Pectate Lyase Genes Abundantly Expressed During the Infection Regulate Morphological Development of Colletotrichum camelliae and CcPEL16 Is Required for Full Virulence to Tea Plants

ABSTRACT Colletotrichum camelliae is the dominant species causing foliar diseases of tea plants (Camellia sinensis) in China. Transcriptome data and reverse transcription-quantitative PCR (qRT-PCR) analysis have demonstrated that the pectate lyase genes in C. camelliae (CcPELs) were significantly upregulated during infectious development on tea plants (cv. Longjing43). To further evaluate the biological functions of CcPELs, we established a polyethylene glycol (PEG)-mediated protoplast transformation system of C. camelliae and generated targeted deletion mutants of seven CcPELs. Phenotypic assays showed that the genes contribute to mycelial growth, conidiation, and appressorium development. The polypeptides encoded by each CcPEL gene contained a predicted N-terminal signal peptide, and a yeast invertase secretion assay suggested that each CcPEL protein could be secreted. Cell death-suppressive activity assays confirmed that all seven CcPELs did not suppress Bax-induced cell death in tobacco leaf cells. However, deletion of CcPEL16 significantly reduced necrotic lesions on tea leaves. Taken together, these results indicated that CcPELs play essential roles in regulating morphological development, and CcPEL16 is required for full virulence in C. camelliae. IMPORTANCE In this study, we first established a PEG-mediated protoplast transformation system of C. camelliae and used it to investigate the biological functions of seven pectate lyase genes (CcPELs) which were abundantly expressed during infection. The results provided insights into the contributions of pectate lyase to mycelial growth, conidial production, appressorium formation, and the pathogenicity of C. camelliae. We also confirmed the secretory function of CcPEL proteins and their role in suppressing Bax-induced cell death. Overall, this study provides an effective method for generating gene-deletion transformants in C. camelliae and broadens our understanding of pectate lyase in regulating morphological development and pathogenicity.

during the infection (3). As a result, we performed reverse transcription-quantitative PCR (qRT-PCR) assays to further confirm their expression patterns. Total RNA was extracted from tea leaves at 3,6,24,36,48,72, and 96 h after inoculation with a highly virulent isolate of C. camelliae (LS_19) (27). qRT-PCR analyses showed that expression of CcPELs, including CcPEL2, CcPEL6, CcPEL9, CcPEL16, CcPEL25, CcPEL26, and CcPEL33, was induced and significantly upregulated during infection (Fig. 1). The expression patterns of CcPEL2, CcPEL9, CcPEL16 and CcPEL33 were comparable, showing considerable transcript accumulation with peaks at 48 or 72 hpi (hours postinoculation). In addition, with the development of infection, the general expression levels of the four genes gradually increased (Fig. 1). In contrast, CcPEL6 and CcPEL25 showed high expression at the early stage of infection (3 or 6 hpi), and expression levels significantly decreased in the later stage. The expression pattern of CcPEL26 was different from that of the other six genes, exhibiting a slight transcript accumulation at different stages of infection (Fig. 1). Notably, compared with those of the other genes, the overall expression levels of CcPEL25 and CcPEL26 were lower at each stage of infection. These results suggest Functional Characterization of CcPELs mSphere that the seven CcPELs probably play important roles in the process of C. camelliae infecting tea plants.
