key: cord-0789231-9bv5enqu
authors: Yan, Shaomin; Wu, Guang
title: Potential 3‐chymotrypsin‐like cysteine protease cleavage sites in the coronavirus polyproteins pp1a and pp1ab and their possible relevance to COVID‐19 vaccine and drug development
date: 2021-04-29
journal: FASEB J
DOI: 10.1096/fj.202100280rr
sha: 99b67bb737850f3469b818115e48e6c615b8f553
doc_id: 789231
cord_uid: 9bv5enqu

Coronavirus (CoV) 3‐chymotrypsin (C)‐like cysteine protease (3CL(pro)) is a target for anti‐CoV drug development and drug repurposing because along with papain‐like protease, it cleaves CoV‐encoded polyproteins (pp1a and pp1ab) into nonstructural proteins (nsps) for viral replication. However, the cleavage sites of 3CL(pro) and their relevant nsps remain unclear, which is the subject of this perspective. Here, we address the subject from three standpoints. First, we explore the inconsistency in the cleavage sites and relevant nsps across CoVs, and investigate the function of nsp11. Second, we consider the nsp16 mRNA overlapping of the spike protein mRNA, and analyze the effect of this overlapping on mRNA vaccines. Finally, we study nsp12, whose existence depends on ribosomal frameshifting, and investigate whether 3CL(pro) requires a large number of inhibitors to achieve full inhibition. This perspective helps us to clarify viral replication and is useful for developing anti‐CoV drugs with 3CL(pro) as a target in the current coronavirus disease 2019 (COVID‐19) pandemic.

and then cleaves three sites: between nsps1/2, nsps2/3, and nsps3/4 with the LXGG recognition motif. 17 3CL pro -also called the main protease (M pro )-is released from nsp5 through autocleavage. 18 Using recognition motifs such as XXXLQAXXX and XXXLQSXXX, 19 3CL pro cleaves the remaining nsps. 20 3CL pro is a desirable target for developing wide-spectrum anti-CoV drugs and drug repurposing [21] [22] [23] [24] because (a) the nsps cleaved by 3CL pro influence the formation of the doublemembraned vesicles (DMV), 18, [25] [26] [27] [28] [29] [30] [31] (b) 3CL pro shares significant sequence identity and 3D structure among CoVs, 32, 33 (c) human proteases do not have identical cleavage specificity, 8 and (d) 3CL pro inhibitors are unlikely to be toxic to humans. 8 However, 3CL pro has the potential for industrial application because its autocleavage occurs near the membrane 34 or in the cytoplasm; [35] [36] [37] therefore, its efficacy is higher than that of other proteases. 35 Much information on the 3D structure, 38 mechanism, 39 and substrate specificity 40 of 3CL pro has been elucidated. For example, 3CL pro has at least three crystal structures 41 : the wild-type active dimer (wt-dimer) with one active site per subunit 1, 35 ; the monomeric form or the G11A, R298A, and S139A mutants that cannot dimerize [42] [43] [44] [45] ; and the superactive octamer form. 3 Therefore, two strategies have been employed to develop 3CL pro inhibitors based on their active sites and dimerization. [46] [47] [48] So far, however the cleavage sites by 3CL pro and cleaved nsps remain controversial. For instance, a study based on 24 completely sequenced CoV genomes predicted 11 cleavage sites between nsp2, nsp3, nsp4, nsp5, nsp6, nsp7, nsp9, nsp10, nsp11, nsp12, and nsp13, 49 but found that the cleavage site between nsp7/8 in pp1a and the cleavage site between nsp7/9 in pp1ab are the same. 49 Here, there is no nsp8. 3CL pro has 11 cleavage sites, that is, between nsp4/5, nsp5/6, nsp6/7, nsp7/8, nsp8/9, nsp9/10, nsp10/11, nsp12/13, nsp13/14, nsp14/15, and nsp15/16. 26, [49] [50] [51] Here, there is no cleavage site between nsp11/12.

Sometimes, a cleavage site by 3CL pro is defined between nsp10/12. 10, 21 Here, there is no nsp11.

Although these discrepancies seem trivial, they are essential for 3CL pro action and are instrumental for developing 3CL pro inhibitors. In this perspective, we closely explore these inconsistent cleavage sites by 3CL pro and their nsps, and discuss their possible relevance to coronavirus disease 2019 (COVID-19) vaccine and drug development.

