Evaluation of backcross-derived maintainer lines for stigma Exsertion and associated agronomic traits in hybrid rice

Introduction

Hybrid rice is an important technology for increasing rice productivity, but its commercial value depends not only on heterosis but also on the ability to produce hybrid seed efficiently and economically. In the three-line hybrid rice system, cytoplasmic male sterile lines serve as female parents, while corresponding maintainer lines are required to maintain and improve these femaleline backgrounds. Floral traits that improve pollen reception, stigma exposure and outcrossing therefore have direct practical relevance in parental-line improvement. Among floral traits, stigma exsertion is one of the most important determinants of outcrossing ability. Stigma exsertion refers to the protrusion of one or both stigmas outside the spikelet after anthesis and spikelet closure. Exserted stigmas remain available for pollen capture for a longer period and can improve cross-pollination under hybrid seed production conditions. Classical studies on rice floral biology reported genetic variability for floral characters influencing outcrossing (Virmani & Athwal, 1973) [1] , while later syntheses emphasized that pistil traits, including stigma exsertion, are among the most useful traits for improving hybrid seed production (Marathi & Jena, 2015) [2] . Recent literature continues to support this breeding priority. Ashraf et al., (2024) [3] reviewed the role of floral traits and breeding technologies in hybrid rice production and emphasized that floral architecture, stigma traits and flowering behavior influence hybrid seed production success. Guo et al., (2024) [4] described stigma exsertion rate as a key trait affecting the outcrossing ability of sterile lines and summarized molecular and quantitative genetic progress in stigma-exsertion improvement. Studies on grain shape and stigma exsertion have also shown that floral and spikelet structure can influence stigma exposure and outcrossing rate (Tan et al., 2023; Zhu et al., 2023) [5, 6] .

Stigma exsertion is not a simple single-gene trait. It is generally considered a quantitative floral trait controlled by multiple loci and influenced by genetic background, stigma size, style length, glume morphology, flowering behavior and panicle architecture. QTL studies have confirmed that multiple genomic regions contribute to stigma exsertion rate in rice, and that high stigma exsertion can be reconstructed through pyramiding of favourable loci (Tan et al., 2022; Yang et al., 2024) [7,8] . This genetic complexity makes phenotypic screening of breeding material important, especially when the objective is to identify practical parental-line candidates rather than only map genomic regions. Backcross-derived maintainer lines provide a useful context for such evaluation because they are expected to retain desirable maintainer-line backgrounds while carrying improved floral or agronomic attributes. However, high stigma exsertion alone is not sufficient for breeding use. A line with high SE% but poor spikelet fertility or poor yield-related performance may be valuable as a donor but may not be suitable for direct advancement. Therefore, stigma exsertion must be interpreted together with fertility, grain filling and agronomic traits. The present study was conducted to evaluate BC2 backcrossderived maintainer lines and related genotypes for SE% and associated agronomic traits. The objectives were to quantify variability for stigma exsertion, identify high-SE maintainerline derivatives, assess the relationship of SE% with fertility and yield-related traits, and develop a balanced selection interpretation for hybrid rice parentalline improvement.

Materials and methods 2.1 Experimental material and field evaluation

The experimental material consisted of 32 rice genotypes comprising four parental lines, five check varieties and 23 BC2 backcross-derived maintainer lines. The parental lines included 14RR1, 14RR2, IR79156B and IR58025B. The check varieties included CO-51, US-314, GONTRA BIDHAN-3, VNR-2228 and 27P63. The backcross-derived maintainer lines consisted of BCN and BCP series lines developed in the hybrid rice breeding programme. The field experiment was conducted during Kharif 2025 at ICAR-Indian Institute of Rice Research, Rajendranagar, Hyderabad, India. The experiment was laid out in a randomized block design with two replications. Standard crop management practices were followed to raise a healthy rice crop. Observations were recorded on representative competitive plants from each genotype and replication.

Trait recording

Observations were recorded for days to 50% flowering (DFF), plant height (PH), panicle length (PL), number of productive tillers (NPT), number of grains per panicle (NGP), number of filled grains per panicle (NFG), number of unfilled grains per panicle (UFG), spikelet fertility% (SF), thousand grain weight (TGW), single plant yield (SPY) and stigma exsertion percentage (SE%). Rice trait observations were recorded following standard rice descriptor principles described in the Standard Evaluation System for Rice (IRRI SES, 2013). Stigma exsertion was recorded after anthesis from representative panicles. Spikelets were examined and classified as spikelets with no stigma exsertion, single-stigma exsertion or dual-stigma exsertion. Total SE% was calculated as the percentage of spikelets showing either single or dual stigma exsertion. For distributional interpretation, genotypes were grouped into 10% SE classes from 0-10% to 90-100% (Ying and Zhang 1989) [9]

