NIa-Pro of sugarcane mosaic virus targets Corn Cysteine Protease 1 (CCP1) to undermine salicylic acid-mediated defense in maize

Papain-like cysteine proteases (PLCPs) play pivotal roles in plant defense against pathogen invasions. While pathogens can secrete effectors to target and inhibit PLCP activities, the roles of PLCPs in plant-virus interactions and the mechanisms through which viruses neutralize PLCP activities remain largely uncharted. Here, we demonstrate that the expression and activity of a maize PLCP CCP1 (Corn Cysteine Protease), is upregulated following sugarcane mosaic virus (SCMV) infection. Transient silencing of CCP1 led to a reduction in PLCP activities, thereby promoting SCMV infection in maize. Furthermore, the knockdown of CCP1 resulted in diminished salicylic acid (SA) levels and suppressed expression of SA-responsive pathogenesis-related genes. This suggests that CCP1 plays a role in modulating the SA signaling pathway. Interestingly, NIa-Pro, the primary protease of SCMV, was found to interact with CCP1, subsequently inhibiting its protease activity. A specific motif within NIa-Pro termed the inhibitor motif was identified as essential for its interaction with CCP1 and the suppression of its activity. We have also discovered that the key amino acids responsible for the interaction between NIa-Pro and CCP1 are crucial for the virulence of SCMV. In conclusion, our findings offer compelling evidence that SCMV undermines maize defense mechanisms through the interaction of NIa-Pro with CCP1. Together, these findings shed a new light on the mechanism(s) controlling the arms races between virus and plant.

Conversely, pathogens have devised multiple stratagems to bypass plant immunity.A common tactic is the suppression of PLCP activities, often achieved through secretion of effectors or the deployment of endogenous inhibitors to stifle host defenses [4,5,12,[16][17][18][19][20][21].For example, the oomycete pathogen Phytophthora infestans has been shown to secret cystatin-like effectors (i.e., EPIC1 and EPIC2B) to inhibit the function of C14 protease in tomato and potato.Moreover, the C14 protease in tomatoes can be targeted by the P. infestans effector AvrBlb2, preventing its secretion into the apoplast and subsequent activation of a defense response [5].In the fungal maize pathogen Ustilago maydis, the effector protein Pit2 impedes the induction of SA-mediated plant defense [20].Pit2 operates as a substrate mimic, releasing an inhibitory peptide upon cleavage by apoplastic PLCPs [22].In citrus plants, the Sec-delivered effector 1 (SDE1) from Huanglongbing-associated bacteria has been verified to inhibit the functions of immune-related PLCPs [23].However, to date, our knowledge on how plant viruses regulate PLCP activities is still limited.Though a geminivirus tomato yellow leaf curl virus (TYLCV) was reported to interact with the tomato PLCP CYP1 and suppress its activity via its encoded V2 protein, the roles of PLCPs during viral infections in plants are still largely ambiguous [24,25].
Plant viruses pose significant threats to global crop production.Given their compact genomes, viruses heavily rely on host factors and proteins adept at suppressing plant defenses to ensure successful life cycles.Potyviruses, belonging to the most expansive group of plantinfecting RNA viruses, have a widespread presence, infecting numerous crops, leading to considerable agricultural setbacks [26,27].Potyviral genomes consist of single-stranded, positivesense RNA that encodes a large polyprotein.Additionally, a P3N-PIPO, resulting from viral RNA polymerase slippage, is expressed within the P3 cistron [28,29].This large polyprotein undergoes post-translational processing to yield ten mature proteins, including P1, HC-Pro, P3, and others [28,30,31].

Virus targets plant protease to suppress immunity
As the main protease of the potyviruses, NIa is responsible for the processing and cleavage of all proteins except for P1 and HC-Pro [32,33].Potyviral Nla is a protein that possesses two functional domains: an N-terminal VPg domain of 22 kDa and a C-terminal protease domain of 27 kDa (referred to as NIa-Pro) [32,33].The VPg protein is typically covalently linked to the 5' end of the virus and facilitates the stability of the viral genome [33].Additionally, the suboptimal cleavage between VPg and NIa-Pro is crucial to regulating their protein concentrations, and the slow release of VPg from the aggregated NIa protein is necessary for tobacco etch virus (TEV) infectivity [34].VPg can regulate the structure and function of NIa-Pro, thereby adjusting the virus' enzyme activity to different stages of infection [35].Although NIa-Pro possesses a cysteine residue in its active site, it shares structural motifs with eukaryotic cellular serine proteases [36].Consequently, it is classified as a cysteine protease (MEROPS Clan PA, family C4) and is related to the 3C proteases of picornaviruses [37][38][39][40].NIa-Pro is multifaceted, exhibiting nonsequence-specific DNase activity, binding to NIb for virus replication, and acting as a viral pathogenicity determinant [40,41].It also enhances aphid vector growth and reproduction and has been identified to trigger the Ry-mediated disease resistance in potato [32,36,[41][42][43][44][45][46][47].A recent proteomic study revealed that the NIa-Pro of TEV interacts with 76 host proteins associated with plant stress responses, metabolism, and photosynthesis [48].However, the precise mechanisms underlying NIa-Pro's function in virus infection remain elusive.
In this study, we demonstrated that SCMV infection in maize augments PLCP activity and specifically upregulates the expression of the PLCP gene, Corn Cysteine Protease 1 (CCP1).We determined that CCP1 imparts resistance against SCMV infection.Furthermore, we revealed that SCMV NIa-Pro interacts with CCP1, inhibiting its activity, which in turn diminishes CCP1's resistance and promotes SCMV infection.We also presented evidence highlighting the significance of two residues (K230 and D234) of NIa-Pro in its interaction with CCP1 and in SCMV's virulence.

SCMV infection induces CCP1 gene expression and activates PLCP activity
To ascertain whether PLCPs play a role in SCMV infection in maize, we initially probed the expression levels of PLCPs using our previous RNA sequencing dataset available in the Genome Sequence Archive (accession number CRA001815) from the Beijing Institute of Genomics Data Center (http://bigd.big.ac.cn/gsa) [60].Our analysis revealed that the expression of the CCP1 gene (Corn Cysteine Protease 1, Zm00001d016446) was significantly elevated at 3 and 9 days post SCMV inoculation (dpi).The CCP2 gene (Corn cysteine protease 2, Zm00001d020636) also showed heightened expression at 3 dpi but not at 9 dpi.In contrast, SCMV infection did not influence the expression levels of other genes in the maize PLCP family (S1 Fig) .Consequently, our attention was centered on CCP1.
The upregulation of CCP1 in SCMV-infected maize plants was further validated using quantitative reverse transcription-PCR (RT-qPCR) at 5, 7, 10, 12, and 14 dpi (Fig 1A).Immunoblotting with a specific anti-CCP1 antibody indicated a downregulation of CCP1 precursor (preCCP1) during viral infection at 5 and 7 dpi, while an upregulation was observed at 10 and 12 dpi.This pattern suggests potential post-translational regulation of CCP1 in SCMVinfected maize plants.
Considering previous reports that PLCPs' proteolytic activity can sometimes diverge from their gene expression and protein abundance [18,23,61], we explored how SCMV infection influences PLCP protease activities.Using DCG-04, a biotinylated derivative of E-64 that binds exclusively to active forms of cysteine proteases, we conducted activity-based protein profiling (ABPP) [62,63].The ABPP analyses revealed a temporary surge in PLCP activity at 7 dpi, which subsequently returned to the activity levels observed in mock-inoculated plants by 10 dpi (Fig 1D and 1E).Collectively, our findings demonstrate that SCMV infection alters maize PLCPs across gene expression, protein abundance, and protease activity.

