Genome-wide identification of the CYP82 gene family in cucumber and functional characterization of CsCYP82D102 in regulating resistance to powdery mildew

Plant materials and stress treatment

Portions of this text were previously published as part of a preprint (https://doi.org/10.21203/rs.3.rs-1196334/v1) (Wang et al., 2021). Cucumber 9930 cultivated in North China was selected as plant material (donated by Prof. Huang Sanwen, Institute of Vegetable Research, Chinese Academy of Agricultural Sciences). The 9930 seeds were placed in a clean culture dish with wet filter paper. After the cotyledons are opened, they were moved to the soil (vermiculite: nutrient soil = 1: 1) in the greenhouse at 22 ± 4 °C with a photoperiod of 16 h light and 8 h dark. All materials, including roots, stems, and leaves were immediately frozen in liquid nitrogen and stored at −80 °C until total RNA isolation.

Four-week-old cucumber plants were exposed to abiotic stresses and defense-modulating plant hormones. Experimental solutions were prepared using methanol as a solvent for 100 μM JA, while distilled water was used as a solvent for 100 μM salicylic acid (SA), 100 μM abscisic acid (ABA), 100 μM ethylene (ETH), 250 mM NaCl, and 20% polyethylene glycol (PEG 6000). At specified time intervals (0, 3, 12, 24, and 48 h) after treatment, plantlets from each treatment group and control samples were collected. The collected samples were swiftly frozen in liquid nitrogen and preserved at −80 °C.

Identification of CYP82 subfamily members in Cucumis sativus L.

The CYP82 subfamily was identified and analyzed based on the complete cucumber genome. Pertinent genomic and protein data for cucumber were retrieved from the database (http://cucurbitgenomics.org/). The Hidden Markov model for the P450 structural domain (PF00067) was acquired from the Pfam database (http://pfam.xfam.org/) in this study. Subsequently, the Simple HMM Search program within the TB tools software was employed to search the cucumber genome and predict the IDs of the CYP82 gene family. These sequences were further validated using the Protein BLAST function in the NCBI database (https://blast.ncbi.nlm.nih.gov/Blast.cgi). The cucumber (Chinese Long v2) genome, CDS, transcripts, polypeptides, and a 2,000 bp segment upstream of the promoter region, was retrieved from the database, spanning from the translation initiation site (ATG). ClustalX was utilized to analyze the amino acid sequences of the target CYP82 sequences. Additionally, the neighbor-joining (NJ) method parameter in MEGA7.0 was employed, utilizing pairwise deletion and 1,000 bootstrap replicates for multiple alignments in constructing a phylogenetic tree among the CYP82 family.

Next, to identify all potential P450 genes in the cucumber genome, a BLAST search against the cucumber genome was conducted using 245 AtCYP450s genome sequences downloaded from Cytochrome P450 database (https://drnelson.uthsc.edu/plants/). Based on the 230 cucumber P450 gene sequences obtained from the P450 homepage, and by comparing them with the downloaded Arabidopsis sequences to find a sequence with higher consistency, each gene is classified and named according to international standards.

Conserved domain, motif identification, and gene structure analysis

The conserved motif of cucumber P450 was analyzed using the online tool MEME (http://meme-suite.org/tools/meme). The maximum number of motifs was 10, and the optimized motif width was 10–100 amino acid residues. The rest of the parameters are default (Hu et al., 2015). The location information of the gene and the conservative area was confirmed and then plotted using the Perl SVG package.

Phylogenetic analysis

The full-length amino acid sequences of gene family members derived from closely related species were used for phylogenetic analysis. The MEGA program 7.0 performed multiple sequence alignments of the obtained genes. The phylogenetic tree was generated using the Neighbor-Joining (NJ) technique in MEGA and a bootstrap test conducted 1,000 times (Kumar, Stecher & Tamura, 2016).

Gene replication

The potential replication genes were identified using the duplicate MCScanX software’s gene classifier program (Wang et al., 2012). The genome’s all coding gene protein sequences were aligned consuming blastP, and the alignment consequences were used as input files for the MCScanX Software for reproduction gene projection. The gene was identified as a reproduction gene pursuant to e-value < 1e⁻⁵ or e-value < 1e¹⁰.