Establishment of PEG-mediated protoplast transformation system of C. camelliae. To investigate the biological functions of CcPELs in C. camelliae, it is necessary to generate targeted gene deletion mutants of the CcPELs. The PEG-mediated protoplast transformation system has been widely used for genetic transformation in plant-pathogenic fungi (28)(29)(30). Thus, we attempted to establish a PEG-mediated protoplast transformation system for generating deletion mutants in C. camelliae. The C. camelliae strain LS_19 was used as the wild-type strain. Fifty mg/mL hygromycin B (HygB) or 100 mg/mL G418 completely inhibited mycelial growth of LS_19, suggesting its extreme sensitivity to hygromycin B and G418 ( Fig. 2A and B). Therefore, 50 mg/mL HygB or 100 mg/mL G418 was used for transformant selection. Protoplast yield was related to the incubation time in the mixed lytic enzyme generation, which was low, 0.41 6 0.01 Â 10 7 protoplasts/mL at 2 hpi, and increased with the extension of incubation time. At 7 hpi, the yield was the highest, up to 3.66 6 0.06 Â 10 7 protoplasts/mL (Fig. 2C). To verify the efficiency of the PEG-mediated protoplast transformation system, a vector carrying green fluorescent protein (GFP) tag was transformed into a protoplast of the LS_19 strain. The resulting transformants were confirmed by fluorescence microscopy: GFP fluorescence Functional Characterization of CcPELs mSphere could be observed in the hyphal cells of nearly 50% transformants (Fig. 2D). The results proved that the high-efficiency PEG-mediated protoplast transformation system can be used for genetic transformation and generating transformants lacking the target genes.
Targeted gene geletion of CcPELs in C. camelliae. We generated targeted gene deletion mutants of seven CcPELs via a homologous recombination strategy with the PEG-mediated protoplast transformation system. The entire open reading frame of each CcPEL in the wild-type strain LS_19 was replaced, respectively, by a HygB resistance cassette. The gene replacement events in the null mutants were confirmed by PCR analysis using two primer pairs ( Fig. S1; Table S1).
Pectate lyase contributes to appressorium development. To investigate the role of CcPELs in appressorium development, conidia of each strain were allowed to germinate on hydrophobic coverslips. By 24 hpi, about 81% of conidia produced by the wild-type strain LS_19 germinated and formed melanized appressoria with various shapes at the tips of germ tubes. However, targeted deletion of CcPEL6, CcPEL16, and CcPEL26 resulted in decreased appressorium formation. For the DCcPEL6-50 and DCcPEL6-56 mutants, only about 49% and 54% of conidia, respectively produced appressoria on hydrophobic surface. About 40% and 34% of conidia from the DCcPEL16-1 and DCcPEL16-2 mutants, respectively, could form appressoria. Additionally, the rate of appressorium formation in the DCcPEL26-5 and DCcPEL26-6 mutants significantly declined, being only around 21% and 29%, respectively ( Fig. 5A and B). These results suggest that CcPEL6, CcPEL16, and CcPEL26 are crucial for appressorium formation. In addition, the appressoria produced by the DCcPEL6 and DCcPEL25 mutants were more melanized compared to those produced by LS_19 (Fig. 5A). Conidia of the DCcPEL25-18 mutant could produce two appressoria at one germ tube, and conidia from the DCcPEL33-12 mutant germinated from cells at both ends and formed appressoria at the tips of two germ tubes (Fig. 5A). These results demonstrated that pectate lyase contributes to appressorium development in C. camelliae.
