Identification of Microsatellite (SSR) and RAPD Markers Linked to Rice Blast Disease Resistance gene in Rice (Oryza sativa L.)

School of Biosciences and Technology, Plant Biotechnology Division, Vellore Institute of Technology University. Vellore, India. Center for Plant Molecular Biology, Tamil Nadu Agricultural University, Coimbatore, India. Vellore, India. Center for Plant Breeding and Genetics, Tamil Nadu Agricultural University, Coimbatore, India. Centre for Plant Protection Studies (Entomology), Tamil Nadu Agricultural University, Coimbatore, India. Centre for Plant Protection Studies (Plant Pathology), Tamil Nadu Agricultural University, Coimbatore, India.

cultivars with one or two resistance genes named Pi-, followed by a different letter for each gene.Some of these genes have several alleles, such as loci Pi-z, Pi-ta with two alleles and Pi-k with five alleles.It has been proposed that Pi3-(t) is allelic to Pi-5(t), as both confer similar resistance spectra to a variety of blast lineages.Liu et al. (2002) demonstrated that the resistance genes Pi-9(t) and Pi-2(t) were tightly linked and allelic.
Several blast resistant varieties have been and are being released regularly.They lose resistance due to the change in the race pattern of the blast fungus.Fungicides fail to control blast remarkably as it is too static to deal with the dynamic relationships between plants and disease that are deeply tied to the surrounding ecology.Resistance is considered durable when it remains effecttive in a cultivar despite widespread cultivation in an environment favoring the disease.
In different patho-systems, durable resistance is variously controlled by single genes, multiple genes with cumulative effects and polygenes.The resistance may be either complete or incomplete (Parlevliet, 1988).Some upland cultivars such as the traditional African cultivars, Moroberekan and OS6, have been cultivated for many years in large areas in West Africa without high losses of blast.Five resistance genes have been identified in African cultivar Moroberekan.These cultivars have been widely used as resistance donors in breeding programs.
Molecular markers are useful tools for monitoring gene introgressions and to detect polymorphism among species.The use of molecular markers can help in estimating the overall genetic variability, visualize the proportion of the genome introgressed from the donor, identify the genes related to the increase in the phenotypic value of analyzed traits, and then allow marker assisted selection in subsequent generations of these introgression lines (Brondani et al., 2003).
However, the presence or absence of the associated molecular marker could indicate at a very early stage, the presence or absence of the desired target gene.Selection based on genetic information was retrieved through the application of molecular markers.Marker assisted selection (MAS) could be employed to enhance plant breeding efforts and to speed up the creation of new cultivars.A molecular marker very closely linked to the target gene can act as a tag, which can be used for indirect selection of the gene(s) in a breeding programme.This process is referred to as MAS in breeding (Witcombe and Hash, 2000;Asins, 2002;Liu et al., 2004;Zhang et al., 2005).In this paper, attempts were made to reveal the random amplified polymorphic DNA (RAPD) and simple sequence repeats (SSR) markers linked to rice blast disease resistance by single marker analysis (SMA)

Hybridization program
Crossing was effected with the leaf blast resistant cultivar, Moro-berekan, as a pollen parent and the susceptible cultivar, White Ponni as a ovule parent.Moroberekan (A West African cultivar) is known for its durable resistance to rice blast disease (Wang et al., 1994) and the susceptible variety, White Ponni (selection from the cross derivative, Taichung 65/2 x Mayang Ebos -80) is known for its fine grain quality and as a moderate yielder (45 quintals per hectare).The genotypes were raised in nursery beds with three staggered sowings at 10 to 15 days interval to ensure synchronized flowering to enable hybridization.

Development of mapping population
The crossed seeds thus obtained were raised in the field and selfed.The trueness of the hybrids were confirmed both by morphological and molecular markers.A total of 220 selfed seeds of an identified F1 plant were collected and raised as F2s by single seed descent method to generate segregating population.Leaves were collected from all the 220 F2 individuals for DNA extraction.

Biometric observations recorded
In the segregating population (White Ponni / Moroberekan), the following biometric observations viz., days to 50% flowering, plant height, presence or absence of pigmentation at the tip of the spikelets / Glume (GPT) / culm region, number of tillers per plant, number of productive tillers per plant, panicle length and grain yield per plant were recorded.

