Subscribe Now Subscribe Today
Research Article
 

Rice Breeding for High Yield by Advanced Single Seed Descent Method of Selection



Malinee Janwan, Tanee Sreewongchai and Prapa Sripichitt
 
Facebook Twitter Digg Reddit Linkedin StumbleUpon E-mail
ABSTRACT

Rice improvement for high yield is important to increase productivity of the crop. The success of breeding program depends on the choice of best parents and selection method. A research was conducted by applying Single Seed Decent (SSD) with Rapid Generation Advance (RGA) for speeding up the breeding cycle and to select elite line at F7 in 2012. A total of 271 recombinant inbred lines (RILs) were obtained in this program. Augmented design in RCBD with 3 replications was performed using standard check varieties (PTT1, CNT1, SPR60, RD31 and RD41). Three lines were observed with significantly higher yield than the best check variety, CNT1. However, only one line was significantly higher for number of filled seed per panicle than the best parent, CH1. All ten top-yielding lines had significantly higher filled seed per panicle than CNT1. Plant height of mostly the top-ten high-yielding lines ranged between 110 to 120 cm and days to 50% flowering all early more than KDML 105. Positive trangressive segregation was observed for 11 of the traits evaluated; however, the frequency was higher for plant height, days to 50% flowering and number of panicles. The result of correlation analysis revealed highly significant and positive correlation between yield and all the eleven traits under study. Stepwise regression analysis identified panicle weight, number of panicle, days to 50% flowering, seed-setting rate and flag leaf length as traits contributing for linear increase in yield. These traits could be considered as critical criteria for selecting high-yielding lines in rice breeding programs.

Services
Related Articles in ASCI
Search in Google Scholar
View Citation
Report Citation

 
  How to cite this article:

Malinee Janwan, Tanee Sreewongchai and Prapa Sripichitt, 2013. Rice Breeding for High Yield by Advanced Single Seed Descent Method of Selection. Journal of Plant Sciences, 8: 24-30.

DOI: 10.3923/jps.2013.24.30

URL: https://scialert.net/abstract/?doi=jps.2013.24.30
 
Received: January 04, 2013; Accepted: April 05, 2013; Published: June 21, 2013



INTRODUCTION

Food shortage is one of the important problems for global food crops. Rice breeding for high yield is one of the important factors to increase rice production as a solution of food shortage. Breeding method for rice high yield include conventional hybridization and selection, F1 hybrid breeding, ideotype (ideal plant type) breeding and enhancement of photosynthesis (Jeon et al., 2011). The achievement of breeding program consisted the choice of parents for the potential of crosses in order to develop superior line (one parent should be selected on the basis of proved performance in the area of projected use and the second parent must complement the first in character under improvement) and the appropriate selection method (Chandraratna, 1964; Allard, 1966; Briggs and Knowles, 1967). Single Seed Descent (SSD) and Rapid Generation Advance (RGA) have been applied in many conventional breeding programs to speed up breeding cycle. SSD was used to develop cold-tolerant RILs. The results revealed higher spikelet fertility for the cold-tolerant lines (51-81%) than the cold-sensitive (7%) and the cold-tolerant (73%) parents (Jena et al., 2012). The efficacy of SSD with RGA was speeding up the breeding cycle, increasing the number of favorable genotypes and reducing breeding costs (Maruyama, 1987). Furthermore, SSD can be used to produce wide range of trait variation and high level of trangressive segregation (Moon et al., 2003). The segregation of individuals in the F2 or a later generation of a cross that shows a more extreme development of a character than either parent is referred to as trangressive segregation Trangressive segregation is the segregation of F2 or a later generation that reveals higher or lesser character than parents (Grant, 1975). Advantage of transgressive segregation is particularly attractive as a mechanism for rapid evolution of plant and animal (Rieseberg et al., 2003). Trangressive segregant lines can be applied to breeding program improving for new cultivar or germplasm; especially, quantitative traits is particularly important for crop improvement.

The objective of the study was to develop high yielding rice varieties by using single seed descent method.

MATERIALS AND METHODS

Plant material: The plant materials used in this research include KDML105 and CH1 (New Plant Type (NPT) for high yield potential) as parents and five cultivated Thai varieties, PTT1, CNT1, SPR 60, RD31 and RD41.