CCP1 confers resistance to SCMV infection
To elucidate the role of CCP1 in SCMV infection, we suppressed its expression in maize using virus-induced gene silencing (VIGS).We inserted a 210 bp fragment with high sequence specificity for CCP1 into a cucumber mosaic virus (CMV, genus Cucumovirus in the family Bromoviridae)-based VIGS vector (S2 Fig) [64].Maize seedlings treated with CMV-GUS served as negative controls.By 10 dpi, the plants treated with CMV-CCP1 exhibited no notable developmental aberrations.Subsequently, all the maize plants were subjected to challenge inoculation with SCMV.Relative to control plants, those with silenced CCP1 manifested pronounced leaf chlorosis and a dwarfed growth post SCMV infection (Fig 2A).RT-qPCR data showcased that, in the first SCMV-systemically infected leaves, the expression level of CCP1 in CMV-CCP1 silenced plants diminished by approximately 60% at 10 dpi when compared to SCMV-infected CMV-GUS control plants (Fig 2B).However, the level of SCMV genomic RNA in the CCP1-silenced plants was augmented by about 1.7-fold compared to the controls (Fig 2B).The accumulation levels of CMV genomic RNA were similar in both groups (S4 Fig) .Immunoblotting findings paralleled these results, revealing decreased CCP1 protein levels but heightened SCMV CP protein levels in the CCP1-silenced plants (Fig 2C and 2E).Moreover, ABPP demonstrated notably diminished PLCP protease activities in these plants (Fig 2D and 2E).
To further ascertain the role of CCP1 in SCMV infection, we co-expressed CCP1 or a mutant CCP1 with three substitutions in its catalytic residues (termed CCP1m) and SCMV RNA in maize protoplasts.Protoplasts co-transfected with mRFP and SCMV RNA were used as a control.At 16 hours post transfection (hpt), we harvested the co-transfected protoplasts and subjected them to RT-qPCR analysis.Our analysis has revealed a 1.3-fold increase in the expression levels of CCP1 or CCP1m-overexpressed protoplasts (S5A Fig) .Additionally, a substantial reduction of 0.5-fold in the accumulation of SCMV RNA was observed in the CCP1-overexpressed protoplasts, but not in the CCP1m-overexpressed protoplasts, compared to the protoplasts co-transfected with mRFP and SCMV RNA (S5B Fig) .We have also examined the expression of SA marker genes ZmPR1 and ZmPR5 in protoplasts expressing CCP1, and found a significant increased for both (S5C and S5D Fig) .Collectively, our evidence underscores the defensive role of CCP1 proteinase activity against SCMV infection in maize.

CCP1 modulates SA signaling pathway
A recent report has shown that the PLCPs CP1A and CP2 can directly regulate SA-dependent signaling in maize [65].Prompted by this, we sought to discern if CCP1 similarly modulates SA signaling by examining SA content and the transcription of SA-responsive pathogenesisrelated genes (PRs) in both silenced and non-silenced maize.Upon CCP1 suppression, no discernible growth alterations were observed relative to CMV-GUS controls (Fig 3A).Yet, SA content in CCP1-silenced plants was reduced by approximately 50%, while the levels of JA, ABA, and IAA remained consistent (Figs 3B and S3).Concurrently, the transcription levels of ZmPR1 and ZmPR5 in CCP1-silenced plants decreased by around 65% compared to controls (Fig 3C and 3D).These observations hint at CCP1's involvement in both SA accumulation and PR gene expression in maize.
In light of the up-regulation of CCP1 expression by SCMV infection and the subsequent accumulation of SA, we further inspected the effect of SCMV infection on SA signaling in CCP1-silenced plants compared to CMV-GUS controls.Our findings revealed that, while SCMV infection augmented SA accumulation and PRs expression in both sets of plants, the levels in CCP1-silenced plants remained significantly lower than in controls (Fig 3B -3D).Since previous studies have demonstrated that SA can impede SCMV infection in maize [59], one can speculate that the suppression of SCMV infection by CCP1 is mediated by the modulation of SA signaling pathways.