Promoter cis-element analysis

Intercepting the upstream sequence (2.0kb) of each CsCYP450 gene annotation file in Cucurbitaceae database (http://cucurbitgenomics.org/), you then proceeded to scan the cis-acting elements from those sequences. Cis-acting elements prevalent in genes were identified, including CAAT-box, TATC-box, TATA-box, and cis-acting elements with specific functions were only shown (Lv et al., 2020; Trapnell et al., 2010; Wang et al., 2020).

The expression analysis of CYP82 genes

Cucumber transcriptome data (PRJNA80169) was acquired from the Cucurbit Expression Atlas within the Cucurbit Genomics Database (CuGenDB). A detailed analysis of the expression profiles of 12 CYP82 genes was performed using the FeatureCounts R package, and these patterns were subsequently visualized as heatmaps utilizing TBtools (Yang et al., 2022). Concurrently, heatmaps representing the expression patterns of CYP82 genes in response to PM treatment were generated using cucumber transcriptome data (PRJNA321023) (Yang et al., 2022).

Quantitative reverse transcription PCR (qRT-PCR) expression analysis

According to the results of combining CYP82 family with RNA-seq analysis, 10 candidate genes were selected for qRT-PCR analysis. Primers for qRT-PCR were designed by using Primer Premier 5.0 software (Table S1). The cucumber Actin gene was used as a control for normalization between samples. Total RNA was extracted from each samples using RNA pure Plant Kit (Cwbio, Beijing, China) according to the manufacturer’s instructions. First-strand cDNA synthesis was performed using an oligo (dT) primer and 2 g of total RNA in a 20-L reaction volume, according to the manufacture’s instruction for the Super RT cDNA Synthesis Kit (Cwbio, Beijing, China). The integrity of RNA was assessed by 1% agarose gel electrophoresis, qRT-PCR was carried out using a Ul’tra SYBR Mixture (WithRox) kit (Cwbio, Beijing, China) with a Fast Real-Time PCR System (Applied Biosystems 7500, USA). The PCR mixture consisted of 1 μL cDNA template, 1 μL 10 μM forward primer, 1 μL 10 μM reverse primer, 12.5 μL Ul’tra SYBR Mixture, and 9.5 μL RNase-free water. The PCR verification pictures are shown in Fig. S1. The recommended conditions for PCR were used as follows: 95 °C for 10 min, followed by 40 cycles of 95 °C for 30 s, 58 °C for 30 s, and 72 °C for 30 s. All reactions were performed in triplicate for each sample. The relative expression analysis of each gene were calculated using the 2^−ΔΔCT method (Duan et al., 2020).

Subcellular localization of CYP82D102

The site Plant-mPLoc 2.0 (http://www.csbio.sjtu.edu.cn/bioinf/plant-multi/), was used to make predictions regarding subcellular localization (Ai et al., 2020). This bioinformatic investigation disclosed that CsCYP82D102 predominantly localizes within the endoplasmic reticulum (ER). For subsequent experimental validation, the coding sequence (CDS) of CsCYP82D102 was successfully cloned into the pCAMBIA1302-GFP vector, omitting stop codons, precisely at the NcoI and BglI. sites. The GFP-CYP82D102 fusion construct was introduced into Arabidopsis protoplast cells. In this assay, the Cauliflower Mosaic Virus (CaMV) 35S and pHB-GFP vectors were employed as controls, with the fusion construct serving as the experimental treatment. The protoplast preparation method is detailed in Shen et al.’s work (Shen et al., 2014). The samples were then observed using a Leica TCS SP5-II confocal fluorescence microscope. The primers employed for vector construction can be located in Table S1.

Plasmid construction and generation of transgenic plants

To generate CsCYP82D102 overexpressing (OE) plants, the complete cDNA fragment of CsCYP82D102 (1617 bp) was cloned into the binary expression vector pBI121, which is driven by a CaMV 35S promoter. Agrobacterium tumefaciens strain LBA4404 was employed for plant transformation. The desired fragment was amplified through the Golden Gate seamless cloning technique for RNAi analysis (Yan et al., 2012). It was then inserted into the vector and delivered into cucumber using LBA4404. The procedures for generating OE and RNAi plants followed a similar protocol, with modifications based on the method described by Liu et al. (2023) for genetic transformation in cucumber. These transgenic lines were planted in a greenhouse self-pollination to obtain T1 generation seeds.