Field experiment: The cross between KDML105 and CH1 was performed to produce F1 seed. Growing F1 plant and selfing to produce F2 generation were done in nursery. The F2 plants were grown under long day condition and photoperiod insensitive plants were selected. After that, the selected F2 plants were self-pollinated until F6 lines by SSD with RGA by growing individual plant in the small trays as 4x3 cm. The F7 population was grown for preliminary yield trial by using Augmented Design in RCBD with four replications compared with the five standard check varieties and the parental lines.

The data were collected as Plant Height (PH), flag leaf length (FL), Number of Tillers (TN), Effective Tillers (ET), days to 50% flowering (DFF), Number of Panicles (PN), Panicle Length (PL), Panicle Weight (PW), filled seeds/panicle (FP), spikelets/panicle, seed-setting rate (SR), 1,000 grains weight (TW), Harvest Index (HI) and yield/plant (YP). The breeding procedure was conducted at the Department of Agronomy, Kasetsart University. The preliminary yield trial was conducted at Khlongsamwa district, Bangkok. Test of significance were analyzed using IRRISTAT for windows version 5.0, phenotypic correlation coefficients and stepwise regression were analyzed using MSTATC and seed-setting rate criteria was according to Department of Agriculture of Thailand (Department of Agriculture, 1988).

RESULT AND DISCUSSION

Selection of photoperiod insensitive lines from the F2 population: The number of the selected lines for photoperiod-insensitivity and sensitivity were 288 and 784, respectively from the total F2 population evaluated. The ratio of photoperiod sensitive to photoperiod insensitive lines was tested by Chi-square (χ2 = 1.99); it fitted the 3:1 segregation ratio and revealed that photoperiod sensitivity was controlled by a dominance gene while photoperiod insensitivity was controlled by a recessive gene.

Yield, yield component and agronomic characters: Three lines i.e., 76-1-1-1-1-1, 181-1-1-1-1-1 and 103-4-1-1-1-1 performed significantly higher in yield with 70.64, 61.38 and 57.69 g plant-1, respectively than the best check (CNT1, 44.84 g plant-1).

Table 1: Means for yield, yield components and some agronomic characters for the top-ten high-yielding lines
Image for - Rice Breeding for High Yield by Advanced Single Seed Descent Method of Selection
PH: Plant height, DFF: Days to 50% flowering, PW: Panicle weight, ET: Effective tillers, PN: No. of panicles, FP: Filled seeds/panicle, TW: 1,000 grains weight, YP: Yield/plant

Table 2: No. of lines for seed-setting rate in RILs derived from KDML 105xCH1
Image for - Rice Breeding for High Yield by Advanced Single Seed Descent Method of Selection

The line 103-4-1-1-1-1 performed significantly higher in number of filled seed per panicle than the best parent (CH1 exhibited 242 seed/panicle). All the top-ten high-yielding lines had significantly higher number of filled seeds per panicle than CNT1. Plant height of mostly the top-ten high-yielding lines ranged between 110 to 120 cm and they were earlier in flowering than KDML 105 (Table 1). From the top-ten high-yielding lines, two lines had hundred percent effective tillers. In terms of seed-setting rate (Table 2), the 32 lines gave the higher seed-setting (>90%) than best check (CH1). However, most of the RILs (191 lines) showed the seed-setting rate in range of 75-89%, which were similar to CH1 (87.02%). The proposed yield component of NPT had low tillering capacity with 8-10 tillers when transplanted and 200-250 filled seed/panicle (Virk et al., 2004). Seventy percent of the top-ten high-yielding lines were similar to NPT for filled seed per panicle but not for tillering capacity. In direct seeded conditions, most unproductive tillers and excessive leaf area may initiate unitary shading and a reduction sink size (Dingkuhn et al., 1991). Although KDML 105 is the best for cooking quality in the world (Lanceras et al., 2000), photoperiod sensitivity and susceptibility to lodging limited the use of KDML 105 in rice improvement for high yield. The result of this research showed that KDML 105 had heavy panicle weight (6.52 g) and high number of spikelets/panicle (297.8 spikelets/panicle) while CNT1 has 162.9 spikelets/panicle. However, partly sterile character (63.30% seed-setting rate) of KDML 105 could lead to decreased yield potential. Therefore, the inheritance of semi-dwaft, photoperiod insensitivity and fertile character (87.02% seed-setting rate) of CH1 is important to increase rice yield potential.

