Innate immune evasion mediated by picornaviral 3C protease: Possible lessons for coronaviral 3C‐like protease?

Summary

Severe acute respiratory syndrome coronavirus‐2 is the etiological agent of the ongoing pandemic of coronavirus disease‐2019, a multi‐organ disease that has triggered an unprecedented global health and economic crisis. The virally encoded 3C‐like protease (3CL^pro), which is named after picornaviral 3C protease (3C^pro) due to their similarities in substrate recognition and enzymatic activity, is essential for viral replication and has been considered as the primary drug target. However, information regarding the cellular substrates of 3CL^pro and its interaction with the host remains scarce, though recent work has begun to shape our understanding more clearly. Here we summarized and compared the mechanisms by which picornaviruses and coronaviruses have evolved to evade innate immune surveillance, with a focus on the established role of 3C^pro in this process. Through this comparison, we hope to highlight the potential action and mechanisms that are conserved and shared between 3C^pro and 3CL^pro. In this review, we also briefly discussed current advances in the development of broad‐spectrum antivirals targeting both 3C^pro and 3CL^pro.

Abbreviations

ACLY

ATP‐citrate synthase

ASCs

apoptosis‐associated speck‐like protein containing a caspase recruitment domain

avSGs

antiviral stress granules

CARD

caspase activation and recruitment domain

cGAS

cyclic GMP‐AMP synthase

Covid‐19

coronavirus disease‐2019

CVA6/A16/B3

Coxsackievirus‐A6/A16/B3

CYLD

cylindromatosis

CALCOCO2

coiled‐coil domain‐containing protein 2

DAMPs

danger associated molecular patterns

DCP1/2

decapping protein ½

DDX6

DEAD‐box RNA helicase‐6

DHAV

duck hepatitis A virus

DHX36

DEAH‐box helicase 36

dsRNA

double‐stranded RNA

eIF4E/F/G

eukaryotic initiation factor 4E/F/G

EDC3/4

enhancer of mRNA decapping protein‐3/4

EMCV

encephalomyocarditis virus

ERAV

equine rhinitis A virus

EV‐A71/D68

enterovirus‐A71/D68

FMDV

foot‐and‐mouth disease Virus

GBF1

golgi‐specific brefeldin A‐resistance guanine nucleotide exchange factor

G3BP1

Ras‐GTPase‐activating SH3 domain binding protein 1

GSDMD

gasdermin D

HAV

hepatitis A virus

HDAC2

histone deacetylase 2

HRV

human rhinovirus

hnRNP M/K

heterogeneous nuclear ribonucleoprotein M/K

IFN‐I

type‐I interferon

IKKε

IκB kinase‐ε

IRES

internal ribosomal entry sites

IRF3/7/9

interferon regulatory factor‐3/7/9

ISGs

interferon‐stimulating genes

K63‐linked

lysine‐63‐linked

M protein

membrane/matrix protein

MAVS

mitochondrial antiviral signaling protein

MDA5

melanoma differentiation‐associated protein‐5

MERS‐CoV

Middle East respiratory syndrome‐CoV

MEX3C

Mex‐3 RNA binding family member C

mRNPs

messenger ribonucleoproteins

NDP52

nuclear dot 10 protein 52

NLRP3

NLR family PYD containing protein‐3

N proteins

nucleocapsid proteins

NEMO

NF‐κB essential modulator

NFκB

nuclear factor‐κB

NLRs

(NOD)‐like receptors

NSP5

nonstructural protein 5

ORF1a/3a/3b/6/8b

open reading frame 1a/3b/6/8b

PAMPs

pathogen‐associated molecular patterns

PDCoV

porcine deltacoronavirus

PDCD6IP

programmed cell death 6‐interacting protein

PEDV

porcine epidemic diarrhea virus

PFAS

phosphoribosylformylglycinamidine synthetase

PKR

protein kinase R

PLEKHM1

pleckstrin homology domain and RUN domain containing M1

PRRs

pathogen recognition receptors

PRRSV

porcine reproductive and respiratory syndrome virus

PLpro

papain‐like protease

PV

poliovirus

RIP1

receptor‐interacting serine/threonine‐protein kinase 1

RLRs

RIG‐I‐like receptors

RdRp

RNA‐dependent RNA polymerase

SARS‐CoV‐2

severe acute respiratory syndrome coronavirus‐2

SNAP29

synaptosomal‐associated protein 29

ssRNA

single‐stranded RNA

STAT2

signal transducer and activator of transcription 2

STX17

Syntaxin 17

SQSTM1

sequestosome 1

SVV

Seneca Valley virus

TAB1

TAK1 binding protein‐1

TAILS

terminal amine isotopic labeling of substrates

TBK1

TANK binding kinase‐1

TLRs

toll‐like receptors

TRAF3

TNF receptor associated factor‐3

TRIM25

tripartite motif containing 25

TRIF

(TIR)‐domain‐containing adapter‐inducing interferon‐β

TRMT1

tRNA methyltransferase 1

UMP

uridine monophosphate

UTR

untranslated region

VAMP8

vesicle‐associated membrane protein 8

VPg

viral protein genome‐linked

XRN1

5′‐3′ exoribonuclease 1

2A^pro

2A‐protease

3CL^pro/3C^pro

3C‐like protease/3C protease

3D^pol

3D polymerase

1. INTRODUCTION

Coronavirus disease‐2019 (Covid‐19) is a multiple organ disease that has posed an unprecedented health and economic threat worldwide since its emergence in late 2019. Covid‐19 is caused by a novel virus strain, ¹ , ² namely severe acute respiratory syndrome‐coronavirus‐2 (SARS‐CoV‐2) ³ , categorized within the family Coronaviridae. It can infect various hosts and target multiple organs through the body. ⁴ Coronaviruses are a family of enveloped, single‐stranded RNA (ssRNA) viruses with positive polarity that can cause respiratory, enteric, cardiovascular, and central nervous system diseases. ⁵ , ⁶ This family of RNA viruses features the second largest genome size (27−31 kb) found to date right after planarian nidovirus (∼41.1 kb). ⁷ Together with SARS‐CoV and Middle East Respiratory Syndrome‐CoV (MERS‐CoV) that caused SARS and MERS outbreaks in 2003 and 2012, respectively, SARS‐CoV‐2 belongs to the genus betacoronavirus. The genome of betacoronavirus encodes more than 20 proteins, including four major structural proteins (i.e., a spike (S) protein that binds to the cell receptor and mediates fusion between virus and cell membrane, a small envelope (E) protein, a highly hydrophobic membrane (M) protein, and a nucleocapsid (N) protein that interacts with viral RNA to form a helical nucleocapsid structure), two cysteine proteases (i.e., a papain‐like cysteine protease (PL^pro) and a 3‐chymotrypsin‐like cysteine protease (3CL^pro, also known as the main protease, M^pro) that processes viral polyproteins into individual functional proteins, a helicase required for unwinding double‐stranded RNA (dsRNA), a RNA‐dependent RNA polymerase (RdRp) that catalyzes the replication of RNA from RNA template, and other enzymes such as endo‐ and exonucleases essential for viral nucleic acid metabolism. ⁸

Among these proteins, SARS‐CoV‐2 proteases play a vital role in viral replication and transcription, thereby being recognized as attractive antiviral targets for Covid‐19 treatment. ⁹ , ¹⁰ Of the two known CoV proteases that are encoded by open reading frame 1a (ORF1a), 3CL^pro [corresponding to nonstructural protein 5 (NSP5)], which is highly conserved among all CoV 3CL^pro, has been identified to be structurally analogous to the 3C^pro of picornaviruses (3CL^pro is named after the picornaviral 3C^pro). ¹¹ , ¹² Despite subtle structural differences in the active sites, 3CL^pro and 3C^pro share a similar chymotrypsin‐like tertiary structure with a catalytic triad (or dyad) site containing a cysteine nucleophile (Figure 1). Moreover, both of the enzymes have a strong preference for glutamine (Gln) at the P1 position of their targets, the most key determining factor for their substrate recognition. The conserved active sites of 3C^pro and 3CL^pro have been confirmed by high‐resolution three‐dimensional structural analysis. Therefore, it is proposed to serve as an attractive target for the design of broad‐spectrum antiviral drugs. ¹³ , ¹⁴ , ¹⁵ Picornaviruses are small, non‐enveloped viruses containing a positive‐sense, ssRNA genome with a length of 7.0–8.5 kb. This family comprises 29 genera, including Apthovirus (e.g., foot‐and‐mouth disease virus, FMDV), Cardiovirus (e.g., encephalomyocarditis virus, EMCV), Enterovirus (e.g., poliovirus, PV; coxsackievirus A16/B3, CVA16; CVB3; enterovirus‐A71/D68, EV‐A71; EV‐D68), Rhinovirus (e.g., human rhinovirus, HRV), and Hepatovirus (e.g., hepatitis A virus, HAV) genera. ¹⁶ Picornavirus genomic RNA at its 5′ end is covalently linked to a small viral protein (VPg, also known as 3B) that serves as a primer for the initiation of viral RNA replication. Further, instead of a cap structure, the genome of picornaviruses possesses an element termed internal ribosome entry site (IRES) in their 5′‐untranslated region (UTR), which is necessary for initiating a cap‐independent translation of viral RNA. The viral genome of picornaviruses contains one open reading frame encoding a single viral polyprotein that undergoes proteolysis by two viral proteases, 2A^pro and 3C^pro, with the latter being responsible for the majority of the maturation cleavage events of viral polyprotein similar to coronaviral 3CL^{pro 17} . In addition to processing viral polyprotein, picornaviral proteases also target cellular proteins to evade the human immune surveillance and facilitate viral infection. ¹⁸

FIGURE 1.

