2026/04/27 by Sen Li, Shuai Yin, Tiantian Pei +10 · 1 voice
Agricultural and Biological Sciences · Biochemistry, Genetics and Molecular Biology · #Chromosomal and Genetic Variations #Plant Molecular Biology Research #Plant tissue culture and regeneration
paper · doi:10.1111/jipb.70276
openalex publication_date 2026/04/27 · openalex created_date 2026/04/28 · openalex updated_date 2026/07/28
Editing the cucumber peroxidase gene CsPOD7 establishes a haploid inducer line capable of inducing maternal haploids. Haploid induction (HI), an essential component of doubled haploid (DH) breeding, has gained significant attention for the genetic improvement of crops due to its unique capacity to fix recombinant haplotypes within just two generations while substantially accelerating breeding cycles, providing distinct advantages over conventional crossbreeding methods (Li et al., 2025). Although haploid plants can be generated through in vitro culture of microspores or anthers and megaspores or ovules, this method suffers from low efficiency and is technically challenging, limiting its widespread application. By contrast, in vivo techniques for HI have improved considerably in recent years. Several key genes have been successfully used to induce haploid production in maize (Zea mays), such as MATRILINEAL (MTL)/PHOSPHOLIPASE A1 (PLA1)/NOT LIKE DAD (NLD), CENTROMERIC HISTONE 3 (CENH3), DOMAIN OF UNKNOWN FUNCTION 679 MEMBRANE PROTEIN (DMP), and the peroxidase gene ZmPOD65 (Ravi and Chan, 2010; Kelliher et al., 2017; Zhong et al., 2019; Jiang et al., 2022). Among these, mutations in DMP homologs have triggered maternal HI in both monocot and dicot species, including maize, cucumber (Cucumis sativus), Arabidopsis (Arabidopsis thaliana), tomato (Solanum lycopersicum), watermelon (Citrullus lanatus), and Brassica napus (Zhong et al., 2019; Li et al., 2022; Zhong et al., 2022; Chen et al., 2023; Tian et al., 2023; Yin et al., 2023; Li et al., 2025). However, MTL/PLA1/NLD is conserved only in monocots, and CENH3-mediated HI has been successful in only a limited number of crops (Ravi and Chan, 2010; Li et al., 2025), thereby greatly restricting the development and application of DH technology in dicots. By contrast, ZmPOD65, which has been used to trigger in vivo maternal HI in maize, is also highly conserved in dicots. It remains unclear whether mutations in ZmPOD65 homologs could be used for HI in dicots. Here, we identified five genes homologous to ZmPOD65 in the cucumber genome. Among these, CsPOD7 (CsaV31G008260) is most similar to ZmPOD65 (Figure S1); their encoded proteins share 59.55% amino acid sequence identity (Table S1). RT-qPCR indicated that CsPOD7 is specifically expressed at extremely high levels in mature anthers (Figure S2). To generate CsPOD7-edited plants, we designed a clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated nuclease 9 (Cas9) construct using the pKSE402 vector as a backbone. This construct expresses two single guide RNAs (sgRNAs) targeting the first exon of CsPOD7 (Figure 1A). The construct also contains a separate enhanced green fluorescent protein (EGFP) marker gene expression cassette driven by the cauliflower mosaic virus (CaMV) 35S promoter to facilitate subsequent haploid seed identification based on green fluorescence (Yin et al., 2023). Mutating CsPOD7 creates a haploid inducer line in cucumber (A) Diagram of the CRISPR/Cas9 mutagenesis vector targeting CsPOD7. (B) Diagram of the CsPOD7 genomic locus, showing the positions of the sgRNAs used to target CsPOD7 and the sequences of the WT and mutants. Gray blocks, coding regions; red lines, target regions (T1, T2). The target sequence is shown in black, substitutions are highlighted in blue, and the protospacer adjacent motif (PAM) is shown in red. (C) Table summarizing the WT sequence and the mutations in two representative plants edited in the CsPOD7 target region. (D) Representative photographs of fruits and seeds from self-pollinated WT and Cspod7-1 plants. White arrowhead indicates a normal seed; red arrowhead indicates an aborted seed. (E, F) Seed number (E) and seed phenotypes (F) of WT and Cspod7 (Cspod7-1 and Cspod7-2) fruits (n = 35). Values are means ± standard deviation (SD). The significance of differences was analyzed by two-tailed Student's t tests (***P < 0.001). (G) Representative photographs of seeds from diploid (D) and putative haploid (H) plants viewed under fluorescent light. +, positive control; −, negative control. (H) Ploidy analysis by flow cytometry. (I) HIR of WT and Cspod7 (Cspod7-1 and Cspod7-2) plants obtained by selfing or crossing. (J–O) Representative photographs of diploid and haploid plants showing different organs. Scale bars, 1 cm (G), 1.5 cm (M–O), 5 cm (D, J, L), 10 cm (K). We generated two homozygous Cspod7 mutants in the cucumber cultivar “changchunmici” (wild type, WT) background (Figure 1B). One mutant carried a 12-bp substitution, resulting in three non-synonymous amino acid changes and one synonymous change. The other mutant carried a 1-bp substitution, resulting in a single non-synonymous amino acid change (Figure 1C). We detected significantly lower peroxidase activity in Cspod7 compared to WT plants (Figure S3). Additionally, off-target analysis of the CsPOD7 target sites