CCP1 interacts with SCMV NIa-Pro
To investigate if SCMV directly counteracts CCP1 to allow efficient infection, we employed yeast two-hybrid (Y2H) assays to identify SCMV protein(s) that might interact with CCP1.The CCP1 belongs to the subfamily 7 of PLCPs, which is homologous to the RD19 in A. thaliana [17].Precursor CCP1 protein contains an N-terminal signal peptide (SP), an autoinhibitory pro-domain and the mature protease domain (Fig 4A).Firstly, we used the full-length CCP1 (preCCP1) to examine interactions with individual SCMV proteins, however, no interactions were observed in the transformed yeast cells.Considering that the protease domain of PLCPs was employed in the protein interaction assays in the maize-U.Maydis and citrus-Huanglongbing systems, we utilized the protease domain of CCP1 (aa residue 137-371, referred to as mCCP1) as a bait protein for Y2H to screen for interactions with each of the SCMV proteins [20,23].Remarkably, we discovered that NIa-Pro is a potential interactor of mCCP1 (Fig 4B).On the other hand, neither the precursor of CCP1 (aa residue 1-371, preCCP1) nor the signal-peptide (SP) deletion mutant of CCP1 (aa residue 19-371, iCCP1) interacted with NIa-Pro in yeast cells (Figs 4A and S6).While there is no direct research evidence on the activation of CCP1, taking the example of a PLCP RD21 in A. thaliana [6,66], we speculate that CCP1 requires two steps to remove the signal peptide and auto-inhibitory domain, ultimately activating to its mature protease form.
Validating this interaction in Nicotiana benthamiana leaves, we confirmed that 3Flag-NIa-Pro interacts with both mCCP1-3Myc and preCCP1-3Myc (Figs 4C and S7).Leaves coexpressing mRFP-3Myc or GUS-3Myc and 3Flag-NIa-Pro served as negative controls (Figs 4C and S7).Furthermore, in vivo interaction between CCP1 and NIa-Pro was examined using a luciferase complementation assay (LCI).Co-expression of Nluc-mCCP1 or Nluc-preCCP1 and Cluc-NIa-Pro in N. benthamiana leaves led to a positive Luc activity (Figs 4D and S8).Moreover, in vitro pull-down assays provided evidence of a direct interaction between NIa- CCP1 contains a signal peptide (SP), an autoinhibitory prodomain (pro-domain) and a mature protease domain.Position of the catalytic triad (Cys, His and Asn) in the mCCP1 is indicated.Different forms of CCP1 are named as preCCP1, iCCP1, and mCCP1, respectively.B) Pair-wise direct Y2H assay was conducted to examine the interaction between mCCP1 and the individual SCMV-encoded proteins.Yeast cells grown on the QDO selective medium are the cells with a positive protein-protein interaction.Yeast cells grown on the DDO medium are the transformed cells.Yeast cells transformed with the empty vector (BD) are used as a negative control.BD, Gal4 Binding Domain; AD, Gal4 activation domain.The AD-T+BD-53 serves as the positive interaction control.C) Co-immunoprecipitation (Co-IP) assay was conducted to verify the interaction between mCCP1 and NIa-Pro in N. benthamiana leaves.Mutant mCCP1 C161A,H303A,N330A has three alanine substitutions at its catalytic triad.3Flag-NIa-Pro was co-expressed with mCCP1-3Myc or mCCP1 C161A,H303A,N330A -3Myc in N. benthamiana leaves through agro-infiltration.Leaf tissues were harvested at 48 hours post agroinfiltration (hpai) for Co-IP analysis.Samples of plants co-expressing 3Flag-NIa-Pro and mRFP-3Myc were used as a negative control.Co-IP assay was performed using an anti-Flag affinity agarose gel.Protein samples (Input) and the immunoprecipitated protein samples (Flag-IP) were analyzed through immunoblotting assays using an anti-Flag or an anti-c-Myc antibody.D) Luciferase (Luc) complementation imaging (LCI) analysis of the interaction between NIa-Pro and mCCP1 in N. benthamiana leaves.The indicated plasmid pairs were transiently co-expressed in N. benthamiana.NLuc co-expressed with CLuc or CLuc-NIa -Pro, and mCCP1-NLuc co-expressed with CLuc, were used as negative controls.The luciferase activity was measured using a luminometer at 48 hpai.E) In vitro pull-down assay using a GST-tagged NIa-Pro to immunoprecipitate TF-tagged mCCP1 or mCCP1 C161A,H303A,N330A .GST-NIa-Pro was used as a bait in this assay and TF-mCCP1 or TF-mCCP1 C161A,H303A,N330A was used as a prey.The input and the pull-down protein samples were analyzed through immunoblotting assays using an anti-GST or an anti-His antibody.Please note that the His Abs did show unspecific binding for the GST and GST-NIa-Pro proteins.*, the protein bands corresponding to the expected TF-mCCP1.Purified GST was used as a negative control.https://doi.org/10.1371/journal.ppat.1012086.g004 Pro and mCCP1 (Fig 4E).These results confirm that NIa-Pro interacts with the protease domain of CCP1.
To better understand the mode of interaction, we asked whether NIa-Pro directly binds to the catalytic triad in CCP1.To this end, we generated a mutant mCCP1 C161A, H303A, N330A where all residues of the catalytic triad were exchanged by Alanine.Strikingly, co-immunoprecipitation (Co-IP), GST Pull-down, Y2H, and LCI assays showed that mCCP1 C161A, H303A, N330A did not interact with NIa-Pro (Figs 4C, 4E, S9 and S10).In addition, a fusion protein with VPg at the N terminus of NIa-Pro, known as NIa during SCMV infection in plant, also interacted with mCCP1 but not with mCCP1 C161A, H303A, N330A (S11 Fig) .These results suggest that the catalytic triad of CCP1 is critical for the interaction with NIa-Pro.To exclude the possibility of protein conformational changes caused by catalytic triad residues substitution of CCP1, we predicted the 3D-structure of mCCP1 and mCCP1 C161A, H303A, N330A separately.The results suggest that the mutations have no effect on the protein conformation (S12 Fig).
To study the localization pattern, NIa-Pro was tagged at its C-terminus with GFP (NIa-Pro-GFP) and transiently expressed in N. benthamiana leaves through agroinfiltration.As expected, NIa-Pro was distributed in the cytoplasm and nucleus [46] (S13 Fig) .Immunoblotting findings suggested that fusion protein NIa-Pro-GFP retains its integrity when compared to the control GFP (S14 Fig) . Moreover, A. thaliana Aleurain (the marker for lytic vacuole) and HDEL (the marker for endoplasmic reticulum) were used for an elaborate investigation into the localization of CCP1.Similar to the observation of ortholog A. thaliana RD19, a perfect colocalization between CCP1 and Aleurain was discerned (S15A Fig) [67].Also, a conspicuous colocalization of CCP1 with HDEL (S15B Fig) suggested that CCP1 is indeed situated on the endoplasmic reticulum initially, and may subsequently be directed to the lytic vacuole for activation.
We further co-expressed NIa-Pro-GFP and CCP1-mRFP for colocalization investigations in N. benthamiana and maize, revealing that partial colocalization within the cytoplasm of NIa-Pro and CCP1 occurs in the initial stages following agrobacterium infiltration (prior to 48 hours post-infiltration) or bombardment (S16A and S16C

NIa-Pro inhibits CCP1 protease activity
To elucidate the potential functional link between these two proteins, we investigated the impact of NIa-Pro on the stability of CCP1.The protein levels of preCCP1 or mCCP1 were assessed in N. benthamiana leaves co-expressing GUS and NIa-Pro, and no significant difference was observed in these samples (S17 Fig) , indicating that NIa-Pro does not affect the stability of CCP1 in vivo.Given that NIa-Pro interacts with CCP1 through the protease domain, we asked whether it could inhibit CCP1 activity.Two assays were used to measure the proteolytic activities of CCP1 in the presence of NIa-Pro.In all these assays, a specific chemical inhibitor of PLCPs, E-64 was used as a positive control [62,68].
First, we examined the in vitro effect of NIa-Pro on CCP1 proteolytic activity using recombinant proteins purified from E. coli.Recombinant mCCP1 was expressed in E. coli with an Nterminal TF-tag, which was removed by thrombin cleavage and subsequent gel filtration.Concurrently, GST-tagged NIa-Pro or GST proteins were produced in E. coli.After pretreatment with purified GST, GST-NIa-Pro, or E-64, the relative activity of CCP1 was measured using a fluorescein-labeled casein substrate.When cleaved by PLCP, the casein substrate releases a fluorescent signal that can be quantified with a fluorometer.The results showed that the proteolytic activity of CCP1 was strongly inhibited by E-64 or GST-NIa-Pro pretreatment compared to GST pretreatment (Fig 5A).The mutant mCCP1 C161A, H303A, N330A exhibited no protease activity (Fig 5A ), confirming the importance of the catalytic triad for CCP1 protease activity.When GST-NIa-Pro (0.5 μM) was added to the reaction, the proteolytic activity of CCP1 decreased by approximately 72% (Fig 5A).Its inhibitory effect was significant, though weaker than that of E-64, which reduced protease activity by 90%.
Next, we conducted ABPP in a semi-in vitro assay using recombinant NIa-Pro protein purified from E. coli and CCP1 expressed in plant tissues.The results demonstrated that CCP1 could be labeled by DCG-04, suggesting it is an active enzyme.As expected, a circa 0.65-fold reduction in CCP1 activity was observed with the addition of purified GST-NIa-Pro (1.0 μM) compared to the control (Fig 5B).The addition of E-64 completely abolished the DCG-04 labeling.
We then ascertained whether NIa-Pro could inhibit PLCPs protease activities in maize.A heterologous foxtail mosaic virus (FoMV, genus Potexvirus in the family Alphaflexiviridae) vector was used for NIa-Pro expression [69].Coding sequences of full-length SCMV NIa-Pro fused to 3Flag were inserted into a FoMV-based expression vector to produce FoMV-3Flag-NIa-Pro, with FoMV-3Flag-GFP set as a control.Total proteins were extracted from FoMVinfected maize leaves for analysis.The results showed that, though the accumulation level of 3Flag-NIa-Pro was significantly lower than that of 3Flag-GFP, PLCP activity was markedly reduced in NIa-Pro-expressed (FoMV-3Flag-NIa-Pro) plants compared to the control (FoMV-3Flag-GFP) plants (Fig 5C and 5D).Collectively, NIa-Pro plays a role in suppressing the protease activity of PLCPs.