Artificial screening for leaf blast disease reaction in parents and F3 families
Artificial screening for rice blast disease was done in the year 2005, in Paddy Breeding Station, Coimbatore, Tamilnadu, in a specially constructed screen house with good irrigation facilities fitted with mist blowers, which can spray water in a fine mist inside the chamber.Subsequently, the seedlings were misted 4 to 5 times at intervals.The screen house was maintained at 32 to 37°C (day temperature) and 94 to 96% relative humidity (RH) for the potential disease occurrence.The rate of sporulation increases with increase in relative humidity provided with lower night temperature with minimum of 25°C.Inoculations with M. grisea Hebert (Barr) were performed 3 weeks after sowing by spraying with conidial suspensions.For artificial rice blast screening, the method by Sallaud et al. (2003) was followed.Isolation of rice blast fungus was done from paddy breeding station, Coimbatore.Artificial culturing of the rice blast fungal inoculum is shown stepwisely in Figure 1.For the spray method, 30 ml of a 50,000 conidia.ml - suspension with 0.5% gelatin were sprayed on each tray.The observation on the disease incidence was recorded, when the susceptible check was severely infected by blast.Observations were recorded from 20 plants in each F3 family following Standard Evaluation System (SES, 2002) on 0-9 scale at 25th day after sowing.The resistant check used was IR 64.Observations were recorded in plants, when they were at third leaf stage.The screen house, disease symptoms and variations of occurrence of the disease in incidence in F3 population is shown in Figure 2. The grade and criterion based on standard evaluation system is as follows score 0 -no lesions observed; score 1 -small brown specks of pin point size or larger brown specks without sporulating centre; score 3 -small roundish to slightly elongated necrotic grey sporulating spots about 1 to 2 mm in diameter with a distinct brown margin; score 5 -narrow or slight elliptical lesions, 1 to 2 mm in breadth, more than 3 mm long with brown margin; score 7 -broad spindle shaped lesion with yellow, brown or purple margin; score 9 -rapidly coalescing small, whitish, greyish or bluish lesions without distinct margins.
The potential disease incidence percent (PDI %) was worked out Figure 1. Isolation and sporulation of rice blast fungus (M.grisea) for artificial disease screening.A, Collection of typical, young, greenish rice blast symptom from 15 days old rice seedling in a moist chamber; B and E, after surface sterilization with 0.1% HgCl2, the symptom alone is half cut and placed in PDA or OMA medium; C and F, five days after inoculation of the symptom, kept at 24 to 26°C, immersed mycelial growth appears; D and G, eleven days after inoculation, kept at 24 to 26°C, individual colony with pale white mycelial growth appears; H, subculturing was done after eleventh day up to 3 to 4 subcultures from the growing mycelial tips.J and K, an effused colony, thinly hairy, turning from olivaceous brown to greyish brown with immersed mycelium and wedge shaped centre was seen.L, full grown rice blast (M.oryzae) colony with typical colony morphology; M, Inoculated conical flasks maintained at 27°C at 90% RH inside fluorescent incubator for 15 days; N, observation of rice blast spores at 15 -20 days after inoculation in Leitz (Flovert FS) microscope under the magnification of 100x.O, disease incidence after spraying of artificial rice blast sporulated inoculum with spore concentration adjusted to 50,000 conidia.ml - suspension with 0.5% gelatin spores (approximately) 2 to 3 times inside the screen house for disease induction.
using the formula of McKinney (1923): Sum of numerical rating / number of leaves observed x 100 / maximum disease grade.

Isolation of DNA
Fresh leaf samples collected from 15 days old seedlings of parental genotypes and the segregating population was used for isolation and purification of total genomic DNA following the method of McCouch et al. (1988).A total of 50 decamer primers supplied by Operon Technologies Inc., Alameda, California, USA were used in the study for parental genotyping.Seven RAPD primers which are highly polymorphic were used to genotype the F2 mapping population which produced 12 polymorphic markers namely, OPAL 16940, OPAL 181054, OPBA 15458, OPBA 15742, OPBD 12680, OPBD 12600, OPBD 121584, OPM 9350, OPBB 5194, OPBB 5258, OPBB 5700 and OPBB 13755.The details of RAPD markers along with the sequences are given in Table 1.

