Molecular and Morphological Characterisation of Back Cross Generations for Yield and Blast Resistance in Rice

Blast caused by Pyricularia oryzae is one of the major constraints limiting rice production globally. Study was undertaken at TNAU, India using a set of selected genotypes including cultivated varieties and (Near Isogenic Lines) NILs and they were screened for blast resistance both genotypically and phenotypically. Genetic diversity was studied among the genotypes and those which are diverse in blast resistance but had similar morphological and quality traits were selected as parents for breeding programme. Selected genotypes were crossed and advanced to further generations up to BC4F1 by marker assisted back crossing. Genetic variability and heritability parameters were assessed among these segregating generations for yield and resistance traits and their trend in each generation was analysed. The disease resistance traits showed high variability in segregating generations as the parents were selected based on the diversity in disease reaction and therefore gives scope for improvement through selection. High heritability coupled with low genetic advance was found in all characters in the segregating generations except single plant yield and panicle length. *Corresponding author: Dr. Divya Balakrishnan, Scientist (Plant Breeding and Genetics), National Professor Project, Crop Improvement Section, Directorate of Rice Research-ICAR, Rajendranagar-30, Hyderabad, India, Tel: +918374499907; E-mail: divyabalakrishnan05@gmail.com or divyab0005@gmail.com Received January 18, 2016; Accepted August 15, 2016; Published August 25, 2016 Citation: Balakrishnan D, Robin S, Rabindran R, Joel AJ (2016) Molecular and Morphological Characterisation of Back Cross Generations for Yield and Blast Resistance in Rice. J Rice Res 4: 169. doi:10.4172/2375-4338.1000169 Copyright: © 2016 Balakrishnan D, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.


Introduction
Rice (Oryza sativa L.) being one of the prime staple crops of the world has a crucial role in the global food security and contributes for the livelihood of majority of the Asian population. Rice production and productivity are constrained by biotic and abiotic stresses in the growing areas. Rice blast caused by the fungus Pyricularia oryzae has been recognized as one of the most serious diseases and is distributed across 85 countries globally. Host plant resistance has been recognized as an important strategy to manage blast disease compared to chemical control measures. More than 100 blast resistance genes have been identified in rice [1,2] but effective and durable use of these reported genes has been limited because of evolution of new virulent races of the fungus [3]. Enhancing host plant resistance through pyramiding of multiple resistance genes is one of the useful strategies to avoid frequent breakdown of resistance. In addition to overlapping resistance, it could provide cross protection by minimizing the race evolution in the fungus. Combining many resistant traits or resistant genes without appropriate study of their effect on the genetic background may derive undesirable genotypes.
The application of molecular tools in identification of resistance sources and selection of desirable genotypes in segregating generations can accelerate the development of blast-resistant cultivars in an adapted genetic background. Simple Sequence Repeats (SSRs) linked to report resistance have been used for screening the parents and to identify the polymorphism between the genotypes because of their quantity and high polymorphism among rice varieties [4][5][6][7][8]. Rice genetic diversity on the basis of morphological characterization has been studied previously and used in selection of parents in breeding programs [9][10][11][12][13].
Along with this, genetics of resistance traits and their effect on yield is to be assessed to obtain high yielding resistant lines. Study on inheritance and genetic basis of the resistance traits is of high importance, in initial segregating generations or prior to crop improvement programs. It will be helpful in implementing appropriate breeding strategy, and precise screening of advanced generations. Variability parameters for different morpho-agronomic as well as resistance traits were estimated in the segregating generations. It is evident that assessing the variability in the segregating lines under varying environmental conditions is essential for the improvement of popular varieties with stable yield potential. Study was conducted to improve blast resistance in popular Indian rice varieties through marker assisted backcross breeding approach [14]. The parental genotypes were screened for identifying suitable genotypes to develop a high yielding variety with blast resistance. Each segregating populations generated from selfing and back crossing of the selected cross were phenotyped for yield and resistance traits. Information on the phenotypic data of different generations was subjected to analysis of variability and heritability parameters and the trend in each generation was assessed.