Crystal structures and superposition of picornaviral 3C protease (3C^pro) and coronaviral 3C‐like protease (3CL^pro). (a) Ribbon overlay of the picornaviral 3C^pro structures of poliovirus (PV; PDB 1L1N), foot‐and‐mouth disease virus (FMDV; PDB 2BHG), and human rhinovirus (HRV; PDB 1CQQ). (b) Ribbon overlay of human coronaviral 3CL^pro structures of severe acute respiratory syndrome‐coronavirus (SARS‐CoV; PDB 2Q6G), Middle East Respiratory syndrome‐CoV (MERS‐CoV; PDB 4YLU), SARS‐CoV‐2 (PDB 6M2N), and HCoV‐HKU1 (PDB 3D23). (c) A side‐by‐side comparison of PV 3C^pro and SARS‐CoV‐2 3CL^pro with the two domains of the chymotrypsin‐like fold highlighted and the active site catalytic residues labeled and highlighted (red). (d) Close‐up images of the active site catalytic residues of PV 3C^pro and SARS‐CoV‐3CL^pro are shown

Given the common characteristics of 3C^pro and 3CL^pro, we postulate that SARS‐CoV‐2, like picornaviruses, is capable of regulating host innate antiviral processes through the catalytic activity of its 3CL^pro. The delay or inhibition of multiple host antiviral machineries would allow effective viral growth and subsequently optimal release and infection. Here we will recapitulate some of the scenarios on how picornaviruses utilize its 3C^pro to target major host antiviral mechanisms.

1.1. Structural and functional similarities between picornaviral 3C^pro and coronaviral 3CL^pro

Early during the viral replication cycle, the positive‐sense ssRNA (+ssRNA) genomes of picornaviruses and coronaviruses are translated into one or more polyproteins, which include integrated viral protease domains. Maturation cleavage events mediated by virally encoded proteases in both backgrounds are indispensable for virus replication. The picornaviral 3C^pro (working in concert with the 2A^pro) mediates the majority of viral cleavage events including autocleavage from the 3D polymerase (3D^pol) domain of the virus. Similarly, coronaviral 3CL^pro is responsible for at least 11 maturation cleavages of the viral replicase polyproteins, including its own autoproteolytic cleavage. Beyond these requisite viral polyprotein cleavages, both 3C^pro and 3CL^pro target and cleave host cellular proteins. Due to the importance in their respective viral backgrounds, these proteases have been extensively studied as primary targets for viral inhibition for over 30 years.

The picornaviral 3C^pro is a chymotrypsin‐like cysteine protease comprised of two β‐barrel domains of six antiparallel strands which surrounds a conserved Cys‐His‐Asp/Glu catalytic triad (Figure 1). ¹⁹ , ²⁰ The protease preferentially cleaves after a P1‐Gln with greater cleavage site variability in the other cleavage site residue positions. ²¹ During picornaviral replication, the 3CD^pro precursor, which is able to process the P1 structural precursor but lacks polymerase activity, is cleaved to release 3C^pro and viral RdRP 3D^{pol 22} . The conversion of 3CD^pro to 3C^pro plays a critical role in facilitating the transition and regulation from viral translation to replication. ²³ , ²⁴ Structurally and functionally analogous to the picornaviral 3C^pro, the coronaviral 3CL^pro is an approximately 300 residue, 3‐domain protease. Domains 1 and 2 comprise the substrate‐binding and enzymatic active sites of the protease with dimerization driven by interactions between the structurally unique and largely helical domain 3. ^{25 , 26} Domains 1 and 2 of 3CL^pro form a chymotrypsin‐like fold comprised of antiparallel β‐barrels housing the His‐Cys catalytic dyad residues. ²⁶ , ²⁷ The antiparallel β‐barrel conformation within domains 1 and 2 surrounding and forming the active site of 3CL^pro shares structural similarity to the core structure of the picornaviral 3C^pro, albeit with subtle differences in strand positioning (Figure 1). Unlike the picornaviral 3C^pro (monomer with only two catalytic domains), an attached helical third‐domain in 3CL^pro facilitates dimerization of the protease, an essential event for its enzymatic activity and viral replication (Figure 1C, Domain 3). ²⁸ While targeting the third domain of 3CL^pro serves as a valid strategy to disrupt its dimerization and catalytic activity, any inhibitors identified or designed to do so likely have no impact on picornaviral 3C^pro activity since dimerization is not necessary for its function. In addition, among coronaviruses, there is considerably more structural conservation for the chymotrypsin‐like fold of 3CL^pro compared to that of the picornaviral 3C^pro. Based on large part to their structural and functional similarities between the 3C^pro and 3CL^pro, it remains unclear to what extent these proteases share common host cellular targets during viral replication and what role these potentially shared cleavages play in viral replication.

2. THE SUBVERSION OF HOST DEFENSE MECHANISMS BY picornaviral 3C^pro

2.1. Targeting type I interferon signaling pathway

The effectiveness of an antiviral innate immunity depends on the accurate recognition of viral moieties, known as the pathogen‐associated molecular patterns (PAMPs), by pattern‐recognition receptors (PRRs) composed of at least three classes: retinoic acid‐inducible gene‐I (RIG‐I)‐like receptors (RLRs), Toll‐like receptors (TLRs), and nucleotide‐binding oligomerization domain (NOD)‐like receptors (NLRs). ²⁹ Upon RNA viral infection, dsRNAs often accumulate in cells in the form of the viral genome or its replication intermediates. The dsRNAs can be recognized by cytosolic viral RNA sensor, RLRs (e.g., RIG‐I (encoded by Ddx58 gene) and melanoma differentiation‐associated proteins (MDA5, encoded by Ifih1 gene)), and/or endosomal viral RNA sensor (e.g., TLR3) to initiate type I interferon (IFN‐I) immune response. While RIG‐I preferentially binds to shorter dsRNA (<1‐2 kb) bearing 5′‐triphosphate group, MDA5 primarily recognizes long dsRNA. ³⁰ Similar to picornavirus, SARS‐CoV‐2 has a long RNA genome. It is therefore expected that SARS‐CoV‐2 RNA favorably binds to MDA5 rather than its paralog RIG‐I. Indeed, MDA5 has been previously shown to be the specific PRR that recognizes murine coronavirus RNA. ³¹ Interaction between viral RNA and MDA5 forms MDA5 filaments along dsRNAs, which brings together neighboring caspase activation and recruitment domain (CARD) in close proximity to induce oligomerization and activation of the adapter mitochondrial‐antiviral signaling protein (MAVS). ³² Activated MAVS then transmits the signals to its downstream transcription factors, interferon regulatory factor‐3/7 (IRF3/7) and nuclear‐factor‐κB (NF‐κB), through TANK‐binding kinase‐1 (TBK1) and IκB kinase‐ε (IKKε). The homo‐ or hetero‐dimerized IRF3/7 subsequently translocate to the nucleus and induce the expression of IFN‐I‐associated genes (Ifna and Ifnb1), which could further activate the Janus kinase/signal transducers and activators of transcription (JAK/STAT) signaling cascade to trigger the expression of antiviral genes, called interferon‐stimulated genes (ISGs). ³⁰ Please see Figure 2 for the details.

FIGURE 2.

