The Cotton Ghplp2 Positively Regulates Plant Defense against Verticillium Dahliae by Modulating Fatty Acid Accumulation and Jasmonic Acid Signaling Pathway

Yutao Zhu College of Science, China Agricultural University Xiaoqian Hu College of Science, China Agricultural University Yujiao Jia College of Science, China Agricultural Department Linying Gao College of Science, China Agricultural University Yakun Pei College of Science, China Agricultural University Zhaoyue Ge College of Science, China Agricultural University Xiaoyang Ge CAAS CRI: Chinese Academy of Agricultural Sciences Cotton Research Institute Fuguang Li Chinese Academy of Agricultural Sciences Cotton Research Institute Yuxia Hou (  yuxiacau@163.com ) China Agricultural University https://orcid.org/0000-0002-7170-131X


Introduction
In-plant, lipids play a critical role in cell membrane components, sustainable energy storage, and signaling transduction in response to biotic and abiotic stress (Li, et al. 2016;Lim, et al. 2017;Wang 2004). The accumulation of hydrolysis products via lipid acyl hydrolase (LAH) activity in lipid metabolism affects pathogenesis and resistance mechanisms in plant-microbe interactions (Shah 2005).
In this study, we report the identi cation and characterization of the V. dahliae-induced patatin-like protein gene GhPLP2 in cotton (Gossypium hirsutum). The transcriptional expression patterns of GhPLP2 were investigated in response to biotic and abiotic stress. Sequence analyses showed that GhPLP2 had conserved catalytic dyad residues, which is critical for the LAH activity of patatins. The recombinant GhPLP2 protein further con rmed its LAH activity in vitro. Reduced HR response during V. dahliae elicitor infection were observed in GhPLP2-silenced plants, suggesting the role of GhPLP2 in HR-like cell death signaling. A potential role of GhPLP2 in positively regulating plant resistance to V. dahliae was examined by overexpression in Arabidopsis and virus-induced gene silencing (VIGS) in cotton plants. Transgenic Arabidopsis plants exhibited higher accumulation of linoleic acid, α-linolenic acid, and jasmonic acid, which were decreased in GhPLP2-silenced cotton plants. Besides, the expression of genes involved in the jasmonic acid synthesis pathway and defense response is positively correlated with the expression of GhPLP2. Together, we showed that GhPLP2 positively regulates defense against V. dahliae by mediating fatty acids metabolism and activation of the JA signaling pathway.  with BoxShade (http://www.ch.embnet.org/software/ BOX_form. html). The phylogenetic tree was constructed with the neighbor-joining method using MEGA 7 with bootstrap values from 1000 replicates indicated at the nodes, and motifs were annotated using MEME (http://meme-suite.org/tools /meme) (Bailey and Elkan 1994;Kumar, et al. 2016). The homology model of GhPLP2 was generated using SWISS-MODEL, and three-dimensional models were analyzed and visualized using EzMol, Version 1.20 (Bordoli, et al. 2009;Reynolds, et al. 2018 Phospholipases substrates were 1,2-distearoyl-sn-glycerol-3-phosphoethanolamine, 1,2-distearoyl-snglycerol-3-phosphocholine, or L-α-phosphatidylglycerol (Aladdin ®, China). The reaction mixture comprised 50 mM Tris-HCl (pH 8.0), 10 mM CaCl 2 , 0.05% (v/v) Triton X-100, 500 µg substrate and 10 µg puri ed protein in a nal volume of 600 µl. Reactions were performed at 37 ℃for 1 h and termination by adding 500 µL chloroform/methanol (2:1, v/v). The product fatty acids released from the substrates were separated and analyzed according to the method described before (Camera, et al. 2010). Free fatty acids are methylated in 1 mL 0.5 mol/L KOH-methanol at 60℃ for 2 h. Then 1 mL hexane containing 0.01% butylated hydroxytoluene (BHT) and 0.1 mg methyl nonadecanoate was added. After shaking and standing for strati cation, the supernatant containing fatty acid methyl esters (FAMEs) was separated. Quantitation of individual FAMEs was analyzed by gas chromatography (GC) equipped with an Agilent column (USA) (30 m by 0.25 mm, 0.25 um lm) and a ame ionization detector (FID).