Physiological index detection of the transgenic cucumber

Enzyme activities associated with reactive oxygen species (ROS), including superoxide dismutase (SOD), peroxidase (POD), and catalase (CAT), were determined following established protocols. Additionally, malondialdehyde (MDA) content was quantified using previously described methods. Marker gene expression in the SA and MeJA signaling pathways was analyzed using qRT-PCR. Detection of superoxide anion (O₂.⁻) and hydrogen peroxide (H₂O₂) accumulation, as well as cell death in cucumber leaves, was performed via histochemical staining with 3,3′-diaminobenzidine (DAB) and nitroblue tetrazolium (NBT) respectively (Daudi & O’Brien, 2012; Jambunathan, 2010).

Statistical analyses

Three biological replicates and three technical replicates were executed to uphold the reliability of our findings. The gathered data for comprehensive analysis employing data processing software such as Origin 8.0 and DPS 9.5. Subsequently, the results were presented as the mean ± standard deviation (SD). Noteworthy distinctions between the treatment and control groups were assessed using Student’s t-tests, with statistical significance set at P < 0.05. Additionally, a higher level of significance (P < 0.001) was considered exceptionally significant. Moreover, the amassed data underwent further scrutiny through analysis of variance (ANOVA).

Identification and physicochemical properties analysis of CYP82 protein in Cucumis sativus L.

After removing the redundant genes, 12 CYP82 family members were identified in the cucumber genome (Table S2). The protein sequences encoded by these genes were further confirmed by NCBI and SMART screening to contain four conserved motifs, namely the FxxGxRxCxG heme-binding protein domain, I-helix, K-helix, and PERF/W motif. The size of cucumber P450 protein was between 317 and 1,056 aa, the theoretical isoelectric point (pI) of the encoding protein was 6.85–9.52, and the molecular weight was between 36,767.9 and 65,839.3 kDa. The subcellular localization predicted that all the CYP82 genes were located in the endoplasmic reticulum. These results suggest that CsCYP82 plays a significant role in cucumber’s cellular structure and function.

Analysis of gene structure, conserved domain, and motif of CYP82 subfamily in Cucumis sativus L.

Based on the alignment results of the coding sequence to the genomic sequence of the cucumber CYP82 gene family, the exon numbers of these CYP82 genes varied from 1 to 10, revealing the structural diversity of the P450 gene family (Figs. 1A–1C). Of the cucumber CYP82 gene family, the Csa3G852560 gene contained the most exons. Further analysis showed that most of the same-family members had similar distribution characteristics regarding exon length and number. Conserved motif analysis of P450 in cucumbers is shown in Fig. 1C. A total of 10 conserved motifs were identified and named motif 1 to motif 10, respectively (Table S3). Six motifs were found in more than 90% of the genes. Except for Csa3G853140 and Csa3G852619, all CYP82 gene members contained motif 4: FGAGRRICPG (FxxGxRxCxG), which was annotated as the core functional domain (the heme-binding domain). The heme-binding domain contains critical cysteine residues, which are axial ligands of heme. Motif 3 was annotated as a helix K domain (EXXR) that plays an important role in P450 protein folding. Motif 3 was annotated as an I-helix conserved domain involved in oxygen binding and oxygenation. This result shows that motif 3 is highly conserved during cucumber development. Motif 6 was annotated as a PERF domain and motif 7 was annotated as a PKG motif. Members of the same gene cluster shared similar conserved domains, whereas different gene clusters had specific conserved domains. In conclusion, the highly conserved motif and similar gene structure of CYP82 in the same family further support the phylogenetic analysis’s close evolutionary relationship and reliability.

