VdPLP, A Patatin-Like Phospholipase in Verticillium dahliae, Is Involved in Cell Wall Integrity and Required for Pathogenicity

The soil-borne ascomycete fungus Verticillium dahliae causes vascular wilt disease and can seriously diminish the yield and quality of important crops. Functional analysis of growth- and pathogenicity-related genes is essential for revealing the pathogenic molecular mechanism of V. dahliae. Phospholipase is an important virulence factor in fungi that hydrolyzes phospholipids into fatty acid and other lipophilic substances and is involved in hyphal development. Thus far, only a few V. dahliae phospholipases have been identified, and their involvement in V. dahliae development and pathogenicity remains unknown. In this study, the function of the patatin-like phospholipase gene in V. dahliae (VdPLP, VDAG_00942) is characterized by generating gene knockout and complementary mutants. Vegetative growth and conidiation of VdPLP deletion mutants (ΔVdPLP) were significantly reduced compared with wild type and complementary strains, but more microsclerotia formed. The ΔVdPLP mutants were very sensitive to the cell-wall-perturbing agents: calcofluor white (CFW) and Congo red (CR). The transcriptional level of genes related to the cell wall integrity (CWI) pathway and chitin synthesis were downregulated, suggesting that VdPLP has a pivotal role in the CWI pathway and chitin synthesis in V. dahliae. ΔVdPLP strains were distinctly impaired in in their virulence and ability to colonize Nicotiana benthamiana roots. Our results demonstrate that VdPLP regulates hyphal growth and conidial production and is involved in stabilizing the cell wall, thus mediating the pathogenicity of V. dahliae.


Introduction
Verticillium dahliae, a soil-borne, filamentous fungal plant pathogen with wide distribution around the world, can attack many important crops such as cotton, soybean, potato, sugar beet and sunflower, causing significant crop losses annually [1,2]. It is a significant threat to cash-crop production. It also produces microsclerotia, dormant structures that can survive for many years in the soil and be induced to germinate by plant root exudates [3,4]. Hyphae colonize the plant root surface, and then form a narrow penetration peg that can penetrate the epidermal cells [5]. Hyphae grow intracellularly and intercellularly through the root cortex, enter the xylem vessels and colonize the xylem of the hypocotyl and leaves. Ultimately, water flow in the plant tissue is disrupted, causing the wilt symptoms, stunting, chlorosis and necrosis [4,6]. The molecular mechanisms underlying V. dahliae pathogenesis are complex and still not clear [4][5][6][7][8], so functional analyses of growth-and pathogenicity-related genes are crucial in developing new strategies to control V. dahliae. Meanwhile, PLPs have been shown to exhibit a PLA 2 -like activity [36,37]. PLPs show difference in substrate specificity. PLPs have broader substrate preference, while cytosolic PLA 2 s show increased substrate specificity for arachidonic acid-containing lipids [36][37][38]. PLPs also do not contain a lid-like structure and are limited to interfacial activation [38]. Phospholipases have been studied in some plant pathogenic fungi, such as Fusarium graminearum [14] and M. oryzae [25,26,28]. The phospholipase C (FgPLC1) is considered to be closely related to regulation of development, stress responses and pathogenicity of F. graminearum [14]. cAMP phosphodiesterase PdeH regulates the intracellular cAMP levels, CWI and pathogenicity of M. oryzae [25,26,28], while the function of the patatin-like phospholipase (encoded by VdPLP, VDAG_00942) remains unclear in V. dahliae. The relations among VdPLP, cell wall integrity and fungal virulence are unknown. In this study, to investigate the function of VdPLP, we generated a deletion mutant (∆VdPLP) of V. dahliae and characterized its growth and development, the stress tolerance and integrity of the cell wall, root penetration ability, and virulence in Nicotiana benthamiana.

