Kiwifruit is a functional dioecious plant. Male plants only form male flowers and female plants only form female flowers. The pistil on the male flower is not developed, and the pollen on the female flower is completely aborted (Caporali et al., 2019). The female plants used in this study were Psa high-resistant variety ‘Jinkui’, and the male plants were 3 materials screened by our team with high resistance, medium resistance and high sensitivity to Psa, respectively. For artificial hybridization, the pollen of male plants was collected first, and then the female plants were artificially pollinated and bagged. Once the fruit had ripened, the seeds were collected, treated, and then sown to obtain the hybrid. The six hybrid populations of this study were constructed in 2018, with a total of 4351 plants. JA, JB and JM were three BC₁ hybrid populations, and A₁F₂, D₁F₂ and R₁F₂ were three F₂ populations. The population materials were planted in the kiwifruit base of Chibi Shenshan Xingnong Technology Co. LTD. and grew well.

The field resistance of six populations was investigated in April 2022 according to the reported method (Wang et al., 2019) with a few modifications. The disease resistance grade was simplified from five (Highly Resistant (HR), Resistant (R), Moderate (M), Susceptible (S) and Highly Susceptible (HS)) to three (HR, M and HS). In brief, plants with no visible lesions on the leaves are HR plants; plants with some lesions on the leaves are M plants, and plants with many lesions on the leaves and almost all leaves have lesions are HS plants.

The pathogen Psa (C48) was cultured in LB liquid medium at 25°C with shaking at 200 rpm for 24 hours. Then the bacteria were collected by centrifugation, and the concentration of Psa bacteria was diluted to 1×10⁸ cfu/mL with sterile water for infection. After the kiwifruit leaves fall in winter, six healthy kiwifruit branches with a diameter of about 0.8 cm were collected from each plant. According to the reported method (Wang et al., 2019, Wang et al., 2023), the branches were washed with running water and then were cut into 10 cm sections. The ends of the branches were sealed with paraffin to reduce the loss of water. Next, the branches were disinfected with 75% alcohol in the ultra-clean bench. After the alcohol volatilizes, punching was performed with a puncher at a distance of about 1.5 cm from the end of the branch to remove the epidermis. The 10 μL diluted Psa solution was inoculated at the wound for infection, and sterile water was inoculated as a control. Each plant was infected with six branches and repeated three times. Once the bacterial liquid had been completely absorbed in the wound, the branches were placed in a tray covered with soaked filter paper. The tray was placed in an artificial climate chamber and cultured at 20°C, 12 h light/12 h dark, relative humidity of 80%. The incision symptoms were observed every three days, and the lesion length was measured at 40 days after inoculation.

In the R₁F₂ population, nine extremely HR plants and nine extremely HS plants to Psa in field investigation and in vitro branch identification were screened for transcriptome sequencing. Leaf samples of new shoots were taken, frozen in liquid nitrogen and stored at -80°C for later use. Total RNA was extracted using the Total RNA Extraction Kit (Aidlab, Beijing, China) according to the manufacturer’s instructions. The purity, concentration and integrity of RNA samples were detected by NanoDrop, Qubit 2.0 and Agilent 2100. Qualified RNA of each three extremely high-resistant plants was mixed into one sample, and finally a total of three high-resistant samples were obtained, namely HR1, HR2, and HR3. Similarly, the RNA of each three extremely high-susceptible plants was mixed into one sample to obtain HS1, HS2 and HS3. Then, the RNA was sequenced by the Illumina platform in BioMac Biotechnology Co., Ltd. The sequencing reads were assembled and mapped to the kiwifruit reference genome (http://kiwifruitgenome.org) using StringTie and HISAT software (Rao, 2024; Thakur, 2024). Applying DESeq2, differentially expressed genes (DEGs) were annotated by calculating fragments per kilo-base of transcripts (FPKM) with fold change ≥ 2 and FDR < 0.01, and KEGG enrichment analysis was performed (Liu et al., 2021). TFs in DEGs were screened and classified according to gene annotation. Heat maps of DEGs were visualized by the HeatMap Illustrator tool of TBtools (Chen et al., 2023).