The 48-amino acid inhibitor motif is sufficient for function of NIa-Pro to interact with CCP1 and suppress its activity
To pinpoint the essential amino acid residues responsible for suppressing CCP1 activity, we predicted the structure of SCMV NIa-Pro using its complete amino acid sequence in the RCSB PDB (http://www.rcsb.org/).The analysis revealed that NIa-Pro contains an N-terminal protease domain (pdNIa-Pro, 1-194 aa) (Fig 6A and 6B).As observed in crystal structures of several potyviral protease, the C-terminus of SCMV NIa-Pro (195-242 aa) is disordered (Fig 6B) [70,71].Previous studies have discovered that the C-terminus of NIa-Pro encoded by TEV, genus Potyvirus binds to its active site [70].The full-length form of NIa-Pro encoded by TVMV (tobacco vein mottling virus, genus Potyvirus) exhibits significantly lower activity than its Cterminus-deleted form, suggesting that the C-terminus of NIa-Pro might potentially exert inhibitory effects on its activity.The predicted structure of SCMV NIa-Pro also suggested that the speculated inhibitory motif is surface-localized, potentially facilitating its association with CCP1.
Subsequently, we divided NIa-Pro into two regions based on its predicted structure: the protease domain (pdNIa-Pro, 1-194 aa) and the speculated inhibitory motif (imNIa-Pro, 195-242 aa) (Fig 6A).An LCI assay confirmed that imNIa-Pro (195-242aa) alone could interact with mCCP1 (Fig 6C).Further analysis of the amino acid sequences of NIa-Pro from various potyviruses showed that the speculated inhibitory motif is present in other potyviruses, with several residues being highly conserved across multiple NIa-Pros (S18 Fig) .While W199 and W203 are strictly conserved in the aligned Potyvirus species, residues N206, K230, and D234, although not strictly conserved, are conserved in the majority of potyviruses.To identify the specific amino acid(s) responsible for the NIa-Pro-CCP1 interaction, we introduced single point mutations (W199A, W203A, N206A, K230A, and D234A) into the inhibitory motif.The LCI assay and Y2H results revealed that residues K230 and D234 are both vital for the NIa-Pro-CCP1 interaction (Figs 6D and S19).
These findings suggest that the inhibitor motif alone is sufficient for the NIa-Pro-CCP1 interaction, with residues K230 and D234 being essential.We hypothesized that the inhibitor motif alone could interfere with CCP1 activity.To test this, we used purified recombinant protein GST-imNIa-Pro and GST-NIa-Pro K230A, D234A for protease activity experiments.The results showed that GST-imNIa-Pro reduced the proteolytic activity of CCP1 by approximately 80%, a more significant reduction than the full-length NIa-Pro (Fig 6E).The mutated NIa-Pro K230A, D234A did not significantly affect CCP1 activity (Fig 6E).

The essential role of NIa-Pro's key residues in SCMV infection
To elucidate the significance of the inhibitor motif and the key residues of NIa-Pro in SCMV infection, we incorporated the mutations K230A and D234A into the NIa-Pro of the SCMV-GFP infectious clone, resulting in the variant pSCMV-NIa-Pro K230A, D234A -GFP.Concurrently, we excised the inhibitor motif of NIa-Pro from the SCMV-GFP infectious clone, yielding the clone pSCMV-NIa-Pro IMD -GFP.Agrobacterium cells harboring pSCMV-NIa-Pro K230A, D234A -GFP, pSCMV-NIa-Pro IMD -GFP, and pSCMV-GFP were then infiltrated into N. benthamiana leaves.By 5 days post-agroinfiltration (dpai), GFP fluorescence was discernible in areas infected with SCMV-GFP or SCMV-NIa-Pro K230A, D234A -GFP (Fig 7A).Notably, the GFP fluorescence intensity in areas infected with SCMV-NIa-Pro IMD -GFP was markedly diminished compared to regions infected with SCMV-GFP and SCMV-NIa-Pro K230A, D234A -GFP (Fig 7A).RT-qPCR analyses further corroborated that the relative accumulation of SCMV mRNA in areas infected with SCMV-NIa-Pro IMD -GFP was substantially reduced compared to those infected with SCMV-GFP or SCMV-NIa-Pro K230A, D234A -GFP (Fig 7B).
Subsequent inoculation of maize seedlings with crude extracts from SCMV-infected N. benthamiana leaves revealed pronounced mosaic symptoms in maize plants infected with SCMV-GFP by 8 dpi.In contrast, the virulence of the two mutant SCMV on maize was The predicted 3D structure of SCMV NIa-Pro shows a protease domain (Cartoon shape) and an inhibitor motif (the blue line indicated).C) LCI assay of the interaction between mCCP1 and pdNIa-Pro or imNIa-Pro in N. benthamiana leaves.The Agrobacterium strains carrying the indicated constructs were infiltrated into N. benthamiana leaves.D) LCI assay for identification of the amino acid within the imNIa-Pro essential for interaction with mCCP1.The Agrobacterium strains carrying the indicated constructs were infiltrated into N. benthamiana leaves.NLuc co-expressed with CLuc or CLuc-NIa-Pro-, and mCCP1-NLuc co-expressed with CLuc, were used as negative controls.The luciferase activity was measured using a luminometer at 48 hpai.E) ABPP analysis to estimate the inhibitory effect of different form NIa-Pro on CCP1 protease activity.preCCP1-3Flag or preCCP1 C161A,H303A,N330A (preCCP1 m )-3Flag was transiently expressed in N. benthamiana leaves.Then the Flag-enriched proteins were incubated with 5 μM E-64 (as a positive control) or 1.0 μM purified GST-NIa-Pro, GST-NIa-Pro K230A, D234A (NIa-Pro m ), GST-imNIa-Pro and GST (as a negative control), followed by immunoblotting assay using a streptavidinconjugated with horseradish peroxidase (HRP).PreCCP1 m -3Flag was used as an inactive control.ImageJ software was used to estimate the detection signal intensity.https://doi.org/10.1371/journal.ppat.1012086.g006At the same time, we measured the levels of CCP1 accumulation and PLCPs enzyme activity in maize plants.The results showed that, compared with the significant upregulation of CCP1 transcriptional levels in maize plants infected with SCMV-GFP, there were no significant changes in CCP1 expression levels in maize plants infected with the SCMV-NIa-Pro K230A, D234A -GFP, which were comparable to those in the mock group (Fig 8C).Additionally, we detected the accumulation levels of active CCP1 and found that CCP1 was significantly activated in maize infected with SCMV-GFP, whereas there was no significant induction observed in the SCMV-NIa-Pro K230A, D234A -GFP infected group (Fig 8D ).
Analysis of active PLCPs also revealed a significant increase in maize plants infected with SCMV-GFP at 8 dpi, whereas there was no such increase in the SCMV-NIa-Pro K230A, D234A -GFP -infected group (Fig 8D).Lastly, measurement of SA hormone levels and the expression of SA marker genes showed that there was no significant induction in SA levels or the upregulation of PR1 and PR5 in the SCMV-NIa-Pro K230A, D234A -GFP-infected group, despite the increase observed in the SCMV-GFP-infected group (Fig 8E and 8F).