Picornaviruses evade type I interferon immune response via the function of 3C protease (3C^pro). Binding of picornaviruses to their respective receptors facilitates their entry into the cells and release of the 5′‐viral protein genome‐linked‐containing genomic RNA into cytoplasm. Long double‐stranded RNA generated during the replication process binds to MDA5, exposing its CARD and allowing homotypic CARD‐CARD interactions with its downstream adapter, MAVS. Subsequently, MAVS triggers the expression of IFN‐I genes (Ifnb1 and Ifna in dendritic cells) and ISGs for antiviral purposes through the activation of transcription factor IRF3/7 and NF‐κB. To facilitate a robust signaling, more efficient detection of dsRNA occurs in antiviral stress granules. Targets of viral encoded 3Cpro are indicated. CARD, caspase activation and recruitment domain; CTD, C‐terminal binding domain; G3BP1, Ras GTPase‐activating protein‐binding protein 1; IFN‐I, type‐I interferon; IKKε, inhibitor of nuclear factor‐κB (IκB)‐kinase‐ε; IRF3/7, interferon regulatory factors‐3/7; ISGs, interferon‐stimulating genes; MAVS, mitochondrial antiviral signaling protein; MDA5, melanoma‐differentiation associated protein‐5; NF‐κB, nuclear factor‐κB; NEMO, NF‐κB essential modulator; P, phosphate‐group; RIG‐I, retinoic acid‐inducible gene‐I; TBK1, TANK binding kinase‐1; TLR3, Toll‐like receptor 3; TRAF3, TNF‐receptor associated factor‐3; TRIF, Toll/IL‐1 receptor domain‐containing adapter‐inducing interferon‐β

To antagonize human antiviral innate immune response, picornaviruses target critical components of the RLR signaling pathway for degradation. It was reported that infections with Seneca Valley virus (SVV), PV, EV‐A71, EV‐D68, CVB3, CVA16, CVA6, HRV‐1A, EMCV, and HAV cleave MDA5, MAVS, RIG‐I, IRF7, and/or IRF9 through the actions of 3C^pro, leading to a disruption of RLR‐mediated IFN‐I immune responses. ³³ , ³⁴ , ³⁵ , ³⁶ , ³⁷ , ³⁸ , ³⁹ In addition to 3C^pro, studies have found that MDA5 and MAVS can also be targeted by 2A^pro upon PV, EV‐A71, CVB3, and HRV‐1A infections. ³⁴ , ⁴⁰ , ⁴¹ Moreover, both TAK1 binding protein‐1 (TAB1) and NF‐κB essential modulator (NEMO, an adapter protein bridging the canonical IKKα/β kinases and the noncanonical kinases TBK1/IKKε via the TANK adapter ⁴² ) are cleaved by 3C^pro following EV‐A71, FMDV and HAV infections, resulting in reduced production of IFN‐I. ⁴³ , ⁴⁴ , ⁴⁵ In addition, FMDV utilizes its 3C^pro to inhibit STAT2 function, a component of the IFN‐stimulated gene factor 3 complex, which is also mirrored by the 3CL^pro of PDCoV (Table 1, yellow highlighted row). ⁵⁷ , ⁵⁸ Another interesting finding associated with this topic is 3C^pro‐induced cleavage of Toll/IL‐1 receptor (TIR)‐domain‐containing adapter‐inducing interferon‐β (TRIF) during CVB3, EV‐A71, and HAV infections. ³⁶ , ⁴⁷ , ⁴⁸ TRIF is an adapter protein mediating type I IFN antiviral response downstream of the endosomal TLR3 (a viral RNA sensor) ¹¹⁴ and cleavage causes the loss of its function in host defense.

TABLE 1.

Picornaviruses and coronaviruses proteinase targets [Correction added on 17 February 2021, after first online publication. Table 1 was updated in this version.]

	Proteinase	Virus	Gene symbol	Full gene name	Cleavage site	Implication to host	Ref
Host defense and inflammation	3C	CVB3, RV, PV, EV‐A71, SVV	RIG‐I	Retinoic acid‐inducible gene‐I	N/A	Dampened cytokines production; enhanced viral propagation	33, 39
	2A	PV, EV‐A71, CVB3	MDA5	Melanoma differentiation‐associated protein 5	N/A		37, 40
	2A, 3C, 3CLSP	CVB3, HRV, EV‐A71, PV, SVV, PRRSV	MAVS	Mitochondrial antiviral signaling	Q148↓, Q209↓, Q251↓, Q265↓, Q268↓		34, 36, 38, 40, 41, 46
	3C	EV‐D68, EV‐A71, CVB3, HAV, SVV	TRIF	TIR domain‐containing adapter inducing interferon beta	Q159↓, Q190↓, Q312↓, Q554↓, Q653↓		36, 47, 48
	PL	SARS‐CoV‐2	IRF3	Interferon regulatory factor 3	G270↓		49
	3C	EV‐D68, EV‐A71	IRF7	Interferon regulatory factor 7	Q167↓, Q189↓		35, 50
	3C	EV‐A71	IRF9	Interferon regulatory factor 9	N/A		51
	2A, 3C	EV‐A71, CVB3	NLRP3	NACHT, LRR, and PYD domain‐containing protein 3	Q225↓226G, G493↓494L		52, 53
	3CL	SARS‐Cov‐2	NLRP12	NACHT, LRR, and PYD domain‐containing protein 12	Q238↓, Q938↓		49
	3C	EMCV, SVV	TANK1	TRAF family member‐associated NF‐κB activator‐1	E272 and Q291		54
	2A	EV71	IFNAR	Interferon alpha and beta receptor subunit 1	N/A		55
	3C	HRV, PV	C3	Complement C3	N/A		56
	3C, 3CL	FMDV, PDCoV	STAT2	Signal transducer and activator of transcription 2	Q685↓, Q758↓		43, 44, 57, 58, 59, 60, 61
	3C, 3CL, 3CLS	FMDV, HAV, PEDV, PRRSV	NEMO	NF‐κB essential modulator	Q304↓, Q383↓, Q231↓, E349↓350S
	3C	EV‐D68, EV‐A71, CVA6, CVA16	TAK1	Transforming growth factor‐β‐activated kinase 1	Q360↓361S		37, 45, 49
	3C, 3CL	EV‐A71, SARS‐CoV‐2	TAB1	TAK1 binding protein 1	Q414↓415G, Q451↓452S, Q132↓133S, Q444↓445S
	3C	EV‐A71	TAB2	TAK1 binding protein 2	Q113↓114S
	3C	EV‐A71	TAB3	TAK1 binding protein 3	Q173↓174G, Q343↓344G
	3C	CVB3, PV	RIP1, RIP3	Receptor interacting protein‐1, ‐3	Q134↓, Q430↓, R118↓119I		62, 63
	3C	EV‐A71	ZAP	Zinc‐finger antiviral protein	Q369↓370G		64
	3C	EV‐A71	GSDMD	Gasdermin‐D	Q193↓194G	Inhibition of pyroptosis & promotion of viral growth	65
Autophagy	2A	CVB3	p62/SQSTM1	Sequestosome‐1	G241↓	Disruption of cellular autophagy pathway	66, 67, 68, 69, 70
	2A	CVB3	NBR1	Neighbor of BRCA1 gene	G402↓, E682↓
	3C	CVB3	CALCOCO2/NDP52	Calcium binding and coiled‐coil domain containing protein‐2; Nuclear dot 10 protein 52	Q139
	3C	CVB3	SNAP29	Synaptosomal‐associated protein 29	Q161↓
	3C	CVB3	PLEKHM1	Pleckstrin homology domain and RUN domain Containing M1	Q668↓
Cellular Integrity	2A	CVB3	DYSF	Dysferlin	N/A	Disruption of muscle membrane repair	71
	2A	CVB3, CVB4	DMD	Dystrophin	Human: (G589↓; G2,435↓)Mouse: (G591↓; G2,428↓)	Muscular dystrophy, especially heart muscle	72
	2A	HRV2	CK‐II	Cytokeratin K8	G15↓	Disruption of cytoplasmic cytoskeleton	73
	3C	PV	MAP‐4	Microtubule‐associated protein 4	Q188↓189G	Collapse of microtubules, disruption of host protein dynamics and interactions	74
Transcription	3C	PV	Oct‐1	Octomer binding transcription factor‐1	Q330↓331G	Disruption of polymerase activity, affecting gene transcription	75
	3C	PV	La/SSB	Lupas autoantigen/Sjogren syndrome antigen B	Q358↓359G		76
	2A	PV	TBP	TATA‐binding protein	Q18↓19G, Q104↓105S		77, 78, 79, 80
	3C	PV	TFIIIA/C	Transcription factor for polymerase III A	N/A		81, 82
	3C	PV	CREB	cAMP response element‐binding protein	Q172↓173G
	2A	CVB3	SRF	Serum response factor	G327	Disruption of SRF‐dependent gene expression (e.g. cardiac contractile and regulatory factors)	83
	3C	PV	PTB/p52/hnRNAP‐I	Polypyrimidine tract‐binding protein	Q148/152/321A	Disruption of normal RNA metabolism (e.g. splicing)	84
	3C	PV	hnRNP M, K	Heterogenous nuclear ribonucleoprotein M, K	G389↓390E, Q364↓365G		85
	2A	PV	Gemin3	Gemin3	G463↓		86
	3C	CVB3	TDP‐43	Transactive response DNA‐binding protein‐43	Q327↓328A		87
	3C	PV	p65‐RelA	Nuclear factor (NF)‐κB p65 subunit	Q480↓481G	Disruption of p65‐dependent gene expression	88
Translation and RNA turn over	3C	CVB3	AUF1	(AU)‐rich element RNA‐binding factor 1	N/A	Disruption of host mRNA turn over	89
	2A, 3C	PV, CVB3, HAV, EMCV, FMDV, DHAV	PABP	Poly(A)‐Binding protein	Q537↓, Q367↓368G, M487↓, G488↓, Q437↓438G	Host protein translation shut‐off, disruption of SG formation	90, 91, 92, 93
	3C	FMDV	eIF4A	Eukaryotic elongation factor 4A	G674↓675R, R481↓482G		94, 95, 96, 97, 98, 99
	L, 2A, 3C	FMDV, PV, CVB3, EV‐A71, RV	eIF4G‐I, ‐II	Eukaryotic elongation factor 4G‐I/4G‐II
	3C	PV, CVB3, HRV	eIF5B	Eukaryotic elongation factor 5B	Q478↓479G		100
	2A	CVB3	DAP5	Death‐associated protein 5	G434↓		101
	3C	PV	PCBP2	Poly(rC)‐binding protein‐2	Q253↓, Q306↓, S254↓		102
	3C	PV	DCP1a	mRNA‐decapping enzyme 1A	N/A	Disruption of P‐body foci; disruption of 5’→3′ guanosine cap removal and Xrn1‐Dcp1‐dependent antiviral pathway	103
Others	3C	CVB3	IKBα	Inhibitor of κBα	Q249↓250G	Constitutive activation of NFκB and apoptosis	104
	2A, 3C	CVB3	GAB1	Growth factor receptor bound 2‐associated binding protein‐1	G175↓, G436↓, G238↓	Disruption the assembly of protein complexes for intracellular signaling	105
	2A	HRV, PV	Nup62, Nup98, Nup153,	Nucleoporins	A103↓, G177↓, G201↓, G218↓, G247↓, A298↓,	Temporal blocking of nucleocytoplasmic trafficking	106, 107
	L, 3C	CVB3, PV, EMCV, FMDV, ERAV	G3BP1, G3BP2	Ras‐GTPase activating‐SH3‐binding protein‐1	Q325↓, E284↓	Disruption of SG and innate immune response	108, 109, 110, 111, 112, 113
	3C	CVB3, PV	GBF1	Golgi‐specific brefeldin A‐resistance guanine exchange factor‐1	Q1,297↓1298S	Disruption of host secretory pathway	63
			ACLY	ATP citrate lyase/synthase	Q777↓778A	Disruption of fatty acid biosynthesis
			p115/USO1	General vesicular transport factor p115	Q832↓833G	Disruption of the host peripheral membrane recycling process between cytosol and Golgi
			ALIX/PDCD6IP	Programmed cell death 6‐interacting protein	Q728↓729S	Disruption of cellular apoptosis
			PFAS	Phosphoribosylformylglycinamidine synthetase	Q472↓473G	Disruption of host ATP, and L‐glutamine biosynthesis