Analysis of replication events of cucumber CYP82 subfamily

The process of gene duplication stands as a pivotal evolutionary mechanism, fostering genetic diversity and the emergence of novel functions, thereby significantly contributing to adaptation and speciation. Gene duplications are categorized as segmental, tandem, proximal, singleton, and dispersed. Among these categories, the prevalence of tandem duplication accounted for 38.6% (76 instances), while fragment duplication represented 19.3% (38 instances), proximal duplication constituted 9.1% (18 instances), decentralized duplication encompassed 33% (65 instances), and no single-copy duplication was observed (refer to Fig. 1D). The findings underscored that tandem replication acts as the primary driving force behind the amplification of the CsCYP450 gene family. This particular mode of replication is crucial for sustaining an extensive gene family, enabling rapid expansion or contraction in response to environmental variations, thereby enhancing genetic complexity, especially under favorable conditions (Yu et al., 2017). All members of the CYP82 subfamily are located on chromosome 3 of cucumber. Except for Csa3G852560, the others are tandem replication. Moreover, the chromosomal distribution pattern of CsCYP450 provides strong indications that the proliferation of CsCYP450 genes in cucumbers is significantly influenced by tandem replication.

Evolutionary analysis of related species

The genetic relationships within the cucumber P450 gene family were examined, and a phylogenetic tree was generated using the results from the multiple sequence alignment. Subsequent to the phylogenetic analysis, the 165 P450 genes were categorized into two major groups: the A-type and non-A-type P450 gene families (Fig. S2). The A-type P450 gene family was found within the CYP71 clan, while the non-A-type P450 gene family was subdivided into seven distinct clans, denoted as CYP85, CYP86, CYP72, CYP97, CYP711, CYP51, and CYP710. The phylogenetic relationships of the CYP82 protein family were explored using full-length amino acid sequences from cucumber, garden pea, opium poppy, sweet basil, soybean, and Arabidopsis (Fig. 2). It was observed that the cucumber CYP82 is in close proximity to soybean, garden pea, peppermint, and sweet basil in terms of phylogenetic relationships.

The P450 genes of cucumber were compared with CYP82 subfamilies in Arabidopsis, tomato, soybean, maize, rice, poplar, grape, and moss (Table S4). CYP82 has evolved in the cucumber genome compared with Arabidopsis. There are 34 CYP82 family members in grapes and 24 in soybeans. Some studies have shown that CYP736, CYP83, and CYP82 are very similar, and they strongly induce Phytophthora sojae to infect cucumber hypocotyls (Guttikonda et al., 2010). Nevertheless, in the genomes of cucumber, soybean, and grape, CYP82 exhibits a greater number of P450 members, whereas it is not present in the genomes of rice and moss.

Collinearity analysis

Collinearity was originally used to describe the locations of genes on the same chromosome. It now refers to the conservation of gene types and relative order in different species derived from the same ancestral type. Collinearity analysis can identify linear homologous genes among species, annotate protein-coding genes, and discover evolutionary events. The collinearity of species was constructed using McScanX and plotted using the Circos software (Fig. S3). The CYP82 subfamily genes of cucumber, melon, and Arabidopsis genomes were jointly analyzed to study their collinear genetic relationships Collinearity analysis showed that zero pairs of collinear genes were identified between the genomes of Arabidopsis and cucumber. There were three collinear gene pairs between cucumber and melon (Fig. S2). These results suggest that the functional differentiation of these genes may have occurred in cucumbers and melons during evolution.

Analysis of cis-acting elements of CsCYP82 promoter

To better understand the potential regulatory mechanism of CsCYP82 during cucumber growth and development, the cis-regulatory element in the promoter regions of CsCYP82 was identified in this study. The upstream sequence (2.0 kb) of all CsCYP82 translation initiation sites was scanned and the potential role of CsCYP82 expression elements was predicted using the Plant Care tool (Fig. 3).

In addition to the TATA-box, CAT-box, and other specific elements, the CsCYP82 promoter contains various cis-regulatory elements related to light signal response, tissue and organ development, hormone response, defense, and stress. Many regulatory elements related to the light response were detected in the CsCYP82 gene, such as ACE, ATCT-motif, and CAG-motif. Most genes contain at least two box4 elements, and there are six Box4 elements in the promoter region of the Csa3G852640 gene, suggesting that some DNA modules may be involved in the light response.

Among the elements related to plant growth and development, nine specific regulatory elements were identified, including those related to cell differentiation (CAT-box), endosperm expression (GCN4-motif), and zein metabolism regulation (O₂⁻site). Csa3G852580, Csa3G852560, and other genes were identified as cis-acting regulatory elements involved in regulating circadian rhythm, which means that their function may be affected by day length.