Statistical analysis

Analysis of variance was performed to test genotypic differences among the 32 genotypes. Phenotypic correlation coefficients were estimated using Pearson's correlation approach (Pearson, 1895) [10] to examine trait associations. Path coefficient analysis was performed following Dewey and Lu (1959) [11] to partition correlations into direct and indirect effects. Statistical analyses were carried out using STAR version 2.0.1 (Statistical Tool for Agricultural Research, International Rice Research Institute), Microsoft Excel 365 and R version 4.4.2 using R Studio 2024.12.0+467 for figure generation.

Results

Significant genotypic variability for stigma exsertion and associated traits

Analysis of variance revealed significant differences among genotypes for all traits studied, including SE% (Table 1). The genotype mean square was highly significant for DFF, PH, PL, NPT, NGP, NFG, UFG, SF, TGW, SPY and SE%, indicating the presence of substantial variability among parental lines, checks and backcross-derived maintainer lines. The significance of genotypic variation for SE% was particularly relevant because this trait is directly associated with outcrossing ability. However, significant variation for fertility and yield-related traits indicated that high SE% should be evaluated together with reproductive and agronomic performance rather than used as an isolated selection criterion. (81.66%). However, high SE% was not always associated with high spikelet fertility or single plant yield. Therefore, these lines should be considered high-SE sources requiring further validation for fertility, outcrossing ability and agronomic performance (Table 2).

Frequency distribution further clarified the structure of variation (Figure 1

High stigma exsertion did not always coincide with fertility and yield superiority

The high-SE genotypes showed important differences in reproductive and yield-related performance (

Trait associations relevant to stigma exsertion-based selection

Correlation analysis revealed that SE% had a significant positive association with PL (r=0.3355 * * ) and TGW (r=0.4582 * * ), whereas its association with SPY was negligible and nonsignificant (r=-0.0113) (Table 3). Similarly, SE% showed weak and non-significant associations with NPT, NGP, NFG, UFG, SF% and DFF. The correlation heat map further confirmed that SE% was more closely associated with panicle length and thousand grain weight than with direct yield expression (Figure 2). This result is important for breeding interpretation because it indicates that SE% should be treated primarily as an outcrossingrelated floral trait rather than a direct yield trait.

Discussion

The present study demonstrates that backcross-derived maintainer lines possess substantial exploitable variation for SE%. The observed range of 30.45-86.87% indicates that the evaluated material allowed clear differentiation between low, moderate and high stigma-exsertion types. The presence of eight genotypes in the 80-90% class is particularly useful for hybrid rice breeding because high stigma exsertion improves stigma exposure and increases the opportunity for pollen capture in female-line seed production. The importance of stigma exsertion in hybrid rice is well established. Virmani and Athwal (1973) [1] showed that rice floral traits influencing outcrossing are genetically variable. Marathi and Jena (2015) [2] later emphasized that pistil traits are critical for improving hybrid seed production. More recent reviews and genetic studies have reinforced the same breeding logic, identifying stigma exsertion rate as a key determinant of outcrossing ability in sterile lines and an important target for molecular and phenotypic improvement (Ashraf et al., 2024; Guo et al., 2024; Yang et al., 2024) [3,4,8] . A major practical finding of the study is that high SE% did not always coincide with high fertility or yield. The high-SE parental lines 14RR1 and 14RR2 recorded SE values above 70%, confirming their usefulness as donor parents for stigma exsertion improvement. Among the backcross-derived maintainer lines, BCN 16.2 recorded the highest SE% but had low spikelet fertility and low single plant yield, while BCN 31 also expressed high SE% but showed poor spikelet fertility. These genotypes may be useful as high-SE sources, but their direct advancement would require caution. Among the high-SE derived lines, BCN 26 showed comparatively better balance between SE%, spikelet fertility and yield-related performance. This distinction between high-SE donor value and advancement value is important for breeding decisions. The weak association between SE% and SPY further clarifies the biological role of stigma exsertion. SE% is not a direct yield trait under maintainer-line plot evaluation. Instead, it is an outcrossing-related floral trait that may improve hybrid seed production when expressed in the corresponding female-line background. Grain yield per plant depends more directly on filled grains, spikelet fertility, productive tillers and grain number. Therefore, selection for SE% should be integrated with fertility and agronomic performance rather than used as a standalone criterion. The significant positive association of SE% with PL and TGW suggests that stigma exposure may be linked with structural attributes of panicle and spikelet development. This interpretation is consistent with recent reports showing relationships between stigma exsertion, grain shape and floral organ morphology (Tan et al., 2023; Zhu et al., 2023) [5,6] . However, the moderate magnitude of these correlations indicates that PL and TGW cannot replace direct measurement of SE%. Direct phenotyping remains necessary for accurate identification of high-SE lines. The supporting divergence analysis also indicated that SE% was the largest contributor to total divergence among the evaluated traits. This confirms that SE% was not a marginal descriptor but a major differentiating trait in the material. Nevertheless, full diversity analysis was not retained as a central section in the manuscript because the primary objective was to identify and interpret high-SE maintainer-line candidates. Treating divergence as supporting evidence keeps the article focused and avoids converting the study into a general diversity paper. Overall, the most useful outcome is the identification of traitspecific breeding material. BCN 16.2, BCN 22.2, BCN 17.2, BCN 26 and BCN 31 are useful high-SE sources. BCN 31 should be used cautiously due to low fertility. Such lines should be evaluated further for flowering synchrony, actual outcrossing rate, seed set under hybrid seed production conditions and performance of corresponding male sterile-line backgrounds.