Abbreviations: CVA6/A16/B3, Coxsackievirus‐A6/A16/B3; CALCOCO2, coiled‐coil domain‐containing protein 2; DHAV, duck hepatitis A virus; EMCV, encephalomyocarditis virus; ERAV, equine rhinitis A virus; FMDV, foot‐and‐mouth disease virus; GBF1, golgi‐specific brefeldin A‐resistance guanine nucleotide exchange factor; GBF1, Ras‐GTPase‐activating SH3 domain binding protein 1; GSDMDM, gasdermin D; HAV, hepatitis A virus; HRV, human rhinovirus; hnRNP M/K, heterogeneous nuclear ribonucleoprotein M/K; MAVS, mitochondrial antiviral signaling protein; MDA5, melanoma differentiation‐associated protein‐5; NDP52, nuclear dot 10 protein 52; NBR1, neighbor of BRCA1; NLRP3, NLR family PYD containing protein‐3; NEMO, NF‐κB essential modulator; NFκB, nuclear factor‐κB; PDCoV, porcine deltacoronavirus; PEDV, porcine epidemic diarrhea virus; PFAS, phosphoribosylformylglycinamidine synthetase; PLEKHM1, pleckstrin homology domain and RUN domain containing M1; PRRs, pathogen recognition receptors; PRRSV, porcine reproductive and respiratory syndrome virus; RIP1, receptor‐interacting serine/threonine‐protein kinase 1; SARS‐CoV‐2, severe acute respiratory syndrome coronavirus‐2; SSV, Seneca Valley virus; STAT2, signal transducer and activator of transcription 2; TAB1, TAK1 binding protein‐1; TRIF, (TIR)‐domain‐containing adapter‐inducing interferon‐β

Note: Color coding correspond between virus and cleavage site)

In addition to directly targeting components of the IFN‐I pathway, 3C^pro can also modulate the function of proteins that participate in the regulation of this pathway. For instance, the 3C^pro of EV‐A71 downregulates miRNA‐526a, consequently leading to increased expression of cylindromatosis (CYLD, a target of miRNA‐526a). ¹¹⁵ CYLD is a deubiquitinating enzyme negatively regulating the function of RIG‐I by removing K63‐linked polyubiquitin chains from RIG‐I. ¹¹⁶ Ubiquitination is a post‐translational modification required for RIG‐I activation, ¹¹⁷ and deubiquitination results in its inactivation. Together, available evidence reveals that 3C^pro plays a key role in the efforts of picornaviruses in counteracting the host antiviral immune response by cleaving or inactivating essential adapter proteins in the RIG‐I/MDA5‐MAVS and/or the TLR3‐TRIF signaling pathways (Figure 2).

Like picornaviruses, both SARS‐CoV and MERS‐CoV trigger a limited IFN‐I response. ¹¹⁸ , ¹¹⁹ , ¹²⁰ A mouse model infected with SARS‐CoV demonstrated that a significant delay in IFN production contributes to disease progression and severity. ¹²¹ Using different model systems including cells and ferrets infected with SARS‐CoV‐2 and post‐mortem lung tissues from COVID patients, a recent study showed that SARS‐CoV‐2 infections elicit low levels of IFN‐I and no activation of TBK1 and ISGs. ¹²² Of note, it was found that SARS‐CoV‐2 is highly sensitive to IFN‐I, suggesting an important role for IFN‐I in antiviral defense. ¹²² , ¹²³ Although experimental data on SARS‐CoV‐2 is still limited, current evidence from SARS‐CoV and MERS‐CoV research revealed that CoVs develop different strategies to overcome the host innate immunity. For example, the PL^pro of SARS‐CoV, which has deubiquitinating activities, acts as an IFN‐I antagonist by removing ubiquitin chains from IRF3 and through preventing the phosphorylation of IRF3 ¹²⁴ , ¹²⁵ . It was also discovered that the ORF3b, ORF6, and N proteins of SARS‐CoV inhibit production and action of IFN‐I. ¹²⁶ Moreover, M protein of SARS‐CoV was shown to physically associate with RIG‐I, TBK1, IKKε, and TRAF3 and inhibit gene transcription of IFN‐I. ¹²⁷ Most interestingly, several coronaviruses, including SARS‐CoV‐2, porcine deltacoronavirus (PDCoV), and porcine epidemic diarrhea virus (PEDV) have been reported to cause the cleavage of TAB1 or NEMO through their individual 3CL^pro (Table 1, yellow highlighted row), ⁴⁹ , ⁵⁹ , ⁶⁰ suggesting an important role for 3CL^pro in antagonizing the host antiviral innate immune response.