Initially, 13 probable cleavage sites with dipeptides Q/S(G) were predicted for 3CL pro from infectious bronchitis virus (IBV), and cleavage sites 10, 11, 12, and 13 were located in F2 polyprotein (pp1b). 20 Subsequently, it was revealed that these 13 cleavage sites include 2 cleavage sites-sites 3 and 4-flanking 3CL pro for autocleavage. 20 Thus, 3CL pro cleaves 11 sites. 20 These findings seem to show that 3CL pro is located in nsp3 in IBV, but 3CL pro is actually located in nsp2 in IBV per the current annotation (accession no. NC_001451.1). In addition, nsp8 is positioned from nucleotide (nt) 12313 to nt 12381 in pp1a, but is absent in pp1ab, where it merges with nsp9 from nt 12313 to nt 12354, and from nt 13354 to nt 15131.

Shortly after, seven cleavage sites were experimentally identified in ORF1b from murine CoV mouse hepatitis virus (MHV), 52 which generated eight nsps rather than the five nsps generally accepted in pp1b.

Later, it was accepted that PL pro and 3CL pro in HCoV-229E RNA-directed RNA polymerase ORF1A (accession no, 464694) have 12 cleavage sites, 1 site cleaved by PL pro and 11 sites cleaved by 3CL pro , generating 13 nsps. 50 Here, the number of nsps in pp1a is more than that currently accepted-pp1a has 11 nsps from nsp1 to nsp11 and pp1ab has 16 nsps because ORF1b encodes 5 nsps from nsp12 to nsp16. 15 In addition, HCoV-229E has two copies of PL pro and generates two nsps, whereas most CoVs have a single PL pro copy and generate three nsps.

Eventually, 3CL pro cleaves 9 sites rather than 11 cleavage sites, 2 of which form its own flanking N-and C-terminal autoprocessing sites. 38, 53 This is plausible because 3CL pro is located in nsp5, and its own N-and C-terminal autoprocessing sites are located at the cleavage sites between nsp4/5 and nsp5/6. In other words, 3CL pro can trans-cleave five sites in pp1a-between nsp6/7, nsp7/8, nsp8/9, nsp9/10, and nsp10/11 and trans-cleave four sites in pp1b-between nsp12/13, nsp13/14, nsp14/15, and nsp15/16.

Of the frequently referenced CoVs, nsps were clearly annotated in nine CoVs ( Table 1 ). As shown in Table 1 , the identified nsps were inconsistent across different CoVs. For example, nsp11 is absent but merges with nsp12 in MHV (A59 C12 mutant) and SARS-CoV (Tor2). In another example, nsp8, nsp14, nsp15, and nsp16 are absent from avian IBV (Beaudette), porcine epidemic diarrhea virus (PEDV) CV777, and two SARS-CoVs (CUHK-W1 and TW1), but their nsp8 and nsp9 merge.

This inconsistency is interesting because the sizes of nsps cleaved by 3CL pro are highly conserved among different groups of CoVs, whereas the sizes of nsps cleaved by PL pro are irregular. 49 Table 1 shows the existence of 16 nsps for several CoVs. This is interesting because 15 cleavage sites are required to generate 16 nsps. However, PL pro cleaves 3 sites and 3CL pro cleaves 11 sites-which sums up to 14 cleavage sites-generating 15 nsps only. Indeed, 15 nsps are often mentioned, 54 and therefore, the generation of 15 nsps is arithmetically correct. Intriguingly, a cleavage site is missing from such calculations. No predictable cleavage sites were observed between nsps10/11 and nsps11/12 in CoVs using the NetCorona 1.0 webserver. 19 Additionally, the cleavage site between nsp11/12 in SARS-CoV, MERS-CoV, and SARS-CoV-2 cannot be found in literature although that between nsp10/12 is mentioned. 21 pp1ab is a fusion between nsp11 and nsp12 when a ribosomal frameshifting occurs between ORF1a and ORF1b 55-57 ; thus, it is likely that the missing cleavage site is between nsp11/12, which is the cleavage site between pp1a and pp1b. Hence, a convincing explanation is that nsp11 appears only when pp1a exists, in which case the ribosomal frameshifting does not occur. Consequently, there are 15 nsps for pp1ab, but the co-existence of pp1a and pp1ab provides 16 nsps. Here, the function of nsp11 remains unclear?

nsp14, nsp15, and nsp16 are absent from the four CoVs (Table 1) . Meanwhile, a comparison based on the data 58, 59 shows the absence of nsp11 and nsp16, and total 14 nsps in both SARS-CoV and SARS-CoV-2. 10 The absence of nsp16 in pp1ab draws our attention, leading us to explore nsp16 in the next section.