Conclusion

Overall, 14RR1 and 14RR2 were confirmed as high-SE parental donors, with SE values above 70%. Among the backcrossderived maintainer lines, BCN 16.2, BCN 22.2, BCN 31, BCN 17.2 and BCN 26 expressed high SE% and may serve as useful sources for improving stigma exsertion. The trait SE expressed a positive correlation with the traits panicle length and thousand grain weight. Since, high SE% did coincide with high spikelet fertility or single plant yield, the selected lines require further validation for fertility, outcrossing rate and hybrid seed production performance before use in parentalline improvement.

References

1. Virmani SS, Athwal DS. Genetic variability in floral characteristics influencing outcrossing in Oryza sativa L. Crop Science. 1973;13(1):66-
2. Marathi B, Jena KK. Floral traits to enhance outcrossing for higher hybrid seed production in rice: Present status and future prospects. Euphytica. 2015;201:114.https://doi.org/10.1007/s10681-
3. Ashraf H, Ghouri F, Baloch FS, Nadeem MA, Fu X, Shahid MQ. Hybrid rice production: A worldwide review of floral traits and breeding technology, with special emphasis on China. Plants. 2024;13(5):578.https://doi.org/10.3390/plants13050578
4. Guo H, Liang Y, Lv J, Su X, Ren G, Gao F. Research progress on the trait of stigma exsertion in rice. Plants. 2024;13(23):3404.https://doi.org/10.3390/plants13233404
5. Tan Q, Chen S, Gan Z, Lu Q, Yan Z, Chen G, et al. Grain shape is a factor affecting the stigma exsertion rate in rice. Frontiers in Plant Science. 2023;14:1087285.https://doi.org/10.3389/fpls.2023.10872 85
6. Zhu X, Gou Y, Heng Y, Ding W, Li Y, Zhou D, et al. Targeted manipulation of grain shape genes effectively improves outcrossing rate and hybrid seed production in rice. Plant Biotechnology Journal. 2023;21:395407.https://doi.org/10.1111/pbi.13959
7. Tan Q, Bu S, Chen G, Yan Z, Chang Z, Zhu H, et al. Reconstruction of the high stigma exsertion rate trait in rice by pyramiding multiple QTLs. Frontiers in Plant Science. 2022;13:921700.https://doi.org/10.3389/fpls.2022.921700
8. Yang H, Zhou Y, Li P, Liu E, Sun P, Ao Y, et al. Genetic analysis and fine mapping of QTLs for stigma exsertion rate in rice. Rice. 2024;17:74.https://doi.org/10.1186/s12284-024-00752-6
9. Ying CS, Zhang SW. Studies on the character of stigma exsertion among some Oryza species. Chinese Journal of Rice Science. 1989;3:62-66.
10. Pearson K. Notes on regression and inheritance in the case of two parents. Proceedings of the Royal Society of London. 1895;58:240-242.
11. Dewey DR, Lu KH. A correlation and path-coefficient analysis of components of crested wheatgrass seed production. Agronomy Journal. 1959;51(9):515-
12. International Rice Research Institute. (n.d.). STAR: Statistical Tool for Agricultural Research. IRRI Education. Retrieved June 30, 2026, fromhttps://education.irri.org/sciencecourses/basicexperimental-design-and-data-analysisusing-star/
13. Mahalanobis PC. On the generalized distance in statistics. Proceedings of the National Institute of Sciences of India. 1936;2:49-55.
14. R Core Team. R: A language and environment for statistical computing (Version 4.4.2). R Foundation for Statistical Computing; 2024.https://www.R-project.org