The importance of virus‐induced cytoplasmic aggregates, termed antiviral stress granules (avSGs), ¹²⁸ in RLRs‐mediated innate immunity has been increasingly recognized. Upon viral infection, dsRNAs are generated and accumulate in the cytoplasm, which activate the viral RNA sensor protein kinase R (PKR) and cause phosphorylation of eukaryotic initiation factor 2α and consequent formation of avSGs. Together with key molecules (i.e., Ras‐GTPase‐activating protein SH3 domain‐binding protein 1 (G3BP1) and T cell intracellular antigen 1), avSGs are formed by recruiting multiple antiviral effectors, such as RLRs, Pumilio, DEAH‐box helicase 36, Mex‐3 RNA binding family member C, tripartite motif containing 25, OAS and RNaseL. ¹²⁸ , ¹²⁹ , ¹³⁰ , ¹³¹ , ¹³² The close‐proximity of 5′‐triphosphate containing ssRNA and dsRNA with RLRs within the compact avSG compartment facilitates a more robust RLR‐mediated IFN‐I responses to suppress viral replication. ¹²⁸ , ¹²⁹ , ¹³⁰ , ¹³¹ , ¹³² To bypass this, picornaviruses, including PV, EMCV, CVB3, FMDV, and EV‐D68, utilize 3C^pro to cleave G3BP1 and block avSG formation to prolong viral survival. ¹⁰⁸ , ¹⁰⁹ , ¹¹⁰ , ¹¹¹ , ¹³³

CoVs were also found to be able to modulate the formation of avSGs. It was reported that MERS‐CoV accessory protein 4a prevents PKR activation by directly binding viral dsRNA, thereby inhibiting avSG formation allowing for effective viral replication. ¹³⁴ , ¹³⁵ However, the role of protein 4a in avSG formation appears to be cell type‐specific. MERS‐CoV mutant with deletion of 4a gene still impedes avSG formation in certain cell types, ¹³⁴ , ¹³⁵ suggesting that additional proteins (possibly viral proteases) encoded by MERS‐CoV are required to antagonize avSGs. In a recent report, Grogan and colleagues ⁸ utilized an affinity purification mass spectrometry proteomics approach to screen for cellular proteins that interact with individual SARS‐CoV‐2 proteins, including 3CL^pro. Using both wild‐type and catalytically inactive (C145A) 3CL^pro, they identified two high‐confidence interactions with the histone deacetylase 2 (HDAC2) and tRNA methyltransferase 1 (TRMT1), respectively. Of interest, HDAC2 has been previously shown to be required for IFN‐I signaling through histone modification. ¹³⁶ , ¹³⁷ Remarkably, in addition to 3CL^pro, other SARS‐CoV‐2‐encoded proteins were also found to interact with cellular proteins involved in regulating host innate immunity, including the core avSG protein G3BP1 ⁸ , a known antiviral protein that induces the innate immune response. ¹²⁹ , ¹³⁸

Clinical evidence revealed that SARS‐CoV‐2 infections predispose patients to subsequent bacterial or other viral infections, which are associated with poor outcomes including increased severity and fatality. ¹³⁹ , ¹⁴⁰ Based on what is known about the role of 3C^pro during picornaviral infections, at least two mechanisms may contribute to the secondary infections. First, TRIF is not only involved in antiviral but also participates in antibacterial host defense mechanisms. ¹⁴¹ Disruption of TRIF during the initial phase of viral infection could render the host more susceptible to bacterial invasions. Second, recent studies identified G3BP1 as a critical component in cyclic GMP‐AMP synthase (cGAS)‐mediated innate immune responses against DNA viruses. ¹⁴² , ¹⁴³ cGAS is a cytosolic DNA sensor that detects viral DNA to activate the IFN‐I pathway. ¹⁴⁴ Cleavage of G3BP1 is thus expected to weaken this mechanism and leave the patients at higher risk from a secondary DNA viral infection. A proposed model that picornaviruses (probably CoVs as well) induce two phases of infection (or co‐infection) and disease progression is depicted in Figure 3.

FIGURE 3.

A proposed mechanistic model for the secondary infection after the initial picornavirus or coronavirus infection. During the first phase of infection, proteins (e.g., TRIP and G3BP1) that are involved in a broad range of host defenses (i.e., against not only RNA viral infection, but also bacterial and DNA viral invasion) are targeted by 3C protease or possibly 3C‐like protease for degradation, rendering the patients more vulnerable to secondary infection. Combinatorial infection during the late phase could result in increased disease severity and mortality

2.2. Targeting NLR family PYD containing protein‐3 inflammasome pathway

Similar to RLRs, NLRs are cytosolic PRRs that sense intracellular PAMPs and/or danger‐associated molecular patterns (DAMPs) to induce pro‐inflammatory cytokine production and inflammatory cell death (termed pyroptosis). ¹⁴⁵ Under stress, NLRs initiate the formation of a large multiprotein complex, called inflammasome. The best‐studied inflammasome is the NLR family PYD containing protein‐3 (NLRP3) inflammasome, which consists of the sensor NLRP3, the adapter apoptosis‐associated speck‐like protein containing a caspase recruitment domain (ASC), and the effector caspase‐1. Upon sensing viral PAMPs and DAMPs generated during infection, NLRP3 oligomerizes and recruits pro‐caspase‐1 through the adopter protein ASC to form NLRP3 inflammasome. Subsequently, pro‐caspase‐1 undergoes self‐cleavage and activation, promoting maturation and secretion of pro‐inflammatory cytokine interleukin‐1β (IL‐1β) and IL‐18, and inducing cleavage of the pro‐pyroptotic factor gasdermin D (GSDMD). ¹⁴⁶ The resulting N‐terminal cleavage product of GSDMD then creates pores on cell membrane, triggering pyroptotic cell death and facilitating release of inflammatory cytokines (IL‐1β and IL‐18). ¹⁴⁷ Similar to other types of cell death, pyroptosis limits viral replication by eliminating infected cells.

As NLRP3 plays a vital role in antiviral response, numerous viruses, including picornaviruses, have adopted strategies to counteract its functions. ¹⁴⁵ It was reported that NLRP3‐, ASC‐, and caspase‐1‐deficient mice infected with EV‐A71 display more severe disease phenotype and higher virus titers as compared to wild‐type control mice, suggesting a defensive role for the NLRP3 inflammasome against EV‐A71 infection. ⁵² Further investigation identified that, to overcome the antiviral immune response, EV‐A71 has evolved to inactivate the NLRP3 inflammasome by directly cleaving NLRP3 through the proteolytic activities of 2A^pro and 3C^{pro 52} In addition, EV‐A71 3C^pro also targets its effector GSDMD for cleavage at a site (Gln193‐Gly194) different from that (Asp275‐Asp276) mediated by caspase‐1. The cleavage products of GSDMD by 3C^pro fail to stimulate pyroptosis; hence, further enhancing EV‐A71 replication. Similarly, following CVB3 infection, NLRP3^‐/‐ mice present more rapid disease progression and increased viral loads when compared with wild‐type mice. To evade the host defense, CVB3 3C^pro directly targets NLRP3 and its upstream signaling molecules (RIP1/RIP3, receptor‐interacting protein 1/3) for degradation. ⁵³ , ⁶² , ¹⁴⁸

It is increasingly recognized that early, temporary activation of NLRP3 inflammation has an antiviral role by clearing virus and infected cells, whereas prolonged and extreme activation is harmful, causing disease‐related immunopathology. For example, during hepatitis C viral infection, activation of NLRP3 inflammasome is associated with enhanced inflammation and tissue damage. ¹⁴⁹ Persistent activation of NLRP3 during CVB3 infection has also been linked to pathological outcome. ¹⁵⁰ , ¹⁵¹ It is well documented that SARS‐CoV‐2 infection leads to cytokine storm, featured by excessive production and secretion of pro‐inflammatory cytokines and chemokines, contributing significantly to disease severity of Covid‐19. ^{5 , 122 , 152} Several proteins encoded by SARS‐CoV, including E, ORF3a, and ORF8b, have been shown to activate the NLRP3 inflammasome. ¹⁵³ , ¹⁵⁴ , ¹⁵⁵ Recent evidence has also revealed that NLRP3 inflammasome is activated in response to SARS‐CoV‐2 infection. ¹⁵⁶ It is speculated that overactivation of the inflammasome may be responsible, at least in part, for the observed cytokine storm in Covid‐19 patients.

The NLRP inflammasome pathway and viral manipulation are summarized in Figure 4.