Discussion
In this study, we found that SCMV infection enhances CCP1 activation and induces SA signaling pathway activation in maize plants.The activated CCP1 mediates maize resistance to SCMV and is associated with SA signaling pathway, whereas the precise regulatory mechanism remains unknown.Conversely, SCMV-encoded NIa-Pro interacts with CCP1 to inhibit its activity, thereby weakening resistance to SCMV.Thus, we proposed a working model to illustrate these findings (Fig 9).
PLCPs are known to play important roles in regulating plant defense against a broad range of microbial pathogens including bacteria, fungi and oomycetes [7,23,66,[72][73][74].In this study, we have demonstrated the upregulation of CCP1 in maize upon SCMV infection, and this upregulation contributes to the resistance against SCMV infection.We found that the expression of CCP1 is up-regulated upon SCMV infection in maize leaves.In line with this, PLCP activity was induced early after SCMV infection.This is supported by the analyses of published transcriptome data from several virus-infected plants, which revealed that the expression of PLCPs can be up-regulated upon virus infection in various species (S1 Table) [75][76][77][78][79][80].Through knock-down assays using CMV-based VIGS vector, we have shown that CCP1 plays a role in maize resistance to SCMV infection.Given the resistance roles of PLCPs in fungal or bacterial infected plants, it is suggested that the PLCPs-mediated resistance is conserved for against diverse biotic stresses [9,16,81].
The roles of plant proteases in disease resistance have been documented, yet the mechanisms underpinning these resistances remain elusive [18,82].Some studies suggest that PLCPs might directly hydrolyze pathogen components.For instance, papain was found to curb the growth of Phytophthora palmivora in papaya [83].Additionally, PLCPs have been noted to cleave plant peptides, initiating host defense responses [9].In maize, SA prompts the activation of PLCPs, which subsequently triggers SA-dependent defense mechanisms [12].The activation of SA-induced PLCPs results in the release of Zip1, a bioactive immune signaling peptide derived from its propeptide precursor.This peptide acts as an activator of SA biosynthesis [65].Notably, the Zip1-mediated activation of SA signaling necessitates the apoplastic proteases CP1A, a member of the PLCP Subfamily 1, and CP2 from Subfamily 8.In related research, increased activities of PIP1 and RCR3 in tomatoes post benzothiadiazole treatment, an SA analog [18].The potential of other PLCPs to modulate SA signaling is yet to be ascertained.Our findings show that silencing CCP1 expression in maize markedly reduces the levels of endogenous SA and the expression of SA-responsive genes ZmPR1 and ZmPR5.Given that SA content rises in SCMV-infected maize plants and that activated SA signaling plays a crucial role in maize's resistance against SCMV infection [59], we postulate that CCP1-mediated resistance to SCMV is moderated via SA signaling regulation in maize.
Our findings in this study suggest that CCP1 contributes to SA-associated defense in maize, which in turn is important for plant resistance to SCMV.However, it still remains to be answered how CCP1 interferes with SCMV infection on the cellular level.Besides the induction of SA-signaling, PLCPs have also been reported to be involved in the induction of plant cell death responding to pathogen infection and leaf senescence [11,13,14].Silencing CCP1 does not activate PLCP-mediated SA signaling, thus the virus replicates in higher level and causes more severe symptom.
For infection to be compatible, plant PLCPs need to be inhibited by pathogen effectors [20].Ample evidence reveals that various pathogens, including bacteria, fungi, oomycetes, and nematodes, have evolved to produce effectors that hinder PLCP activities, thus aiding their infection in plants [5,7,16,23,66,72,74,84].While PLCPs are targeted by specific effectors, the inhibiting effectors do not possess conserved motifs, indicating that these effectors have evolved independently to disrupt the activity of particular PLCPs [23].In our investigation, we found that the localization pattern of CCP1 is similar to that of vacuolar proteases that are synthesized as preproteins in the rough endoplasmic reticulum and transiently transported to the vacuole through the endomembrane secretion system [85], ultimately exerting its protease activity in the vacuole.In addition, we found that SCMV NIa-Pro constrains CCP1 protease activity through its interaction with the active protease domain (mCCP1, aa residue 137-371), but not the CCP1 precursor (preCCP1, aa residue 1-371).Furthermore, when co-expressed with CCP1, NIa-Pro is found to be co-localized with CCP1 in the lytic vacuole during the later stage, thereby effectively inhibiting CCP1 activity.This observation aligns with prior findings pertaining to the U. maydis effector Pit2 and maize cystatin CC9, both of which interact with PLCP protease domains [12,20].Similarly, the Huanglongbing (HLB)-associated bacterium Candidatus Liberibacter asiaticus (CLas) effector SDE1 targets citrus PLCPs through interactions with conserved protease domains [23].A most recent study uncovered the crucial role of PLCP TaRD21A in wheat against wheat yellow mosaic virus (WYMV, genus Bymovirus in the family Potyviridae) through the release of a small peptide, while the NIa protein encoded by WYMV suppresses the activity of TaRD21A to facilitate infection in susceptible wheat [86].Thus, direct binding of effector proteins to plant PLCP protease domains appears to be a recurrent tactic employed by microbial pathogens to suppress the activity of these immune proteases.
PLCPs possess a conserved catalytic triad composed of three amino acid residues: Cys, His, and Asp.We observed that a mutant CCP1, with all the catalytic triad residues substituted, not only lost its protease activity but also had no ability to bind with NIa-Pro.This highlights the importance of the catalytic triad in the interaction between CCP1 and NIa-Pro.One theory posits that NIa-Pro impedes CCP1 protease activity by targeting this catalytic triad.Another theory proposes that CCP1 identifies NIa-Pro as a substrate, which would explain the requirement for an active catalytic triad for binding.Our data support that NIa-Pro could bind to CCP1, but cannot cleave CCP1, thus obstructing the CCP1 catalytic triad.This idea also resonates with the U. maydis Pit2 effector, regarded as a "false substrate", which suppresses the activity of various PLCPs [22].In addition, it appears that the protein stability of NIa-Pro in samples co-expressing CCP1 and NIa-Pro is seemingly affected, implying the potential degradation of NIa-Pro as a substrate of CCP1 (S21 Fig) .However, further investigation is required to confirm this possibility.
We further found that the 48 aa inhibitor motif is sufficient for function of NIa-Pro to interact with CCP1 and suppress its activity (Fig 6).The predicted structure of NIa-Pro also suggested that the inhibitory motif is surface-localized, potentially facilitating its association with CCP1.Evidence for the function of the 48 aa inhibitor motif is given by the mutational analyses that resulted in: i) loss of NIa-Pro-CCP1 interaction in LCI, ii) loss of protease inhibition of recombinant protein in the in-vitro assays and iii) loss of virulence, i.e. a SCMV mutant pSCMV-NIa-Pro IMD -GFP which the 48 aa inhibitory motif was deleted.This line of evidence is further supported by the finding that the inhibitor efficiency of 48 aa inhibitor motif is higher than full length NIa-Pro.Amino acid sequence alignment of NIa-Pros belonging to the different potyviruses shows about 74.44% identity and there are some conserved residues in the inhibitory motif.Our further study indicated that among the five residues, only two residues K230 and D234 were crucial for the NIa-Pro-CCP1 interaction.The mutant NIa-Pro K230A, D234A resulted in the loss of inhibitor activity of NIa-Pro, suggesting that K230 and D234 were crucial for the function of NIa-Pro.Strikingly, we found that a mutant SCMV-NIa-Pro K230A, D234A -GFP also showed the significantly reduced virulence compared to WT SCMV.
To our knowledge, this is the first report of viral-encoding protease utilize the inhibitory motif to suppress PLCPs, a hub in host immunity to promote virus infection.The most intriguing questions are about whether the inhibitor motif of NIa-Pro could be processed upon SCMV infection, and how the inhibitor motif is transported and targets the activated PLCPs to function.
Our study provide evidence that CCP1 plays positive roles in maize resistance to SCMV infection.In the early stages of virus infection, we found that both transcriptional and protein levels of CCP1 show an increasing trend, and the overall enzymatic activity of PLCPs increases.This correlates with our previous findings that SCMV infection leads to an increase in SA accumulation.The virus counters SA-mediated resistance by interacting with and inhibiting the enzymatic activity of key PLCPs involved in SA pathways.This is consistent with the observed inhibition of PLCP activity in the later stages of virus infection, allowing the virus to infect efficiently.This suggests that CCP1 does not trigger PCD.Since SCMV infection maize plants results in mosaic symptoms but not leaf cell death, there might be a balance between the upregulation of CCP1 due to SCMV infection and its inhibition by SCMV NIa-Pro.Altogether, these discoveries unveil the significant involvement of plant PLCPs in fighting viral infections and uncover a previously undisclosed mechanism of virulence employed by potyviruses that infect plants.