RAPD analysis
The basic procedure of the RAPD reaction was as follows: 1x reaction buffer, 0.1 mM of each dNTP, Mg 2+ 2 mM, Taq polymerase 1.25 units, template DNA 20 ng, primer 25 ng, each reaction volume is made up to 25 µl with nuclease water.The amplification reactions were carried out by the following profile: 94°C, 5 min for one cycle; 94°C, 1 min; 37°C, 1 min; 72°C, 2 min for 40 cycles; then 72°C, 5 min.The amplification products were analyzed by electrophoresis in 1.5% agarose gels with ethidium bromide.The results were visualized and documented in a gel documentation system (Alpha Imager TM 1200, Alpha Innotech Corp., California, USA).

Microsatellite marker analysis
Microsatellite marker analysis was done based on the protocol reported by McCouch et al. (2002).A set of 400 SSR primers from Research Genetics Inc., USA and from M/s Sigma Aldrich Inc., USA was used for PCR amplification using DNA from parents to identify polymorphic primers.The polymorphic primers were surveyed on the individuals of F2 population.PCR amplifications were performed in 96-well plates on thermal cycler PTC-100 TM MJ Research Inc., USA (or) Eppendorf gradient thermal cycler.The PCR reaction was done in volumes of 15 µl containing 20 ng of DNA, 0.2 µM of each forward and reverse primers, 100 µM of dNTPs, 10 mm Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2, 0.01% gelatin and 0.5 unit of Taq DNA polymerase.The general polymerase chain reaction (PCR) profile is as follows: 94°C for 5 min followed by 35 cycles of 1 min at 94°C, 1 min at 56 to 67°C, specific to the primer pairs used and 1 min at 72°C with final extension of 5 min at 72 and 4°C for infinity to hold the sample.The patterns of amplified products were resolved by silver staining following the procedure described by Panaud et al. (1996).The details of SSR primers used in this study are listed in Table 2.

Data analysis for the F2 mapping population
The basic statistics viz., mean, standard deviation and coefficient of variation were worked out for the six quantitative traits along with disease reaction and pigmentation trait.A standard analysis of variance (one way ANOVA) was used to evaluate mean differences among the three classes defined by each SSR marker locus and two classes defined by RAPD markers for different quantitative traits to establish phenotype-marker association as suggested byBeckmann and Soller (1986) and Edwards et al. (1987) using the general linear model (GLM) procedure of the Statistical Analysis System (SAS) program (SAS Institute Inc., 1990).A significant Ftest (P < 0.01) indicated segregation of marker locus genotypic classes with phenotype.Significance at the P = 0.05 level was considered suggestive of a quantitative trait loci (QTL) at the marker locus.Simple regression analysis was also performed according to the method described by Haley and Knott (1992) using regression coefficient as a function of unknown QTL parameters.Then, stepwise regression was done to identify the most probable marker combinations associated with the trait of interest with their contributions towards total phenotypic variance.Each quantitative trait was treated as dependent variable and various marker genotypes (scored 1, 2 and 3 for SSR and 0 and 1 for RAPD) as independent variables.Segregation ratios for the SSR and RAPD marker classes were tested for the expected 1:2:1 and 3:1, respectively with chi-square test.Multi-point analysis was carried out to confirm the order of the SSR markers and construct a primary map of White Ponni /Moroberekan using MAPMAKER / EXP MS-DOS 3.0.The Kosambi mapping function was used in the analysis (Kosambi, 1944).

Phenotypic traits and their variations in F 2 mapping population
Out of five phenotypic traits recorded in the 220 individual plants of the mapping population, parents showed a high level of variation for pigmentation, days to flowering, plant height (Figure 3), panicle length and grain yield per plant.White Ponni recorded phenotypic values higher than Moroberekan for days to flowering, plant height, panicle length and grain yield per plant.The genotype Moroberekan possessed the glume pigmentation trait (GPT) character, whereas White Ponni lacked it.The frequency distribution of quantitative traits in F 2 population recorded is illustrated in the histogram in Figure 4.The GPT trait was recorded in the F 2 mapping population depicted in Figure 5.
The F 2 families exhibited a wider variation from parents for all the traits studied.Wider variation among the F 2 families for plant height was observed as it ranged from 48.00 to 125 cm with a mean height of 81.66 cm.Days to flowering among the F 2 individuals ranged from 71 to 126 days with a mean of 99.78 days.Panicle length deviated from 11.00 to 27.00 cm with the mean length of 20.59 cm.Grain yield per plant ranged from 19.54 to 100.26 g with a mean yield of 45.81 g.Phenotypic values of some of the traits studied in the F 2 population exceeded the parental values indicating the presence of transgressive segregation in all the traits.The mean, standard deviation and coefficient of variation for the four quantitative traits viz., plant height, days to flowering, panicle length, grain yield per plant and a qualitative trait viz., purple pigmentation (GPT) trait are given in Table 3.

