CHARACTERIZATION OF SOME RICE GENOTYPES FOR FERTILITY RESTORING GENES USING RAPD AND SSR MARKERS

o meet the demand of increasing population and maintain selfsufficiency the present rice production needs to be increased by 30% by the year 2020 (Singh et al., 2012). The task is quite challenging and the options available are very limited in view of plateauing trend of yield in high productivity areas, decreasing and degrading land and scarcity of water and labour (Santhanalakshmi et al., 2010). Rice (Oryza sativa L.) is one of the major crop species in which hybrids are used commercially (Selvaraj et al., 2011). It is a major staple food that feeds more than one-half of the human population. During the past decade, hybrid rice technology has emerged as one of the most practical and acceptable approaches to achieve this target (Tiwari et al., 2011). It has not only contributed to food security but also benefited the environment (Sreedhar et al., 2011). The current level of rice production will not meet future demand either in national or international level. Egypt has to be increase the average national rice yield by about 25-30% to meet the demand of increasing population. Yield of traditional rice cultivar is now considered to have reached the plateau. So, achieving further yield improvement is now unsatisfactory. Exploitation of heterosis through hybrid rice technology is an important strategy to increase grain yield beyond the present ceiling. Identification of effective maintainers and restorers are of great importance in any hybrid rice breeding program based on cytoplasmic male sterility (CMS) (Bijral et al., 1991). It is a common phenomenon in higher plants characterized by sterile, non functional pollen grains due to the alterations in mitochondrial genome. The sterility can be restored by crossing CMS lines with the restorer lines containing fertility restorer genes (Rf). In rice, wild abortive CMS (WA-cms) source has been utilized extensively in commercial hybrid rice seed production duo to its stability, excellent out crossing potential and the availability of broader genetic base for the restorer lines (Virmani et al., 1994). Restorer lines play an important role in successful hybrid rice development. They are detected conventionally through test cross procedure by crossing rice germplasm lines (male parents) with the sterile CMS lines (female parents) and the F1's are evaluated for the pollen and spikelet fertility.

o meet the demand of increasing population and maintain selfsufficiency the present rice production needs to be increased by 30% by the year 2020 (Singh et al., 2012).The task is quite challenging and the options available are very limited in view of plateauing trend of yield in high productivity areas, decreasing and degrading land and scarcity of water and labour (Santhanalakshmi et al., 2010).Rice (Oryza sativa L.) is one of the major crop species in which hybrids are used commercially (Selvaraj et al., 2011).It is a major staple food that feeds more than one-half of the human population.During the past decade, hybrid rice technology has emerged as one of the most practical and acceptable approaches to achieve this target (Tiwari et al., 2011).It has not only contributed to food security but also benefited the environment (Sreedhar et al., 2011).The current level of rice production will not meet future demand either in national or international level.Egypt has to be increase the average national rice yield by about 25-30% to meet the demand of increasing population.Yield of traditional rice cultivar is now considered to have reached the plateau.So, achieving further yield improvement is now unsatisfactory.Exploitation of heterosis through hybrid rice technology is an important strategy to increase grain yield beyond the present ceiling.Identification of effective maintainers and restorers are of great importance in any hybrid rice breeding program based on cytoplasmic male sterility (CMS) (Bijral et al., 1991).It is a common phenomenon in higher plants characterized by sterile, non functional pollen grains due to the alterations in mitochondrial genome.The sterility can be restored by crossing CMS lines with the restorer lines containing fertility restorer genes (Rf).In rice, wild abortive CMS (WA-cms) source has been utilized extensively in commercial hybrid rice seed production duo to its stability, excellent out crossing potential and the availability of broader genetic base for the restorer lines (Virmani et al., 1994).Restorer lines play an important role in successful hybrid rice development.They are detected conventionally through test cross procedure by crossing rice germplasm lines (male parents) with the sterile CMS lines (female parents) and the F 1 's are evaluated for the pollen and spikelet fertility.
Molecular marker assisted identification with high power of genetic resolutions has emerged as a robust technique for cultivar fingerprinting, identity profil-T ing, estimating and comparing genetic similarity, and variety protection.Several types of molecular marker such as RAPD (Wang and Lu, 2006;Ichii et al., 2003), SSR (Yashitola et al., 2002;Nandakumar et al., 2004) have been used in this term.These markers are highly polymorphic and easy to detect.The high polymorphism means that these markers can be used in germplasms that is closely related (Ni et al., 2002).Recently, a fairly dense SSR map of rice has been published (McCouch et al., 2002).Agronomical important genes can provide useful information for plant breeders.In this line, our objective was to study the maintenance and restoring ability of the pollen parents over different cytoplasmic sources, and to detect an important Rf locus in rice using RAPD and SSR markers which could be used in hybrid rice breeding program depending on molecular assisted selection (MAS).