Validation of the secretory function of CcPEL proteins. SignalP version 5.0 (https://services.healthtech.dtu.dk/service.php?SignalP-5.0) predicted that all seven CcPELs possess a putative signal peptide (SP) at the N terminus (Fig. 6A). We performed yeast invertase secretion assays to validate the secretory function of the putative SPs via the yeast strain YTK12 and pSUC2 vector. Strain YTK12 is invertase-deficient, and pSUC2 vector contains an invertase gene but lacks methionine and a SP sequence (32). The predicted sequences of each SP were cloned into the vector pSUC2, and then transformed into cells of the yeast YTK12 mutant strain. Compared with the positive control carrying pSUC2-Avr1bSP vector, the predicted SP of CcPELs mediated the complementation of YTK12 strain growing on YPRAA medium (Fig. 6B). In addition, the invertase enzymatic activity could be further detected by the reduction of 2,3,5-triphenyltetrazolium chloride (TTC) to insoluble red-colored 1,3,5-triphenylformazan (TPF). Culture filtrates from each YTK12 strain carrying pSUC2-CcPELs exhibited a red color with TTC treatment. In contrast, the YTK12 strain carrying pSUC2 empty vector, as the negative control, was not able to grow on YPRAA medium, and its culture filtrates remained colorless after TTC treatment (Fig. 6B). These results suggested that the CcPEL proteins carry a functional secretory SP. Seven CcPELs cannot suppress Bax-induced cell death in Nicotiana tabacum. Bax, as a pro-apoptotic member of the B cell lymphoma/leukemia (Bcl-2) family, triggers programmed cell death (PCD) (33). When fungal effectors are secreted into plant cells, they may exert its death-suppressive function. Therefore, we assayed the ability of CcPELs to suppress Bax-induced PCD, which resembles the defense-related hypersensitive reaction (HR) in plant cells (34). In the tobacco transient expression assay, we overexpressed these seven CcPELs lacking their N-terminal SPs in Nicotiana tabacum leaves 24 h before infiltrating Bax into the agroinfiltration sites. Unexpectedly, the HR induced by infiltrating Bax into the each CcPEL transient overexpression area was not suppressed (Fig. 7A), even though Western blot analysis of agroinfiltrated leaves showed that these seven CcPELs were actually expressed (Fig. 7B). The results indicate that the seven CcPELs did not suppress Bax-induced cell death in N. tabacum.
CcPEL16 plays a role in the pathogenicity of C. camelliae. Although seven CcPEL proteins may not function as an attenuator of HR response in tobacco, the role of these CcPELs in virulence of C. camelliae toward the host Ca. sinensis needs to be further analyzed. We performed a pathogenicity analysis of CcPEL deletion mutants by inoculating Longjing43 abraded leaves with conidial suspensions. By 4 dpi (days postinoculation), large necrotic lesions were observed at the inoculation sites of leaves inoculated with the wild-type strain LS_19, while much smaller lesions on inoculated leaves were seen for the DCcPEL16-1 and DCcPEL16-2 mutants (Fig. 8). However, there were no significant differences in lesion diameter between the other mutants and the LS_19 strain, indicating that the deletion of these genes did not affect the pathogenicity of C. camelliae. The results suggested that CcPEL16 plays a role in the pathogenicity of C. camelliae on tea leaves.

DISCUSSION
C. camelliae is one of the dominant pathogens causing foliar diseases in tea plants and can always be isolated from the diseased leaves (5,35). In our previous studies, we have revealed the lifestyle characteristics and gene expression patterns of C. camelliae based on transcriptome data (3). As a hemibiotrophic pathogen, C. camelliae initially develops biotrophic hyphae inside host plant cells and then transitions to the necrotrophic stage (36). During infection, plant-colonizing fungi secrete effectors to directly dampen the plant immune system or redirect host processes, facilitating fungal growth (37). We have predicted many secreted proteins as potential candidate effectors that are significantly upregulated during C. camelliae infection of Ca. sinensis, including PCWDEs such as the carbohydrate esterases and PELs (3). These can induce the host innate immune defenses during infection, such as callose deposits and PCD (38). Although PELs have been