Rice blast screening of the 220 F 3 families
Two hundred and twenty F 3 families (Twenty plants from each of the corresponding 220 F 2 individual plants) were Figure 2. Experimental setup for artificial rice blast disease screening and variations of disease incidence observed in F3 population and parents.A and B, Well equipped artificial screen houses for rice blast screening with prepared trays beds; C, artificial screen house equipped with mist blowers to maintain RH at 95% and leaf wetness; D, incidence of rice blast disease indicated by specks and burnt up appearance, inside the screen house; E, variations observed in the degree of infection within the F3 families namely, 169, 167-susceptible; 168-moderately resistant; 166-moderately susceptible; F, resistant F3 families of the mapping population; G, closer view of rice leaf blast disease at earlier stage; H, highly susceptible parent, White ponni (P1) and the resistant parent, Moroberekan (P2).Twenty seven F 3 families were moderately susceptible (MS) with the mean PDI percent ranging from 56.67 to 77.41 and 5.10 to 6.97.Forty nine F 3 families were highly susceptible (S) with the mean PDI percent ranging from 79.26 to 99.99 and the mean scores ranging from 7.13 to 9.00 (Figure 6).The mean PDI percentage and the mean disease scores recorded in the F 3 families and parents are given in Table 4.

Mapping survey of the F 2 segregating population by RAPD markers
A total of 50 RAPD primers were surveyed on parents to identify polymorphic markers between them.Seven RAPD primers which showed distinct polymorphism were used to genotype the F 2 mapping population which produced 12 markers namely, OPAL 16 940 , OPAL 18 1054 , OPBA 15 458 , OPBA 15 742 , OPBD 12 680 , OPBD 12 600 , OPBD 12 1584 , OPM 9 350 , OPBB 5 194 , OPBB 5 258 , OPBB 5 700 and OPBB13 755.The banding pattern of F 2 individuals for the primer OPBD 12 is shown in Figure 7. F 2 analysis results indicated that, out of 12 markers, 11 fit to the expected Mendelian segregation ratio of 3:1 and the remaining marker deviated from the Mendelian segregation (Table 5).

Mapping survey using micro-satellite markers
A subset of 57 SSR primers from a total of 400 polymor-phic primer pairs was surveyed on 220 F 2 individuals to establish their segregation pattern.Out of 52 primer pairs used, the amplified products of 5 primers could not be scored.Only 52 SSR primer pairs produced unambiguous amplified products.Maximum number of markers were scored on chromosome 7 (10 primers) followed by chromosome 1 (9 primers).Multiple copy fragments were observed in some of the primers.Only a single polymorphic fragment was scored among the multiple fragments.The segregation pattern of SSR marker RM 451 in the F 2 individuals is shown in Figure 8. Out of 52 marker loci, 44 (84.61%) fit into the expected segregation ratio of 1:2:1 based on χ 2 -test at 0.05% probability value (Table 6).Out of 8 markers which deviated from the Mendelian segregation ratio, 5 (9.61%) exhibited segregation distortion towards Moroberekan and 3 (5.77%)skewed towards White Ponni.The overall allele frequency for the 44 loci showed an overabundance of heterozygotic alleles.

Construction of preliminary linkage map using SSR markers
A set of 52 SSR primer pairs which produced polymorphic phic markers between parents was surveyed on 220 F 2 individuals.The segregation data were used for preliminary genetic linkage map construction.A total of 12 markers remained unlinked.Out of these 12 marker loci, 9 (RM 168, RM 280, RM 411, RM 472, RM 492, RM 551, RM 566, RM 4584 and RM 5102) did not show significant deviation from the expected Mendelian segregation ratio indicating their independent assortment.The markers RM 566 and RM 2878, skewed towards White Ponni and the marker RM 81a skewed towards Moroberekan When two-point analysis was done involving 52 SSR markers and 12 RAPD markers, 12 SSR markers remained unlinked.The RAPD markers linked with either SSR marker or with other RAPD markers.Out of 12 RAPD markers, only 3 (OPBD 12 600 , OPBB 5 258 and OPBB 5 194 ) were found to be linked with SSR markers and the remaining markers paired themselves.OPBD 12 600 and OPBB 5 194 was linked with RM1309 on chromosome 8 but these two RAPD markers paired between themselves making the situation complex to assign the RAPD markers on the specific chromosome.OPBB 5 194 was linked with RM 7285 on chromosome 8 and OPBB 5 258 was linked to RM1048 on chromosome 7.The other 9 RAPD markers linked themselves making the situation more complex to place the RAPD markers on specific chromosome.The multipoint analysis carried out on the segregating 52 marker loci using MAPMAKER program resulted in the construction of linkage map with 40 loci.The linkage map thus constructed involving 40 SSR loci is depicted in Figure 9.