Generation of Transgenic Plants
The PCR product with Sal I/Spe I restriction site was inserted into a modi ed pCAMBIA 1300 vector harboring a hygromycin phosphor-transferase (hptII) gene and the green uorescent protein (GFP) ). The construct was introduced into Agrobacterium tumefaciens strain GV3101 by freeze-thaw method, and transformation of Arabidopsis was performed by the oral dip method (Clough and Bent 1998). Transgenic Arabidopsis seeds were screened on MS plates containing 25 mg/mL hygromycin B and T3 homozygous lines displaying 100% hygromycin resistance were used for further experiments.

Construction of VIGS Vectors and Agrobacterium-mediated VIGS
The VIGS transient expression methods were followed by Gao (Gao, et al. 2011). The silenced fragments of GhCLA1(Cloroplastos alterados 1) and GhPLP2 were ampli ed from cotton cDNA and inserted into the TRV:00 vector to generate the TRV: GhCLA1 and TRV: GhPLP2 vectors. Then plasmids of TRV: GhCLA1 and TRV: GhPLP2 were transformed into A. tumefaciens strain GV3101 by heat shock (Dong, et al. 2007). TRV:GhCLA1 plants were used as positive controls. Two weeks after in ltration, when GhCLA1-silenced plants showed clear signs of albinism in leaves, the e ciency of GhPLP2 was evaluated by RT-PCR. GhUBQ7 was ampli ed as internal control with primers qUBQ-F/R. The primers used in vector construction were listed in Supplementary Table S1 V. dahliae Inoculation and Disease Investigation Two weeks after VIGS, when true leaves of GhCLA1-silenced cotton plants showed clear signs of albinism, inoculation with V. dahliae (10 7 conidia/mL) was performed previously reported ). Plant disease index (DI) was monitored as the following formula according to Wang (Wang, et al. 2020): DI = [Σ(n×the number of seedlings at level n)/(4×the number of total seedlings)]×100, n denotes disease level, cotton seedlings were divided into ve levels based on their disease severity after V. dahliae inoculation (level 0, 1, 2, 3, 4). For Arabidopsis plants, four-week-old GhPLP2-transgenic and wild-type (WT) Arabidopsis plants were inoculated with V. dahliae spores as previously described (Gao, et al. 2013).
The controls were dipped in sterilized water. Disease index and symptom classi cation were performed as previously report (Pei, et al. 2019). Data were collected from three independent replicates (n ≥ 30).
V. dahliae recovery assay was conducted at 21 d after infection ). Stem section above the cotyledons was taken from inoculated cotton plants, six slices were transferred onto potato dextrose agar supplemented with kanamycin (50 mg/L) after surface sterilized. V. dahliae biomass was quanti ed following a previously described protocol (Ellendorff, et al. 2009). Inoculated cotton and Arabidopsis plants were harvested at 21 days and 14 days, respectively. DNA was extracted from 100 mg of the ne powder. V. dahliae biomass was determined by qPCR using fungus-speci c ITS1-F and V.dahliae-speci c ST-Ve1-R primers (Fradin, et al. 2011). The GhUBQ-F/R and AtEF1α-F/R primers were used as reference genes. All primers were listed in Supplementary Table 1

Fatty Acid Pro le Analysis and Hormone Quantitation
Four-week-old Arabidopsis and two-week GhPLP21-silenced cotton plants after VIGS were used to analyze the fatty acid pro le and hormone quanti cation. Hormone quantitation was detected as previously reported (Li, et al. 2020). Fresh samples were ground into ne powder in liquid nitrogen. 100 mg powder was extracted with 1 mL of 80% methanol-water (containing 1% formic acid) using ultrasound at 4°C for 10 min and centrifuged at 4°C at 12,000 rpm for 10 min. The supernatant was transferred to a 2mL centrifuge tube containing 50mg primary secondary amine (CNW Technologies GmbH, Germany), dried with nitrogen, and added 100 µL 80% methanol-water (containing 1% formic acid). For fatty acids pro le analysis, plant samples were dried to a constant weight under 90 ℃, and 50 mg of dry powder was weighed. Methyl esteri cation of fatty acids and analysis were performed as described above. GC parameters were as follows: 180°C for 10 min followed by a ramp to 190°C at 1°C min − 1 , holding 190°C for 3 min and then heating up to 220°C with a gradient at 4°C min − 1 , nal temperature was maintained for 3 min.