FIGURE 4.

Picornaviral 3C^pro targets the NLRP3 inflammasome for immune evasion. RNA viruses and other DAMPs could activate NLRP3 inflammasome. Formation of the NLRP3‐dependent inflammasome activates caspase 1, which in turn cleaves pro‐IL‐1β and pro‐IL‐18. GSDMD is also cleaved by caspase 1 and the resulting N‐terminal cleavage products are inserted into the plasma membrane, forming multiple pores and inducing pyroptosis and release of pro‐inflammatory cytokine. Upon picornaviral infection, NLRP3, its upstream signaling proteins (RIP1/RIP3), and its downstream effector GSDMD are all targeted by 3C^pro for degradation. As a result, pyroptosis is inhibited for efficient viral replication. CARD, caspase activation and recruitment domain; DAMPs, danger‐associated molecular patterns; GSDMD, Gasdermin D; LRR, Leucine rich repeat; NLRP3, NLR family and pyrin domain‐containing protein 3; PYD, Pyrin domain; RIP, receptor‐interacting protein; NACHT, NAIP, CIITA, HET‐E and TEP1‐associated families; VPg, viral protein genome‐linked; 3C^pro, 3C protease

2.3. Targeting host RNA degradation components

In eukaryotic cells, posttranscriptional processes (e.g., mRNA surveillance, silencing, translational repression, and degradation) play a central role in the regulation of gene expression and ultimately determine the expression levels of a significant fraction of the transcriptome. Recently, it has become apparent that posttranscriptional processes acting on cytoplasmic messenger ribonucleoprotein complexes (mRNPs) are physically tied and can occur in discrete cytoplasmic entity known as processing (P)‐bodies. These compartments are highly conserved across cells derived from vertebrates, invertebrates, yeasts and plants, containing many enzymes involved in mRNA turnover. To date, P‐bodies have been demonstrated to play critical roles in general mRNA degradation, nonsense‐mediated mRNA decay, adenylate‐uridylate‐rich (AU‐rich) element‐mediated mRNA degradation at the 3′ UTRs, and miRNA‐induced RNA silencing pathway. ¹⁵⁷

In eukaryotes, the degradation process is initiated firstly by removal of the poly(A) tail by deadenylases. Following deadenylation, mRNAs are subjected to exonucleolytically degradation starting from their 3′‐end by the exosome, a multimeric complex with 3’→5′ exonucleases. In the meantime, the guanosine cap structure at the 5′‐end is removed by several decapping enzymes and coactivators, including decapping protein 1/2 (DCP1/2), DEAD‐box RNA‐helicase‐6 (DDX6), enhancer of mRNA decapping protein‐3 (EDC3) and EDC4, rendering the mRNA susceptible to 5’→3′ degradation by the major cytoplasmic exoribonuclease 1 (XRN1). ¹⁵⁸

RNA degradation has emerged as an important antiviral host defense mechanism. ¹⁵⁸ The most widely studied viruses in the context of XRN1 are positive‐stranded RNA viruses from the Flaviviridae family, including Dengue virus, West Nile viruses, hepatitis C virus, and Yellow Fever virus. XRN1 acts as an antiviral factor by degrading their genomic RNA. ¹⁵⁷ In addition, DCP1/2‐XRN1 were shown to have critical antiviral functions against both picornaviruses and negative‐stranded cytoplasmic RNA viruses. During infection, these viruses generate various RNA intermediate species with defined structures (long/short dsRNA and ssRNA) inside their replication complexes. Both long ssRNA and dsRNA could induce the re‐localization of both DCP1/2 and XRN1 nucleases into the viral replication complexes for viral RNA degradation. ¹⁵⁹ This further supports the prior observation that the presence of XRN1 and DCPs potentially poses a threat to enterovirus RNA. To conquer this antiviral activity, PV utilizes its 3C^pro to disperse and destabilize P‐bodies by directly targeting DCP1a for degradation. ¹⁰³ The interplay between the cellular RNA degradation pathway and the viruses is illustrated in Figure 5.

FIGURE 5.

Dispersion of 5′→3′ RNA degradation components within P‐bodies during picornaviral infection. RNA viruses, including picornaviruses, initiate viral replication in a discrete compartment within cytoplasm, generating various RNA species with defined signatures. These includes 5′ppp, 5′p‐ssRNA, dsRNA and viral mRNA. Both picornavirus and coronavirus are able to generate long ssRNA and dsRNA and trigger the translocation of 5’→3′ RNA degradation components, including DCP1, DCP2, and XRN1, into the viral replication complex for degradation of associated viral RNA species. Picornaviral 3C^pro cleaves DCP1, resulting in increased viral particles and infectivity. DCP1/2, decapping protein‐1/2; XRN1, 5’→3′ exoribonuclease‐1; dsRNA, double‐stranded RNA; 3C^pro, 3C protease

While both the genomic and subgenomic SARS‐CoV‐2 mRNAs contain 5′‐cap structure, previous studies have shown that inhibiting several eukaryotic initiation factors family proteins (eIF4E, eIF4F, and eIF4G) could impair coronavirus replication, ¹⁶⁰ , ¹⁶¹ highlighting the importance of cap‐dependent translation in SARS‐CoV‐2 mRNA synthesis. Thus, it is not surprising that SARS‐CoV‐2 would intervene the function of host decapping enzyme DCP1/2 and XRN1, possibly through its 3CL^pro. However, to date, antiviral roles XRN1 in coronaviral mRNA translation have not been reported.

2.4. Targeting host autophagy machinery

Macroautophagy (or autophagy in short) is a conserved intracellular degradation pathway that is essential in maintaining cellular homeostasis by removing unwanted or dysfunctional cellular components. ¹⁶² The process of autophagy is highly regulated by more than 30 “autophagy‐related” proteins and includes three major steps. First, the substrates are sequestered by a crescent‐shaped double‐membrane vesicle called a phagophore. Then, the two ends of the phagophore fuse to form an autophagosome. Finally, autophagosome fuses with a lysosome while the enwrapped cargo is degraded by hydrolysis.

Autophagy plays a significant role in antiviral host defense by directly targeting invading viruses through a process, call virophagy, for clearance. ¹⁶³ , ¹⁶⁴ Virophagy is mediated through the function of autophagy cargo receptors, including sequestosome 1 (SQSTM1)/p62, neighbor of BRCA1 (NBR1), calcium binding and coiled‐coil domain‐containing protein 2 (CALCOCO2)/nuclear dot 10 protein 52 (NDP52), which recruit viral components/particles to autophagosome for degradation. ¹⁶⁵ To evade the antiviral efforts of virophagy, many viruses, including picornaviruses, have evolved to disrupt the function of autophagy receptors. ¹⁶⁶ For instance, SQSTM1/p62 and CALCOCO2/NDP52 are targeted for degradation by CVB3 2A^pro and 3C^pro, respectively. ⁶⁶ , ⁶⁷ Cleavage of SQSTM1/p62 was later confirmed upon PV, HRV‐1A, and EV‐D68 infection. ⁶⁸ Furthermore, NBR1, a homolog of SQSTM1/p62, can also be cleaved by 3C^{pro 69} . Remarkably, it was found that cleavage of SQSTM1/p62 and NBR1 not only causes a loss‐of‐function, but also generates dominant‐negative mutants against the function of native proteins. ⁶⁹ Loss of both SQSTM1/p62 and NBR1 would also impair mitophagy and results in mass production of reactive oxygen species, and IL‐1β through constitutive NLRP3‐independent inflammasome activation (potentially other NLRP family members). ¹⁶⁷ The resulting cleavage fragments of SQSTM1/p62 and NBR1 will accumulate to serve as DAMPs signals and further amplify the inflammatory cascade. Acute respiratory pneumonia due to cytokine storm is a hall mark of SARS‐CoV‐2 infection, whether its 3CL^pro would manipulate host autophagic system in particular to mass produced IL‐1β in Covid‐19 pathogenesis is certainly worth further investigation.

Decreased autophagic flux as a result of blockage of autophagosome‐lysosome fusion is another mechanism adopted by picornaviruses to escape viral RNA/protein degradation. The fusion process is regulated by multiple proteins involved in membrane trafficking, particularly a group of SNAP29 proteins, including syntaxin 17 (STX17), synaptosomal‐associated protein 29 (SNAP29), and vesicle‐associated membrane protein 8 (VAMP8). ¹⁶⁸ It was discovered that upon EV‐D68 and CVB3 infection, the SNAP29 linker protein, SNAP29, is cleaved by 3C^pro, which dissociates the STX17‐interacting domain of SNAP29 from the VAMP8‐binding motif, thereby disrupting the formation of STX17‐SNAP29‐VAMP8 complexes and inhibiting autophagosome‐lysosome fusion. ⁶⁸ , ⁷⁰ Figure 6 summarizes the known mechanism by which picornaviral protease subverts the autophagy for immune evasion.