The total RNA of three HR and three HS plants from the R₁F₂ populations was extracted using the Total RNA Extraction Kit (Aidlab, Beijing, China) according to the manufacturer’s instructions. The quality and concentration of the RNA were detected by agarose gel electrophoresis and Nanodrop spectrophotometer. Reverse transcription was performed using TRUEscript RT MasterMix (OneStep gDNA Removal) (Aidlab, Beijing, China). The cDNA was diluted to 300 ng/μL as a template for qRT-PCR. Primers used in this study are listed in Supplementary Table S6 . The qRT-PCR was performed according to previous reports (Ye et al., 2022). In brief, the total volume of the qRT-PCR mixtures was 10 µL, which included 5 µL of SYBR qPCR Master Mix (Takara, Japan), 0.2 µL of each primer (10 µM), 0.4 µL of cDNA, and 4.2 µL of RNase-free water. The ABI 7500 Sequence Detection System was employed to conduct the reactions under the following thermocycling parameters: an initial enzyme activation step at 95°C for 5 minutes, followed by 40 cycles comprising 10 seconds at 95°C and 20 seconds at 60°C. After the reaction was completed, a melting curve analysis was performed. The 2^−ΔΔCT method was used to calculate the relative gene expression of genes. The AcActin gene (Acc05529.1) was used for the normalization of qRT-PCR data.

AcWRKY75 protein sequence was used for protein blast on NCBI, and the top 50 protein sequences ( Supplementary Table S7 ) with high homology to AcWRKY75 protein were downloaded. These sequences were aligned using the ClustalW multiple method in BioEdit software (Sofi et al., 2022), and then a phylogenetic tree was constructed in MEGA 11 using the maximum likelihood method (Tamura et al., 2021). Twelve WRKY75 proteins with the highest similarity were selected for multiple sequence alignment to analyze the variation of conserved domains. The WRKY domain has a conserved WRKYGQK sequence, and the C2H2 domain comprises CX_4-5-C-X_22-23-H-X₁-H (Yin et al., 2024). According to these characteristics, the location of the domains in the WRKY75 protein was marked. To analyze the variation of the AcWRKY75 gene in HR and HS plants of the R₁F₂ population, the new shoot leaves of three HR plants and three HS plants were sampled, and genomic DNA was extracted following the manual of a modified CTAB Plant DNA Kit (Aidlab, Beijing, China). Then, the DNA was diluted to 300 ng/μL as the amplification template, and the AcWRKY75 gene was amplified by PhantaMax Super-Fidelity DNA Polymerase (Vazyme, Nanjing, China). The target fragment was cloned into the pTOPO-Blunt vector using a Cloning Kit (Aidlab, Beijing, China) and transferred into E.coli DH5α. Single clones containing the target fragments were sent to the company for sequencing. Finally, the sequences were subject to multiple sequence alignment using BioEdit software (Sofi et al., 2022).

Subcellular localization analysis was performed as previously described (Ye et al., 2021). Briefly, the coding sequences (CDS) of the AcWRKY75 gene was amplified, and the part without termination codon was cloned into the pBI121-GFP vector. Then the recombinant vector was transferred into A. tumefaciens strain EHA105 cells. Next, the A. tumefaciens cells were subject to co-infiltration to N. benthamiana leaves. Afterward, the N. benthamiana was cultured in dark condition and light condition for one day, respectively. Finally, the leaves were collected for fluorescence signal detection under a laser confocal microscope (TCS-SP8, Leica, Germany). A red fluorescent protein (RFP) marker was used to indicate the nucleus location by cotransfecting, and the 35S-GFP vector, which emits fluorescence in both the cell membrane and nucleus, was used as a positive control.

To assess the transcriptional activation activity of AcWRKY75, the full-length CDS of the AcWRKY75 gene was amplified and inserted into the pGBKT7 vector, resulting in a fusion with the GAL4-BD domain. The empty pGBKT7 vector and the recombinant plasmid BD-AcWRKY75 were transformed into the yeast AH109 strain and cultured on SD/-Trp medium at 30°C for three days. Following this, the optical density at 600 nm (OD₆₀₀) of the positive clone was diluted to 10⁰,10^-1,10^-2 using 0.9% NaCl solution, and 10 µL of the bacterial solution was spotted onto SD/-Trp, SD/-Trp-His-Ade, and SD/-Trp-His-Ade supplemented with X-α-gal media. After three days of incubation at 30°C, the transactivation activity of the transformants was evaluated based on their growth patterns. Positive clones exhibited growth on all media and turned blue in the SD/-Trp-His-Ade medium containing X-α-gal, while the control grew only on SD/-Trp, indicating that the positive clones possessed transcriptional activation activity.