Fungal Strains, Plant Material and Culture Conditions
V. dahliae strain 991 (V991, wild type (WT)) was kindly provided by Prof. Guiliang Jian from the Institute of Plant Protection, Chinese Academy of Agricultural Sciences (IPP, CAAS). N. benthamiana seedlings were planted in a growth chamber with 16 h of light/8 h of dark and 60-70% relative humidity at 23-25 • C.
For generating VdPLP complementary plasmid pCM-Hyg-VdPLP, the full-length complementary DNA (cDNA) fragment of VdPLP, the TrpC promoter and Nos terminator were amplified from plasmid pCH-GFP and primers C-VdPLP-F/C-VdPLP-R, C-TrpC-F/C-TrpC-R and C-Nos-F/C-Nos-R (Table 1). Plasmid pCM-Hyg carrying the hygromycin B resistance gene (hph) was linearized with restriction enzymes KpnI and XbaI, and the fragments were inserted into plasmid pCM-Hyg digested with In Fusion enzyme.

Confirmation of VdPLP Disruption, Complementation of ∆VdPLP Strains and Screening for GFP-Tagged Strains
VdPLP knockout mutants were cultivated and selected on potato dextrose agar (PDA) plates supplied with geneticin (50 µg/mL). ∆VdPLP-C and GFP-tagged strains were selected in the presence of hygromycin B. After single-spore isolation, all isolates were grown in complete medium (CM, 6 g/L yeast extract, 6 g/L casein acid hydrolysate, 10 g/L sucrose) for DNA extraction. All isolates were further confirmed by genomic PCR with the specific primers. VdPLP was analyzed by genomic PCR with primer pair VdPLP-J-F/VdPLP-J-R (Table 1) for successful homologous recombination and with neo-J-F/neo-J-R for detecting the neo gene. ∆VdPLP-C mutants were selected by genomic PCR with primer pairs VdPLP-J-F/VdPLP-J-R and hyg-F/hyg-R (Table 1) for the hph gene. GFP-labeled strains were examined for the hph gene with primer pair hyg-F/hyg-R and for fluorescence with a confocal laser scanning microscope (LSCM; LSM 700, Zeiss, Jena, Germany) using 488 nm excitation wavelength and band-pass 500 to 550 nm emission filters. Table 1. List of primers used in this study.

Primer Name
Primer Sequence

Phenotype Assays Using Cell Wall-Disrupting Agents Calcoflour White and Congo Red
Susceptibility to cell wall-perturbing agents CFW and CR is generally tested to detect cell wall mutants of mycelial fungi [28,43]. A conidial suspension of each strain (10 µL of 5 × 10 6 /mL) was placed on the center of separate PDA agar plates containing 50 mg/mL CFW or 100 mg/mL CR, or without CR or CFW. Colony phenotypes were observed seven days after inoculation. The experiment was performed three times independently, using 10 plates per strain per treatment.

Quantitative Real-Time PCR Reaction
Cell wall reinforcement and repair in response to cell wall stress is activated by the CWI pathway, a basal cascade pathway triggered by transmembrane sensors in response to cell wall stress [44][45][46]. To examine the potential role of VdPLP in the CWI pathway and chitin synthesis, we analyzed the transcription levels of genes encoding mitogen-activated protein kinase kinase (VdMKK1, VDAG_09823) and mitogen-activated protein kinase (VdMK1, VDAG_09461) for the CWI pathway and chitin synthesis genes VdChi1 (VDAG_10179), VdChi3 (VDAG_08591), VdChi4 (VDAG_00419), VdChi7 (VDAG_01790) in the Vd-wt, ∆VdPLP and ∆VdPLP-C strains, which were grown in liquid CM for 5 days. Total RNA was extracted from each strain using the RNA miniprep kit (Axygen, Union City, CA, USA) according to the user's guide. cDNAs were synthesized with the TOYOBO RT Kit (TOYOBO, Osaka, Japan). The quantitative real-time PCR (qRT-PCR) reactions were performed with the SYBR Fast qPCR kit (KAPA Biosystems, Boston, MA, USA). Specific primers for different genes were designed ( Table 1). The constitutively expressed β-tubulin gene (DQ266153) was used for normalization and amplified with primers VdBt-F/VdBt-R ( Table 1). The qRT-PCR reactions were performed in an ABI7500 Fast PCR thermocycler (Applied Biosystems, Foster City, CA, USA). The experiment was repeated two times.