Single marker analysis (SMA) to identify the RAPD and SSR markers linked to leaf blast resistance and other quantitative traits
One way ANOVA performed on the mean of groups formed based on the individual segregation pattern of 52 SSR marker loci resulted in the identification of 23 SSR markers putatively associated for the 5 quantitative traits and one qualitative trait studied.These markers were spread across the first nine chromosomes from 1 to 9.
A simple regression analysis was also performed by keeping the phenotypic value as the dependent variable and the individual segregation pattern of SSR marker loci as independent variable.A total of 31 linked markers were identified for 6 traits through SMA from the 23 SSR primers, in which some primers were linked for more than one traits studied.Among the 31 linked markers, 3 SSR markers were linked for leaf blast resistance, 6 markers were linked for  7. Out of 12 RAPD markers studied, 7 markers were found to be linked to 3 traits viz., leaf blast resistance (OPAL 16 940 , OPBD 12 680 , OPBB 5 258 and OPBB 5 194 ), pigmentation character / GPT (OPAL 16 940 , OPBD 12 680, OPAL 18 680 , OPBD 12 600 and OPBB 13 755 ).Among the 7 markers, OPAL 16 940 was linked to more than two traits namely, leaf blast resistance, pigmentation character / GPT and grain yield per plant (Table 7).