Gene Expression Pro ling of Jasmonic Acid Signaling Pathway In Arabidopsis and Cotton Plants Total RNA was obtained from non-inoculated plants with an RNA extraction kit [BIOMED GENE TECHNOLOGY (Beijing) CO., LTD.]. The synthesis of cDNA and qRT-PCR assays was conducted as mentioned above. AtEF1a and GhUBQ7 were employed as internal standards genes in Arabidopsis and cotton, respectively. Relative gene expression was calculated using the 2 −ΔΔCt method. Data were presented as the mean ± standard error (SE) of three independent experiments. The primer sequences are listed in Supplementary Table S1.

Data analysis
Data are obtained from three independent replicates per treatment and presented as mean ± standard error. Signi cant differences between groups were analyzed by ANOVA using statistical software IBM SPSS statistics 20 followed by Student's t-test. Asterisks indicates a signi cant difference compared with control (*P < 0.05, **P < 0.01).

Identi cation and sequence analysis of GhPLP2
The full-length GhPLP2 cDNA consists of a 68 bp 5′untranslated region (5′ UTR), 138 bp 3′ UTR, and 1218 bp open reading frame (ORF) encoding a 405 amino acid protein with a theoretical pI of 8.21 and molecular weight of 44.44 kDa. GhPLP2 protein contains the conserved serine hydrolase motif GXSXG at residues 63-67 and a conserved aspartic acid (D) residue at 214 within the patatin domain (residues 21-221) (Fig.S1). The catalytic dyad of serine and aspartic acid residues is critical for the LAH activity of patatin (Rietz, et al. 2010;Rydel, et al. 2003). The homology modeling result shows active sites composed of Ser-Asp catalytic dyad responsible for its LAH activity (Fig.S2). GhPLP2 has no transmembrane domain or signal peptide. The phylogenetic analysis and multiple alignments were generated with plant patatin homologs previously reported. The phylogenetic analysis showed patatins were divided into three groups and GhPLP2 located in group II containing NtPat1 (Cacas, et al. 2005 (Fig. 1a). These genes have been reported to be involved in disease resistance and pathogen-mediated hypersensitive cell death. Consistent with the phylogenetic analysis result, the patatin homologs were divided into three groups based on the conserved motif (Fig. 1b). Group II proteins have the canonical S-D dyad esterase motif constituted by GxSxG and DGG/A in patatin catalytic centers. Also, Group II proteins contain a conserved anion binding element DGGGxxG and proline motif APP. GhPLP2 was classi ed as a group II protein with the conserved catalytic dyad residues.