Plant growth and virus inoculation
Maize inbred line B73 and N. benthamiana plants were grown inside growth chambers maintained at 24/22˚C (day/night) and with a 16/8 h (light/dark) photoperiod.SCMV-BJ isolate is from a previously described source [52] and propagated in maize plants for further uses.SCMV inoculum was prepared by homogenizing SCMV-BJ-infected maize leaf tissues in a 0.01 M phosphate buffer (PB), pH 7.0, immediately prior to inoculation to 8-day-old maize seedlings as described [87].Maize seedlings inoculated with PB were used as the mock-inoculated controls.

Plasmid constructions
For CMV-based VIGS, a 210 bp fragment representing a partial sequence of CCP1 (S2 Fig) was RT-PCR amplified and inserted into the pCMV201-2b N81 vector [64] to generate pCMV201-CCP1 for the CMV-based VIGS.
All the constructs were sequenced before use.Sequences of primers used in this study are listed in S2 Table.

CMV-based VIGS in maize
The CMV-based VIGS assay was performed as reported previously [64].Agrobacterium cultures harboring pCMV101, pCMV301, and pCMV201-CCP1 or pCMV201-GUS were mixed at a 1:1:1 ratio and then infiltrated into N. benthamiana leaves.At 4 dpai, crude extracts from the infiltrated leaves were used to inoculate maize B73 seeds using a vascular puncture inoculation method [64].

FoMV-mediated gene over-expression in maize plants
Inoculation of N. benthamiana and maize seedlings with the foxtail mosaic virus (FoMV) expression constructs was performed as described previously [69].First, agrobacterium cultures harboring FoMV-3Flag-GFP or FoMV-3Flag-NIa-Pro were infiltrated into N. benthamiana leaves.At 4 dpi, infiltrated leaves were ground in 10% (w/v) 0.01 M phosphate buffer, supplemented with 1% (w/v) Celite 545 AW (Sigma-Aldrich).Then crude extracts were used for rub inoculation of maize seedlings.At 10 min post inoculation, leaves were sprayed with tap water, and plants were bagged.Plants were maintained under high humidity and no light for 1 day and finally returned to standard growth conditions for subsequent experiments.

Total RNA extraction and RT-qPCR analysis
Total RNA was isolated from various tissue samples using TRIzol reagent (Tiangen, Beijing, China) followed by a treatment with an RNase-free DNase I (TaKaRa, Dalian, China).Firststrand cDNA (20 μl) was synthesized using 2.0 μg total RNA and an oligo (dT) primer.Tenfold diluted cDNA samples, gene-specific primers and a FastSYBR kit (CWBIO, Beijing, China) were used for qPCR on an ABI 7500 Real Time PCR system (Applied Biosystems, Foster City, CA, USA).The expression level of ZmUbi was used as an internal control and the relative expression of each gene was calculated using the 2 -ΔΔCm method [90].Statistical differences between the treatments were determined using the Student's t-test.

Immunoblotting assay and statistical analyze
Total protein preparation and separation through electrophoresis were the same as described [87].Polyclonal antibody of SCMV CP or CCP1 was used at a dilution of 1:5,000.Monoclonal antibodies of actin and GST (CW0258 and CW0084, CWBIO) were used at a dilution of 1:5,000.Monoclonal antibodies of anti-His (CW0285, CWBIO), anti-Myc and anti-Flag (A5598 and A8592, Sigma-Aldrich, Taufkirchen, Germany) were used at a dilution of 1:10,000.Streptavidin-HRP conjugates (Thermo Scientific) were used at a dilution of 1:5,000.For further quantitative analysis, the accumulation levels of CCP1, SCMV CP, and active PLCPs were normalized to actin control.

