Breeding Cowpea (Vigna unguiculata L.) for High Seed Yield, Protein Content and Harvest Index: Genetic Parameters to Identify Useful Parents

Cowpea (Vigna unguiculata L. Walp.) is a popular grain legume that is versatile, drought-tolerant and nutritious. Indigenous terms like “lobia,” “chowlee,” and many more have been applied to cowpea in India (Singh, 2016; Sharma et al., 2022). Cowpea is cultivated and utilised in a variety of ways and it differ from other crops in that this crop has a wide range of plant type, pod and seed forms as well as a varied growth habit and maturation period. According to Singh (2016), the cowpea grain has a dry weight containing roughly 55-60% carbohydrates, 1% fat and 20-30% protein. Cowpea has the natural capacity to fix nitrogen in even the poorest soils and are resistant to heat and drought. It is a valuable commercial crop and is considered a vital source of protein for both urban and impoverished rural popula-tions. Cowpea leaves and grains are edible by products that are abundant and reasonably priced sources of high-quality protein. All growth phases of the fruits i.e. green pods, fresh or dried seeds, etc., are eaten and young leaves are frequently added to soups (Singh, 2016).  Cowpea grains have a similar nutritional profile to regular beans, however they contain less anti-nutritional and flatulence-causing substances and more folic acid. Global cowpea production was 8.9 million tonnes, occupying 14.9 million hectares (FAOSTAT, 2022). This crop was grown on 55,800 hectares in India, yielding 35,600 tonnes of yield at a productivity of 638.6 kg/ha (INDIASTAT, 2022).  It is mostly grown in the Indian states of Tamil Nadu andhra Pradesh, Gujarat, Rajasthan, Maharashtra and Madhya Pradesh.
       
Shull (1914), provided a lucid explanation of the con-cept of heterosis. The term “heterosis” describes whether an F₁ hybrid is superior or inferior to its parents in one or more characteristics. Taking use of heterosis is a crucial perspective for crop genetic advancement. As growing genotypes across a range of environmental conditions will significantly reduce their potential output, high yielding stable genotypes is the need of the hour (Patel et al., 2021a, Sharma et al., 2022a and Sharma et al., 2024). In self-pollinated crops, hybridization with various mating tech-niques and segregation in subsequent generations are key components of varietal development. The yield and related traits of the ensuing advanced generations are then evaluated in order to determine their suitability as breeding materials (Sharma and Sridevi, 2016, Patel et al., 2021b and Parmar et al., 2024). The most appropriate method for maximising a crop species’ potential for productivity is heterosis breeding, provided that producing F₁ hybrids is both technically and financially possible (Sharma and Shadakshari 2021 and Gandhi et al. 2024a).  The probability of a suitable heterotic combination in F₁ hybrids is extremely low. Only a few number of favourable heterotic combina-tions remain after screening many F₁ crosses (Sharma and Shadakshari, 2021, Sharma et al., 2022b and Gandhi et al., 2024b). Plant breeders frequently argue over the choice of parents for the high frequency of heterotic hybrids that are developed. One crucial factor that was and is utilised to select the parents in order to produce a higher frequency of heterotic hybrids is combining ability (CA). The ability to forecast hybrids performance is where CA is most useful when implemented (Griffings, 1956). Therefore present study was undertaken for identifiying parents which will produce heterotic hybrids for harvest index, protein content and seed yield in cowpea.

Location and genotypes
 
In Summer 2023, at Pulses Research Station, Sardarkrushi- nagar Dantiwada Agricultural University, Sardar-krushinagar, twenty-one crosses were made utilising the line x tester mating design (Kempthorne, 1957) with seven lines and three testers. Therefore, 31 genotypes total-21 crosses, 10 parents including 7 lines and 3 testers were used as the experimental material. These genotypes were asse-ssed using a randomised block design with three replications during the Kharif season of 2023-24. Every entry was planted in a single, 4-meter-long row, with 0.45 metres separating rows and 0.10 metres separating plants within a row. Plant protection measures and agronomic practices were implemented in accordance with guidelines and recommendations.
 
Observations and statistical analysis
 
Observation were recorded on five random plants, at different growth stages of the crop and average values for the traits worked out for 21 F₁s and 10 parents and in each replication on 3 productivity per se traits viz., seed yield per plant, harvest index and protein content. By using Near Infrared (NIR) mass spectroscopy, the protein content was calculated and reported as a percentage. Harvest index was calculated as ratio of seed yield by biological yield and expressed as a percentage. Statistics were applied to the replication-wise mean values of each genotype for a range of characteristics. Using the methods proposed by Panse and Sukhatme (1978), analysis of variance was used to examine the differences between the genotypes for all the features under consideration. The formula for hetero-beltiosis was derived using Fonseca and Patterson’s (1968) methodology. It was quantified as the percentage of the F₁ value’s departure from the superior parent value. The variations resulting from the general combining ability (gca) impacts of three testers and seven lines, as well as the specific combining ability (sca) effects of twenty-one F₁ hybrids, were estimated (Kempthorne, 1957). As a result, overall status regarding the gca effects of the parents and overall status regarding the sca effects of the hybrids and BPH throughout three features were ascertained. Crosses were categorised as HH (both parents in a cross with high overall gca status), HL/LH (one parent with high and the other parent with low overall gca status) and LL (both parents with low overall gca status) based on the overall gca status of the parents.