Lipid acyl hydrolase (LAH) activity assay of GhPLP2 The LAH activity of patatin proteins can hydrolysis membrane lipids into free fatty acids and lysophosphatidic acid, leading to a series of signaling pathways involved in growth development and defense response (Ryu 2004). To further understand the role of GhPLP2 protein, the LAH activity of the recombinant GhPLP2 protein was detected. The recombinant protein of GhPLP2 was produced and puri ed from E. coli BL21 (DE3). GhPLP2 protein was induced with 0.4 mM IPTG at 22 ℃ for 2h (Fig. 2a). GhPLP2 fusion protein was puri ed using Ni columns, and the elution fractions were identi ed by SDS-PAGE (Fig. 2b). When p-NPP was used as the substrate, the His-GhPLP2 protein exhibited high LAH activity (97 nmol min − 1 mg − 1 ); however, the His alone was no enzyme activity (Fig. 2c). Besides, when phospholipids phosphocholine (PC), phosphoethanolamine (PE), phosphatidylglycerol (PG) were used as substrates, the release of free fatty acids was quanti ed. Puri ed GhPLP2 protein released fatty acids from these phospholipids and exhibited variant enzyme activities in different phospholipids substrates (Fig. 2d). The results indicate that GhPLP2 functions as a lipid acyl hydrolase to release free fatty acids in plants.

Subcellular Localization of GhPLP2 protein
To determine the subcellular localization of GhPLP2 protein, we examined the root tissues of GhPLP2transgenic Arabidopsis with GFP by Confocal Laser Scanning Microscopy (Fig. 3). The result indicated that GhPLP2 was located in the cell wall or the plasma-membrane ( Fig. 3d-f). To further clarify the location of GhPLP2, the seedlings were treated with 0.8 M mannitol for 10 min. After plasmolysis, the result revealed that GhPLP2 was distributed in both cell wall and plasma membrane ( Fig. 3g-i)..

Induction of GhPLP2 by Various Stresses
Expression of the GhPLP2 gene in cotton was examined by qRT-PCR. GhPLP2 was most abundant in the cotton root (Fig. 4a), which was the rst physical barrier in cotton plants against V. dahliae infection. The change of GhPLP2 expression in response to various stresses was investigated. The expression of GhPLP2 was upregulated in response to infection with V. dahliae or F. oxysporum. GhPLP2 expression was signi cantly increased at 0.5 h, 1 d, 5 d, and 7 d after inoculation with V.dahliae (Fig. 4b). With the infection of F.oxysporum, GhPLP2 expression upregulated at 0.5 h, 3 d, 5 d, and 7 d (Fig. 4c). We further investigated the expression of GhPLP2 after treatment with the defense-related signaling molecules JA and ethylene (ET). GhPLP2 expression was extremely higher at 3 h, and another peak appeared at 24 h after JA treatment (Fig. 4d). In contrast, ET induced GhPLP2 expression increased at 0.5 h, with a maximum level observed at 3 h (Fig. 4e). The treatment with 2.5% (w/v) PEG 6000 markedly increased gene expression at 1h and returned to normal level at 24 h (Fig. 4f).
Enhanced disease susceptibility of GhPLP2-silenced cotton Plants to V. dahliae Infection The GhPLP2 gene was silenced in cotton plants to clarify its function by virus-induced gene silencing (VIGS) (Gao, et al. 2013). The cotton gene GhCLA1, which is involved in chloroplast development, was used as a positive control (Gao, et al. 2011). To assess the gene silencing e ciency, the expression of GhPLP2 was monitored by semi-quantitative RT-PCR and qRT-PCR in TRV:00 and TRV: GhPLP2 cotton true leaves after two weeks of VIGS ( Fig. 5g and Fig. S3). The results indicated that GhPLP2 expression was effectively reduced in GhPLP2-silenced cotton plants.