MATERIALS AND METHODS
Four CMS lines namely, G46A (Gambica cytoplasmic source), D297A (Dissi), Yimi15A (Dian type) and V20A (Wild abortive cytoplasmic source) and ten elite cultivated rice genotypes namely, Sakha 106, Giza 182, Giza 178, Giza 181, Sakha 101, Wita 4, Wita 12, Gz.1368-5-4, IET 1444 and Sakha 105 rice cultivar were comprised the materials for present study.The pollen parents were planted in plot with a spacing of 20x20 cm.Before crossed, all CMS plants were tested separately by pollen sterility to ensure 100% pollen sterility of CMS plants.The Line x Testers design was used according to Kempthorne (1957).The generated 40 crosses along with their parents were grown in randomized complete block design with three replications.Thirty day old seedlings were transplanted in a single seedling per hill; each experimental unit consisted of 5 row of 3 m length with the spacing of 20 cm.x 20 cm.All the recommended agronomic package of practices was followed.In each entry, five plants were randomly selected from each replication and the observations were recorded for spikelets fertility percentage, No. of filled grains/ panicle, No. of panicles/ plant, 1000-grain weight (g) and grain yield (kg/m 2 ) following the Standard Evaluation System for Rice (IRRI, 1996).

Classification of pollen parents
The pollen parents were classified into four categories; Maintainers (M), Partial Maintainers (PM), Partial Restorer (PR) and Restorer (R) according to the scale of Manuel and Rangaswamy (1993) (Table 1).

Plant materials and DNA isolation
Seven rice genotypes, including one cytoplasmic male-sterile line (GA46A), three restorer lines (Giza 181, Giza 182 and Giza 178) and their F 1 hybrids combinations were used in this study.Genomic DNA was isolated from young leaves according to Doyle and Doyle (1990) procedure.The quantity and quality of DNA was assessed with Nano Drope.

RAPD and SSR analyses
Twelve randomly decamer primers were tested.PCR was carried out in 25 μl reaction mixture containing 30 ng of template DNA, 200 μM of each dNTPs, 10x Taq buffer PCR (10 mM Tric-HCl, 50 mM KCl (pH 8.8), 2.5 mM of MgCl 2 , 1.2 U of Taq polymerase, and 1 μM each of the RAPD primers using a MJ research thermal cycler.Amplification reaction was started by initial denaturation of template DNA at 94C for 4 min and followed by 40 cycles at 92C for 1 min, 35C for 1 min, and 72C for 2 min.Final extension was at 72C for 5 min.PCR amplified products were separated by electrophoresis in 1.5% agarose gels at 70 V in 0.5x TBE buffer.Gels were stained with ethidium bromide and imaged using Biometra (UV-solo model) gel documentation system.Each reaction was repeated twice and only reproducible bands were considered for analysis.
Ten microsatellite markers (SSRs) were used in this study.The sequences of the polymorphic primer pairs are presented in Table (2).Polymerase chain reaction (PCR) was performed in a volume of 20 μl reaction mixture containing 30 ng of template DNA, 200 μM of each dNTPs, 10x Taq buffer PCR (10 mM Tric-HCl, 50 mM KCl (pH 8.8), 2.5 mM MgCl 2 , 1 U of Taq polymerase and 0.2 μM each of the SSR primer pair (Metabion International AG, Germany) using a MJ research thermal cycler.Amplification reactions were initiated by 5 min pre-denatutation at 95C and followed by 35 cycles each at 94C for 1 min, 55C for 1 min, and 72C for 2 min.A final extension step at 72C for 7 min was performed.PCR amplified products were separated by electrophoresis in 2% agarose gels at 70 V in 0.5x TBE buffer.Gels stained with ethidium bromide and then were imaged using Biometra (UV-solo model) gel documentation system.