Table 3: No. of trangressive segregant lines for eleven traits in RILs derived from KDML 105xCH1
Image for - Rice Breeding for High Yield by Advanced Single Seed Descent Method of Selection
PH: Plant height, DFF: Days to 50% flowering, PW: Panicle weight, ET: Effective tillers, PN: No. of panicles, FP: Filled seeds/panicle, TW: 1,000 grains weight, YP: Yield/plant

Good agrobotanical characteristics in the superiority of potentials of the hybrid (DTPMFe+) is a results of the richness genetic variability in local rice germplasm (Oziegbe and Faluyi, 2008). KDML 105 is a thai local rice varieties that shows superior lines is as a result of the richness genetic variability in parents.

Evaluation of transgressive segregation: The evaluation of transgressive segregation (Table 3) revealed eleven traits had positive trangressive segregant lines. Higher number of positive trangressive segregant lines was observed for plant height (171), days to 50% flowering (110) and number of panicles (65). Phenotypic distribution of RILs derived from KDML 105xCH1 for yield/plant and traits for high number of positive trangressive segregant lines were controlled by polygenes (Fig. 1). The higher number of positive trangressive segregant lines were observed for plant height, days to 50% flowering and number of panicle. The finding was in agreement with Kjaer et al. (1991) who reported high number of positive trangressive segregant lines for plant height and days to heading in barley. The parents had comparable values for yield/plant and number of panicle. However, plant height and days to 50% flowering had contrasting values between CH1 and KDML 105 (Table 3, Fig. 1). The frequency of transgressive segregant implicate to genetic diversity that similarity phenotypic of parental and the frequency of recombinants with the number of recombination round (Kuczynska et al., 2007). Number of panicle is yield component, breeding program can be used trangressive segregant lines in this trait for germplasm or selection superior lines.

Phenotypic correlation coefficients and stepwise regression: The estimates of phenotypic correlation coefficients (Appendix Table 1) showed highly significant (p<0.01) and positive correlation between yield and the eleven traits. The results of phenotypic correlation coefficients between yield and filled seed per panicle (r = 0.622**) had highly significant and positive correlation. Mulugeta et al. (2012) revealed that yield had positive and significant association with filled seed per panicle (r = 0.847**). Highly significant and positive correlations were observed between effective tillers and all yield component traits (number of panicle, filled seeds/panicle and 1,000 grains weight). Highly significant and positive correlations were also observed between panicle weight and seed-setting rate; these traits also had highly significant and positive correlation with filled seeds/panicle and 1,000 grains weight.

Image for - Rice Breeding for High Yield by Advanced Single Seed Descent Method of Selection
Fig. 1(a-d): Phenotypic distribution of RILs derived from KDML 105xCH1 for yield/plant, (a) Traits for high number of positive trangressive segregant lines i.e., plant height, (b) Days to 50% flowering, (c) Number of panicles and (d) Black and white arrows indicate the mean values of the parental lines, CH1 and KDML 105, respectively

Appendix Table 1: Phenotypic correlation coefficients between 11 traits and yield/plant in RILs derived from KDML 105xCH1, * ,**Significant at the 0.05 and 0.01 probability levels, respectively
Image for - Rice Breeding for High Yield by Advanced Single Seed Descent Method of Selection

Table 4: Stepwise regression for selection of traits
Image for - Rice Breeding for High Yield by Advanced Single Seed Descent Method of Selection
* ,**Significant at the 0.05 and 0.01 probability levels, respectively, R2 = 0.843

Stepwise regression analysis was performed to determine the traits contributed to yield. The result revealed that five out of the eleven traits had significant linear relationship with yield. The traits that had linear relationship include panicle weight, number of panicle, days to 50% flowering, seed-setting rate and flag leaf length (Table 4). The following model was obtained:

Image for - Rice Breeding for High Yield by Advanced Single Seed Descent Method of Selection

The R2 explained 84.3% from the total variations relative to yield. Similar to this finding, Ghaffar and Ghorbanali (2012) reported linear relationship of number of panicle with yield and 56.4% R2.

CONCLUSION

The three superior lines performed significantly higher for yield than the best check (CNT1). Single seed descent method can be used for quantitative traits selection because elite lines can selected, wide range of trait variation and high transgressive segregation are produced. Higher number of transgressive lines was obtained for three traits, including plant height, days to 50% flowering and number of panicle. Panicle weight, number of panicle, days to 50% flowering, seed-setting rate and flag leaf length can be considered as critical criteria for yield improvement in segregating generations of rice. Seesang et al. (2013) revealed that the number of panicle and flag leaf length could be used for selection criteria of high yielding in inbred genotypes.

ACKNOWLEDGMENT

The authors acknowledge the Graduate school, Kasetsart University for financial support.