In summary, this section indicates that (a) nsp11 exists in pp1a only but not in pp1ab, and there is no cleavage site between nsp11/12; (b) the 11 3CL pro cleavage sites are only applicable to pp1ab; and (c) the function of nsp11 is yet to be determined.

nsp16 is an RNA cap-modifying enzyme, 60 and forms a complex with nsp10. [60] [61] [62] Therefore, it plays the role of 2′-O-methyltransferase (2′-O-MTase) in CoVs, which was first found in the nsp16 from feline CoV (FCoV) FIPV 79-1146 63 ; in the FASTA format of the FCoV genome (accession no, DQ010921), the mRNA for nsp16 is positioned at nt 19307-nt 20206, whereas the mRNA for S-protein is positioned at nt 20206-nt 24564, thus, their ORFs overlap at a single nt, 20206. In the graphics format of this FCoV genome, the mRNA for nsp16 is positioned at nt 19307-nt 20209, whereas the mRNA for S-protein is positioned at nt 20206-nt 24564, thereby, their ORFs overlap at three nucleotides, 20206-20209. Table 2 lists 39 frequently referenced CoV genomes, where ORF1ab overlaps the mRNA for S-protein in 22 CoVs ( † in the last column in Table 2 ), that is, the mRNA for pp1ab overlaps the mRNA for S-protein in some groups of CoVs. This feature is more remarkable in CoVs from avian, bat, feline, swine as well as MERS-CoVs. The overlapping can be as large as 58 nts for MERS-CoV (the penultimate row in Table 2 ).

The overlap of ORF1ab on the subgenomic mRNA in CoVs is intriguing because the size of proteins translated from subgenomic mRNA is theoretically that of non-overlapping coding regions. 64 Most subgenomic mRNAs are structurally polycistronic but functionally monocistronic, and thus their translation begins only from the 5′ of most ORF in viral proteins. 65 In the 1990s, the S-protein gene in equine isolate Berne virus (BEV), a torovirus, 66 overlapped the replicase gene (ORF1b). 64 Therefore, it was proposed that a motif, UGUUUAGU, directs the synthesis of the S-protein gene. 67 Subsequently, this S-protein gene (mRNA 2) was found to have a short non-contiguous leader coming from the 5′ terminus of the BEV genome. 68 Here, the overlapping results from a heterologous RNA recombination. 68 In contrast to torovirus, CoVs have a common 5′ leader sequence, 29 which protects the capped mRNAs from nsp1induced endonucleolytic cleavage, causing the accumulation of SARS-CoV mRNAs and proteins. 69 In CoV lifecycle, the production of pp1ab occurs in the first phase of translation, whereas S-protein production by subgenomic mRNA occurs in the second phase. 70 Between these two phases is the replication of minus-and plus-strand RNA in DMV. It is also unclear whether such overlapping affects nsp16 and S-protein functions, and the fusion of S-protein with a leader sequence, which includes the transcription-regulating sequence (TRS). This leads to the question of the selective advantage of CoV with the leader sequence in its subgenomic mRNAs. 65 As the RNA-dependent RNA polymerase (RdRp), nsp12, 71 catalyzes leader-body fusion, we closely explore nsp12 in the next section.

In summary, this section focuses on the overlapping of the S-protein mRNA by the nsp16 mRNA because several mRNA vaccines are based on the S-protein mRNA, [72] [73] [74] whereas the overlapping could be a potential source for heterologous RNA recombination.

RdRp (nsp12) is a target for developing anti-CoV drugs. 16, 75 Usually, nsp7 and nsp8 act as nsp12 cofactors. 76 Remdesivir, currently authorized by the FDA for emergency use, 77 was originally designed to target the polymerases in HIV and hepatitis C virus (HCV). 78 The first high-resolution cryo-electron microscopy structure of SARS-CoV-2 full-length nsp12 has 932 residues, 79 which is the size of joined nts 13442-13468 and 13468-16236 (SARS-CoV-2, Table 1 ). This illustration again confirms that nsp11 does not exist as a single entity in pp1ab; nsp11 only appears when ribosomal frameshifting does not occur, whereas ribosomal frameshifting is the only way to generate the joined nsp12, RdRp.