shown to be involved in enhancing pathogenicity and Based on transcriptome data of C. camelliae LS_19, seven genes encoding PELs were identified in C. camelliae. qRT-PCR analyses further confirmed that CcPEL expression is induced and significantly upregulated during infection (Fig. 1), which suggested that the seven CcPELs probably play important roles in the process of C. camelliae infection of tea plants. Gene deletion mutants of the CcPEL had to be generated to further analyze their molecular biological functions. Thus, we needed to establish an efficient genetic transformation system for C. camelliae. For filamentous fungi, two common methods used for genetic transformation are Agrobacterium tumefaciens-mediated transformation (ATMT) and PEG-mediated protoplast transformation (39). Compared with the complicated ATMT system, the PEG-mediated protoplast transformation system is particularly simple and efficient (40). It requires only a moderate amount of soluble wall-collapse enzymes, which has the advantages of simple operation, system maturation, and low instrument Functional Characterization of CcPELs mSphere requirements. Therefore, it has been maturely applied in a variety of filamentous fungal pathogens, such as M. oryzae, Fusarium graminearum, and Colletotrichum spp. (41). In this study, we explored and optimized the PEG-mediated protoplast transformation system for C. camelliae. The optimum time for enzymatic hydrolysis was 7 h, with the young mycelia germinated from conidia under the proper osmotic buffer (Fig. 2C). The successful establishment of a PEG-mediated protoplast transformation system provides favorable conditions for the illumination of CcPEL biological functions in C. camelliae. We generated targeted gene deletion mutants of seven CcPELs, including CcPEL2, CcPEL6, CcPEL9, CcPEL16, CcPEL25, CcPEL26, and CcPEL33, respectively, via a homologous recombination strategy. Phenotypic analysis showed that all CcPELs except for CcPEL33 are involved in vegetative growth (Fig. 3); CcPEL2, CcPEL9, CcPEL25, and CcPEL26 are required for conidia production (Fig. 4); CcPEL6, CcPEL25, and CcPEL26 are involved in regulating conidial morphology (Fig. 4C); and CcPEL6, CcPEL16, and CcPEL26 are crucial for appressorium formation (Fig. 5). The results indicated that pectate lyases play important roles in the morphogenesis of C. camelliae. Surprisingly, few studies have reported the roles of PELs in the morphological development of pathogenic fungi. In C. gloeosporioides, the in vitro growth rate of pelB mutants showed no significant difference on glucose or Na polypectate media (22). In Alternaria brassicicola, colony sizes were similar for the Dpl1332-1 and wild-type strains on nutrient-rich PDA and minimal mineral agar supplemented with pectin, indicating that the pectate lyase-coding gene PL1332 was dispensable for vegetative growth (42). In C. magna, two transformed isolates of pel were generated and selected for PEL activity and pathogenicity analyses, but lack of phenotypic analysis (43). However, the roles of pelB in C. gloeosporioides, PL1332 in A. brassicicola, and Cmpel in C. magna in conidia production or appressorium development have not been studied. In this study, we first revealed the critical roles of PEL genes in vegetative growth, conidiogenesis, and appressorium development in C. camelliae. All seven CcPEL proteins in C. camelliae could be secreted into the plant cells (Fig. 6), suggesting their potential roles as effectors. A pathogenicity test on abraded Ca. sinensis leaves demonstrated that CcPEL16 contributes to the virulence of C. camelliae (Fig. 8). Although PELs have been extensively reported to be involved in pathogenesis in phytopathogenic fungi (19,(21)(22)(23)(24)(25)(26), the role of PEL16 in pathogenicity is rarely reported. In P. capsici, the pectate lyase gene PcPL16 showed strong expression levels during infection, and overexpressing PcPL16 triggered strong cell death in pepper and tobacco leaves (25). The silenced line 16-S11 (PcPL16) produced significantly smaller lesions on pepper leaves compared with the aggressive isolate SD33, suggesting that PcPL16 makes an important contribution to virulence (25). In this study, we first demonstrated that CcPEL16 plays a critical role in the virulence of C. camelliae against tea plants.