The vectors PBI121 and TRV1/2 were used for overexpression and virus-induced gene silencing (VIGS), respectively. The recombinant plasmids were separately transferred into A. tumefaciens strain EHA105. Transient overexpression in kiwifruit leaves was performed using a previously described method (Lu et al., 2024). Briefly, the OD₆₀₀ of A. tumefaciens infection solution was adjusted to 0.8 with MES buffer (10 mM MgCl₂, 10 mM MES, 150 mM AS, pH 5.6), and then injected into the leaves of kiwifruit variety ‘Jinyi’. Each infection solution was injected into the same area of different regions. The infected plants were cultured in the dark for one day and then cultured at 25°C with a 16h light/8h dark photoperiod for another day. Then, the leaves were collected and placed in -80°C for the AcWRKY75 gene expression level analysis. Two days post-injection, the Psa solution was diluted with sterile water containing 10 m M Mg Cl2 to OD600 = 0.5 and then added with 0.03% Silwet L-77. The infection site was then brushed twice with this bacterial solution. Afterward, the plants were promptly positioned beneath a plastic dome at 20°C, maintaining a photoperiod of 16 hours of light followed by 8 hours of darkness. Two weeks post-injection, images of the infected leaves were collected and the disease index was calculated using ImageJ (V1.8.0). The number of Psa in the infected leaves was quantified according to a reported described method (Zhao et al., 2024). Briefly, 0.1 g leaves were transferred to a test tube containing 2.0 ml of 0.9% sodium chloride and gently ground with quartz sand. Subsequently, the suspension was subjected to a series of dilutions (10⁰, 10⁻¹, 10⁻² and 10⁻³), and 100 µL of each sample was taken and cultured in LB medium at 28°C for 72 h. Once the colony appeared, the number of bacteria was counted. Assays were repeated at least three times for each infected sample.

The 2 kb promoter fragments of 14 ABA-related Bet v 1 family genes were isolated from the A. chinensis genome (http://kiwifruitgenome.org) using TBtools (Chen et al., 2023), and the W-box was found. The AcBet V1 (Acc27163) gene promoter contains four W-box (GGTCAA) elements. The W-box (-230 bp to-236 bp) closest to the start codon was ligated into the pAbAi vector, and the AcWRKY75 coding region was ligated into the PGADT7 vector. The bait vector pAcBet V1-AbAi was transformed into the yeast Y1H-Gold strain and tested self-activation with SD/-Ura medium under different concentrations of AbA (0, 100, 300, 500, 800 ng/mL). Yeast transformation was performed according to the manual of the Y1HGold Yeast One-hybrid System kit (Coolaber, Beijing, China). The pGADT7-p53+p53-AbAi was a positive control, and the pGADT7+pAcBet V1-AbAi was a negative control.

To obtain different resistance materials of Psa, we constructed six hybrid populations with Psa high resistance variety ‘Jinkui’ as the female parent and varieties with different Psa resistance levels as the male parents. To screen out the Psa resistance segregating populations, the resistance of lignified branches in each population was identified. The results showed that the resistance segregation of JA, JB, A₁F₂ and D₁F₂ populations was not obvious, and most plants were HR. While JM and R₁F₂ populations were separated and R₁F₂ population had the most obvious plant resistance separation ( Supplementary Figure S1 ). The disease resistance of the six populations to Psa in the field was further investigated, and the disease resistance was divided into high resistance (HR), medium resistance (M) and high sensitivity (HS) according to the disease spots on the leaves ( Figure 1A ). The statistical results indicated a clear separation in disease resistance within the R₁F₂ population ( Supplementary Figure S2 ), which aligned with the identification results of branch resistance. These findings suggest that the R₁F₂ population had the most obvious Psa resistance segregation among the six populations, making it the most suitable population for mapping canker disease resistance genes.

To further screen the individual plants with extremely HR and HS to Psa from the R₁F₂ population, the individual plants showing HR and HS to Psa in the field were re-evaluated in isolated branches ( Figure 1B ). The HR variety ‘Jinkui’ was used as the HR-positive control, and the HS variety ‘Hort 16A’ was used as the HS-negative control. Based on the lesion length, the plants of the R₁F₂ population were classified into HR, MR and HS. For example, R₁F₂-319 is identified as an extremely HR plant, and R₁F₂-408 is identified as an extremely HS plant ( Figure 1B ). Nine plants exhibiting extremely HR were selected to construct HR pools, and nine plants exhibiting extremely HS were also chosen to construct HS pools for transcriptome sequencing.