Protein interaction assays
Y2H assay was performed using the Matchmaker Yeast Two-Hybrid System as instructed (Clontech Laboratories, Mountain View, CA, USA).Briefly, specific combinations of the AD and BD vectors were co-transformed into yeast cells and then allowed to grow on the SD/-Leu/-Trp DO (DDO) medium or the SD/-Leu/-Trp/-Ade/-His DO (QDO) medium at 30˚C.The pGADT7-T/pGBKT7-53 (AD-T/BD-53) plasmid was used as a positive control, and pGADT7-mCCP1/pGBKT7 (AD-mCCP1/BD) was used as a negative control.
For in vivo co-immunoprecipitation (Co-IP) assay, plasmid pGD-3Flag-NIa-Pro, pGD-mCCP1-3Myc or pGD-mCCP1 C161A,H303A,N330A -3Myc were individually transformed into Agrobacterium cells and then heterologously expressed in the leaves of 4-week-old N. benthamiana plants through infiltration.The infiltrated leaves were harvested at 72 hours post agro-infiltration (hpai) and grounded (1 g tissue per sample) in an extraction buffer (50 mM Tris-HCl, pH7.5, 150 mM NaCl, 1 mM EDTA, 0.1% Triton X-100 and 10% glycerol) supplemented with 1 mM phenylmethyl sulfonyl fluoride and a protease inhibitor cocktail as instructed (Sigma-Aldrich, Taufkirchen, Germany).The crude leaf extracts were centrifuged, filtered through 0.45-μm filters, and checked for proteins concentrations through the Bradford assay [91].Equal amount of protein in a sample (about 1.2 ml) was mixed with 40 μl of anti-Flag M2 Affinity gel (A2220, Sigma-Aldrich) and then incubated for 3 h at 4˚C on a rotary incubator to immunoprecipitate the target protein.The precipitates were washed four times with an ice-cold IP buffer followed by four times with an ice-cold PBS buffer.The resulting samples were individually solubilized in a loading buffer, boiled for 5 min, and then analyzed by immunoblotting assay using specific antibodies.
For LCI assay, equal amounts of Agrobacterium cultures containing different cLUC and nLUC construct pairs were co-transformed into N. benthamiana.The plants were harvested at 48 hpai and sprayed with a 1 mM luciferin solution (Promega, Beijing, China), and then examined and imaged using a low-light cooled charge-coupled-device camera (Lumazone 1300B, Teledyne Princeton Instruments).
For in vitro pull-down assay, GST-NIa-Pro, TF-mCCP1 or TF-mCCP1 C161A, H303A, N330A was expressed in E. coli strain BL21 (DE3) cells followed protein purification.The purified TF-mCCP1, TF-mCCP1 C161A, H303A, N330A , GST and GST-NIa-Pro were mixed in specific combinations in glutathione resins (50 μl per reaction, Thermo Scientific, USA) and then incubated for 3 h at 4˚C.The resins were rinsed with an IP buffer containing 20 mM Tris-HCl, 200 mM KCl, 0.1 mM EDTA, 0.05% Triton X-100, pH6.0, boiled for 10 min in a 2 × SDS sample buffer, and then subjected to gel electrophoresis followed by immunoblotting using an anti-GST or an anti-His antibody.

Protease activity assay
Plasmids including pGD-3Flag-mCCP1, pGD-3Flag-mCCP1 C161A,H303A,N330A or pGD-3Flag-mSbCP1 were expressed heterologously in 4-week-old N. benthamiana.At 48-72 hpai, the harvested leaf samples (1 g per sample) were ground individually in liquid nitrogen and then homogenized in a 5 ml of ice-cold 50 mM Tris buffer, pH7.2, containing 0.2% polyvinylpyrrolidone (PVP) and 5 mM mercaptoethanol.Then, the resulting samples were applied to immunoprecipitate Flag-tagged proteins with addition of anti-Flag M2 Affinity gel as previously described in Co-IP assay.After immunoprecipitation, the enriched proteins were washed four times with the IP buffer and then twice in the PBS buffer.
Protease activity was measured using a fluorimetric substrate (Z-Phe-Arg-7-amido-4-methylcoumarin, Sigma-Aldrich), which leads to release of fluorescence at 460 nm when cleaved by protease activity [92].Specific concentrations of purified TF-mCCP1 or the anti-Flag M2 Affinity Gel-enriched 3Flag-mCCP1 was mixed with the purified GST protein (negative control), GST-NIa-Pro or E-64 (positive control), and then incubated at 4˚C for 2 h.After incubation, each protein mixture (90 μl) was combined with 10 μl of 10 μM substrate followed by the measurement of absorbance at 460 nm using a microplate reader (SpectraMax i3x, Molecular Devices, USA).

Prediction of protein structure
The structures of proteins were predicted based on the complete amino acid sequence, utilizing the online structure prediction software SWISS-MODEL (https://swissmodel.expasy.org/interactive) in combination with the PDB protein structure database (https://www.rcsb.org/).

Activity-based protein profiling
For the total PLCPs activity in maize, total extracts from the 1 st -systemic leaves of SCMV or mock-inoculated plants were incubated with a final concentration of 2 μM DCG-04 (Medkoo, USA) for 4 h at room temperature, followed by precipitation with 100% ice-cold acetone.Samples were centrifuged at 12,000×g, washed with 70% acetone, then centrifuged again.Precipitated products were re-suspended in 50mM Tris buffer (pH 6.4) and either used directly for immunoblotting using Streptavidin-HRP conjugates (Thermo Scientific).
For the semi-in vitro ABPP assay, total leaf extracts from CCP1-3Flag or CCP1 C161A,H303A, N330A -3Flag was pretreated with either 5 μM E-64 or 1.0 μM GST-NIa-Pro or GST proteins.After pretreatment, the samples were incubated with 2 μM DCG-04 for 4 h at room temperature, followed by 3 min centrifugation at 3,000 g and then two rinses in PBS buffer.The resulting protein samples were individually mixed with a 2 × SDS sample loading buffer, boiled for 10 min, and then analyzed through immunoblotting assays using a Streptavidin-HRP conjugate as instructed.

S21 Fig. Full-length of CCP1 (preCCP1
) affects the stability of NIa-Pro in planta.pGD-preCCP1-3Myc or pGD-mCCP1-3Myc was co-infiltrated with pGD-3Flag-NIa-Pro in N. benthamiana leaves.pGD-GUS-3Myc co-expressed with pGD-3Flag-NIa-Pro was used as control.At 3 dpi, infiltrated leaves were sampled and analyzed for the accumulation levels of 3 Flag-NIa-Pro via immunoblotting assays.The strength of detection signal was analyzed using the ImageJ software.Three independent experiments were conducted with at least three biological replicates per treatment. (TIF)

Fig 1 .
Fig 1. SCMV infection up-regulates the expression of CCP1 and alters the activities of maize PLCPs.A) The expression of CCP1 in maize was determined using quantitative reverse transcription PCR (RT-qPCR) at 5, 7, 10, 12 and 14 days post inoculation (dpi) of SCMV (gray bars).The plants inoculated with phosphate buffer (Mock, black bars) were used as controls.Statistical differences (*, P < 0.05; **, P < 0.01) were evaluated by Student's t test analysis.B) The accumulation of CCP1 precursor (preCCP1) in maize was determined by immunoblotting assay using antibody specific to CCP1 or SCMV CP at 5, 7, 10 and 12 dpi.C) The relative amount of preCCP1 was quantified by software ImageJ, normalized to actin control.And the lanes of 5 dpi Mock were set to 1.0.Data were presented as means ± SE (n = 4).Statistical differences (***, P < 0.001) were evaluated by two-tailed Student's t test analysis.D) The protease activities of maize PLCPs were determined by an activity-based protein profiling (ABPP).Only the active forms of PLCPs were labeled with DCG-04 and then detected using a streptavidin-conjugated with horseradish peroxidase (HRP).E) The relative accumulation of active PLCPs was quantified by software ImageJ, normalized to actin control.And the lanes of 5 dpi Mock were set to 1.0.Data were presented as means ± SE (n = 4).Statistical differences (***, P < 0.001) were evaluated by two-tailed Student's t test analysis.Three independent experiments are conducted.https://doi.org/10.1371/journal.ppat.1012086.g001

