Generation of Micro-mutants for Yield and Component Traits in Horse Gram [Macrotyloma uniflorum (Lam.) Verdc.]

Horse gram [Macrotyloma uniflorum (Lam.) Verdc.] is an autogamous diploid legume (2n=20, 22 and 24) with massive nutrient significance. It is a climate resilient crop, widely cultivated up to an altitude range of 1800 m (Chahota et al., 2013) above sea level. This multipurpose legume has been domesticated during pre-historic period around 2000 BC in India (Mehra, 2000). At present, horse gram is widely grown as a grain legume in India, Sri Lanka Nepal, Malaysia, Mauritius, West Indies and Myanmar (Asha et al., 2006). The nutraceutical properties of horse gram were quoted in ancient Indian medicine which serves as a remedy for many diseases (Prasad and Singh, 2015). It was identified as potential food legume for future owing to its nutrient and medicinal importance coupled with drought tolerance.
       
Horse gram is a photosensitive crop with long maturity duration of 100-180 days (Morris, 2008). It is widely cultivated in drought prone areas in India with less input resources (Jansen and Westphal, 1989). Limited scientific efforts were made on evolving horse gram varieties suitable for modern agriculture at global level. The replacement of horse gram cultivable areas with commercial crops was a major reason for the decline in genetic diversity (Chahota et al., 2013). Induced beneficial mutations would circumvent bottleneck situations of self-pollinated crops including horse gram (Mehraj et al., 1999; Bolbhat and Dhumal, 2009). Irradiations tend to induce varying extent of physiological damages, chromosomal aberrations and gene/point mutations in the plant material (Swaminathan, 1968). The modified genetic factors will be transferred to subsequent generations resulting in evolution of novel variants. Therefore, an attempt was made using gamma rays (G), electron beam (EB) and G+EB combination for generation of mutants with desirable agronomic traits in horse gram.

A photo-sensitive cultivar “CRIDA 1-18R” was utilized for irradiation to generate wide variability for yield and its attributing traits in horse gram. The seed materials were obtained from Central Research Institute for Dryland Agriculture, Telangana, India. The well filled seeds were equilibrated to 12% moisture content before carrying out irradiations at Bhabha Atomic Research Centre, Mumbai, India.
       
The treated seeds (M0) with different doses viz., 100 Gy, 200 Gy, 300 Gy and 400 Gy of gamma rays, electron beam and its combination (G+EB) were sown in a randomized block design with two replications. Plant injuries were recorded in field conditions at all phases of crop growth. All the mutant plants (M₁) were harvested on a single plant basis.
       
Based on the mutagenic sensitivity studies, the desirable doses viz., 200 Gy and 300 Gy of gamma rays, 100 Gy and 200 Gy of electron beam and 100 Gy of G+EB combination were forwarded to M₂ generation during 2018. Each individual plant was raised as progeny rows by adopting spatial pattern of 30 × 15 cm. Traits viz., days to first flowering, days to maturity, plant height (cm), number of primary branches per plant, pod length (cm), number of clusters per plant, number of pods per plant, number of seeds per pod, hundred seed weight (g), biological yield (g) and single plant yield (g) were recorded on randomly selected 150 normal looking plants similar to control.
       
The superior genotypes exhibiting higher yield than CRIDA 1-18R were forwarded to M₃ generation during 2019 following parent - progeny row basis. Similar spacing methodology and observations were followed for mutant progenies. The data was recorded on randomly selected normal looking mutant progenies similar to check in each family.
       
Data recorded on M₂ and M₃ population were utilized to determine the degree and direction of induced novel micro-mutants. In this study, the variance from control was considered as error / environmental variance (𝛔2e). The observed variations in mutant population were taken as phenotypic variance (𝛔2p). The genotypic variance (𝛔2g) is measured by the difference between phenotypic and error variance. The estimates of genotypic coefficient of variation (GCV), phenotypic coefficient of variation (PCV) (Burton, 1952), broad sense heritability (H²) (Lush, 1940) and genetic advance as percent of mean (GAM) (Johnson et al., 1955) were calculated using variances. Skewness (β₁) and Kurtosis (β₂) provides information on nature of trait distribution in a segregating population. The frequency pattern was estimated for M₂ and M₃ population using SPSS Software as per formula given by Snedecor and Cochran (1967).

