules. A plausible reason could be that warhead peptidomimetics generally show more potent inhibitory activities, typically in the nanomolar range, compared to small molecules. In addition, warhead peptidomimetics have been successfully developed into antiviral drugs, exemplified by the hepatitis C virus protease inhibitors Boceprevir and Telaprevir. ^[64]

The peptidomimetic inhibitors covered in this review bear a P1 5‐ of 6‐membered lactam glutamine‐mimicking residue first described by Agouron Pharmaceuticals in 1999 (WO 99/57135). ^[46] Reported P2 residues ranged from leucine to cyclohexylalanine and phenylalanine, suggesting that the SARS‐CoV S2 subsite could accommodate bulky residues with 6‐membered side‐chains. However, a 2020 research paper by Pfizer revealed that a P2 leucine exhibited 11‐ and 25‐fold IC₅₀ improvement over cyclohexylalanine and phenylalanine respectively, suggesting that leucine was highly preferred at the P2 position. ^[18] Indeed, Pfizer utilised a P2 leucine in their phosphate prodrug of PF‐00835231 (6 b) which entered phase 1 clinical trials in September 2020 (ClinicalTrials.gov identifier: NCT04535167). Other plausible P2 residue substitutions include norvaline and t‐butylalanine (Figure 24). The P3 position is more accommodating to different moieties including indoles (Figures 6, 10 and 18), amino acid residues like threonine (Figure 3), valine (Figures 5 and 8) and naphthylalanine (Figures 14 and 22). The longest peptidomimetic inhibitor in this review constituted 4 residues (Figure 8) but unfortunately, no IC₅₀s were reported as it would be interesting to study the correlation of peptidomimetic length to inhibition potency.

Small molecule inhibitors constituted 10 of the 24 patents (Table 1). In contrast to the peptidomimetic inhibitors, their reported IC₅₀s or K _is were generally inferior (close to or above 1 μM), with the exception of two patents filed by Fulcrum Pharmaceuticals and the National Health Research Institutes, Taiwan (Table 1; Figures 4 and 7 respectively). The former described a small molecule containing two boronic acid moieties (K _i 22 nM) and the latter, a biphenyl sulphone (IC₅₀ 300 nM). In our view, both warrant further investigations into their drug development potential.

We believe this review serves to bolster the knowledge of existing coronavirus 3CL^pro inhibitors reported beyond academic journals and the general structural motifs required for binding and inhibiting SARS‐CoV‐2 3CL^pro. We are hopeful that some of these inhibitors will be successfully developed and approved as drugs for treating COVID‐19.

Abbreviations

3CL^pro

3‐chymotrypsin‐like protease

CoV

coronavirus

COVID‐19

coronavirus disease 2019

EC₅₀

half‐maximal effective concentration

FDA

Food and Drug Administration

GABA

gamma aminobutyric acid

IC₅₀

half‐maximal inhibitory concentration

MERS

Middle East Respiratory Syndrome

M^pro

main protease

NMDA

N‐methyl‐D‐aspartate

SARS

severe acute respiratory syndrome

WHO

World Health Organization

Conflict of interest

The authors declare no conflict of interest.

Biographical Information

Brian Chia obtained his B.Sc. with first‐class honours and a Ph.D. from the University of Adelaide followed by postdoctoral training at Cambridge University, UK. In 2005, he joined the Institute of Molecular and Cell Biology before becoming a group leader at the Experimental Drug Development Centre, Agency for Science, Technology and Research (A*STAR). He has co‐authored more than 50 research papers and is an inventor/co‐inventor of 8 granted or pending patents involving small molecules, peptides and peptidomimetics with antibacterial, anticancer, antifungal and antiviral properties.

Biographical Information

Weijun Xu obtained his B.Sc. with first‐class honours in Biochemistry and a Ph.D. from the University of Queensland followed by postdoctoral training at the University of California, San Diego. He lectured at the Singapore Polytechnic School of Chemical and Life Sciences for 8 years before his current position at the Experimental Drug Development Centre, Agency for Science, Technology and Research (A*STAR) as a senior research fellow. His research interests include computational biology, modelling of protein−ligand and protein−protein interactions including GPCRs, proteases and kinases inhibitors.

Biographical Information

Pearly S. Ng is a medicinal chemist who has been involved in the development of various novel small molecule therapeutics for infectious diseases in the past decade. Prior to being a medicinal chemist, she spent 6 years as a research chemist in various companies working on diversity‐oriented synthesis and parallel synthesis technologies. She obtained her Ph.D. in organic chemistry from Nanyang Technological University and has co‐authored several research publications in medicinal and organic chemistry. She is also a co‐inventor of 3 patent applications.

C. S. B. Chia, W. Xu, P. Shuyi Ng, ChemMedChem 2022, 17, e202100576.