Among the elements related to plant hormones, 11 hormone response regulatory factors were identified, which were related to abscisic acid (Jayakannan et al., 2015; Martín et al., 2011), GA (GARE-motif, P-box, TATC-box), IAA (AuxRR-core, TGA-box, TGA-element), SA (TCA-element, SARE), and MeJA (CGTCA-motif, TGACG-motif). Among these, the ABA-related response elements are widely distributed. The promoter regions represented by the Csa3G852580 gene was predicted to contain four ABRE elements, which might be involved in the ABA signaling pathway, and SA and GA elements were enriched in most CYP82 promoters, indicating that SA and GA may induce them.

Among the biotic and abiotic stress-related elements, ARE elements were found in most gene promoter regions. Four ARE elements were found in Csa3G852600, and three ARE elements were found in Csa3G853160. ARE is necessary for cis-acting element-based anaerobic induction and may directly affect the antioxidant capacity of the gene. ERE elements have been found in the promoter regions of some genes, such as Csa3G852590 and Csa3G852640. In addition, a few genes were rich in WUN-motif, LTR, and DRE response elements, suggesting that they may be interested in wound response, low temperature, and osmotic stress response mechanisms. These results show that CYP82s may significantly affect the response to stress, hormones, and light exposure.

Expression profiles of CYP82 genes in Cucumis sativus L.

P450 genes assume vital functions within a wide array of biological pathways, with their distinct expression patterns in specific tissues intricately connected to the diverse roles these genes perform in various tissue types. A previous study comprehensively examined the tissue-specific expression of CYP82 genes in cucumber, encompassing roots, stems, leaves, male and female flowers, ovaries, and tendrils through an RNA-seq data analysis. Remarkably, Csa3G852610, Csa3G852560, Csa3G852600, and Csa3G852580 demonstrated consistent expression across all examined tissues, underscoring their significant contributions to cucumber development (Fig. 4A). Furthermore, Csa3G852590 and Csa3G852640 exhibited heightened transcriptional levels in tendrils compared to other tissues, whereas Csa3G852630 displayed relatively elevated expression in the roots. These distinctions suggest diverse roles for CYP82 genes in the development of cucumber.

Furthermore, qRT–PCR analysis was conducted on six distinct CYP82 genes within six different tissue types, namely female flowers, male flowers, leaves, roots, stems, and tendrils (refer to Fig. 4B). The qRT–PCR results exhibited consistent trends when compared to the transcriptome data. Notably, Csa3G852600, Csa3G852610, and Csa3G852590 displayed notably higher expression levels in tendrils in comparison to the other examined tissues. Conversely, the expression of Csa3G852580 in leaves, female flowers, and stems exceeded that observed in roots and tendrils. Csa3G853140, on the other hand, showed pronounced expression in roots but the lowest expression in stems.

In order to investigate the involvement of CYP82 genes in responding to biotic stress, the TPM values of CYP82 genes were collected from the RNA-seq data following powdery mildew (PM) infection and subsequently constructed heatmaps (Fig. 4C). During the infection of Sphaerotheca fuliginea (PM treatment), the expression profiles of three CYP82 genes exhibited significant alterations in cucumber varieties. Specifically, Csa3G852610, Csa3G852630, and Csa3G853160 demonstrated increased expression levels in response to PM treatment in both susceptible and resistant varieties, implying their pivotal roles in conferring powdery mildew resistance. These findings suggest a significant role for CYP82 genes in cucumber disease resistance during infection.

Furthermore, the expression patterns induced by PM inoculation of the selected genes were also assessed (refer to Fig. 4D). Following PM inoculation, it became evident that the expression patterns in Csa3G852630 and Csa3G853160 closely resembled each other, with the induced expression of these genes peaking at 24 h post-inoculation. In contrast, the expression of Csa3G852640 remained unaltered in response to PM. These findings indicate that Csa3G853160 exhibits a more rapid response to PM in cucumber when compared to other CYP82 genes. These results provide evidence that Csa3G853160, Csa3G852610, and Csa3G853160 may play a role in conferring resistance to PM.