DISCUSSION
The rice blast disease caused by M. grisea (Hebert) Barr.
(asexual form known as P. grisea (Cooke) Sacc.), is one of the most serious fungal disease which is widespread and threatening the world rice production (Yu et al., 1996).Numerous races of the fungus are prevalent.Blast resistance genes, commonly called Pi as genes, providing a broad spectrum of resistance against the most prevalent races can be extremely valuable in rice breeding efforts (Fjellstrom, 2003).
Choosing parents is one of the most important steps in any breeding program.No selection method can extract good cultivars if the parents used in the program are not suitable.Therefore, emphasis was given to choose appropriate parents in order to obtain useful segregants.Breeders have different approach to choose parents and have achieved this in different ways.In common, inclusion of at least one locally adapted, popular cultivar as parent (White Ponni) will largely help to ensure the recovery of a high proportion of progenies with adaptation and quality that are acceptable to farmers.The selection of parents would be in such a way that each parent should have the ability to complement the weakness of other.Apart from this, the union of genetically dissimilar parents like Moroberekan would give better recombinants in the segregating progenies.
Selection of parents for the development of mapping population also depends on the performance of parents based on the earlier reports.White Ponni was a moderate yielder with 45 quintals / hectare; superior grain quality for cooking purpose and it is widely used in Tamil Nadu state, India and Southern Indian states.The variety suffers heavy yield losses due to leaf blast and panicle blast.Introgression of resistance genes for leaf blast  might favour the variety to sustain yield losses.
Moroberekan is an upland japonica rice cultivar with apparently durable resistance that has been cultivated for many years in large areas of West Africa without losses to blast disease (Wang et al., 1994).Five resistance genes have been identified in African cultivar Moroberekan.Wang et al. (1994) located two major blast resistance loci, Pi5(t) and Pi7(t), located on chromosomes 4 and 11, respectively, through restriction fragment length polymorphism (RFLP) analysis of the Moroberekan/CO39 recombinant inbred line (RIL) population.Using this same RIL population, Naqvi and Chattoo (1996) and Naqvi et al. (1995) identified and mapped two more major resistance loci, Pi10(t) and Pi157, on chromosomes 5 and 12, respectively.Chen et al. (1999) mapped the blast resistance gene Pi-44(t) using bulk segregant Amplified Fragment Length Polymorphism (AFLP) analysis.
The cultivar has been widely used as a resistance donor in many resistance breeding programs (Wang et al., 1994).It has been suggested that this stable resistance may be due to a combination of complete and partial resistance factors.The preliminary mapping effort by Wang et al. (1994) supported the view that a combination of complete and partial resistance factors may confer stable resistance, and provide a first step in allowing a more efficient utilization of these particular genes in future breeding efforts.The results obtained based on the natural disease screening for partial leaf blast resistance indicated that Moroberekkan was a highly resistant genotype with a highly significant, low mean disease score and PDI%, respectively.Contrastingly, White Ponni was found to be a highly susceptible variety, as evidenced by the highly significant, higher mean disease reaction score and PDI % recorded.
Results obtained pertaining to the artificial screening for durable blast resistance also indicated that the genotype, Moroberekan recorded a highly significant lower mean disease score and mean PDI%.Contrastingly, White Ponni was found to be highly susceptible as evidenced by the highly significant, blast reaction scores and the PDI% obtained (Table 4).
Artificial screening showed variations for different scores of reaction, indicating the involvement of more than one resistance gene.Genetic analysis of blast resistance has indicated that many cultivars carry multiple genes for resistance.In this study, the F 3 population did not show a good fit to the expected genetic ratio (1:2:1) based on Chi-square test in the F 3 generation.Earlier studies indicated that either F 3 populations or backcross were used to confirm the expected genetic ratios (Kiyosawa, 1984;Yu et al., 1987).The number of resistant families (110) was higher than expected.The reasons for segregation distortion might be due to the presence of several resistance genes in a single cultivar, as the cultivar Moroberekan consisted of more than two resistance genes (Chen et al., 1999) which imparts more resistance.The increase in the number of resistance genes would decrease the proportion of plants susceptible to the isolates or the blast pathogen, thus making reliable detection of the segregation ratios more difficult (Yu et al., 1987).
The increased resistance or the increased level of dominance to leaf blast disease reaction observed in this study may be due to the presence of more than two resistance genes in the resistant parent, Moroberekan or might be due to the disease escape during screening.
Besides SSR markers, which usually produce single locus markers, RAPD markers could remain as an alternative marker system considering their ability to produce multi-locus markers.Seven RAPD primers viz., OPAL 16, OPAL 18, OPBA 15, OPBB 5, OPBB 13, OPBD 12 and OPM 9 were used to screen the 220 F 2 individuals of White Ponni / Moroberekan as they generated more number of markers which were clear and polymorphic than the other polymorphic RAPD primers.
The utility of RAPD analysis for genetic map construction was not fully exploited because of its inherent weakness in the technique.RAPD markers are preferred to Inter Simple Sequence Repeat (ISSR) and AFLP markers as their development costs and their running costs per data point were comparatively lower.The amount of DNA required is also lower when compared to other marker systems and conveniently agarose gel electrophoresis can be used instead of acrylamide, which might be carcinogenic if handled improperly.
Among the 12 RAPD markers scored, none of the markers deviated from the expected Mendelian segregation ratio of 3:1, except OPBB 5 258 which skewed more towards Moreoberekan.The RAPD markers showed the overabundance of Moroberekan alleles.Since RAPD markers are of dominant in nature, they lack the power to discriminate the heterozygotic alleles from the dominant alleles in the F 2 mapping population.This was evident from the overabundance of resistant alleles (Moroberekan) rather than the recessive alleles (White Ponni).
The technical efficiency and multiplex potential of SSRs makes them preferable for many forms of high throughput mapping, genetic analysis by diversity studies and marker assisted plant improvement as they are co-dominant and multi-allelic (Coburn et al., 2002;Sharopova et al., 2002).
In the present investigation, among the two marker systems, SSR markers are preferred because of their abundance, technical simplicity, stable inheritance, high level of polymorphism and wide coverage of the whole genome (Chen et al., 1997).Nguyen et al. (2006) developed 190 F 2 progeny / F 3 families from a cross between Danghang-Shali (susceptible), an indica rice cultivar and Hokkai 188 (resistant), a japonica rice cultivar and they selected 126 F 2 progeny/F 3 families randomly for mapping