PELs can manipulate host immune responses as the pathogen-associated molecular patterns (PAMPs) or damage-associated molecular patterns (DAMPs) (44). For example, VdPEL1 in V. dahlia was secreted into plant cells to degrade pectin in the roots, resulting in the release of the plant cell wall fragments (DAMPs) and triggering defense responses (19). Here, cell death-suppressive activity assays showed that CcPEL16 did not suppress Bax-induced cell death in N. tabacum leaves (Fig. 7), indicating that it did not trigger the plant defense responses in tobacco. However, many studies have shown that functional effectors contributing to pathogen virulence strongly induce plant cell death in tobacco leaves, such as F. graminearum Fg12, Phytophthora sojae PsCRN63, and P. sojae Avh241 (45)(46)(47). Therefore, we further performed pathogenicity analyses in host tea leaves to determine the role of CcPELs in C. camelliae virulence. When susceptible tea leaves were inoculated with conidial suspensions, the DCcpel16 mutants caused significantly reduced necrotic lesions compared to the wild-type strain LS_19 (Fig. 8). The result indicates that CcPEL16 as a virulence factor is essential for the pathogenicity of C. camelliae against tea plants. The attenuated virulence of the DCcPEL16 mutants may be due to the significant decrease in appressoria as the infection structure. Considering the induced high expression during infection (Fig. 1) and Functional Characterization of CcPELs mSphere the presence of a functional secretory SP (Fig. 6), we speculated that CcPEL16 in C. camelliae may act as an effector or PAMP to elicit immune defenses in its own host Ca. sinensis. In subsequent studies, we need to confirm whether tea plants can recognize PEL degradation products which do or do not act as DAMPs. In addition, the PAMP activity of CcPEL16 protein needs to be further confirmed in the future.
Conclusions. In this study, we found that seven CcPEL genes in C. camelliae were significantly upregulated expressed during infection. To further confirm their biological functions, we generated targeted gene deletion mutants of seven CcPELs via a homologous recombination strategy with an efficient PEG-mediated protoplast transformation system in C. camelliae. Phenotypic analysis revealed that pectate lyase genes are involved in vegetative growth, conidiogenesis, and appressorium development in C. camelliae. All seven CcPEL proteins contained SPs at the N terminus and could be secreted. Furthermore, deletion of CcPEL16 impaired the virulence of C. camelliae against Ca. sinensis. However, the roles of CcPEL16 protein as a candidate effector needs to be further confirmed in the future.

MATERIALS AND METHODS
Fungal strains and culture conditions. The wild-type C. camelliae strain LS_19 was previously isolated from tea plants in China (27). Strain LS_19 and the CcPEL mutants were cultured on PDA (20% potato, 2% D-glucose, and 1.5% agar) medium at 28°C in the dark.
RNA isolation and quantitative RT-PCR analyses. Total RNA was extracted from tea leaves at 3, 6, 24, 36, 48, 72, and 96 h after inoculation with LS_19 using FastPure Universal Plant Total RNA Isolation kit (Nanjing Vazyme Biotech Co., Ltd., China), and then the first strand of cDNA was synthesized with reverse transcription by HiScript II Q RT SuperMix for qPCR (1gDNA wiper) (Nanjing Vazyme Biotech Co., Ltd., China). qRT-PCR analysis was performed with ChamQ SYBR qPCR Master Mix (Nanjing Vazyme Biotech Co., Ltd., China) using the Bio-Rad CFX96 Real-Time System. The primers used to amplify the seven CcPELs and CcNEW1 used for qRT-PCR assays are listed in Table S1. The relative quantification of each transcript was calculated as previously described (28,48). All qRT-PCR analyses were conducted in three replicates for each sample and three biological replicates were maintained.
Generation of CcPEL deletion mutants. All CcPEL sequences were obtained from the genome data (GenBank: JAELVL010000001.1) of C. camelliae. The primer pairs used to construct the CcPEL gene replacement vectors, the HPH gene cassette, and the upstream and downstream flanking fragments are listed in Table S1. The fused fragments of each CcPEL gene were cloned into the vector pKOV21 using the CloneExpress Multis reaction system (Nanjing Vazyme Biotech Co., Ltd., China) (49). The gene replacement vectors were then transformed into protoplasts of the wild-type strain LS_19 to generate null mutants. All gene deletion mutants were confirmed by PCR assays.