FIGURE 6.

Subversion of host autophagy through picornaviral 3C^pro. Schematic diagram depicted the molecular mechanism for the initiation of host autophagy pathway upon the presence of RNA virus for the clearance of viral‐associated molecules. Picornaviruses utilize its own encoded 2A^pro (not shown here) and 3C^pro to cleave key components such as p62, NBR1, SNAP29 and PLEKHM1 to facilitate a more robust replication. ATGs, autophagy‐related genes; DFCP1, double FYVE‐containing protein‐1, FIP200, focal adhesion kinase family interacting protein of 200kD; NBR1, neighbor of BRCA1; SNAP29, synaptosomal‐associated protein‐29; PLEKHM1, Pleckstrin homology and RUN domain containing M1; p62, also known as sequestosome 1 (SQSTM1); STX17, Syntaxin 17; ULK, Unc‐51‐like kinase‐1; UVRAG, UV radiation resistance‐associated gene protein; VAMP8, vesicle‐associated membrane protein‐8; WIPI2, WD‐repeat domain phosphoinositide‐interacting protein‐2; 2A^pro, 2A‐protease; 3C^pro, 3C protease

Similar to the observations made with several picornaviruses, it was recently demonstrated that SARS‐CoV‐2 infection impairs autophagic flux by blocking autophagosome fusion with a lysosome and induction of autophagy reduces viral replication. ¹⁶⁹ This finding is consistent with an earlier report with MERS‐CoV infection by the same research group. ¹⁷⁰ Although the mechanism responsible for decreased autophagosome‐lysosome fusion remains largely unclear, it was previously shown that over‐expression of the membrane‐associated PL^pro of SARS‐CoV and MERS‐CoV is sufficient to suppress the fusion between autophagosome and lysosome, ¹⁷¹ whether CoV 3CL^pro has a role in regulation of autophagic flux and cargo recognition in general as some picornaviral 3C^pro does has not been explored and requires further investigations. The list of substrates of 3C^pro and 3CL^pro is summarized in Table 1.

2.5. High‐throughput identification of cellular substrates of 3C^pro

A proteomics‐based quantitative method, termed N‐terminomic terminal amine isotopic labeling of substrates (TAILS) (Figure 7), ¹⁷² has been exploited to globally search for novel cellular substrates of picornaviral proteases. ⁶³ , ⁸⁵ TAILS, developed by the Overall laboratory at the University of British Columbia, uses an unbiased negative selection approach to identify neo‐N and ‐C termini (named N‐terminomic and C‐terminomic TAILS, respectively). ¹⁷² , ¹⁷³ , ¹⁷⁴ , ¹⁷⁵ This state‐of‐the‐art approach has several advantages, including quantitative, highly sensitive, and concurrent identification of both the substrates and the cleavage sites, ¹⁷⁶ and has been used to analyze substrates of many types of proteases. ¹⁷⁷ , ¹⁷⁸

FIGURE 7.

Schematic workflow of TAILS N‐terminomics screening of 3C^pro and 3CL^prosubstrates. Schematic diagram depicts the TAILS workflow and scheme for identification of 3C^pro or 3CL^pro substrates. In brief, protein samples from whole cell lysates were subjected to in vitro cleavage by either recombinant purified WT 3C^pro/3CL^pro or mutant (C147A) 3C^pro/(C145A) 3CL^pro, followed by N‐terminal enrichment using TAILS (left panel). Samples were then combined and subjected to pre‐TAILS shotgun‐like mass‐spectrometry analysis after complete digestion with trypsin. The exposed amine groups of N‐termini generated by the trypsin digestion were then removed by covalently coupling to a high‐molecular weight polyaldehyde polyglycerol polymer. This process allowed for selection via negative enrichment of blocked N termini (middle panel). Peptides were subsequently identified and quantified using high‐resolution mass spectrometry (indicated in the right panel). The resultant high‐confidence candidate substrates were determined through the analysis of the quantified heavy/light (H/L) ratio of dimethylation‐labeled semitryptic neo‐N terminus peptides. They will be subjected to further validation through similar in vitro cleavage assay by 3C^pro/3CL^pro, followed by immunoblotting using specific antibodies; TAILS, terminal amine isotopic labeling of substrates; 3C^pro, 3C protease; 3CL^pro, 3C‐like protease

Jan and colleagues conducted N‐terminomic TAILS on HeLa cell or mouse cardiomyocyte extracts subjected to incubation with purified recombinant PV or CVB3 3C^pro, respectively, to identify cleaved neo‐N‐terminal peptides (see Figure 7 for the detailed procedure). ⁶³ , ⁸⁵ A list of high confidence candidate proteins was generated for 3C^pro, including peptides corresponding to previously reported substrates at the known cleavage sites (i.e., poly(A) binding protein (PABP), ⁹⁰ G3BP1 ¹¹⁰ , and TAR DNA‐binding protein 43 (TDP‐43) ⁸⁷ ). Among them, a subset of candidate targets of PV 3C^pro has been validated in vitro and under viral infection, which include four common protein substrates of both PV and CVB3 3C^pro (i.e., heterogeneous nuclear ribonucleoprotein M (hnRNP M), hnRNP K, RIP1, and phosphoribosylformylglycinamidine synthetase [PFAS]), programmed cell death 6‐interacting protein (PDCD6IP, also known as ALIX), general vesicular transport factor p115, ATP‐citrate synthase (ACLY), Golgi‐specific brefeldin A‐resistance guanine nucleotide exchange factor 1 (GBF1). Studies through gene‐silencing or expression of a non‐cleavable mutant form of these substrates have demonstrated a pivotal role for these proteins in regulating viral replication/propagation. ⁶³ , ⁸⁵

Given the effectiveness of the N‐terminomic TAILS in identifying the cellular targets of 3C^pro, this unbiased proteomics approach is currently being utilized to analyze the cellular targets of MHV and SARS‐CoV‐2 proteases, including 3CL^pro. It is anticipated that identification of the full repertoire of host substrates of CoV proteases will provide a more comprehensive understanding of viral tropism, interaction with host cells, and pathogenesis, as well as assist in the development of novel antiviral drugs.

2.6. Current 3C^pro and 3CL^pro inhibitors and future challenges

As discussed earlier, both picornaviral 3C^pro and coronaviral 3CL^pro are chymotrypsin‐like cysteine proteases with conserved substrate specificity (P1‐P1′ and P4 cleavage sites of the substrates are highly conserved between two enzymes) and active sites. ¹⁷⁹ Owing to these similarities, efforts have been made to explore the potential of developing broad‐spectrum antiviral compounds. ¹³ , ¹⁴ , ¹⁵ The fact that no known human homologs further increases the feasibility of this strategy.

Rupintrivir (AG‐7088, a protease inhibitor originally developed for HRV to treat common cold) and/or its derivatives/analogs have been found to possess antiviral activities against a wide range of picornaviruses (i.e., HRV, PV, EV‐A71, EV‐D68, CVB, CVA, HAV, and FMDV) ¹⁸⁰ , ¹⁸¹ , ¹⁸² , ¹⁸³ and coronaviruses, including SARS‐CoV. ^{14 , 184} Furthermore, it was reported that dipeptidyl aldehyde (GC373), α‐ketoamide (GC375), and dipeptidyl bisulfite adduct (GC376), are able to inhibit both picornaviral 3C^pro and coronaviral 3CL^pro activities and block viral replication. ¹³ These studies point to the possibility of developing broad‐spectrum antiviral therapeutics against SARS‐CoV‐2. However, rupintrivir failed to show efficacy in natural HRV infection conditions during a clinical phase II trial. ¹⁸⁵