Fig 2 .
Fig 2. Knockdown of CCP1 in maize plants enhanced the accumulation level of SCMV.A) The CCP1-silenced plants showed stronger leaf chlorosis and plant stunting compared with the CMV-GUS control plants at 10 d post challenge inoculation with SCMV.Bar = 5 cm.B) RT-qPCR analysis of the relative accumulation levels of CCP1 and SCMV RNA in the SCMV first systemically infected leaves at 10 dpi.C) Immunoblotting analysis of CCP1 and SCMV CP accumulation levels.Actin was used as gel loading control.Band intensities were measured using software ImageJ.Numbers indicate the accumulation levels of CCP1 and SCMV CP normalized to actin control.D) Activity-based protein profiling (ABPP) analysis of the protease activities of maize PLCPs in the SCMV first systemically infected leaves at 10 dpi.Actin was used as gel loading control.Band intensities were measured using software ImageJ.Numbers indicate the accumulation levels of active PLCPs normalized to actin control.E) The relative accumulations of CCP1 (left), SCMV CP (middle) and active PLCPs (right) were quantified by software ImageJ, and the mean of CMV-GUS+SCMV was set to 1.0.Error bars represent the means ± SE.Statistical differences (*, P < 0.05; **, P < 0.01; ***, P < 0.001) were evaluated by two-tailed Student's t test analysis.Three independent experiments are conducted.https://doi.org/10.1371/journal.ppat.1012086.g002

Fig 3 .Fig 4 .
Fig 3. Knockdown of CCP1 in maize plants compromises SA biosynthesis and decreases ZmPR1 and ZmPR5 expressions.A) Appearance of maize plants silenced for CCP1 gene expression at 7 dpi.B) The contents of SA in the CMV-CCP1-inoculated or the CMV-GUS-inoculated maize plants at 7 d post challenge inoculation with SCMV or Mock.C) and D) Relative expression of ZmPR1 (C) and ZmPR5 (D) in the CMV-CCP1-inoculated or the CMV-GUS-inoculated maize plants.Three independent experiments were conducted with at least three biological replicates per treatment.Error bars represented the means ±SE.Significant differences between CMV-GUS and CMV-CCP1 infected plants are indicated (*, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001) were evaluated by Unpaired Student's t test analysis.https://doi.org/10.1371/journal.ppat.1012086.g003 Fig), whilst in the latter phase of expression (from 48 to 72 hours post-infiltration), colocalization within the lytic vacuole can be observed in ~30% N. benthamiana cells (S16B Fig).

Fig 5 .
Fig 5.NIa-Pro inhibits CCP1 activity in vitro and vivo.A) Relative proteolytic activity of CCP1 was measured by digestion of a fluorescent casein substrate after pre-treatment with 5 μM E-64, 0.5 μM recombinant GST-NIa-Pro or GST (as negative control).In this assay, mutant mCCP1 C161A, H303A, N330A with three alanine substitutions at its catalytic triad was used as an inactive control.Fluorescence at 485/530 nm (excitation/emission) was measured.The results are presented as means ± standard deviation (n = 3).Statistical differences between treatments were determined using the unpaired t-test.****, P < 0.0001.B) ABPP was conducted to estimate the inhibitory effect of NIa-Pro on CCP1 protease activity.preCCP1-3Flag was transiently expressed in N. benthamiana leaves.Then the Flag-enriched proteins were incubated with 5 μM E-64 (as a positive control) or 1.0 μM purified GST-NIa-Pro followed by immunoblotting assay using a streptavidin-conjugated with horseradish peroxidase (HRP).ImageJ software was used to estimate the detection signal intensity.C) Expressing NIa-Pro via FoMV vector exhibited reduced PLCPs activity in maize.Active proteases were enriched using streptavidin beads and detected using streptavidin-HRP conjugates.Numbers indicate the accumulation levels of CCP1 and SCMV CP normalized to actin control.D) The relative accumulations of FoMV-overexpressing proteins (left), and active PLCPs (right) were quantified by software ImageJ, and the mean of FoMV-3Flag-GFP was set to 1.0.Error bars represent the means ± SE.Statistical differences (*, P < 0.05; **, P < 0.01) were evaluated by two-tailed Student's t test analysis.Three independent experiments are conducted.https://doi.org/10.1371/journal.ppat.1012086.g005

Fig 6 .
Fig 6.Identification of the key amino acids of NIa-Pro for interaction with CCP1.A) Schematic representation of NIa-Pro in SCMV genome.SCMV protein: P1: the first protein; HC-Pro: Helper Component-proteinase; P3: the third protein; 6K1: the first 6K protein; CI: cytoplasmic inclusion protein, 6K2: the second 6K protein, VPg: viral genome-linked protein, NIa-Pro: nuclear inclusion protein a-proteinase, NIb: nuclear inclusion protein b, CP: coat protein, P3N-PIPO: the N-terminal half of P3 fused to Pretty Interesting Potyvirus Open Reading Frame.B)The predicted 3D structure of SCMV NIa-Pro shows a protease domain (Cartoon shape) and an inhibitor motif (the blue line indicated).C) LCI assay of the interaction between mCCP1 and pdNIa-Pro or imNIa-Pro in N. benthamiana leaves.The Agrobacterium strains carrying the indicated constructs were infiltrated into N. benthamiana leaves.D) LCI assay for identification of the amino acid within the imNIa-Pro essential for interaction with mCCP1.The Agrobacterium strains carrying the indicated constructs were infiltrated into N. benthamiana leaves.NLuc co-expressed with CLuc or CLuc-NIa-Pro-, and mCCP1-NLuc co-expressed with CLuc, were used as negative controls.The luciferase activity was measured using a luminometer at 48 hpai.E) ABPP analysis to estimate the inhibitory effect of different form NIa-Pro on CCP1 protease activity.preCCP1-3Flag or preCCP1 C161A,H303A,N330A (preCCP1 m )-3Flag was transiently expressed in N. benthamiana leaves.Then the Flag-enriched proteins were incubated with 5 μM E-64 (as a positive control) or 1.0 μM purified GST-NIa-Pro, GST-NIa-Pro K230A, D234A (NIa-Pro m ), GST-imNIa-Pro and GST (as a negative control), followed by immunoblotting assay using a streptavidinconjugated with horseradish peroxidase (HRP).PreCCP1 m -3Flag was used as an inactive control.ImageJ software was used to estimate the detection signal intensity.

Fig 9 .
Fig 9. Proposed model of NIa-Pro targeting Corn Cysteine Protease 1 (CCP1) for efficient infection of SCMV in maize.CCP1 is a maize PLCP which confers resistance to SCMV infection via activation of SA signaling pathway.NIa-Pro is the main viral protease of SCMV.NIa-Pro can interact with CCP1 and inhibit its activity to counteract CCP1-mediated resistance and facilitate SCMV infection in maize plants.https://doi.org/10.1371/journal.ppat.1012086.g009