Penetration Ability Assay
Penetration was assayed as described previously [47,48]. About 20 µL of 2 × 10 6 /mL conidial suspension of each strain were placed on cellophane laid on PDA plates, then incubated at 25 • C. After 3 days, the cellophane was removed from the plates, which were then incubated for 7 more days at 25 • C, and hyphae were observed. This assay was done three times, with five plates per strain each time.
For microscopically observing initial infection, roots of N. benthamiana seedlings with 6-7 true leaves were inoculated with GFP-labeled Vd-GFP or ∆VdPLP-GFP strains [8]. After 3 days, the roots were vertically cut and observed on temporary slides with LSCM.

Pathogenicity Assays, Microscopic Observation of Initial Infection and Fungal Biomass Quantification
For pathogenicity assay experiments, roots of 10 N. benthamiana plants were inoculated with conidia of strain Vd-wt, ∆VdPLP or ∆VdPLP-C as previously described [39]. The disease symptoms were recorded at 8, 10, and 12 days post inoculation (dpi), and the mean of the disease grade was calculated as an indicator of the severity of the disease. Wilt symptoms were classified into five grades: 0, no disease symptoms; 1, wilting of fewer than two leaves; 2, wilting of three to five leaves; 3, wilting of more than five leaves; and 4, death or near death of plants [39,49].
Colonization ability of strains was assessed by quantifying the fungal biomass in N. benthamiana using qRT-PCR. DNA was extracted from infected plant stems using the DNAsecure Plant Kit (TIANGEN, Beijing, China). The actin gene of N. benthamiana was selected as an internal standard [8] and amplified with primer pair Nb-actin-F/Nb-actin-R (Table 1). ITS1 and ITS2 regions of the ribosomal RNA genes (Z29511) of V. dahliae were amplified with primer pair Vd-F/Vd-R (Table 1), which was used to quantify the fungal DNA in the mixed DNA samples. All qRT-PCR reactions were done as previously reported [39].

Statistical Analysis
Significant difference of data among the groups was analyzed using Duncan's multiple range test (p-value < 0.05) and SPSS Statistics 17.0 software (SPSS, Chicago, IL, USA).

Deletion and Complementation of VdPLP in V. dahliae
The VdPLP gene was replaced by the neo cassette in the Vd-wt strain. With the targeted gene replacement, geneticin resistance was generated. Transformations were confirmed by genomic PCR; homologous transformation events had occurred in 3 of 30 independent transformants of the Vd-wt strain. VdPLP deletion mutants (∆VdPLP-1 and ∆VdPLP-4) were randomly selected for further functional analysis ( Figure 1B,C). For complementation, the In-fusion-resulting plasmid was introduced into ∆VdPLP-1 and ∆VdPLP-4 strains by protoplast transformation (Figure 1D), resulting in complementary strains ∆VdPLP-1-C and ∆VdPLP-4-C.

Radial Growth of Mycelia was Significantly Reduced in ∆VdPLP Mutants
The effects of VdPLP disruption on hyphal growth were defined by growing Vd-wt and the mutants on MM agar with different carbon source (sucrose, pectin, starch, galactose or xylose) and comparing the phenotypes of each strain. Vegetative growth of ∆VdPLP mutants was severely retarded compared with that of Vd-wt, ∆VdPLP-1-C and ∆VdPLP-4-C strains, and colony diameter of the ∆VdPLP mutants was significantly smaller than that of Vd-wt, ∆VdPLP-1-C and ∆VdPLP-4-C strains ( Figure 2B). These results suggested that the VdPLP contributes to the hyphal growth of V. dahliae on different carbon sources. Values represent the mean ± standard deviation (SD) from three independent replicates, Different letters (a-c) above the bars represent the significant differences among the groups (p-value < 0.05) and the data were analyzed using Duncan's multiple range test.

