Most of the cultivated rice varieties are susceptible to salinity but rice germplasm do have sources for salt tolerance trait (Flowers et al., 2000). Traditional breeding efforts made to introgress complex traits like salinity tolerance have met with limited success and only a few salt-tolerant varieties (~30 cultivars in 12 different plant species) have been developed and released for commercial cultivation. It has been even more difficult to introgress desirable traits into Basmati rice [aromatic, varietal Group V (Glaszmann, 1987)] due to the complex nature of Basmati rice grain quality traits and its poor combining ability with the other rice genotypes (Khush and dela Cruz, 1998). Application of molecular marker technology in linkage mapping and molecular dissection of the complex agronomical traits such as salinity can greatly enhance the efficiency and accuracy of breeding process. In this paper, we report the genetic and molecular structure of a F3 segregating population derived from a cross between CSR 10 X Taraori Basmati for salinity tolerance using Inter Simple Sequence Repeat (ISSR) markers.

Crosses were made between CSR10 (semi-dwarf, salt-tolerant variety developed from cross between Damodar and Jaya) and Taraori Basmati (tall-stature, aromatic, salt-sensitive) and the hybrid nature of the F₁ plants was confirmed by morphological traits and SSR marker analysis (data not shown). A population of 272 F₂ plants was raised by single seed descent method of which 130 were used in the present study. The F₂ seed along with parental genotypes were germinated in 30 mM NaCl supplemented Yoshida solution (Yoshida et al., 1976) and were scored for salt tolerance on the basis of seedling growth and leaf injury on a 1-9 scale as per the standard evaluation system (IRRI, 1988). The two parental rice varieties differed significantly for salinity tolerance; CSR10 and HBC19 scored 1.5 and 8.2 respectively. The F₃ population score ranged from 1.72 to 8.45 with a mean value of 5.308. The data suggested a good fit ( chi² = 7.765, p= 0.01) to a normal distribution as tested using 'Z' statistics; chi² test (Figure 1). As the chi² estimate test the null hypothesis regarding normality but does not indicate against any departure from normality therefore, the population was subjected to detect skewness and kurtosis as described by Mishra et al. (1998). The coefficient of skewness (g₁= -0.0478) was less than the standard deviation ( sigma 6/n = 0.2148), which indicates normal distribution of the population. But the curve was found platykurtic as Kurtosis (g₂ = -0.0925), which was much less than the standard deviation ( sigma 6/n = 0.4296) showing a flat topped negative distribution. The F₃ population showed more widely dispersed frequencies to the two extremes than their concentration towards the mid point. Some plants in this population were even more tolerant or susceptible (transgressive segregants) than their parents. It indicates that segregation of such stress related genes/QTLs may results in to new combinations with enhanced tolerance or sensitivity to salinity and that this should be ideal for linkage mapping studies.

Eleven F₃ plants each in categories of most salt-tolerant and most salt-sensitive were selected for ISSR marker analysis using the methods as described by Blair et al. (1999). Out

of 100 primers (UBC set #9, 801-900; John Hobbs, NAPS Unit, University of British Columbia, Vancouver, V6T 1Z3 Canada) used for DNA amplification in two parental rice varieties, 41 primers successfully amplified the loci but good amplification and clear banding profiles were obtained for 26 primers only. A total of 149 bands ranging from 200 bp to 3530 bp were scored for the two rice varieties and 22 selected CSR10 x HBC19 segregating F₃ lines using 26 ISSR primers. Number of bands per primer ranged between 4 (UBC824) and 11 (UBC891) with an average of 5.73 bands per primer. Out of 149 bands, 60 were polymorphic of which 36 and 20 bands were specific to CSR10 and HBC19 respectively and remaining four bands were observed in some of the segregating F₃ plants only. UBC primers 807 and 823 showed the maximum polymorphism (80.0%) between the parental rice varieties.

ISSR primers with di-nucleotide repeat motifs and 5'-anchored end amplified more number of bands (7.0 bands/primer) compared to 3'-anchored dinucleotide repeat primers (5.4 bands/primer), but 3'-anchored dinucleotide repeat primers revealed higher level of polymorphism (2.6 polymorphic bands/ primer) compared to 5'- anchored dinucleotide repeat primers (1.43 polymorphic bands/ primer). This might be expected because 5' anchored primers lack selective nucleotide at the critical 3' end. The three UBC primers with tri-nucleotide motifs (nonanchored UBC 864 and 866; anchored mixed UBC 899) showed polymorphism (41.1 %) comparable to that obtained using di-nucleotide repeat based primers (43.4 %). The similarity coefficient between the two parental genotypes was 0.611 (UPGMA, "Sahn" subprogram of NTSYS-PC; Rholf 1993). Selected salt tolerant genotypes showed an average similarity of 0.748 with CSR10, which was higher than the similarity (0.635) with HBC19. However, selected salt-susceptible plants showed more or less equal similarity with CSR10 (0.674) and HBC19 (0.642). The Principal Component Analysis (PCA) using the ISSR database showed the scattering of 22 selected F₃ genotypes not only between the two parental lines but also away

from them (Figure 2).

While distribution of majority of these polymorphic bands were more or less equal in the segregating lines irrespective of their salt-tolerance potential, but some of the bands did show skewed distribution. Fourteen of the 36 CSR-specific polymorphic bands amplified using UBC primers 823, 825, 826, 840, 848, 849, 853, 864, 866, 884, 889 and 890, were present at high frequencies (54.5-90.9%) in the selected salt-tolerant F₃ plants compared to that (9.1-45.5%) in the sensitive ones. Such polymorphic bands stand greater chances of having a linkage with the genomic DNA sequences, which may have significant effects on salt tolerance and should be the target for further studies.

We thank K.R. Gupta (Rice Research Station, Kaul, Haryana, India) for providing us the rice material and Rockefeller Foundation (New York, USA) for the research grant (RF2000FS#023).

References

Blair, M. W., O. Panaud and S. R. McCouch, 1999. Inter-simple sequence repeat (ISSR) amplification for analysis of microsatellite motif frequency and fingerprinting in rice (Oryza sativa L.). Theor. Appl. Genet. 98: 780-792.

Flowers, T. J., M. L. Koyama, S. A. Flowers, C. Sudhakar, K. P. Singh and A. R. Yeo, 2000. QTL: their place in engineering tolerance of rice to salinity. J. Exp. Bot. 51: 99-106.

Glaszmann J. C. (1987). Isozyme and classification of Asian rice varieties. Theor. Appl. Genet 74: 21-30.

Khush, G. S. and N. dela Cruz, 1998. Developing Basmati rices with high yield potential. Cahiers Options Mediterraneennes 24: Rice quality. A pluridisciplinary approach. (CD-ROM computer file) CIHEAM, Paris

Mishra, B., R. K. Singh and V. Jetly, 1998. Inheritance pattern of salinity tolerance in rice. J. Genet. Breed. 52: 325-331.

Rohlf, F. J., 1993. NTSYS-PC: Numerical Taxonomy and Multivariate Analysis System. Version 1.8, Exeter Software, Setauket, New York.

Yoshida, S., D. A. Forno, J. H. Cock and K. A. Gomez, 1979. Laboratory manual for physiological studies of rice. International Rice Research Institute, Los Banos, The Philippines.