Profiles of CsCYP82D102 genes expression under different treatments

Because CsCYP82D102 (GenBank accession number: XP_004147821.1) gene reacts strongly and quickly in the face of SA treatment, it was selected for further experiment. Expression analyses were conducted in different plant tissues to investigate the function of CsCYP82D102 in cucumbers. The expression levels of CsCYP82D102 were assessed in cucumber leaves, stems, flowers, and roots. The findings revealed that CsCYP82D102 exhibited the highest expression in the roots, followed by the leaves, while the expression in the stems was the lowest. Upon treatment with GA₃, ETH, and ABA, the expression of CsCYP82D102 decreased to varying extents compared to control plants. Conversely, exogenous SA and MeJA treatments resulted in an upregulation of CsCYP82D102 expression. Specifically, SA treatment remarkably increased the relative expression level of CsCYP82D102 by more than 1633-fold at 48 h. MeJA treatment slightly elevated the relative expression of CsCYP82D102 by more than 238-fold at 24 h (Figs. 5A–5H). These findings suggest that CsCYP82D102 may participate in the SA and MeJA signaling pathways.

Overexpression of CsCYP82D102 in cucumber enhances resistance to PM

To investigate the function of the P450 family gene, CsCYP82D102, in cucumbers, transgenic plants overexpressing this gene under the control of the 35S promoter were generated. Two independent T1 generations were selected for functional characterization. Genomic PCR analysis confirmed the integration of the transgene, while expression analysis verified its transcription (Figs. S4A–S4C). Subcellular localization showed that CYP82D102 was located in the endoplasmic reticulum, consistent with the prediction (Fig. 5I). Wild-type(WT) leaves exhibited average lesion areas of approximately 29.7% at 7 dpi upon powdery mildew inoculation. In contrast, the two independent CsCYP82D102 overexpression lines displayed significantly reduced necrotic areas, with average lesion areas of approximately 13.4% and 5.7%. At 15 dpi, the lesion area of WT leave was 70.8%. The lesion area of OE#1 and OE#3 was 32.2% and 27% respectively (Fig. 6B). Subsequently, the levels of reactive oxygen species (ROS) were evaluated in transgenic plants. Notably, WT plant leaves exhibited prominent staining for H₂O₂ and O₂^-, whereas leaves of CsCYP82D102 overexpressing plants displayed relatively weaker DAB and NBT staining (Fig. 6C). Furthermore, under PM stress conditions, all transgenic plants showed enhanced activity of POD, CAT, and SOD compared to WT plants (Fig. 6D). In silent lines, the difference was not significant (Fig. S5). These results indicate that CsCYP82D102 enhances resistance against powdery mildew.

CsCYP82D102 is involved in the SA/MeJA signaling pathway

Considering the changes in resistance levels to powdery mildew (PM) infection, the expression pattern of defense-related marker genes in the defense cascade was examined in CsCYP82D102 overexpression plants. Insights into potential regulatory pathways were aimed to be gained (Fig. 6E). Four widely recognized pathogenesis-related (PR) genes were selected for comparative analysis. Both the SA marker genes, PR1 (pathogenesis-related protein 1) and PR2 (pathogenesis-related protein 2), exhibited significantly lower expression in CsCYP82D102 overexpression plants relative to WT plants, both before and after infection. In contrast, the basal expression of PR5, a gene responsive to JA, was notably higher in the overexpression lines compared to WT plants. Following PM infection, both PR3 and PR4 expression substantially increased, but with lower levels detected in the overexpression plants. These findings suggest that the upregulation of CsCYP82D102 expression in cucumber leads increases the expression of PR5, a key component involved in SA-mediated defense signal transduction. Additionally, the enhanced resistance to powdery mildew observed in the transgenic cucumber plants may be attributed to the upregulation of these genes.

To further elucidate the impact of CsCYP82D102 on signaling transduction, the expression of key regulators within the SA/MeJA signaling pathways was examined. Among them, LOX2 (lipoxygenase 2), which is involved in JA biosynthesis and signaling pathway, exhibited intensified induction in the overexpression plants following infection. JAR1 (jasmonate resistant 1), responsible for JA conjugation to isoleucine and the production of biologically active jasmonate-isoleucine (JA-Ile), displayed significantly higher expression levels in the overexpression plants after infection. Moreover, the expression of defense-marker genes, such as CAT2 and APX, was also modulated in PM-infected plants.