revealed no off-target events (Figure S4). These findings confirm the reliability of CsPOD7 editing. Pollen viability was similar between the WT and Cspod7 mutant. However, the pollen germination rate decreased by approximately 36% in the mutant, accompanied by significantly fewer seeds per fruit, a pronounced drop in the proportion of filled seeds, and a markedly higher proportion of aborted seeds (Figures 1D–F, S5; Tables S2, S3). To evaluate the potential use of Cspod7 mutants for HI, we self-pollinated both WT and Cspod7-1 plants and analyzed the ploidy levels of their progeny using flow cytometry. Three haploid individuals were identified among the 1,739 progeny of self-pollinated Cspod7-1 plants (Figure 1H), whereas no haploids were identified among the progeny of self-pollinated WT plants (Figure 1I). These results indicate that mutating CsPOD7 can induce haploid embryo production in cucumber, providing the first experimental evidence that mutations in a ZmPOD65 homolog can facilitate HI in dicots. To further investigate whether mutating CsPOD7 can induce maternal haploid production in cucumber, we identified T2 plants that were homozygous for both the mutant site at CsPOD7 and the genomic region of the EGFP marker insertion (Yin et al., 2023). We used these plants as the male parents in crosses to seven cucumber accessions with diverse genetic backgrounds: two cultivated varieties (‘CCMC’ and ‘XTMC’) and five inbred lines (Figure 1I; Table S4). Considering that maternal HI mediated by ZmPOD65 inactivation is caused by sperm cell damage and subsequent DNA fragmentation in Zmpod65 mutants, we used green fluorescence as a marker to screen for maternal haploids among the hybrid progeny (Yin et al., 2023). We classified progeny lacking green fluorescence as putative haploids, as confirmed by flow cytometry (Figure 1G, H). The average HI rate (HIR) ranged from 0.16% to 0.2% (Figure 1I). The haploid plants were morphologically similar to diploid plants but were shorter and had smaller leaves and organs (Figure 1J–O). Furthermore, no haploids were detected in the hybrid progeny when the Cspod7 mutant was used as the female parent in crosses to CCMC, XTMC, or line 3407; in crosses of CCMC and XTMC to transgenic lines complemented with a proCsPOD7:CsPOD7 construct; or in control crosses (XTMC × WT and 3407 × WT) (Figures 1I, S6). To confirm their maternal origin, we sequenced the whole genomes of two haploid seedlings. These seedlings carried no paternal single-nucleotide polymorphism (SNP), indicating that crossing with Cspod7 induced maternal haploid production (Figure S7). In addition, when we treated haploid seeds containing 0.5 cm to 1 cm-long radicles with 0.1% (m/v) colchicine and 5% (v/v) DMSO to induce chromosome doubling, DH lines were generated. These results indicate that the Cspod7 mutant can induce maternal haploids in vivo, genotype-independently, in the dicot species cucumber. In summary, we demonstrated that mutating CsPOD7 can induce maternal haploid formation in cucumber, highlighting the feasibility of maternal HI in dicots through mutations in ZmPOD65 homologs. This breakthrough provides a valuable alternative to DMP-based methods for dicots for which DH production remains challenging, significantly expanding the toolbox for HI in dicots. We plan to further validate the feasibility of HI by mutating ZmPOD65 homologs in additional dicot species, such as Arabidopsis. Although the current HIRs obtained by editing CsDMP and CsPOD7 in cucumber are relatively low, we plan to use the Csdmp Cspod7 double mutant or perform combinatorial knockout of CsPOD7 along with other putatively redundant POD family members to determine whether higher-order mutations can further enhance the HIR. This study was supported by the National Natural Science Foundation of China (32573049, 3202013014), the 2115 Talent Development Program of China Agricultural University, and the 111 Project of China Agricultural University. The authors declare no conflicts of interest. S.L., S.Y., X.L., and H.R. conceived and designed the work. S.L. and S.Y. performed the experiments. S.L., T.P., Y.S., L.S., Y.D., and M.A. analyzed the data. S.L. wrote the manuscript. S.L., S.Y., Y.D., X.Z., L.S., S.F., L.Y., X.L., and H.R. revised the manuscript. All authors have read and approved the final manuscript. Additional Supporting Information may be found online in the supporting information tab for this article: http://onlinelibrary.wiley.com/doi/10.1111/jipb.70276/suppinfo Figure S1. Phylogenetic analysis of cucumber peroxidase genes and ZmPOD65 based on protein sequence alignment. Figure S2. Expression analysis of ZmPOD65 homologs in cucumber. Figure S3. Determination of peroxidase activity in WT and Cspod7 mutants. Figure S4. Off-target analysis of CsPOD7. Figure S5. Analysis of pollen viability and germination in Cspod7 mutants. Figure S6. Expression analysis of CsPOD7 in WT and pCsPOD7:CsPOD7 complementation lines. Figure S7. Recombination map of cucumber haploids. Table S1. CsPOD-like genes in cucumber. Table S2. Seed number of WT and Cspod7 in cucumber. Table S3. Seed phenotypes of WT and Cspod7 in cucumber. Table S4. The information on haploids from different crosses Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.