Here, one may ask whether RdRp and nsp11 must merge to function and whether nsp12 alone is nonfunctional. ORF1b can encode a polyprotein only if ribosomal frameshifting from ORF1a into ORF1b occurs 80,81 because ORF1b does not have an independent site for translation initiation. 82 Thus, pp1b does not exist alone, and consequently nsp12 does not exist without nsp11. Hence, the key point for generating RdRp is the −1 ribosomal frameshifting, 82 which is the focus of many studies. 50, [83] [84] [85] This ribosomal frameshifting overlaps two ORFs with 43 nucleotides in HCoV-229E51, with 86 nucleotides in IBV, 83 and with 130 nucleotides in the L-A doublestranded RNA virus of Saccharomyces cerevisiae. 86 The occurrence of ribosomal frameshifting is required for the production of RdRp, 70 which is needed in the second phase of translation. 70 Although the signal for ribosomal frameshifting has been well studied 57, 80, [87] [88] [89] [90] and the RNA pseudoknot is designed as a target for anti-SARS agents, 91, 92 regulation of the ribosomal frameshifting mechanism remains unclarified.

A simple explanation of the mechanism is that the expression of ORF1b is initiated at specific levels relative to pp1a in CoVs, 82 assuming there is a threshold for the occurrence of ribosomal frameshifting based on the ratio of pp1ab to pp1a. This ratio ranges from 1.8% to 1.9% in the L-A doublestranded RNA virus of S. cerevisiae, 86, 93 and from 5% to 10% in HIV-1. 94, 95 For HCoV-229E, the ribosomal frameshifting frequency ranges between 18% and 30%, of which less than 1% can synthesize pp1ab. 55 Thus, the chance of generating pp1ab in CoVs is remarkably lower than that in the L-A double-stranded RNA virus of S. cerevisiae, 86, 93 and HIV-1. 94, 95 Indeed, both pp1a and pp1ab from CoVs are difficult to detect in vivo. 96 This implies that many copies of pp1a, but few copies of pp1ab occur in CoVs. Although 3CL pro is derived from nsp5 in both pp1a and pp1ab, most of 3CL pro is from pp1aderived. Thus, not every 3CL pro has a good chance of cleaving the sites between nsp12/13, 13/14, 14/15, and nsp15/16 in pp1b.

In summary, this section reveals that the 3CL pro significantly outnumbers its cleavage sites in pp1b; thus, a considerable number of 3CL pro inhibitors are required to completely inhibit 3CL pro action.

In this perspective, we attempted to address several controversial issues on the cleavage sites and cleaved nsps in CoV pp1a and pp1ab by 3CL pro . Meanwhile, several questions are raised: (a) what function does nsp11 perform, (b) can the overlap of nsp16 mRNA over the S-protein mRNA affect mRNA vaccine, and (c) can a low ribosomal frameshifting frequency affect 3CL pro activity? The answers to these questions will enrich our understanding of the mechanism of viral replication and benefit the development of anti-CoV drugs with 3CL pro as targets in the current COVID-19 pandemic.

The crystal structures of severe acute respiratory syndrome virus main protease and its complex with an inhibitor

Design of wide-spectrum inhibitors targeting coronavirus main proteases

Three-dimensional domain swapping as a mechanism to lock the active conformation in a super-active octamer of SARS-CoV main protease

Computational modeling of the Bat HKU4 coronavirus 3CL(pro) inhibitors as a tool for the development of antivirals against the emerging Middle East Respiratory Syndrome (MERS) coronavirus

Structure-guided design of potent and permeable inhibitors of MERS coronavirus 3CL protease that utilize a piperidine moiety as a novel design element

Identification, synthesis and evaluation of SARS-CoV and MERS-CoV 3C-like protease inhibitors

Structure of M pro from SARS-CoV-2 and discovery of its inhibitors

Crystal structure of SARS-CoV-2 main protease provides a basis for design of improved α-ketoamide inhibitors

A SARS-CoV-2 protein interaction map reveals targets for drug repurposing

Identification of host cellular protein substrates of SARS-COV-2 main protease

Characterization of a novel coronavirus associated with severe acute respiratory syndrome