Phenotypic analysis. For assessment of the colony growth of each strain, 5-mm diameter mycelial plugs cut from the edge of a 5-day-old colony were placed on PDA plates and incubated at 28°C for 4 days. Conidiation assays of all strains were performed after culturing for 7 days on PDA medium. The conidia produced on PDA plates were scraped off and then filtrated. Next, conidia concentrations were measured using a hemocytometer. The conidial size of each strain was determined using ImageView software under light microscopy. At least 50 conidia were measured. For appressorium development analysis, conidial suspensions with a concentration of 1 Â 10 5 conidia/mL were cultured on hydrophobic coverslips in the dark. The appressoria of each strain were examined under a light microscope at 24 hpi. All experiments were repeated three times with at least three replicates.
SP activity assays. The predicted SPs of CcPEL were fused with the vector pSUC2. Constructed recombinant vectors were transformed into the yeast strain YTK12 as described previously (32). All transformants were cultured on CMD-W medium (6.7 g/L yeast nitrogen base without amino acids, 0.75 g/L tryptophan dropout supplement, 20 g/L agar, sterilization at 121°C for 20 min, addition of 2% sucrose, 0.1% D-glucose) to select positive colonies (8). To detect invertase secretion, the positive transformants were cultured on YPRAA medium (10 g/L yeast extract, 20 g/L peptone, 20 g/L raffinose, 20 g/L agar, sterilization at 121°C for 20 min, addition of 20 mg/mL antimycin A). For the color reaction, the positive transformants were suspended with the suspension buffer (1.5 mM acetic acid-sodium buffer [pH 4.7], 3.5% sucrose solution). After centrifugation of the suspension, the supernatant was added to 0.1% TTC to detect the color and then photographed.
Cell-death inhibition assay. Each CcPEL cDNA sequence (without SP sequence) was cloned into the vector pGR107-GFP. Each constructed vector, respectively, was transformed into A. tumefaciens GV3101. For transient overexpression of CcPELs in N. tabacum, recombinant strains of A. tumefaciens were cultured in LB liquid medium with 50 mg/mL kanamycin and 50 mg/mL rifampicin at 28°C for 24 h. Cells were resuspended and adjusted to an optical density at 600 nm (OD 600 ) of 0.5, and then injected into the leaves of 5-week old N. tabacum plants. A. tumefaciens cells harboring the BAX gene were inoculated into the same sites on the tobacco leaves at 24 hpi. The same leaves were injected with A. tumefaciens carrying an empty vector pGR107-GFP as a control. Cell death symptoms were monitored at 7 dpi and photographed. All tests were performed with three replicates.
Western blot analysis. Total proteins were extracted from the leaf infiltrated area as previously described (47). Protein samples were fractioned by 10% SDS-PAGE gel and electroblotted onto Functional Characterization of CcPELs mSphere polyvinylidene difluoride (PVDF) membrane, then immunoblotted with anti-GFP antibody (Abmart, China) at a dilution of 1:5,000. HiSec horseradish peroxidase-conjugated goat anti-mouse IgG (H1L) (Nanjing Vazyme Biotech Co., Ltd., China) was used as the secondary antibody. The chemiluminescence signals were detected with a FDbio-Femto RCL kit (FDbio Science, China). Pathogenicity assays. Healthy and non-wounded mature leaves collected from a 5-year-old Ca. sinensis cv. Longjing 43 grown in a tea garden in Hangzhou, Zhejiang Province, China, were washed with sterilized water and then inoculated using wound inoculation methods (35,50). Conidial suspensions of the wild-type strain LS_19 at a concentration of 1 Â 10 5 conidia/mL were inoculated on the left sides of leaves, and conidial suspensions of the CcPEL mutants at the same concentration were inoculated on the right sides. Symptomatic lesions and lesion diameters were determined at 4 dpi. Experiments were repeated three times with three replicates for each strain.
Light and fluorescence microscopy. Light microscopy was used to visualize protoplasts, conidia, and appressoria of C. camelliae. Fluorescence microscopy was used to visualize GFP. The exciation and emission wavelengths used were 488 nm and 507 nm.

SUPPLEMENTAL MATERIAL
Supplemental material is available online only.