Conidiation by ∆VdPLP Mutants Decreased Substantially and Microsclerotia Formation Increased
To check whether the deletion of VdPLP affects conidiation and microsclerotia formation, each strain was cultured on Czapek-Dox and BMM plates to enumerate conidia and microsclerotia after 15 days. Mutants ∆VdPLP-1 and ∆VdPLP-4 produced significantly fewer conidia on Czapek-Dox than did the Vd-wt and complementation strains ( Figure 3A). In addition, the ∆VdPLP mutant strains accumulated more melanin on BMM, while Vd-wt, ∆VdPLP-1-C and ∆VdPLP-4-C strains were light-colored and produced no melanin on BMM ( Figure 3B). On BMM, the ∆VdPLP strains formed microsclerotia earlier and in greater numbers than did the Vd-wt and the complementation strains ( Figure 3C,D). The results indicated that deletion of the VdPLP gene accelerated melanin and microsclerotia production in V. dahliae. . Three independent replicates were done; Different letters (a-b) above the bars represent the significant differences among the groups and were analyzed with Duncan's multiple range test (p-value < 0.05).

∆VdPLP Mutants Are Hypersensitive to Cell Wall-Perturbing Agents
To determine whether the VdPLP gene was involved in maintaining cell wall integrity, we tested cell wall-perturbing agents CFW and CR for their effects on growth of the ∆VdPLP mutant strains. On PDA plates without CR or CFW, there was no significant difference between Vd-wt and the complementary ∆VdPLP strains in the colony diameter, but the colony diameter of ∆VdPLP strains showed nearly 50% reduction. On PDA plates containing 100 g/L CR or 50 g/L CFW, the colony diameter of ∆VdPLP was significantly smaller than that of Vd-wt and the complementary ∆VdPLP strains. Vd-wt and the complementary ∆VdPLP strains has no obvious growth defects on PDA plates with 100 g/L CR or 50 g/L CFW compared with growth on PDA without CR or CFW ( Figure 4A,B). However, when 100 g/L CR or 50 g/L CFW was added to the medium, the colony diameter of ∆VdPLP strains was significantly smaller than that of the colony on PDA without CR. These results indicated that ∆VdPLP strains were more sensitive to CR and CFW than the wild type and complementary strains, and VdPLP was responsible for maintaining fungal cell wall integrity in V. dahliae. Three independent replicates were done; Different letters (a-c) above the bars represent the significant differences among the groups and were analyzed with the Duncan's multiple range test (p-value < 0.05).

Genes for the Calcofluor White Pathway and for Chitin Synthesis in the ∆VdPLP Mutant Were Downregulated
To further confirm that VdPLP contributes to the CWI pathway and chitin synthesis, we focused on the regulation of the CWI pathway by profiling transcript levels of the CWI pathway genes VdMKK1 and VdMK1 and chitin synthesis genes VdChi1, VdChi3, VdChi4, and VdChi7. In the ∆VdPLP mutant strains, transcription of VdMKK1, VdMK1, and the chitin synthesis genes all decreased compared with the wild type strain and complementary ∆VdPLP strains ( Figure 5A-F). Downregulation of CWI and chitin synthesis genes indicated that VdPLP participates in the CWI pathway and chitin synthesis.  letters (a,b), and data were analyzed with the Duncan's multiple range test (p-value < 0.05).

Penetration and Fungal Colonization of ∆VdPLP Were Impaired
To investigate possible reasons for the significant virulence defect of ∆VdPLP mutant strains, we analyzed penetration by the Vd-wt, ∆VdPLP and ∆VdPLP-C strains. After a conidial suspension of the respective strains was placed on cellophane membranes, Vd-wt and the ∆VdPLP-C strains penetrated and colonized the membranes, but no hyphae of the ∆VdPLP mutant strains grew ( Figure 6A,B). Thus, the deletion of VdPLP severely hindered the ability to penetrate the cellophane membrane.
GFP-labeled strains ∆VdPLP-4-GFP and Vd-GFP constitutively express GFP; strong green fluorescence thus allows observation of fungal penetration and colonization of plant root tissues using LSCM. Strain Vd-GFP strain colonized N. benthamiana roots, and hyphae penetrated the epidermal and cortical cells. In contrast, few hyphae of the ∆VdPLP-4-GFP strain grew on the root, and the strain penetrated epidermal cells ( Figure 6C). These phenomena indicate that VdPLP contributes to penetration and colonization by V. dahliae.