The genome sequence of the SARS-associated coronavirus

A new coronavirus associated with human respiratory disease in China

The coronavirus replicase

The SARS-coronavirus papain-like protease: structure, function and inhibition by designed antiviral compounds

Targeting SARS-CoV-2 proteases and polymerase for COVID-19 treatment: state of the art and future opportunities

Identification of severe acute respiratory syndrome coronavirus replicase products and characterization of papain-like protease activity

Unique and conserved features of genome and proteome of SARS-coronavirus, an early split-off from the coronavirus group 2 lineage

Coronavirus 3CLpro proteinase cleavage sites: possible relevance to SARS virus pathology

Coronavirus genome: prediction of putative functional domains in the non-structural polyprotein by comparative amino acid sequence analysis

Prediction of the SARS-CoV-2. (2019-nCoV) 3C-like protease (3CLpro) structure: virtual screening reveals velpatasvir, ledipasvir, and other drug repurposing candidates

SARS-CoV-2: an update on potential antivirals in light of SARS-CoV antiviral drug discoveries. Vaccines (Basel)

Repurposing cefuroxime for treatment of COVID-19: a scoping review of in silico studies

Tackling SARS-CoV-2: proposed targets and repurposed drugs

SARS-beginning to understand a new virus

Mechanisms and enzymes involved in SARS coronavirus genome expression

Functional and genetic analysis of coronavirus replicase-transcriptase proteins

A contemporary view of coronavirus transcription

Nidovirus transcription: how to make sense?

The molecular biology of coronaviruses

Coronaviruses post-SARS: update on replication and pathogenesis

Structure of main protease from human coronavirus NL63: insights for wide spectrum anti-coronavirus drug design

Drug design targeting the main protease, the Achilles' heel of coronaviruses

Coronavirus main proteinase (3CLpro) structure: basis for design of anti-SARS drugs

Production of authentic SARS-CoV M(pro) with enhanced activity: application as a novel tagcleavage endopeptidase for protein overproduction

Mechanism of the maturation process of SARS-CoV 3CL protease

Characterization of trans-and cis-cleavage activity of the SARS coronavirus 3CLpro protease: basis for the in vitro screening of anti-SARS drugs

SARS-CoV 3CL protease cleaves its C-terminal autoprocessing site by novel subsite cooperativity

Comprehensive insights into the catalytic mechanism of Middle East Respiratory Syndrome 3C-like protease and Severe Acute Respiratory Syndrome 3C-like protease

Profiling of substrate specificity of SARS-CoV 3CL

Activation and maturation of SARS-CoV main protease

Mutation of Gly-11 on the dimer interface results in the complete crystallographic dimer dissociation of severe acute respiratory syndrome coronavirus 3C-like protease: crystal structure with molecular dynamics simulations

Residues on the dimer interface of SARS coronavirus 3C-like protease: dimer stability characterization and enzyme catalytic activity analysis

Mechanism for controlling the dimermonomer switch and coupling dimerization to catalysis of the severe acute respiratory syndrome coronavirus 3C-like protease

Two adjacent mutations on the dimer interface of SARS coronavirus 3C-like protease cause different conformational changes in crystal structure

Biosynthesis, purification, and substrate specificity of severe acute respiratory syndrome coronavirus 3C-like proteinase

Characterization of SARS main protease and inhibitor assay using a fluorogenic substrate

Long-range cooperative interactions modulate dimerization in SARS 3CL pro

Prediction of proteinase cleavage sites in polyproteins of coronaviruses and its applications in analyzing SARS-CoV genomes

Virus-encoded proteinases and proteolytic processing in the Nidovirales

Characterization of a human coronavirus (strain 229E) 3C-like proteinase activity

Identification of polypeptides encoded in open reading frame 1b of the putative polymerase gene of the murine coronavirus mouse hepatitis virus A59

Genomic characterization of the 2019 novel human-pathogenic coronavirus isolated from a patient with atypical pneumonia after visiting Wuhan

Liberation of SARS-CoV main protease from the viral polyprotein: n-terminal autocleavage does not depend on the mature dimerization mode

Nucleotide sequence of the human coronavirus 229E RNA polymerase locus

An 'elaborated' pseudoknot is required for high frequency frameshifting during translation of HCV 229E polymerase mRNA

A mechanical explanation of RNA pseudoknot function in programmed ribosomal frameshifting