To verify the function of GhPLP2 in the interactions between cotton with V. dahliae, cotton plants were inoculated with V. dahliae. After inoculation, typical disease symptoms appeared at ten days in GhPLP2silenced cotton. At 14 days, leaf chlorosis and necrosis were more severe in GhPLP2-silenced plants (Fig. 5a,d). The plant disease index of silenced plants was higher than that of control plants (Fig. 5h). At 21 days, dark and necrotic vascular bundles of the dissected stems were more apparent in GhPLP2silenced plants (Fig. 5b,e). The fungal renewal cultivation showed that the disease conditions of GhPLP2silenced plants were more serious compared with controls ( Fig. 5c,f). Besides, the fungal biomass of the stems from the GhPLP2-silenced plants was higher than the controls determined by qRT-PCR analysis (Fig. 5i). These results suggest that the silencing of GhPLP2 attenuates cotton resistance to V. dahliae.
Silencing of GhPLP2 compromises the HR symptom triggered by V. dahliae elicitor PevD1 Patatin-like proteins have been shown to participate in hypersensitive response (HR) during avirulent Xanthomonas campestris pv. Vesicatoria infection in pepper plants and response to Tobacco mosaic virus (TMV) in tobacco plants. Therefore, GhPLP2 may regulate resistance to V. dahliae by mediating HR induced by V. dahliae elicitors. To test our hypothesis, we expressed and puri ed an elicitor PevD1 from V. dahliae (Fig. S4), which triggered HR and resistance responses in cotton, Arabidopsis, and tobacco (Bu, et al. 2013;Liu, et al. 2016;). The HR-like symptom was initiated at 24h after injection with PevD1 elicitor, while HR-like cell death was signi cantly reduced in silenced leaves (Fig. 6a). Expression of HR marker genes GhHIN1 and GhHSR203 and callose deposition triggered by PevD1were compromised in GhPLP2-silenced leaves at 24h after infection (Fig. 6b-d). Also, the silencing of GhPLP2 signi cantly compromised the induction of H 2 O 2 by PevD1 (Fig. 6e).
Overexpression of GhPLP2 in Arabidopsis confers enhanced resistance to V.dahliae To further evaluate the role of GhPLP2 in cotton response to V. dahliae, we examined the resistance of wild-type and transgenic Arabidopsis seedlings under V.dahliae infection. Three homozygous transgenic (T3 generation) lines with the highest expression levels of GhPLP2 (L1, L4, and L8) were selected for further experiments (Fig. S5).
Four-week-old transgenic and wild-type (WT) plants were infected with V.dahliae spores by the root dipping method. After inoculation, WT plants showed more serious wilt, yellowish, and necrosis than the transgenic plants at 14 dpi (Fig. 7a); this was consistent with the disease index investigation (Fig. 7b). The fungal biomass of WT plants was remarkably higher compared to the transgenic as determined by qRT-PCR analysis (Fig. 7c). These results indicate that overexpression of GhPLP2 in Arabidopsis plants confers enhanced resistance to V.dahliae. The endogenous LAH activity of crude proteins from different genotypes plants was detected using p-nitrophenyl palmitate (pNPP) according to Lin (Lin, et al. 2011). LAH activity of GhPLP2transgenic Arabidopsis lines was higher than that of WT plants and decreased in TRV: GhPLP2 cotton plants (Fig. S6). To assess the metabolic differences between different genotype plants, fatty acid pro le, basal JA accumulation, and gene expression were determined in non-inoculated plants (Fig. 8).
Compared with WT plants, fatty acid pro le analysis display that the level of polyunsaturated fatty acid linoleic acid (LA, 18:2) and α-linolenic acid (ALA, 18:3) has increased in GhPLP2-overexpression Arabidopsis (Fig. 8a). The basal level of JA in transgenic Arabidopsis plants was also higher than in WT plants (Fig. 8b). Expression of genes involved in JA biosynthesis pathways such as LOX1, LOX2, AOS, OPR3, and JA-responsive marker PDF1.2 in transgenic Arabidopsis was induced compared with the WT plants (Fig. 8c). On the contrary, GhPLP2-silenced cotton plants has decreased accumulation of linoleic acid and α-linolenic acid and the basal accumulation of JA was also compromised in TRV: GhPLP2 cotton plants (Fig. 8d,e). Expression levels of LOX2, AOS, OPR3 and PDF1.2 associated with the JA signaling pathway were downregulated in TRV: GhPLP2 cotton plants (Fig. 8f). The results showed that GhPLP2 contributes to fatty acid accumulation and is involved in JA signaling pathway.