Materials and Methods
In this study, a group of blast susceptible and resistant genotypes (Table 1) were subjected to diversity analysis and they were crossed to obtain F 1 s. The parental genotypes and advanced generations obtained from a selected cross (ADT43 × CT13432-3R) were screened for blast resistance and yield parameters using both genotypic and phenotypic measures. Popular Indian cultivar, ADT 43 was selected as recurrent parent for improvement because of its wide adaptability and acceptability in South India owing to its high yield, short duration and grain quality, but it is highly susceptible to blast disease. CO 39 near isogenic line (NIL), CT13432-3R pyramided with four blast resistance genes [15] developed from CIRAD was selected as resistant donor in efficient of variation (GCV), phenotypic co-efficient of variation (PCV), heritability and genetic advance were calculated by adopting the given formulae [17]. The average variance observed in the genotypes was considered as environmental variance. The genotypic variance of each progeny was estimated by subtracting the estimated environmental variance from the phenotypic variance.
Environmental variance (VE) = Average phenotypic variance of both the parents Phenotypic variance (VP) = VG + VE Genotypic variance (VG) = VP -VE Data on quantitative characters were analysed for variances and significance of treatments. The genotypic co-efficient of variation (GCV) and phenotypic co-efficient of variation (PCV) were estimated. The PCV and GCV values were categorized as follows [18]. The heritability was categorized as follows.

PCV and GCV Category
this study. Parental genotypes and advanced generations were raised during 2009 to 2012 in the experimental plots of the Paddy Breeding Station (PBS), Coimbatore and at Hybrid Rice Evaluation Centre (HREC), Gudalur which is an endemic location for blast disease. Advanced generations from the cross of ADT 43 × CT13432-3R viz., F 2 , F 3 , BC 1 , BC 2 and BC 3 were raised and screened in normal as well as epiphytotic conditions. The data obtained from the segregating generations were assessed statistically for various yield and disease related traits viz., plant height, number of tillers, productive tillers, leaf length, leaf width, panicle length, days to first flowering, days to maturity, filled grains per panicle, total grains per panicle, spikelet fertility, 100 grain weight, single plant yield, lesion number, infected leaf area, potential disease incidence percentage, leaf blast and lesion type (Supplementary Table 1). The experimental details are given in Supplementary Figure 1.

Diversity analysis
The parental genotypes were screened for reported molecular markers linked to blast resistance (Supplementary Table 2 Figure 1). The allelic variation of these markers were surveyed at the parental level was and the polymorphic SSR markers were selected for the further screening of the crosses and segregating populations. Diversity of the parental genotypes were studied using phenotypical data and molecular marker information using NTSYs-pc UPGMA cluster analysis and Jaccard's coefficient [16].

Estimation of genetic parameters
The various genetic parameters like variability, genotypic co-

Estimation of diversity among parental genotypes
The parents ADT 43 and CT13432-3R were grouped into different clusters based on blast resistance parameters, but came under similar cluster on morpho-agronomic trait ( Table 2) information ( Figure  2). The F 1 s of these parents were selected for further advancement, so that recovery of the parental genotype was faster with not much compromise on local adaptability and acceptability. Among the parents, resistant donor CT13432-3R was having alleles of all the Pi genes and the susceptible lines were devoid of those alleles which confirmed the linkage of markers to the resistance genes. Many workers employed the technique of molecular analysis along with morpho-agronomic studies and their association in rice [19][20][21] for the simultaneous improvement of various traits.
Cluster analysis based on SSR marker data on parental genotypes revealed that four susceptible genotype (ADT 43, Improved White Ponni, BPT 5204 and CO39) and moderately resistant genotypes

Genetic advance
Genetic advance was estimated by following formula [17] given as where, h 2 = Heritability in broad sense σ p = Phenotypic standard deviation k = Selection differential (2.06 at 5% selection intensity)

Genetic gain
Expected genetic gain under selection was calculated by the following formula [17] [22]. Similarly eight genotypes of rice were studied for genetic analysis with 85 gene linked SSR markers for rice blast resistance and relatedness among eight genotypes were examined through clustering analysis [23].

