In view of the inherent potential of wild rice species (Oryza sp.) as important genetic resources, a comprehensive study on wild relatives of rice consisting of five different experiments was initiated at National Agricultural Research Centre (NARC), Islamabad, Pakistan during 2005-07. The first experiment comprised characterization' of wild relatives of rice on the basis of qualitative and quantitative traits while the second experiment consisted of studies on interspecific variation of total seed protein in rice germplasm using sodium dodecy1 sulphate polyacrilamide gel electrophoresis (SDSPAGE). Resistance characterization of wild relatives of rice in response to bacterial blight and rice leaffolder, Cnaphalocrocis medinalis (Gn) constituted the third and fourth experiment, respectively. The fifth experiment involved the development and confirmation of F1 wide hybrids resulted from crossing of cultivated varieties with bacterial blight and leaffolder resistant wild relatives of rice.
In the first experiment fourteen wild species of rice along with three commercial cultivars of Pakistan were evaluated for 58 plant parameters. These comprised 27 qualitative and 31 quantitative traits. The material displayed considerable variability for both qualitative and quantitative characteristics except two qualitative traits i.e. leaf sheath hairiness and panicle exsertion. Medium to high variance was observed for culm length, peduncle length, flag leaf length, ligule length, days to flowering initiation, panicle length, percent free length of primary branch to total length, number of secondary branches panicle-1 and number of spikelets panicle-1 whereas for other traits the magnitude of variance was relatively small. Principal Components Analysis (PCA) showed that the first seven principal components had eigenvalues > 1 accounting for 89.2% of total variation. The first and second principal components accounted for 38.5 and 18.2% of the total variation, respectively. The first component was characterized by culm length, number of secondary branches panicle-I, mean length of primary branches, number of spikelets panicle-1, primary branches panicle-1 and both number of nodes and internodes. The second component was heavily weighted by spikelet length, grain length, days to flowering initiation, flag leaf length and distance between spikelets. Combined cluster analysis for both qualitative and quantitative traits clearly grouped wild species and cultivated varieties of rice into different clusters on the basis of their respective genomes.
Variation in seed protein of wild and cultivated rice species was studied through SDS-PAGE in the second experiment. SDS bands were scored as present (1) or absent (0) for protein sample of each genotype. Overall, on basis of total seed protein, the grouping pattern of rice genotypes was mostly compatible with their genome status.
In order to identify resistant sources to virulent isolates of bacterial blight, the third experiment comprising wild rice species and three widely used cultivated varieties of rice in Pakistan was conducted under greenhouse conditions. Adult plants were inoculated with virulent isolates of Xanthomonas oryzae pv. oryzae prevailing in NWFP, Pakistan i.e. Xo-I03, Xo-I07, Xo-139, Xo-143, Xo-304, Xo-35 I and MNR-4. Of all the wild relatives of rice, O. nivara, O. longistaminata and O. grandiglumis showed resistance to all isolates. Remaining wild species showed differential response to the isolates used in the study. These species were resistant to one or few isolates but expressed susceptibility to others.
To search for possible sources of resistance against leaffolder, the fourth experiment consisting of wild rice species, three extensively cultivated varieties of rice in Pakistan together with one susceptible check of cultivated rice was conducted under greenhouse conditions. Most of the tested wild species and the three cultivated varieties showed susceptibility to leaffolder. However, two wild species, O. brachyantha and 0. rufipogon were found resistant to leaffolder with damage rating scale of 1 and comparatively low percent infestation values of 3.7 and 5.3, respectively.
In the fifth experiment, two wild rice species, O. nivara (bacterial blight resistant) and 0. rufipogon (leaffolder resistant) were crossed with three widely cultivated varieties of rice in Pakistan viz. Bas-385, IR-6 and KSK-282 to deploy bacterial blight and leaffolder resistant genes into these cultivars. The hybrid nature of the progenies was confirmed through morphological markers and SDS-P AGE analysis. Most of the qualitative traits of hybrids were observed similar to wild rice species used as male parents. Heterosis was manifested for most of the quantitative traits studied during this experiment. SDS-P AGE analysis of total seed protein displayed variation in number and staining intensities of various bands of parents and their F1's. Respective F1 hybrid plants developed for bacterial blight and leaffolder resistance exhibited strong resistance on artificial inoculation with bacterial blight isolates and leaffolder larvae and thus revealed the dominant nature of resistant genes in 0. nivara and O. rufipogon.
The material under study displayed ample genetic diversity at both morphological, and total seed protein levels. Wild rice species resistant to bacterial blight and leaffolder were identified. In the first ever wide rice hybridization program of the country resistant wild rice species were crossed with cultivated varieties and viable F1 wide hybrids were developed displaying strong resistance reaction against virulent isolates of bacterial blight and leaffolder. This breeding material can be advanced further to have a long term solution to the major disease and insect problems of rice crop in the country.