VdPLP was Required for Fungal Pathogenicity
To verify the role of VdPLP in fungal pathogenicity, we inoculated roots of N. benthamiana seedlings with a conidial suspension of Vd-wt, ∆VdPLP, or ∆VdPLP-C strains. As expected, at 8, 10, and 12 dpi, plants inoculated with either Vd-wt, or the complementary strains developed significant wilting, whereas only a few chlorotic leaves appeared on plants inoculated with ∆VdPLP strains. At 12 dpi, plants inoculated with Vd-wt or ∆VdPLP-C strains were stunted, chlorotic, wilting or dead. In contrast, the seedlings infected with ∆VdPLP mutants exhibited mild symptoms, with interveinal chlorosis on the bottom 1 or 2 leaves, but no necrosis ( Figure 7A,B). Furthermore, when biomass was quantified, less fungal DNA was detected from infected plants inoculated with the ∆VdPLP strains than with Vd-wt or the ∆VdPLP-C strains ( Figure 7C). The decrease in symptoms and biomass suggest that VdPLP disruption reduced the virulence of V. dahliae. Figure 6. Fungal colonization and penetration assay: (A) Vd-wt, ∆VdPLP and ∆VdPLP-C strains grown on cellophane for three days after strains had penetrated a cellophane membrane on the agar plate and membranes were removed; (B) plates after incubation at 25 • C for seven days after cellophane were removed; and (C) fungal colonization of infected roots by the ∆VdPLP-4-GFP mutant strains, Vd-GFP strain, and sterile water (Mock), three days after inoculation.

