Generation and characterization of rice OsCENH3 mutants for haploid induction

Abstract

CRISPR/Cas9-mediated modification of OsCENH3 induces aneuploidy but fails to trigger haploid formation in rice, underscoring limited efficiency and the need for strategy refinement. The centromeric histone H3 variant (CENH3) is essential for kinetochore assembly and accurate chromosome segregation during cell division. Alterations in CENH3 have been shown to trigger haploid induction in various plant species; however, its utility in rice remains largely unexplored. In this study, six OsCENH3 mutant lines were generated through CRISPR/Cas9-mediated genome editing, complemented by a GFP-tagged construct. To assess their haploid induction potential, both self pollination and outcrossing with the cultivar Nipponbare (NIP) were performed. Flow cytometry analysis revealed that none of the mutants produced true haploids. However, two aneuploid individuals (Het-C1-1-1 and Het-C1-5-1) were identified among the hybrid progeny derived from tail domain mutants C1-1 and C1-5, suggesting partial chromosome missegregation. These aneuploid plants exhibited reduced pollen viability, abnormal morphology, and compromised agronomic performance. In addition, significantly elevated rates of embryo/endosperm abortion were observed across different crosses, which far exceeded the frequency of aneuploid production, implying that early-stage chromosomal instability may result in embryo lethality. Collectively, while specific OsCENH3 mutations can induce limited chromosomal instability, their haploid induction efficiency remains low in rice. Further refinement of editing strategies and exploration of favorable genetic backgrounds will be essential for developing effective centromere-based haploid induction systems. This study provides a theoretical framework and technical reference for engineering haploid inducers in rice through centromere manipulation.

Keywords: Aneuploidy; CENH3; CRISPR/Cas9; Haploid induction; Rice.

Fig. 1

Construction of a knockout and fusion-complementation co-expression vector and tertiary structure prediction of mutant proteins. a Schematic representation of the CRISPR/Cas9 target sites within the OsCENH3 gene in rice; b Diagram of the binary vector used for CRISPR/Cas9-mediated gene editing; c Schematic illustration of the fusion-complementation co-expression vector incorporating OsCENH3, GFP, and OsH3.3 domains; d Predicted tertiary structures of OsCENH3 proteins in WT and selected mutants, modeled using the SWISS-MODEL platform

Fig. 2

Expression profiling of OsCENH3 in various rice tissues and expression analysis of mutants. a Expression pattern of OsCENH3 across rice tissues based on ePlant database visualization; b-d Relative OsCENH3 transcript levels measured by qRT-PCR in young panicles (booting stages 4–6), root tips (6–8-day-old seedlings), and mature seeds of different mutant groups: b Domain mutants (C1-1, C1-5, C1-7), transcript levels represent endogenous OsCENH3 expression; c Promoter mutants (C2-1, C2-2), transcript levels represent endogenous OsCENH3 expression; d Fusion-complemented mutant line C3-7, qRT-PCR primers specifically detect transcripts from the GFP-tailswap transgene and do not amplify endogenous OsCENH3. Control samples were taken as calibrator 1. Values represent the mean ± standard error (SE) (n = 3). The asterisks represent significant differences based on the results of the LSD test (ns, not significant; *P < 0.05; **P < 0.01; ***P < 0.001)

Fig. 3

Agronomic traits and pollen fertility of OsCENH3 mutants. a Plant architecture and pollen fertility of OsCENH3 mutants (scale bar for plant architecture = 16 cm; scale bar for pollen fertility = 200 μm); b Seed setting rate of the OsCENH3 mutants compared to WT

Fig. 4

Identification of aneuploids and analysis of embryo/endosperm abortion rates. a Flow cytometry analysis of haploid plants. Diploid plants exhibited a peak at 200, while aneuploid plants displayed a peak between 100 and 200; b Aneuploid plants showed shorter plant height and reduced panicle length (scale bar = 2 cm); c Aneuploid plants exhibited significantly reduced fertility (scale bar = 200 μm); d Frequency of embryo/endosperm abortion. Values are presented as mean ± standard deviation. Different lowercase letters indicate significant differences at P < 0.05 (Duncan test)