Mean performance of M₂ population
 
Both positive and negative shift in mean values from control were registered for the traits studied in M₂ population of CRIDA 1-18R (Table 1). Delay in flowering and maturity duration was observed among mutant population with an exception of EB: 100 Gy (Rudraswami et al., 2006). The shift in mean value for plant height was varying with dose and mutagens employed. Yield attributing traits viz., number of primary branches per plant, pod length, number of clusters per pod, number of pods per plant and biological yield had low mean values in comparison with control. The frequent occurrence of detrimental mutants in a population would be a reason for reduced trait value (Singh et al., 2000). On the contrary, a positive shift in mean was recorded for number of seeds per pod and hundred seed weight in all treatments. The gamma irradiated population had increased mean value (200 Gy: 2.31 g; 300 Gy: 1.21 g) for single plant yield whereas, negative shift in mean from control was observed at EB and its combination (G+EB) treatment as reports in a earlier study (Wani et al., 2011). Though treatments had low mean performance, several novel variants exhibiting superior trait value than control was isolated.
 

 
Transgressive variants in M₂ population
 
Prediction on frequency and magnitude of superior variants would offer scope in identification of mutants with improved trait value (Misra et al., 2008). The mutant expressing trait value higher than the highest value of control was termed as positive transgressive variant (PTV) and the trait value lower than the lowest value of control was termed as negative transgressive variant (NTV). The maximum and average transgressive variation were estimated for respective PTVs and NTVs to determine the extent of variation from check. The lower dose of electron beam can be employed in future breeding programme to evolve early duration lines in horse gram (Table 2). The promising dose for a trait was identified based on simultaneous measure of frequency, average and maximum transgressive variation. The frequent occurrence (13) of superior variants with high positive average (APTV: 7.27 g) and maximum transgressive variation (MPTV: 12.76 g) for single plant yield was exhibited at 100 Gy of EB. Based on three parameters, the enhanced trait improvement would be achieved with the following doses: G (200 Gy) for number of primary branches per plant and biological yield; G (300 Gy) for plant height; G (200 Gy and 300 Gy) and EB (100 Gy) for hundred seed weight; G (300 Gy) and EB (200 Gy) for number of pods per plant; G (200 Gy and300 Gy) for number of seeds per pod; G+EB(100 Gy) for number of clusters per plant; and EB (200 Gy) for pod length.
 

 
Mean performance of M₃ population
 
The genotypes expressing high single plant yield (> 30 g) than CRIDA 1-18R were isolated and forwarded to M3 generation. Traits viz., plant height, pod length, number of clusters per plant and number of pods per plant exhibited an increase in mean value from M₂ to M3 generation (Table 3). Adoption of direct selection on yield proved highly effective in improvement of traits at M₃ generation (Khan et al., 2004). The shift in mean from control decreased in M₃ population compared to its respective M₂ for flowering and duration as reported by Waghmare and Mehra, 2000 earlier. The decline in mean value over control for biological yield in both generations may be a resultant of altered physiological activity or induction of mutations with negative effects. On the contrary, the induction of mutants with positive effects was observed in M₂ and M₃ population for hundred seed weight (Khan and Wani, 2006). Among mutagens, G+EB: 100 Gy recorded the highest mean value for pod length (5.05 cm), number of pods per plant (54.42) and single plant yield (25.68 g) in M₃ generation. All the mutagenic treatments of M₃ population exhibited high single plant yield compared to its respective M₂ except G: 300 Gy. The increased yield potential through adoption of early generation selection was reported earlier by Khan and Qureshi (2006).
 