Discussion
Patatin-like proteins (PLPs) are a signi cant family of lipases with lipid acyl hydrolase(LAH) activity and play an essential role in lipid metabolism during plant defense immunity (Rivas and Heitz 2014). In the present study, we identi ed a novel patatin-like protein gene GhPLP2 from cotton and demonstrated its role in fatty acid metabolism and JA signaling pathway conferring cotton resistance against V.dahliae.
Patatin-related proteins have a conserved S-D catalytic dyad constituted by a serine of esterase motif (GxSxG) and a central aspartic acid, which is essential for the LAH activity (Rydel, et al. 2003). These residues were also identi ed in human Ca 2+ -independent phospholipase A2 (iPLA2s) and many microorganism proteins; their function was con rmed by previous studies (Burke and Dennis 2009; Heller, et al. 2018; Mansfeld 2009). Same as other patatin proteins in plants, GhPLP2 has a canonical esterase motif GxSxG and a conserved S-D catalytic dyad for its lipid acyl hydrolase activity (Fig. 1b). The enzyme activity assay in vitro showed that the fusion protein has LAH activity which for hydrolyzing phospholipids as previously described (Holk, et al. 2002). It suggested that GhPLP2 has functional LAH activity enabling it to release fatty acids from membrane lipids. Bioinformatic analysis and subcellular localization experiments showed that GhPLP2 localized in both cell wall and plasma membrane (Fig. 3), which was consistent with the localization of several PLPs (Holk, et al. 2002;Kim, et al. 2014 Yang, et al. 2007). Further analysis is necessary to clarify the role of PLPs in the interaction between different pathogens and plants. In this work, the GhPLP2 gene was strongly induced at the early stage after infection with V.dahliae (Fig. 4b), followed by a decrease and ). The effector protein PevD1 from V.dahliae could induce typical HR-like necrosis in tobacco and trigger innate immunity in cotton plants (Bu, et al. 2013;). In our study, GhPLP2 silencing has reduced HR phenotype and expression of the HR marker gene during induction of V. dahliae elicitor PevD1 in cotton leaves. The plant immune responses, including H 2 O 2 accumulation and callose deposition, were damaged in silenced leaves (Fig. 6). It suggests that GhPLP2 is required for resistance (R) gene-mediated disease resistance in cotton plants. However, the mechanism of GhPLP2 involved in the HR response induced by the V. dahliae elicitor is still unclear.
Fatty acids are not only major membrane components of the cell, but they also directly or indirectly participate in a series of immune responses in plants (Lim, et  Previous studies have shown that PLPs mediates plant immune responses through regulating FA metabolism and JA signaling pathway. In tobacco, the patatin-like protein was rapidly induced preceding JA accumulation in response to the Tobacco mosaic virus (TMV) (Dhondt, et al. 2010). AtPLAI plays a critical role in maintaining the homeostatic pool of free FA and basal JA, which increase resistance to B. cinerea (Yang, et al. 2007). Overexpression of a patatin-like protein gene OSAG78 in Arabidopsis, increased linoleic acid and linolenic acid amount and induced expression levels of the JA-related defense genes PDF 1.2 and PR4 (Lin, et al. 2011). In this study, GhPLP2 expression was induced upon treatment with exogenous JA (Fig. 4d), suggesting that GhPLP2 is likely to involve in JA-dependent defense pathways. Given the LAH activity of GhPLP2 protein in vivo and in vitro, we analyzed the accumulation of FA and its derivative JA. The data showed that accumulation of C18 FA linoleic acid, α-linolenic acid, and JA are positively correlated with the expression of GhPLP2 (Fig. 8).