Although Remdesivir (a nucleotide analog inhibitor of RdRp, originally developed for the treatment of Ebola virus disease by Gilead Sciences) has been approved for treating severely ill Covid‐19 patients in the United States and the Europe, ¹⁸⁶ its safety and antiviral activity are still under extensive investigations and clinical trials. Similarly, despite some early promising findings showing the effectiveness of chloroquine (a classical anti‐malarial and autoimmune disease drugs) against SARS‐CoV‐2 infection in non‐respiratory cells, ¹⁸⁷ , ¹⁸⁸ recent research has demonstrated that chloroquine fails to block infection of human lung epithelial cells with SARS‐CoV‐2. ¹⁸⁹ In addition, it is known that chloroquine may cause severe cardiotoxicity. ¹⁹⁰ A number of attempts have been made to design drugs targeting 3CL^pro of SARS‐CoV‐2. Different approaches, including laboratory synthesis, virtual screening, drug repositioning, and structure‐based molecular docking have been taken for this purpose. Hilgenfeld laboratory recently elucidated the X‐ray crystallographic structure of SARS‐CoV‐2 3CL^pro and synthesized improved α‐ketoamide inhibitors to target 3CL. ¹² Using computer‐aided drug design, Jin et al. identified a 3CL^pro inhibitor (N3) and determined the crystal structure of its complex with 3CL^pro of SARS‐CoV‐2 ¹⁰ . Through a combined structure‐based virtual and high‐throughput screening, they further showed that six compounds exhibit potent anti‐3CL^pro activities. ¹⁰ Most recently, Dai et al. reported the design and synthesis of two lead compounds (11a and 11b) targeting 3CL^pro and solved the X‐ray crystal structure of these inhibitors bound to 3CL. ⁹ Although these compounds serve as promising drug candidates, their effectiveness in nature infection remain to be investigated. The list of inhibitors targeting 3C^pro and 3CL^pro is summarized in Table 2.

TABLE 2.

Studies of 3C^pro or 3CL^pro inhibitors antiviral compounds [Correction added on 17 February 2021, after first online publication. The references were updated throughout Table 2 and in‐text citations.]

Research phase	Compound	Target	Results	Ref
In vitro validation	TG‐0205221	3CLpro of SARS‐CoV and HCoV‐229E	Reduces SARS‐CoV and HCoV‐229E replication by titer of 4.7 Log10	191
	Flavonoids: Apigenin, luteolin, quercetin, amentoflavone, aueretin, daidzein, puerarin, epigallocatechin gallate, gallocatechin gallate, kaempferol,rhoifolin, pectolinarin, herbacetin, flavonol	SARS‐CoV 3CLpro	Inhibit SARS‐CoV 3CLpro FRET protease assay catalytic activity	192
	Pyrazolone and pyrimidines inhibitors	SARS‐CoV 3CLpro	Show potent inhibitory activities against SARS‐CoV 3CLpro at micromolar range.	193
	Aryl methylene ketones, Mono‐, and difluorinated methylene ketones groups	SARS‐CoV 3CLpro	Improved version is stable and less toxic to cells. Potently inhibits SARS‐CoV 3CLpro at nanomolar range	194
	Heteroaromatic esters and benzotriazole esters derivatives	SARS‐CoV 3CLpro	Show potent inhibitory activities against SARS‐CoV 3CLpro at nanomolar range	195 , 196 , 197
	Boronic	SARS‐CoV 3CLpro	Significantly inhibits SARS‐CoV 3CLpro enzymatic activity in micromolar range	198
	Aza‐peptide epoxides derivatives	SARS‐CoV 3CLpro	Show irreversible inhibition against SARS‐CoV 3CLpro	199 , 200 , 201
	Etacrynic acid derivatives	SARS‐CoV PLpro and 3CLpro	Show more than 70% inhibition on SARS‐CoV at concentration of 100μM	202 , 203
	Peptides aldehydes derivatives	SARS‐CoV and HCoV‐229E 3CLpro	Suppress SARS‐CoV by 4.7 Log10 and HCoV‐229E by 5.2 Log10	191
	Modified version of HIV protease inhibitors	SARS‐CoV 3CLpro	Potent inhibitors against SARS‐CoV 3CLpro but not against HIV protease	204
	Sulfone and dihydroimidazole derivatives	SARS‐CoV 3CLpro	21 derivatives from these two analogs show EC50 less than 50 μM against SARS‐CoV 3CLpro	205 , 206
	Michael acceptor peptidomimetics	SARS‐CoV 3CLpro	Show potent inhibitory against SARS‐CoV 3CLpro	207 , 208
	Lignoids, di‐ and triterpenoid derivatives: Betulinic acid, savinin, ferruginol, pritimererin, tingenone, iguestrin and triterpenoids celastrol	SARS‐CoV 3CLpro	Abietane type diterpenoids are the most robust terpenoids on SARS‐CoV (EC50 = 9.1 μM)	207 , 209 , 211
	Metal conjugated: Zinc‐ or mercuric based	SARS‐CoV PLpro and 3CLpro,	Inhibition is pronounced in Zinc‐conjugated compounds	212 , 213
	α‐Ketoamides:	3Cpro of CVB3, &; HRV, EV‐D68, EV‐A713CLpro of SARS‐CoV, MERS, 229E	Display low toxicity &; low micromolar of EC50 against tested viruses	214
	Pyridyl, pyrazyl and Benzotriazole‐derivatives inhibitors	SARS‐CoV PLpro or 3CLpro	Robust inhibition on SARS‐CoV in vitro within micromolar range	215 , 216 , 217
	Rupintrivir (AG‐7088)	Targeting 3Cpro and 3CLpro encoding viruses	Shows robust activity against SARS‐CoV‐2 in vitro	218 , 219
	Boceprevir, Calpain inhibitors II, and XII	SARS‐CoV‐2 3CLpro	Inhibit SARS‐CoV‐2 in vitro with EC50 less than 5μM	220
	3CLpro‐1	Originally designed for 3Cpro of EV‐A71.	Shows robust efficacy of EC50 200 nM, effective against SARS‐CoV‐2 and MERS‐CoV.	221 , 222
	Isatin derivatives	Targeting 3Cpro and SARS‐CoV 3CLpro	Effectively inhibit SARS‐CoV 3CLpro through noncovalent bonding in low micromolar range	219 , 220
	Anilide derivaties: 2‐chloro‐4‐nitro anilineL‐phenylalanine, 4‐(dimethylamino)benzoic acid	SARS‐CoV 3CLpro	Potent and highly specific inhibitors against SARS‐CoV 3CLpro	225
Computational prediction (docking analysis)	Anti‐HIV‐1 drugs: Indinavir, Darunavir	SARS‐CoV‐2 3CLpro	Docking & binding free energy prediction shows high scores & high binding affinities against SARS‐CoV‐2 3CLpro	226
	Decahydroisoquinoline inhibitors	SARS‐CoV‐2 3CLpro	X‐ray crystallization studies confirmed that these inhibitors fit well into the cleft of 3CLpro	227
In vitro and in vivo validation	Peptides with halomethyl ketone derivatives	SARS‐CoV 3CLpro	Effectively inhibit SARS‐CoV infection, with low cytotoxicity in cells and in mice.	228
Widely tested in animals and now under trial on human for coronavirus disease‐2019	GC376	Targeting 3Cpro and 3CLpro encoding viruses	Shows robust activity against SARS‐CoV, SARS‐CoV‐2, and Norovirus	220

Abbreviations: HRV, human rhinovirus; MERS‐CoV, Middle East respiratory syndrome‐CoV; PLpro, papain‐like protease; SARS‐CoV‐2, severe acute respiratory syndrome coronavirus‐2 3CL^pro/3C^pro, 3C‐like protease/3C protease.

3. CONCLUDING REMARKS

To date, we know very little about the 3CL^pro of SARS‐CoV‐2, especially the molecular mechanism of the pathways by which SARS‐CoV‐2 3CL^pro blocks. Certainly, we cannot rule out that there are some differences but also a lot of similarities among both families. While we are in the process of understanding the structure and functions of SARS‐CoV‐2 3CL^pro, a comparison with the 3C^pro from picornaviruses can provide more insights into the pathogenesis and regulatory mechanisms of Covid‐19. These will serve as a critical foundation for the design of broad‐spectrum anti‐coronaviruses inhibitors, or perhaps anti‐3C/3CL^pro expressing viruses.

CONFLICT OF INTEREST

The authors declare that they have no conflict of interest.

AUTHORS CONTRIBUTIONS

Chen Seng Ng and Honglin Luo designed and outlined the review. Chen Seng Ng prepared the initial manuscript and figures. Christopher C. Stobart prepared the structural figure and relevant information. All authors contributed to revising and writing the final version.

Ng CS, Stobart CC, Luo H. Innate immune evasion mediated by picornaviral 3C protease: Possible lessons for coronaviral 3C‐like protease? Rev Med Virol. 2021;31(5):e2206. doi: 10.1002/rmv.2206

DATA AVAILABILITY STATEMENT

Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.