Means of yield and its component traits
Means of the forty line x tester hybrids (Table 3) indicated worth of genetic variability for the fertility (%), No. filled grains/panicle, No. of panicles/plant, 1000-grain weight (g) and grain yield (kg/m 2 ), which are important in hybrid rice yield.The hybrids V20A x Giza 178 was the most desirable hybrid for high fertility percentage (95.4%)followed by V20A x Wita 12 (94%), D297A x Giza 182 (93.4%) and G46A x Giza 178 (93%).Wide variation among rice genotypes were detected for No. of filled grains/ panicle.G46A x Giza 181 recorded the highest mean value of this trait (221) compared with the other hybrids followed by G46A x Wita 12 (192).V20A x Sakha 101 rice hybrid showed the best combination for No. of panicles/ plant compared with other tested genotypes, the F 1 hybrid rice combination between such parents and G46A recorded the most favorable value for 1000-grain weight.D297A x Giza 182 recorded the best mean value for grain yield (1.04 kg/m 2 ) followed by V20A x Giza 178 (0.98 kg/m 2 ), G46A x Giza 178 (0.95 kg/m 2 ) and D297A x Wita 4 (0.93 kg/m 2 ).The highest performance values of F 1 hybrids concerning grain yield may be attributed to increasing No. of filled grains/ panicle, No. of panicles/plant and 1000-grain weight.

Maintenance and restoring ability
The fertility percentage and category of restorer and maintainer genotypes are shown in Table (3).Maintainers and restorers genotypes were identified according to the scale of Manuel and Rangaswamy (1993) as presented in Table (1).It is clear that Giza 182, Giza 178 and Wita 12 rice parents were identified as restorers for the CMS lines G46A, D297A and V20A (spikelet fertility more than 80%).Contrary, Sakha 106 and Sakha 101 rice cultivars were identified as maintainers for the same aforementioned CMS lines.Giza 181 was identified as restorers for the CMS line G46A and D297A and partial restorer for Yimi 15A and V20A.While, Sakha 105 rice cultivar was identified as maintainers for G46A and D297A.However, the other combinations were identified either partial maintainers or partial restorers which have less of interest in hybrid rice breeding programs.
Results also showed that four testers (Giza 182, Giza 178, Giza 181 and Wita 12) restored the CMS lines G46A and D297A.Also, three testers (Giza 182, Giza 178 and Wita 12) restored V20A while, none of the tested parents restored the CMS line Yimi 15A, indicating that G46A and D297A were easy to restore while, Yimi 15A was hardly to restore.
The difficultly of restoring ability of Yimi 15A may be due to this CMS line belongs to Japonica type and most of restorer lines frequency were belongs to Indica type, suggesting high sterility percentage in F 1 hybrids.These results were, in general, agree with those reported by Murayama and Sarker (2002) who found that F 1 hybrids from crosses between japonica and Indica rice show variable degrees of sterility.
Previous studies showed that among 239 rice genotypes tested for their status in hybrid rice gene pool, 12 restorers and 16 maintainers were identified.Most of genotypes were found to be partial restorers and partial maintainers (Akhter et al., 2008).There are considerable variations in frequency of maintainers for different CMS lines.The average frequency was 5% over the tested crosses.Considerable variability was found in frequency of restorers over different CMS lines and with lines from different sources (Eusebio et al., 2002).

RAPD profiling analysis
Ten informative out of twelve RAPD primers were chosen for DNA profiling (Table 2).A total of 88 reproducible amplifications products were observed and 57 polymorphic bands were scored (64.7%).The lowest and highest numbers of polymorphic products obtained with primers OPK09 and OPK03 with 4 and 13 bands, respectively.The selected CMS line, restorer cultivars and their F 1 hyprids could be uniquely identified by RAPD multilocus amplified profile at the used ten informative RAPD markers.Concerning the restoring linked markers, two primers OPK08 and OPK10 produced fertility restoration linked markers.The markers with the molecular size of 200 and 690 bp for the primers OPK08 and OPK10, respectively (Table 4), were absent in the CMS line whereas found in the selected restorer cultivars and their F 1 hybrids (Fig. 1).