On the size of the active site in proteases. I. papain

Optimization rules for SARS-CoV-2 Mpro antivirals: ensemble docking and exploration of the coronavirus protease active site

Crystal structure and functional analysis of the SARS-coronavirus RNA cap 2′-O-methyltransferase nsp10/nsp16 complex

Biochemical and structural insights into the mechanisms of SARS coronavirus RNA ribose 2′-O-methylation by nsp16/nsp10 protein complex

From SARS to MERS: 10 years of research on highly pathogenic human coronaviruses

Coronavirus nonstructural protein 16 is a cap-0 binding enzyme possessing (nucleoside-2′O)-methyltransferase activity

A 3'-coterminal nested set of independently transcribed mRNAs is generated during Berne virus replication

Continuous and discontinuous RNA synthesis in coronaviruses

Toroviruses: replication, evolution and comparison with other members of the coronavirus-like superfamily

Primary structure and post-translational processing of the Berne virus peplomer protein

Discontinuous and non-discontinuous subgenomic RNA transcription in a nidovirus

SARS coronavirus nsp1 protein induces templatedependent endonucleolytic cleavage of mRNAs: viral mRNAs are resistant to nsp1-induced RNA cleavage

The role of programmed −1 ribosomal frameshifting in coronavirus propagation

Expression, purification, and characterization of SARS coronavirus RNA polymerase

An mRNA vaccine against SARS-CoV-2 -preliminary report

A thermostable mRNA vaccine against COVID-19

SARS-CoV-2 mRNA vaccine design enabled by prototype pathogen preparedness

Overview of targets and potential drugs of SARS-CoV-2 according to the viral replication

One severe acute respiratory syndrome coronavirus protein complex integrates processive RNA polymerase and exonuclease activities

Remdesivir EUA Fact Sheet for Healthcare Providers -updated

Synthesis and antiviral activity of a series of 1′-substituted 4-aza-7,9-dideazaadenosine Cnucleosides

Structure of the RNA-dependent RNA polymerase from COVID-19 virus

An efficient ribosomal frame-shifting signal in the polymerase-encoding region of the coronavirus IBV

The primary structure and expression of the second open reading frame of the polymerase gene of the coronavirus MHV-A59: a highly conserved polymerase is expressed by an efficient ribosomal frameshifting mechanism

Programmed ribosomal frameshifting in decoding the SARS-CoV genome

Characterization of an efficient coronavirus ribosomal frameshifting signal: requirement for an RNA pseudoknot

Mutational analysis of the "slippery-sequence" component of a coronavirus ribosomal frameshifting signal

Characterization of ribosomal frameshifting for expression of pol gene products of human T-cell leukemia virus type I

Ribosomal frameshifting efficiency and gag/gag-pol ratio are critical for yeast M1 double-stranded RNA virus propagation

Signals for ribosomal frameshifting in the Rous sarcoma virus gag-pol region

A three-stemmed mRNA pseudoknot in the SARS coronavirus frameshift signal

Frameshifting RNA pseudoknots: structure and mechanism

Structure and function of pseudoknots involved in gene expression control

Identification of RNA pseudoknotbinding ligand that inhibits the -1 ribosomal frameshifting of SARS-coronavirus by structure-based virtual screening

A drug screening toolkit based on the -1 ribosomal frameshifting of SARS-CoV-2

A -1 ribosomal frameshift in a double-stranded RNA virus of yeast forms a gag-pol fusion protein

Importance of ribosomal frameshifting for human immunodeficiency virus type 1 particle assembly and replication

The human immunodeficiency virus type 1 ribosomal frameshifting site is an invariant sequence determinant and an important target for antiviral therapy

The autocatalytic release of a putative RNA virus transcription factor from its polyprotein precursor involves two paralogous papain-like proteases that cleave the same peptide bond

Potential 3-chymotrypsin-like cysteine protease cleavage sites in the coronavirus polyproteins pp1a and pp1ab and their possible relevance to COVID-19 vaccine and drug development

This study was supported by Scientific Development Fund of Guangxi Academy of Sciences and Key Project of Guangxi Scientific Research and Technology Development Plan (AB17190534).

The authors declare no conflict of interest.

G. Wu designed this perspective and wrote the first draft. Both finalized this manuscript.

https://orcid.org/0000-0001-7642-3972 Guang Wu https://orcid.org/0000-0003-0775-5759