 
Frequency distribution pattern of mutant generations
 
Skewness and kurtosis provide better insight on distribution pattern of variants in a population (Table 4). Directional selection resulted deviations in frequency pattern between mutant generations. Most of the yield attributing traits viz., pod length, number of clusters per plant, number of pods per plant, number of seeds per pod and hundred seed weight had varied distribution pattern with reduction in high order statistic at M₃ generation. Reduction in kurtosis value for flowering (except G: 200 Gy) and maturity denotes the absence of extreme duration types in M₃ population. Adoption of intense selection at M₃ progenies of gamma rays (200 Gy) and its combination (G+EB: 100 Gy) would result in rapid genetic gain for plant height (Roy, 2000) as it showed skewed distribution towards right with leptokurtic curve. High selection intensity can be implemented in gamma rays (200 Gy) and electron beam (100 Gy and200 Gy) to evolve genotypes with increased number of primary branches. The positive skewed distribution indicated the frequent occurrence of variants with decreasing effect for biological yield at both generations. However, the leptokurtic nature of two M₃ population (G: 200 Gy and EB: 100 Gy) would contribute genotypes with increased biomass on adoption of intense selection. All the irradiated treatments of M₃ generation (except EB: 100 Gy) had either non-significant skewness or kurtosis indicating the absence of epistatic interactions for single plant yield (Pooni et al., 1977).
 

 
Extent of variability with H² and GAM
 
High variance over control indicated the potentiality of employed doses in different mutants generation (Table 5).Traits viz., plant height, number of seeds per pod and single plant yield exhibited an increase in variability pattern from M₂ to M₃ generation at all employed doses. Similarly, the release of inherent variability contributed higher GCV and PCV estimate at M₃ generation for traits viz., number of primary branches per plant, number of clusters per plant, number of pods per plant and biological yield with few exceptions. On the contrary, low extent of PCV and GCV was recorded for days to flowering and maturity at both generations (Wani, 2011). Among mutagenic treatments, G: 200 Gy and G+EB: 100 Gy exhibited high values of PCV (21.95% and 21.46%) and GCV (20.47% and 20.19%) for single plant yield at M₃ generation. The increase in variability estimates offer scope for further yield improvement at subsequent generations through effective selection (Muduli and Misra, 2008).
       

 
Selection response can be predicted better by combining the extent of variability parameters with heritability (H²) and GAM (Kharkwal, 2003). The decline in H² and GAM values for days to flowering and maturity at M₃ population indicate the governance of non- additive gene action (Table 6). The decline in H² estimate for number of primary branches at M₃ generation may be resultant of high environmental effects. Traits viz., plant height, pod length, number of clusters per pod, number of pods per plant, number of seeds per pod, biological yield and single plant yield exhibited a gradual increase in both parameters from M₂ to M₃ generation with few exceptions. The increase in H² and GAM values at M₃ generation shows the preponderance of additive gene effects for yield and most of its component traits (Mensah et al., 2005).
 

Direct selection for yield proved promising for all the mutagenic treatments except 300 Gy of gamma rays. Adoption of early generation selection would be beneficial in capturing favorable combination of alleles which eventually disappear in the advanced generations. Among the mutagenic treatments, combination of G+EB: 100 Gy was found promising for single plant yield, number of pods per plant and pod length. The increased mean coupled with high H² and GAM at M₃ population indicates the preponderance of additive gene effects for most of the polygenic traits. Increased variability at M₃ generation provides scope for yield improvement on further selection. The positive kurtosis at lower dose of electron beam (100 Gy) and its combination (G+EB: 100 Gy) treatment indicate the presence of superior variants for single plant yield. The identified high yielding progenies (more than 40 g) will contribute for enhanced yield potential in horse gram.

We acknowledge sincerely the Board of Research in Nuclear Sciences for rendering financial assistance towards this study.