Introduction
Bitter gourd (Momordica charantia) is a nutritionally rich and therapeutically important cucurbitaceous vegetable extensively grown throughout the tropics and subtropics of the world. Higher yield, earliness, gynoecy, fruit size, fruit shape, fruit colour and resistance to diseases and pests are the major breeding objectives in this crop (Rao, Reference Rao, Al-Khayri, Jain and Johnson2021). Crop wild relatives can be exploited for their untapped favourable alleles for improvement of cultivated varieties. A cultivated species M. charantia consists of two botanical varieties viz., M. charantia var. muricata, a wild variety with smaller dark green fruits having markedly sculptured seeds and M. charantia var. charantia, a cultivated botanical variety producing large fusiform fruits which are most preferred in the market (Chakravarty, Reference Chakravarty, Bates, Robinson and Jeffrey1990). Though reported partially cross-compatible with M. charantia var. charantia (Rathod et al., Reference Rathod, Behera, Munshi, Vinod and Jat2019) and harbouring field tolerance to fruit fly (Dhillon et al., Reference Dhillon, Singh, Naresh and Sharma2005) and moderate resistance to powdery mildew (Prasanth et al., Reference Prasanth, Varalakshmi, Venugopalan and Sriram2019), M. charantia var. muricata has not so far been exploited for bitter gourd improvement, especially for the yield and related traits. The present work aimed at studying the cross ability and potential of the wild variety in yield improvement of cultivated forms.
Experimental setup
Parental lines
The M. charantia var. charantia cv. Priyanka, developed through single-plant selection from a local collection from Thiruvalla, Kerala state, India, and released for commercial cultivation by Kerala Agricultural University, India, bearing greenish white spindle-shaped large fruits; and M. charantia var. muricata wild-type accession IC634896, collected from Kottayam, Kerala state, India, and having small, round and dark green fruits, were used as the parental lines. Breeder seeds of the cultivar Priyanka were procured from its breeder at Kerala Agricultural University whereas the seeds of IC634896 were obtained from Regional Station, ICAR-National Bureau of Plant Genetic Resources, Thrissur, India.
Evaluation of cross-compatibility
Priyanka × IC634896 (hybrid 1) and IC634896 × Priyanka (hybrid 2) crosses were made during October–December 2020, under net house conditions, following the standard procedure (Behera et al., Reference Behera, Behera, Bharathi, John, Simon, Staub and Janick2010). Fruit set and germination rate of hybrid seeds were recorded for evaluating the cross ability between the botanical varieties. Seeds extracted were cleaned, shade-dried and treated with hot water (50°C) for 3 h followed by GA3 (60 ppm) for 12 h and mixture of ZnSO4 (1.0%), KNO3 (2.0%) and KCl (1.0%) for 1 h. Fifty seeds each of both parents and hybrids were immediately sown in polybags and germination percentage was calculated 15 days after sowing.
Evaluation of hybrids
Following the treatment mentioned above, F1 seeds were sown in polybags and 10-d-old seedlings were transplanted into pits of 60 × 60 × 30 cm3 size, and raised in a spacing of 2 × 2 m2 during February–April 2021. Hybridity of F1 plants was confirmed using a microsatellite marker McSSR62 polymorphic for the parental lines (online Supplementary Fig. S1) and four representative plants for each cross were used for phenotypic evaluation.
Results and discussion
Of the 30 and 26 crosses made with Priyanka × IC634896 and reciprocal direction, respectively, cent per cent fruit set was obtained, showing the genetic relatedness of the botanic varieties. Though these varieties were reported to be genetically close (Bai and Beevy, Reference Bai and Beevy2012; Bharathi et al., Reference Bharathi, Munshi, Behera, Vinod, Bhat, Das and Sidhu2012), complete cross-compatibility has never been obtained (Rathod et al., Reference Rathod, Behera, Munshi, Vinod and Jat2019). The results suggest that there could be higher genetic variability within M. charantia var. muricata, leading to different levels of success when crossed with the cultivated form. Seeds of hybrid 2 showed only 67.0% germination compared to 100% in seeds of hybrid 1, due to the hard seed coat nature inherited from the female parent M. charantia var. muricata. In the wild parental line also, germination rate was poor (34%). Hard seed coat acts as a physical barrier for moisture absorption and germination of the seed. Seed treatment as given above has enhanced the germination in M. charantia var. muricata and hybrid 2 by nearly 30%.
Evaluation of hybrids
Both hybrids showed intermediate phenotype viz., seeds size, fruit length, fruit breadth and leaf size (Table 1). Hybrid 1 was superior for fruit length and breadth, peduncle length, flesh thickness, fruit weight, number of fruits per plant and fruit yield (Fig. 1). Fruit number in hybrid 1 was 80 compared to 17 in the commercially cultivated variety cum female parent Priyanka. Fruit yield of hybrid 1 (2287.4 g) was more than double of that in Priyanka (1042.3 g) whereas hybrid 2 yielded only 783.7 g. The hybrid 1 plants also flowered earlier compared to the hybrid 2 plants. Fruits harvested from the F1 hybrids of both crosses had similar cooking qualities as the cultivated parent Priyanka. Yield changes with direction of cross confirm the reports that maternal parent has more significant contribution towards the trait expression in offspring (Gopalakrishnan, Reference Gopalakrishnan1986; Devadas, Reference Devadas1993; Sundaram, Reference Sundaram2006).
Performance of M. charantia var. charantia, M. charantia var. muricata and their hybrids for yield traits
Fruit characteristics of (A) Priyanka, (B) IC634896, (C) Priyanka × IC634896 hybrid and (D) IC634896 × Priyanka hybrid.
This study has shown that wild relative M. charantia var. muricata is a potential donor while breeding bitter melons for higher fruit number and consequent yield, especially when used as pollen parent. Since the fruit size in the hybrid was smaller, back crossing with commercial cultivars can improve it, without compromising on the fruit number. For this, mapping the genes and quantitative trait locus contributing to the fruit number and other fruit and yield traits will be highly useful.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1479262123000734.