Discussion
Phospholipases also regulate vegetative growth signals in filamentous fungi [9,13,21,22,50]. The cAMP-specific PDEs can be regulated by some phospholipases, which adjust intracellular cAMP level. cAMP signaling participates in numerous intracellular activities [23,24]. For example, the constitutive activation of cAMP phosphodiesterase, which can activate the cAMP pathway, influences the biosynthesis and structure of the cell wall and membrane [51,52]. Fungal phospholipases affect blastospore development [18] and participate in hyphal extension in the filamentous fungi N. crassa [19], C. cinerea [20], and M. oryzae [25,28]. Due to inhibition of phosphodiesterase activity in V. dahliae, conidial production and virulence was reduced in G protein β subunit gene (VGB) mutants [53]. Similar phenotypic changes were observed in VdPLP deletion mutants. Hyphal growth of ∆VdPLP strains under different carbon sources conditions, and conidia production was drastically reduced compared with the wild type strain and complementation mutants, which suggested that impaired vegetative growth and conidial production caused by VdPLP deletion may be the reason for the virulence decline.
Microsclerotia, melanized survival structures that can survive in the soil for more than ten years in the absence of host plants, play a critical role in the disease cycle of V. dahliae, but the molecular mechanism of their biogenesis is still not clear [54][55][56][57]. VdPKAC1 mutants showed a decrease in fungal virulence and an increase in microsclerotial formation [58]. In contrast, microsclerotia production decreased by V. dahliae class II hydrophobin gene (VDH1) mutants [59], Vayg1 deletion mutants [60], transmembrane mucin Msb (VdMsb) mutants [61] and MAPK mutants [62]. In the present laboratory study, ∆VdPLP mutants produced microsclerotia earlier and in greater numbers than did Vd-wt and the ∆VdPLP-C strains. Based on the above research, opposite results for mutants in different genes suggest that the molecular mechanisms of microsclerotia formation are complex. Regulatory genes such as VdPKAC1, VDH1, Vayg1, VdMsb, and dihydroxynaphthalene (DHN) melanin biosynthesis pathway [57] may regulate the microsclerotia biosynthesis. The loss of VdPLP likely activates microsclerotia production via participating in a signaling pathway, but the specific mechanisms of the transition from vegetative growth to resting structure development require further research. are means ± SD from three independent replicates. Significant differences are indicated by letters (a-d), Duncan's multiple range test (p-value < 0.05) was used to analyze significant difference between each group.
The fungal cell wall plays a critical role in maintaining cell shape, protecting against physical injury, and mediating exchanges between pathogenic fungi and their hosts [62][63][64]. It also provides adhesive properties and protection against host defenses, which is imperative for pathogenicity in fungi [63,65,66]. PdeH are related to cell wall integrity, stress response and virulence in M. oryzae and C. albicans [25,28,[50][51][52]. MoPdeH targets MoMck1 of M. oryzae, one of the components of the MAPK cascade, and regulates cell wall integrity and virulence [28]. Integrity of the cell wall is critical for penetration of the host and thus for fungal virulence [67]. In this study, deletion of VdPLP generated ∆VdPLP mutants that were sensitive to cell wall-perturbing agents, which suggests VdPLP participates in stabilizing the cell wall under stress. Disruption of cell wall integrity in the VdPLP mutants was thus associated with the significantly reduced virulence of the ∆VdPLP mutants.
A small G protein, Rho1, is considered to be the master regulator of CWI signaling. Rho1 integrates signals from the cell surface and regulates many outputs involved in cell wall biogenesis. GTP-bound Rho1 associates with and activate the PKC1 [47]. Polyunsaturated fatty acids like arachidonic acid (AA) released by PLPs may regulate cellular signaling and influence the activity of protein kinases [35]. MAPK pathway can be regulated by protein kinases [30,31]. Thus, PLPs could influence the MAPK pathway and CWI pathway as well. In the citrus postharvest pathogen Penicillium digitatum, disruption of chitin synthase (PdChsVII) impaired CWI integrity, causing reduced virulence [68]. Class I chitin synthases reinforce cell wall integrity under a specific cell wall stress in C. albicans [69]. A myosin motor domain-chitin synthases (MMD-Chs) deletion impaired fungal resistance to host defense mechanisms and colonization by invasive hyphae, possibly due to a decrease in the structural integrity and permeability of the cell wall [70]. In our research, the expression of the MAPK pathway (VdMK1 and VdMKK1) was diminished in the ∆VdPLP mutants. The CWI pathway and chitin synthesis genes were downregulated in the ∆VdPLP strains, suggesting that VdPLP has a critical role in regulating the CWI pathway and chitin synthesis in V. dahliae. Meanwhile, ∆VdPLP mutants cannot penetrate the cellophane membrane. GFP-labeled ∆VdPLP mutants showed impairment in colonization. The results indicate that VdPLP could regulate the MAPK pathway for regulating CWI pathway and chitin synthase and affect the cell wall integrity and virulence of V. dahliae.
Soil-borne V. dahliae must penetrate and colonize the root to cause disease. The colonization of Arabidopsis thaliana [71], cotton [6] and tomato [72] by V. dahliae GFP-labeled strains has been detailed. In our research, the same strategy was used for penetration ability comparison. When the penetration ability of GFP-tagged ∆VdPLP strains was compared with that of Vd-GFP strains, penetration of epidermal cells of the roots by hyphae of ∆VdPLP-GFP strains was severely lowered as was subsequent colonization. In addition, ∆VdPLP strains were impaired in penetration of cellophane, indicating impairment of an early phase of infection. VdPLP may regulate vegetative growth of V. dahliae in plant tissue and affect cell wall stability, and then impair pathogenicity further.
Deletion of VdPLP resulted in severe impairment of vegetative growth on various carbon sources and reduced conidia production. The ∆VdPLP mutants were also severely impaired in their ability to penetrate and colonize roots of N. benthamiana, as shown microscopically and by a decrease in biomass compared with the wild type. The qRT-PCR results suggest that VdPLP participates in the CWI pathway and chitin synthesis. These results revealed the defect of ∆VdPLP mutants to penetrate plant tissue, contributing to the impaired virulence. VdPLP may contribute to the full virulence of V. dahliae by mediating hyphal growth, the cell wall integrity and penetration process, which provides new insights into the molecular pathogenic mechanisms of V. dahliae.

Conflicts of Interest:
The authors have declared that no conflict of interest exists.