REFERENCES

  1. Allard, R.W., 1966. Principle of Plant Breeding. John Wiley and Sons, USA


  2. Briggs, F.N. and P.F. Knowles, 1967. Introduction to Plant Breeding. Reinhold Publishing Corp., America, Pages: 426


  3. Chandraratna, M.F., 1964. Genetics and Breeding of Rice. Longmans, London, Pages: 389


  4. Department of Agriculture, 1988. Rice Data Collection Manual. Department of Agriculture, Bangkok


  5. Dingkuhn, M., F.W.T. Penning de Vries, S.K. de Datta and H.H. van Laar, 1991. Concepts for a new plant type for direct seeded flooded tropical rice. Proceedings of the International Rice Research Conference, August 27-31, 1990, Seoul, Korea, pp: 17-38


  6. Seesang, J., P. Sripicchitt, P. Somchit and T. Sreewongchai, 2013. Genotypic correlation and Path coefficient for some agronomic traits of hybrid and inbred rice (Oryza sativa L.) cultivars. Asian J. Crop Sci.,


  7. Ghaffar, K. and N. Ghorbanali, 2012. Correlation and path coefficient studies in F2 populations of rice. Not. Sci. Biol., 4: 124-127.
    Direct Link  |  


  8. Grant, V., 1975. Genetics of Flowering Plants. Columbia University Press, New York, ISBN: 9780231083638, Pages: 514


  9. Jena, K.K., S.M. Kim, J.P. Suh, C.I. Yang and Y.G. Kim, 2012. Identification of cold-tolerant breeding lines by quantitative trait loci associated with cold tolerance in rice. Crop Sci., 52: 517-523.
    CrossRef  |  Direct Link  |  


  10. Jeon, J.S., K.H. Jung, H.B. Kim, J.P. Suh and G.S. Khush, 2011. Genetic and molecular insights into the enhancement of rice yield potential. J. Plan Biol., 54: 1-9.
    CrossRef  |  Direct Link  |  


  11. Lanceras, J.C., Z.L. Huang, O. Naivikul, A. Vanavichit, V. Ruanjaichon and S. Tragoonrung, 2000. Mapping of genes for cooking and eating qualities in Thai jasmine rice (KDML105). DNA Res., 7: 93-101.
    CrossRef  |  Direct Link  |  


  12. Kjaer, B., V. Haahr and J. Jensen, 1991. Associations between 23 quantitative traits and 10 genetic markers in barley cross. Plant Breed., 106: 261-274.
    CrossRef  |  Direct Link  |  


  13. Kuczynska, A., M. Surma and T. Adamski, 2007. Methods to predict transgressive segregation in barley and other self-pollinated crops. J. Applied Genet., 48: 321-328.
    CrossRef  |  Direct Link  |  


  14. Maruyama, K., 1987. Using rapid generation advance with single seed descent in rice breeding. Proceedings of the International Rice Research Conference, September 21-25, 1987, Hangzhou, China, pp: 253-259


  15. Moon, H.P., K.H. Kang, I.S. Choi, O.Y. Jeong, H.C. Hong, S.H. Choi and H.C. Choi, 2003. Comparing Agronomic Performance of Breeding Populations Derived from Anther Culture and Single Seed Sescent in Rice. In: Advances in Rice Genetics, Khush, G.S., D.S. Brar and B. Hardy (Eds.). Int. Rice Research Institute, USA., pp: 3-5


  16. Oziegbe, M. and J.O. Faluyi, 2008. Comparative agrobotanical characteristics of an enhanced rice cultivar DTPMFe+ and its parents (AWGU-DWARF-W and IJ86-W) Oryza sativa Linn. J. Plant Sci., 3: 116-120.
    CrossRef  |  Direct Link  |  


  17. Rieseberg, L.H., A. Widmer, A.M. Arntz and J.M. Burke, 2003. The genetic architecture necessary for transgressive segregation is common in both natural and domesticated populations. Phil. Trans. R. Soc. Lond. B, 358: 1141-1147.
    CrossRef  |  Direct Link  |  


  18. Seyoum, M., S. Alamerew and K. Bantte, 2012. Genetic variability, heritability, correlation coefficient and path analysis for yield and yield related traits in upland rice (Oryza sativa L.). J. Plant Sci., 7: 13-22.
    CrossRef  |  Direct Link  |  


  19. Virk, P.S., G.S. Khush and S. Peng, 2004. Breeding to enhance yield potential of rice at IRRI: The ideotype approach. Int. Rice Res. Notes, 29: 5-9.
    Direct Link  |  


©  2022 Science Alert. All Rights Reserved