survey.
Linkage map construction of any organism depends on the availability of the segregating markers covering the entire genome.In the present study, 57 SSR primer pairs which gave clear un-ambiguous polymorphic banding during the parental survey were used on 220 F 2 individuals.Of the 57 primer pairs, only 52 primer pairs could be scored, with clear scorable banding pattern.The reduction occurred since the 5 SSR primers surveyed on the F 2 population could not be scored due to their complex banding pattern.
Out of 52 marker loci, 44 (84.61%) fitted into the expected segregation ratio of 1:2:1 based on χ 2 -test at 0.05% probability value.Out of 8 markers which deviated from the Mendelian segregation ratio, 5 exhibited segregation distortion towards Moroberekan and 3 skewed towards White Ponni.61.53% of the individuals had more heterozygotic alleles for 52 markers surveyed.The overall allele frequency for the SSR loci showed an overabundance of heterozygotic alleles and a low level segregation distortion of genetic markers was observed, since the cross combination of distantly related parents were involved.Wang et al. (1994) observed segregation distortion for almost 127 RFLP markers used in the recombinant inbred population of CO 39, an indica variety and Moroberekan, a japonica variety.Sirithunya et al. (2002) reported that out of 121 marker loci, 93 showed significant segregation distortion (p 0.05) in a study conducted to find out the QTLs associated with leaf and neck blast resistance in RIL population of rice.The distortion of markers could be either due to lethal alleles in gametes or competition among gametes and abortion of gamete or zygote or due to the sampling error (Pillen et al., 1992).
In the present investigation, the percentage of segregation distortion (15.38%) was comparatively lower for SSR markers, when compared to the earlier reports (Wang et al., 1994;Sirithuniya et al., 2002).This might be due to the parents involved in the generation of mapping population, as the susceptible parent, White Ponni (Taichung 65/2 x Mayang Ebos -80) consisted of Taichung 65/2 which is a japonica type and the other parent Moroberekan is also a japonica type.Hence the percentage of segregation distortion might be lower as expected since the parents involved were of the same type (japonica type).Although, it is not possible to construct a genetic linkage map covering the entire genome of rice with the available 52 SSR markers, an attempt was made to develop different linkage groups by performing a linkage analysis.Linkage groups were established for all the chromosomes except chromosome 9.Interestingly, no leaf blast resistance gene was reported so far in chromosome 9 (Sallaud et al., 2003).The multipoint analysis carried out on the marker data of the 52 segregating SSRmarkers resulted in the linkage map with 40 loci.
When two point analyses were done involving 52 SSR markers and 12 RAPD markers, it was observed that all of the RAPD markers were either linked with SSR markers or with the RAPD markers.Out of 12 RAPD markers, only 3 R (OPBD 12 600 , OPBB 5 258 AND OPBB 5 194 ) were found to be linked with SSR markers and the remaining markers paired themselves.OPBD 12 600 and OPBB 5 194 was linked with RM1309 on chromosome 8, but these two RAPD markers paired between them making the situation complex to place the RAPD markers on the specific chromosome.The cause for not having the accurate linkage map could be due to the skewness in the population for marker allele distribution.Smaller number of recombinants in a population due to random assortment or non-linkage might be the major limiting factors to accurately assemble linkage groups (Wang et al., 1994).
In this study, an attempt was made to establish the genotype-phenotype association involving SSR and RAPD markers and six plant traits.Though it was possible to have a genetic linkage map of White Ponni/ Moroberekan with 52 SSR markers in the present investigation, the map was very sparse and incomplete.Considering the situation, the QTL analysis for all the six phenotypic characters was carried out using the SMA.The SMA is a good start not only for learning QTL mapping, but also for practical data analysis.Single marker analysis is the method used in earliest studies on QTL mapping (Edwards et al., 1987;Weller et al., 1988).In this, one marker is considered at a time to find the QTL marker association.The SMA can be implemented as a simple ANOVA, linear regression, t-test and maximum likelihood estimation (Haley and Knott, 1992).SMA is simple in terms of data analysis and implementation.It can be performed using common statistical software like Microsoft EXCEL.Marker orders and complete linkage map is not required.
Based on the phenotypic values of the parents, it is evident that Moroberekan is more resistant to leaf blast than White Ponni.White Ponni is more efficient in the single plant yield than Moroberekan along with the other associated traits.Wider variations with transgressive segregation for different traits were observed among the F 2 families indicating the quantitative nature of the traits.The results indicated that number of SSR markers associated with phenotypic traits ranged from 2 (plant height) to 9 (purple pigmentation).Some of the SSR markers were linked for more than one trait in this study.The marker phenotype commonality was found on chromosomes 3, 4, 5 and 6.The detection of common markers for the earlier mentioned parameters suggested that correlated traits might be influenced by common chromosomal regions.
To identify the common QTLs across genetic backgrounds, the results of the present study are compared with similar QTL studies on other populations.In the present study, QTL for leaf blast resistance was detected on chromosome 2 and 4. Similar results were obtained by    Sallaud et al. (2003) who located resistance loci (RL) on chromosome 1, 2, 5, 6, 7, 10, 11 and 12. Wu et al. (2005) identified 10, 6 and 7 QTLs for percent diseased leaf area (percent DLA), lesion number (LN) and lesion size (LS) located on chromosomes 1, 2, 3, 7, 8 and 11 for percent DLA, 1, 2, 3, 8 and 11 for LN, and 1, 2, 3, 8, 10 and 11 for LS, respectively.Major resistance genes for leaf blast resistance were located on chromosomes 2 and 4 as reported earlier by various workers.Wang et al. (1994) reported Pi-5(t); amajor resistance gene was located in chromosome 4 flanked by the markers RG788 and RG864.Fukuoka and Okuno (2001) reported that the resistance gene Pi-21 was located on the chromosome 4 flanked by the markers G271 and G317.Later, Jeon et al. (2003) reported that the same resistance gene Pi-5(t) was located in chromosome 9 and not in chromosome 4. Sallaud et al. (2003) reported that the marker RG520 was located closer to the resistance gene Pi-25(t) located onchromosome 2. The resistant donor used in both studies was Moroberekan.
Practically, the formation of yield is a very complex physiological pathway and genetic dissection of yield seems considerably difficult (Guo et al., 2005).Complex yield character may be regarded as the end-point of a process of which the successive stages are represented by observed yield attributing traits like productive tillers per plant, grains per panicle, percentage seed set and test weight (Piepho, 1995).In the present study, markers linked for yield was detected on chromosomes 5, 7 and 8. Similar results were obtained by Xiao et al. (1998) reported on chromosomes 2, 4, 5, 8 and 12, Brondani et al. (2002) on chromosomes 1, 2, 3, 4, 7 and 11 and Guo et al. (2005) who reported that the QTLs governing the grain yield per plant were located on chromosome 1, 2, 3, 5, 7 and 10.
At present, the phenotypic screening for blast has its own limitation as the exchange of materials between countries would generate more important information on the genetics of blast resistance and race distribution.
Since the pathogen cannot be imported easily from one country into the other participating countries, the exchange is restricted only to seeds, since strict quarantine measures are employed.Nevertheless, numerous varieties have been exchanged between different countries for various experimental purposes, but the gene analysis of the host has not attracted the blast workers.To determine the relationship between genes effective in each country, seeds of F 3 lines or advanced generations of the hybrid between extremely resistant and susceptible varieties might be suitable (Kiyosawa, 1981).