To screen the Psa resistance-related genes of kiwifruit, transcriptome sequencing was performed on the Psa extremely HR and HS plants screened from the R₁F₂ population. A total of 39.99 Gb Clean Data was obtained from six samples. The Clean Data of each sample reached 6.04 Gb, and the percentage of Q30 bases was above 90% ( Supplementary Table S1 ). The Clean Reads of each sample were mapped to the reference genome A. chinensis Red 5. The mapping rates ranged from 74.39% to 83.00%; the uniquely mapped rate ranged from 71.62% to 79.77%. The sequence GC content was above 45% ( Supplementary Table S1 ). These results indicated that the sequencing data of the six samples were of high quality and could be used for downstream analysis. Using false discovery rate (FDR) < 0.01 and Fold Change ≥ 2 as the screening criteria, a total of 581 DEGs were screened. Based on the functional annotation information, we annotated 563 DEGs, among which 403 were up-regulated and 160 were down-regulated ( Figures 2A, B ; Supplementary Table S2 ). Compared with HR plants, most of the DEGs in HS plants were up-regulated. To further analyze the function of DEGs, we performed a KEGG enrichment analysis. The top 20 enriched pathways contained 189 DEGs ( Figure 2C ; Supplementary Table S3 ). Among them, the highest enriched pathways were the plant hormone signal transduction (32 genes, 13.06%) and the plant-bacteria interaction pathway (31 genes, 12.65%) ( Figure 2C ; Supplementary Table S3 ). These results indicated that the genes related to plant hormone signal transduction might play important roles in the Psa resistance of kiwifruit.

Gene classification of the DEGs showed that there were 71 TFs, among which the TF families such as WRKY (12 genes), MYB (12 genes) and ZIP (10 genes) were highly enriched ( Figure 3A ; Supplementary Table S4 ). It is worth noting that all the AcWRKY TFs were up-regulated in HS plants ( Figure 3A ), indicating that WRKY TFs may play an important role in Psa resistance of kiwifruit. Expression heat map analysis showed that AcWRKY18, AcWRKY75 and AcWRKY98 genes were significantly up-regulated ( Figure 3B ). ‘Hongyang’ is a highly susceptible variety of kiwifruit to Psa. Previously, we performed transcriptome analysis on ‘Hongyang’ at 0, 1, 12 and 48 hours post-inoculation (hpi) with Psa. The results showed that 10 of the 12 AcWRKY TFs were significantly up-regulated in 1 hpi, and the rest two displayed no significant differences in the expression levels. In 12 hpi, four AcWRKY TFs were up-regulated and 2 were down-regulated. In 48 hpi, six AcWRKY TFs were up-regulated, and the remaining six showed no obvious differences ( Figure 3C ). Among them, AcWRKY70, AcWRKY18 and AcWRKY75 were significantly up-regulated at 1, 12 and 48 hpi, and AcWRKY18 and AcWRKY75 showed the highest changes in the expression levels ( Figure 3C ). These results indicate that AcWRKY18 and AcWRKY75 genes may play an important regulatory role in the process of kiwifruit resistance to Psa.

To further confirm the induced expression of AcWRKY TFs by Psa, we performed qRT-PCR analysis of 12 up-regulated WRKY TFs in 3 extremely HR and 3 extremely HS plants of the R₁F₂ population. The results showed that 10 of the 12 WRKY TFs were highly expressed in HS plants, which was consistent with the transcriptome results ( Figure 4 ). Importantly, we found that AcWRKY75 was most significantly up-regulated, and its expression level in HS plants was more than 20 times higher than that in HR plants ( Figure 4 ). Therefore, it can be speculated that AcWRKY75 is induced by Psa and plays an important role in the process of Psa infection.

To study the evolutionary characteristics of AcWRKY75, homologous phylogenetic trees were constructed in different species ( Supplementary Figure S3 ). WRKY75 protein is conserved in different species of Actinidia, and is closely related to plants of Juglans and Camellia ( Supplementary Figure S3 ). Multiple sequence alignment showed that AcWRKY75 is a group II WRKY transcription factor with a conserved WRKY domain and a zinc finger motif (C2H2) ( Figure 5A ). In addition, the C-terminus sequences of AcWRKY75 homologous protein are very conserved, while the N-terminal sequences harbor most of the variations ( Figure 5A ). Furthermore, to analyze the sequence polymorphism, we cloned and sequenced the genomic DNA of AcWRKY75 gene in 3 HR plants and 3 HS plants of the R₁F₂ population. The results showed that the DNA sequence of AcWRKY75 gene has obvious polymorphism, and there are many fragment deletions and single nucleotide mutations ( Supplementary Figure S4 ). Most of these mutations are located in introns, and only one single base deletion is located in the second exon ( Supplementary Figure S4 ), which causes a delay in translation termination and an extra amino acid ( Figure 5B ). However, these variations were present in both HR and HS, and no HR or HS-specific variations were found ( Figure 5B ). Whether these mutations affect the function of AcWRKY75 gene remains to be further analyzed.