Estimation of variability parameters
The values of variability parameters like genotypic variance (GV), phenotypic variance (PV), genotypic coefficient of variation (GCV), phenotypic coefficient of variation (PCV), heritability (h 2 ) and genetic advance as per cent of mean for plant height, number of tillers, productive tillers, leaf length, leaf width, panicle length, days to first flowering, days to maturity, filled grains per panicle, total grains per panicle, spikelet fertility, 100 grain weight, single plant yield, lesion number, infected leaf area, potential disease incidence percentage, leaf blast and lesion type pertaining to different segregating populations of the cross ADT 43 × CT13432-3R are presented in Tables 3 and 4. The co-efficient of variations at genotypic (GCV per cent) and phenotypic level (PCV per cent) calculated for the morphological characters observed in various segregating generations are presented in Table 4. In general, the PCV was higher than the corresponding GCV for all the traits under study. The percentage of GCV was less than 20 per cent (low variability) for plant height, leaf length, leaf width, panicle length, days to 50 per cent flowering days to maturity, spikelet fertility, 100 grain weight, single plant yield; 20 to 25 per cent (medium variability) for number of tillers and productive number of tillers; above 25 per cent (high variability) for grains per panicle, lesion number, infected leaf area, potential disease percentage, leaf blast and lesion type.
Selection of parents is one of the most important steps in any breeding program and will helps to extract good cultivars if the parents used in the program were suitable. Therefore, emphasis was given to choose appropriate parents in order to obtain useful segregants. Inclusion of at least one agronomically superior, locally adapted, popular cultivar as parent (ADT 43) in the breeding programme will largely help to ensure the recovery of a high proportion of progenies with adaptation and quality that were acceptable to farmers. The variability in parental genotypes was studied for different morpho-agronomic as well as resistance traits. Mean, standard deviation and variance values for all the traits under study were exhibited less difference in the segregating generations of the cross ADT 43 × CT13432-3R (F 2 , F 3 , BC 1 F 1 and BC 2 F 1 ). But in BC 3 F 1 a reduction in mean plant height, leaf length, panicle length, total grain per panicle and 100 grain weight was observed. Similarly, there was an increase in single plant yield, number of tillers and productive tillers which indicate the maximum restoration of the desirable parent in the third backcross generation as the population mean was more towards the recurrent parent ADT 43 compared to previous generations.
In the segregating generations of the cross ADT 43/ CT13432-3R, co-efficient of variations, PCV and GCV were classified following the low, moderate and high scale [18]   variability for plant height, panicle length, spikelet fertility and 100 grain weight was low as it was evident from low GCV and high variability was exhibited by filled grains per panicle, spikelet sterility and disease parameters like lesion number infested leaf area, potential disease incidence per cent, leaf blast and lesion type. All other characters have shown an intermediate genotypic component of variation. The disease resistance traits showed high variability in segregating generations as the parents were selected based on the diversity in disease response. The high GCV gives an indication of justifiable variability among the genotypes with respect to these characters and therefore gives scope for improvement through selection. In general, the PCV was higher than the corresponding GCV for all the traits under study (Figure 3). The minor variation between values of GCV and PCV shows the limited role of environment in these characters and the heritability was very high for these traits. Selection for improvement of such characters will be rewarding in this situation. Similar results were reported by Bisne et al. [24][25][26][27] in rice.

Estimation of heritability (h 2 ) and genetic advance
Broad sense heritability was estimated for different traits and computed across five generations ( Table 4). The trait such as plant height, number of tillers, days to 50 per cent flowering, number of grains per panicle, spikelet fertility per cent and 100 grain weight had high heritability (h 2 ) values in all the segregating generations. The blast scoring traits viz., lesion number, infected leaf area, potential disease percentage, leaf blast and lesion type also recorded the high h 2 values of 75. 14 (25.65) and all other traits were recorded a lower genetic gain of less than 10 per cent. The heritability percentage was categorized [28]. The genetic advance as per cent of mean was categorized as low, moderate or high [17]. High heritability coupled with low genetic advance was found in all characters in the segregating generations except single plant yield and panicle length. This is the indication of predominance of epistasis and dominant gene action. The traits like number of tillers, spikelet sterility, 100 grain weight, and leaf blast and lesion type recorded simultaneous higher heritability and genetic gain compared to other traits and could be further improved through individual line selection ( Figure 4). Frequency distribution curve of different agronomic yield and resistance traits were made in segregating generations with the normal distribution and given in the Figure 4. Similarly high heritability for most of the yield contributing traits was reported in rice [25,27,29,30].

Conclusion
Clustering of the parental genotypes on morphological traits on yield parameters grouped the major parents under study ADT 43 and CT13432-3R into same cluster but came under different clusters when grouped based on leaf blast score and molecular marker data generated using polymorphic SSR markers. In the segregating populations high variability was exhibited by filled grains per panicle, spikelet sterility and disease parameters like lesion number, infested leaf area, potential disease incidence percent, and leaf blast and lesion type. In general, the PCV was higher than the corresponding GCV for all the traits under study. Traits like number of tillers, spikelet sterility, 100 grain weight, leaf blast and lesion type recorded simultaneous higher heritability and genetic gain compared to other traits indicating the predominance of epistasis and dominant gene action.
Selection of genetically divergent parents and clear understanding of genetics of the selected traits have brought forth gene introgressed lines in adapted genetic background. Any marker assisted breeding programme has to be initiated with a detailed phenotypic and genotypic study of parental genotypes. It is essential to understand the behaviour of reported markers in the novel genetic background of the parents for a successful breeding programme for varietal development. It is also important to understand the genetics of the traits involved and their interactions in advanced segregating generations for the simultaneous improvement of various traits in an existing variety. In this study both the phenotypic and genotypic information was equally employed for the development of a high yielding blast resistant variety with better adaptability and acceptability.