Microsatellite profiling analysis
Six out of the ten SSR loci amplified polymorphic bands in the seven genotypes under study (Table 2).Four markers, RM315, RM288, RM171 and RM179 were monomorphic.The average numbers of bands per primer were 1.9.The number of genotypes possessing a particular locus ranged from 1 (RM315, RM288 and RM179) to 3 for RM302 and RM510 markers.Microsatellite RM302 marker was the most informative locus for genotypic DNA profiling and differentiation.This marker produced two extra bands only in the hybrid genotypes with band size of 495 and 220 bp.The markers RM510 and RM493 could differentiate the CMS line, restorer and their hybrids.RM510 and RM493 produced bands with size of 186 and 128 bp in the restorer cultivar Giza 182 and the hybrid G46A x Giza 182, respectively (Table 4), whereas they produced bands with the size of 165 and 123 bp in the restorer cultivar Giza 181 and the hybrid G46A x Giza 181, respectively, while it amplified bands with the size of 162 and 126 bp in the restorer cultivar Giza 178 and the hybrid G46A x Giza 178, respectively.Non-parental bands were also observed in addition to parental bands in the hybrid between CMS line G46A and the restorer cultivars using primer RM320.In addition to the monomorphic band with the size of 200 bp, this marker produced two extra specific bands to the hybrids with the size of 495 and 220 bp (Fig. 2).
A common feature of the minority of SSR markers is the presence of a nonparental extra band in hybrid individuals, these bands aside from the parent-specific markers indicating hybrid vigour.Xiao et al. (1996) also showed that hybrid vigour has a positive relation with extra banding of DNA.The result from our study suggests that heterosis is positively correlated with the extra DNA banding that appeared in the DNA profile of F 1 and the accumulation of all the parental genes in the F 1 reveals hybrid vigor.In addition the heterosis observed in the F 1 hybrid could be combined alleles associated with hybrid vigor (Young et al., 2004).
Ten microsatellite and twelve RAPD markers were used for DNA profiling of hybrids and their parental lines in this investigation.The results revealed that the SSR loci RM493, RM302 and RM510 and RAPD multiloci OPK08 and OPK10 have the highest efficiency for DNA profiling and discrimination the CMS, restorer lines and their corresponding hybrids.Garg et al. (2006) suggested that some the RM primers were linked to a single Rf gene that possesses co-dominant status can provide a precise and quick alternative to grow out test.F 1 hybrid genotypes were characterized by nonparental band(s) produced by some primers.Similar features were also reported in other markers and plants viz: RAPD codominant markers in soybean (Zheng et al., 2003), rice (Wu et al., 2002), Chrysanthemum hybrids (Huang et al., 2000) and SSR markers in maize heterozygous DH lines (Heckenberger et al., 2002).In addition to the restoring fertility associated marker(s), a single polymorphic marker should suffice to ascertain hybrid seed purity in rice.Also, Nandakumar et al.

SUMMARY
With the objective of identifying restorers and maintainers, to identify fertility restoration linked markers that can distinguish cytoplasmic sources and restorer lines, four male sterile and fertile counterparts of Cytoplasmic Male Sterile (CMS) lines and 10 Restorers (R) lines were characterized.Based on spikelet fertility percentages, ten suggested restorer were identified.Giza 182, Giza 178 and Giza 181 were identified as restorers for the CMS lines G46A and D297A.Ten SSR and 12 RAPD markers were used for the identification of the fertility restoration gene in restorer lines.Among ten suggested restorer lines, only three Egyptian rice cultivars, Giza 182, Giza 178 and Giza 181 used in molecular analysis.Two out of ten SSR markers (RM493 and RM510) could differentiate the CMS line, restorer and their hybrids.RM493 produced bands with size of 123, 128 and 126 bp in the restorer cultivars Giza 181, Giza 182 and Giza 178, respectively.Whereas, the SSR marker RM510 produced bands with size of 165, 186 and 162 bp in the restorer cultivars Giza 181, Giza 182 and Giza 178, respectively.These specific markers were found in their corresponding hybrids with CMS line G46A.In addition, two RAPD primers OPK08 and OPK10 produced fertility restoration linked markers with the molecular size of 200 and 690 bp, respectively.These markers were found only in the restorer cultivars and their respective hybrids with the CMS line G46A.The results suggested effective utilization of SSR and RAPD markers in hybrid test-ing and marker aided heterosis breeding in rice.

(
2004) successfully employed a single restorer gene linked marker assessment for testing genetic purity of hybrid seeds that substantially reduced the time, space and labor.Concisely, concordant with results of the present study, a non-parental extra band is not only useful for fingerprinting the rice hybrids and unambiguous hybrid identity, but could also be used as specific feature for characterization and genetic purity test and detection of off-type seeds in the hybrids based on monomorphic markers.The present study showed that SSR and RAPD markers are quick, effective and results are generally consistent with agronomical traits in the field.Markers identified in the study could be utilized for fertility restoration testing.These markers could be of immense help for hybrid rice seed industry to select appropriate restorer(s) and to detect an important Rf loci in rice.

Table ( 2
): Details of selected primers used in RAPD and microsatellite DNA (SSR) profiling analysis.

Table ( 3
): Mean performance of F 1 hybrids for grain yield and yield related characters, category of restoring ability.