Conclusion
Although different responses have been shown to induce resistance to different groups of pathogens, there are overlapping responses that confer protection against a wide range of diseases (Paul et al., 2000).Thus, it is clear that plant defense is very complex and involves multiple signaling pathways.Nevertheless, better understandings of these complex pathways are now possible by the development of new approaches and genetic tools such as mutants and microarray technology.

Figure 4 .
Figure 4. Frequency distribution of phenotypic traits over the F2 populations of White Ponni / Moroberekan.

Figure 6 .
Figure 6.Frequency distribution of blast disease score over the 220 F3 families of White Ponni x Moroberekan.

Table 1 .
Details of polymorphic RAPD markers used in segregating F2 mapping population.
3 families, 110 were resistant (R) with the mean PDI percentage range of 12.96 to 33.33 and the mean disease reaction scores ranging from 1.17 to 3.00.Thirty four F 3 families were moderately resistant (MR) with the mean PDI percentage ranging from 33.51 to 55.19 and the mean disease scores ranging from 3.02 to 4.97.

Table 2 .
Details of micro-satellite primers used in the F2 mapping population.

Table 3 .
Variation for phenological traits among the F2 mapping population of White Ponni / Moroberekan.

Table 4 .
Leaf blast disease reactions recorded in the F3 families of the cross White Ponni x Moroberekan under artificial condition

Table 4 .
Leaf blast disease reactions recorded in the F3 families of the cross White Ponni x Moroberekan under artificial condition (contd...)

Table 5 .
Segregation pattern of RAPD markers in F2 mapping population.
Figure 8. Segregation pattern of SSR marker RM 451 in the F2 individuals of White Ponni / Moroberekan.chr12 Figure 9. Frame work linkage map of White Ponni/ Moroberekkan F2 population for SSR markers surveyed.

Table 7 .
Summary of SSR and RAPD marker loci putatively associated with phenotypic traits identified by single marker analysis and simple regression analysis.