Analysis of variance for combining ability
 
Analysis of variance for combining ability presented in Table 1 revealed that, the mean square due to lines were significant for protein content; whereas, mean square due to testers was found non-significant for all traits. Significant mean sum of squares due to lines x testers for all the traits suggested that experimental material possessed considerable variability and there were possibilities to improvement of various traits under study through heterosis breeding. A perusal of variance ratio (σ² gca/σ² sca) less than unity suggested the preponderance of non-additive genetic variance for all the characters which further comp-licates the selection of parents but increases the proba-bility of identifying transgressive segregants. Non-additive gene action for these traits were also obtained by Thangaraj (2018) and Debbarma et al., (2022).


     
Trait-wise parental gca effects and hybrid sca and heterosis
 
Variations in the frequency of genes that are passed down to the progeny with the additive effects account for the variations in gca effects between lines and testers. Line PL 4 and tester PL 7 were found to be good general combiner for all the characters having high gca effect. Tester PGCP 12 was good general combiner for harvest index. Harvest index is found to have direct and positive effect of on seed yield per plant. Therefore, these parents could be utilized in future breeding programmes to generate desirable segregates for seed yield and harvest index. Since gca effects are attributed to additive and additive x add-itive gene effects, the above mentioned parents for gca effects have good potential for respective characters and may be used in a multiple crossing programme to synthesize a dynamic population with most of the favourable genes accumulated. 
     
As is true with respect to lines and testers for gca effects (Table 2), the hybrids differed significantly for their sca effects (Table 3). Hybrid GC 7 x GC 4 was desirable specific combiner for seed yield per plant and harvest index. Hybrids viz., GC 3 x GC 4 and GC 7 x PGCP 12 were desirable specific combiner for harvest index and protein content. Hence, these crosses were identified as potential for getting good transgressive segregants for these traits. This result was in agreement with Debbarma et al., (2022) and Joshi et al., (2022).




     
Hybrid viz., EC 724035 × PGCP 12 (Fig 1) and JLS 60 x PL 7 had high better parent heterosis for all the characters except protein content. Cross, GC 3 x GC 4 showed high better parent heterosis for protein content. Therefore, these crosses should be included for further evaluation in the generation advancement for getting good transgressive segregants. However, it was evidenct that characters like protein content may adversely contribute towards the heterosis for seed yield per plant. This may be due to the condition which favoured the development of one component could be adversely affected the other one. Therefore, to obtain good yield with high protein content, desired level of each component should be identified to make an effective selection programme. The results were confirming with those of Ratnakumari et al., (2023) in cowpea.


 
Parental overall gca status and hybrids overall sca and heterotic status
 
Three out of 7 lines viz., JLS 60, EC 72 3909 and PL 4 and two of the three testers viz., PGCP 12, PL 7 displayed high overall gca status and the remaining exhibited low overall gca status (Table 4). Similarly among hybrids, eleven (52.38%) and twelve (57.14%) crosses were classified as having high (H) overall sca and heterotic status, respec-tively and remaining were classified as having low (L) overall sca and heterotic status (Table 5 and 6). Similar results were found in the studies of Bandyopadhyay and Arunachalam (1980), Ramesh et al., (2000) and Keerthi et al., (2016).






 
Relationship of overall parental gca status with hybrids overall sca and heterotic status
 
Requirement of parents with contrasting gca effects to realise higher frequency of hybrids with high overall sca and heterotic hybrids have been observed in many previous studies (Ramesh et al., (2000) and Keerthi et al., (2016)). Results obtained in the present study contrastingly indicated that in Cowpea for the concerned traits high magnitude of heterosis can be produced by either of three combination of parents viz.,  H x H or L x L or H x L / L x H (Table 7). Such results can be attributed to non-additivite gene action found for these traits in our study. So it may not always be possible to identify parents based on gca effects to realise hybrids with high heterosis and before doing so one must consider the gene action governing the trait of interest.

The ability to forecast parental gca impacts on hybrid heter-osis, which would save a significant amount of time and resources, is essential to the effectiveness of any breeding strategy using hybridization. The line PL 4 and tester PL 7 registered as good general combiner for seed yield per plant and harvest index. The crosses GC 7 ´ GC 4 recorded high mean seed yield heterotic effect along with positive significant sca effect for seed yield per plant and harvest index. The perusal of per se performance as well as hete-rotic effect of crosses revealed that, the crosses GC 7 ´ PGCP 12 and GC 3 ´ GC 4 were proved to be best for protein content. Hence, these crosses were identified as potential for getting good transgressive segregants for these traits and suggested for further evaluation in advance generations. Further findings of this study contrast the requirement of parents with contrasting gca effects to realise higher frequency of heterotic hybrids as the present study showed that any combination of parents with regards to gca can lead to heterotic hybrids depending upon the type of gene action governing that trait.

Authors are thankful to Pulses Research Station, Sardarkrushi-nagar Dantiwada Agricultural University, Gujarat, India for providing necessary facilities.

Authors declare there is no conflict of interest to disclose.