Transcriptome and qRT-PCR results have preliminarily confirmed that AcWRKY75 gene was induced by Psa. To further explore the characteristics of AcWRKY75 gene, we performed a subcellular localization analysis in N. benthamiana. The fluorescence signal showed that the AcWRKY75 protein was located in the nucleus ( Figure 6A ). In addition, we examined the transcriptional activity of AcWRKY75 gene. The full-length CDS of the AcWRKY75 gene was fused with the GAL4 DNA-binding domain, and the BD empty vector was used as a negative control. Only yeast cells harboring BD-AcWRKY75 grew well on the SD-Trp/-His/-Ade plate and catalyzed X-α-Gal into blue ( Figure 6B ). These results indicate that AcWRKY75 is a typical transcription factor with transcriptional activation activity. It may bind to the Psa resistance-related gene promoter to regulate its expression, thereby regulating the resistance of kiwifruit to Psa.

To better understand the role of AcWRKY75 in Psa infection, we constructed overexpression and virus-induced gene silencing (VIGS) vectors, and then performed transient transformation in kiwifruit leaves, followed by Psa inoculation ( Figure 7A ). Two days post-transformation, we employed qRT-PCR to verify that the AcWRKY75 gene was significantly up-regulated in the overexpression region and down-regulated in the VIGS region ( Figure 7B ). Two weeks after Psa infection, the overexpression of AcWRKY75 resulted in a significant increase in the symptoms observed in kiwifruit leaves, whereas silencing AcWRKY75 via VIGS led to a reduction in these symptoms ( Figures 7A, C ). Subsequently, we used the plate coating method to count the bacteria on the leaves of the control and different treatments. It found that AcWRKY75 overexpressing leaves showed the highest amounts of bacteria, which was significantly higher than that of water control ( Figure 7D ). However, there were no significant differences between Psa-treated control leaves and AcWRKY75 overexpressed leaves. It may be due to the number of Psa bacteria reaching the upper limit of growth. The amounts of bacteria in the leaves silently expressing the AcWRKY75 gene were lower than that of the Psa-treated control ( Figure 7D ), which was consistent with the statistical analysis of the lesion areas of kiwifruit leaves. These results indicated that AcWRKY75 can reduce the ability of kiwifruit leaves to resist Psa.

It has been reported that hormones are closely related to plant disease resistance (Pieterse et al., 2012). In this study, KEGG enrichment analysis found that DEGs were most significantly enriched in plant hormone signal transduction pathways ( Figure 2C ). Further analysis showed that the DEGs were mainly related to the metabolism of hormones such as auxin (14 genes), abscisic acid (21 genes), zeatin (4 genes), jasmonic acid (4 genes) and salicylic acid (1 gene) ( Supplementary Table S5 ). Expression heat map showed that most genes were up-regulated in HS materials. For example, among the 21 ABA-related DEGs, only four were down-regulated and the rest were up-regulated; among the 14 auxin-related DEGs, 10 were up-regulated ( Figure 8 ). It is worth noting that 14 ABA-related genes belong to the Pathogenesis-related protein Bet V1 family ( Supplementary Table S5 ), indicating that the Bet V1 family proteins play an important role in the infection of Psa in kiwifruit. Promoter response element analysis showed that the Bet V1 family gene’s promoter contained multiple WRKY TF binding sites W-box, and the AcBet V1 (Acc27163) gene promoter contained four W-box ( Figure 9A ). Furthermore, the AcBet V1 gene was highly expressed in HS plants, similar to AcWRKY75 ( Supplementary Table S2 ). Therefore, we speculate that AcWRKY75 may bind to the AcBet V1 gene promoter to regulate the Psa resistance of kiwifruit. Here, we cloned the W-box elements (-230 bp to -236 bp) closest to the start codon on the AcBet V1 gene promoter and ligated into the pAbAi vector, and cloned the AcWRKY75 coding region into the PGADT7 vector for yeast one-hybrid interaction verification. In SD-Ura/-Leu+AbA200 medium, the yeast of pAcBet V1-AbAi+pGADT7-AcWRKY75 grew slightly stronger than pGADT7 empty vector control, indicating that AcWRKY75 can weakly interact with the promoter of AcBet V1 ( Figure 9B ). These results suggest that AcWRKY75 may bind to the promoter of ABA-related AcBet V1 gene and promote its expression, thus regulating the Psa resistance in kiwifruit.