Why low use of germplasm in breeding

Reasons for the underutilization of germplasm include the following: (1) non-availability of reliable information on traits of economic importance; (2) linkage load of undesirable genes and assumed risks; (3) restricted access to germplasm collections and regulations governing international exchange; (4) enhanced role of non-additive genetic variations when diverse germplasm collections are used by the breeders; (5) lack of robust, cost-effective tools to facilitate the efficient utilization of exotic germplasm in crop breeding; and (6) limited exposure to available germplasm and re-circulation of the same genotypes in breeding (Duvick, Reference Duvick1995; Dwivedi et al., Reference Dwivedi, Upadhyaya and Gowda2009; Upadhyaya et al., Reference Upadhyaya, Dwivedi, Ambrose, Ellis, Berger, Smýkal, Duc, Dumet, Flavell, Sharma, Mallikarjuna and Gowda2011a).

Developing representative sets as a means to discover new sources of variation

Core (Frankel, Reference Frankel, Arber, Limensee, Peacock and Stralinger1984) and minicore (Upadhyaya and Ortiz, Reference Upadhyaya and Ortiz2001) collections, which represent diversity of the entire collection of a given species, have been suggested as a gateway to enhance the utilization of germplasm in crop breeding. These subsets are available in chickpea (Upadhyaya and Ortiz, Reference Upadhyaya and Ortiz2001; Upadhyaya et al., Reference Upadhyaya, Bramel and Singh2001), groundnut (Upadhyaya et al., Reference Upadhyaya, Bramel, Ortiz and Singh2002, Reference Upadhyaya, Ortiz, Bramel and Singh2003), pigeonpea (Reddy et al., Reference Reddy, Upadhyaya, Gowda and Singh2005; Upadhyaya et al., Reference Upadhyaya, Reddy, Gowda, Reddy and Singh2006b), pearl millet (Upadhyaya et al., Reference Upadhyaya, Gowda, Reddy and Singh2009a, Reference Upadhyaya, Yadav, Reddy, Gowda and Singh2011e), sorghum (Grenier et al., Reference Grenier, Hamon and Bramel-Cox2001; Upadhyaya et al., Reference Upadhyaya, Pundir, Dwivedi, Gowda, Reddy and Singh2009b), finger millet (Upadhyaya et al., Reference Upadhyaya, Gowda, Pundir, Reddy and Singh2006a, Reference Upadhyaya, Sarma, Ravishankar, Albrecht, Narsimhudu, Singh, Varshney, Reddy, Singh, Dwivedi, Wanyera, Oduori, Mgonja, Kisandu, Parzies and Gowda2010), foxtail millet (Upadhyaya et al., Reference Upadhyaya, Pundir, Gowda, Reddy and Singh2008b, Reference Upadhyaya, Ravishankar, Narsimhudu, Sarma, Singh, Varshney, Reddy, Singh, Parzies, Dwivedi, Nadaf, Sahrawat and Gowda2011c) and proso millet (Upadhyaya et al., Reference Upadhyaya, Sharma, Gowda, Reddy and Singh2011d).

Research carried out to date suggests these subsets to be useful for finding germplasm with agronomically beneficial traits, e.g. resistance to abiotic and/or biotic stress in chickpea (Upadhyaya et al., Reference Upadhyaya, Dronavalli, Dwivedi, Kashiwagi, Krishnamurthy, Pande, Sharma, Vadez, Singh, Varshney and Gowda2013a), finger millet (Kiran Babu et al., Reference Kiran Babu, Thakur, Upadhyaya, Reddy, Sharma, Girish and Sarma2013), groundnut (Upadhyaya et al., Reference Upadhyaya, Dwivedi, Vadez, Hamidou, Singh, Varshney and Liao2014), pigeonpea (Krishnamurthy et al., Reference Krishnamurthy, Upadhyaya, Saxena and Vadez2012; Sharma et al., Reference Sharma, Rathore, Mangala, Ghosh, Sharma, Upadhyaya and Pande2012a), pearl millet (Sharma et al., Reference Sharma, Upadhyaya, Manjunatha, Rai, Gupta and Thakur2013a) and sorghum (Sharma et al., Reference Sharma, Rao, Upadhyaya, Reddy and Thakur2010, Reference Sharma, Upadhyaya, Manjunatha, Rao and Thakur2012b; Vadez et al., Reference Vadez, Krishnamurthy, Hash, Upadhyaya and Borrell2011). Genetically diverse and nutritionally dense germplasm accessions have also been reported in finger millet (Upadhyaya et al., Reference Upadhyaya, Ramesh, Sharma, Singh, Varshney, Sarma, Ravishankar, Narasimhudu, Reddy, Sahrawat, Dhalalakshmi, Mgonja, Parzies, Gowda and Singh2011b), foxtail millet (Upadhyaya et al., Reference Upadhyaya, Ravishankar, Narsimhudu, Sarma, Singh, Varshney, Reddy, Singh, Parzies, Dwivedi, Nadaf, Sahrawat and Gowda2011c) and groundnut (Upadhyaya et al., Reference Upadhyaya, Dronavalli, Singh and Dwivedi2012a, Reference Upadhyaya, Mukri, Nadaf and Singhb).

Population structure, diversity, allele mining and association genetics

Understanding how diversity is structured to unlock its potential for crop improvement is an emerging area made possible by rapid advances in the scale, robustness and reliability of marker technologies and the sharp fall in the unit costs of their deployment. The genomes of several food crops including chickpea, foxtail millet, pigeonpea and sorghum have been sequenced (Hamilton and Buell, Reference Hamilton and Buell2012; Varshney et al., Reference Varshney, Chen, Li, Bharti, Saxena, Schlueter, Donoghue, Azam, Fan, Whaley, Farmer, Sheridan, Iwata, Tuteja, Penmetsa, Wu, Upadhyaya, Yang, Shah, Saxena, Michael, McCombie, Yang, Zhang, Yang, Wang, Spillane, Cook, May, Xu and Jackson2012, Reference Varshney, Mohan, Gaur, Gangarao, Pandey, Bohra, Sawargaonkar, Chitikineni, Kimurto, Janila, Saxena, Fikre, Sharma, Rathore, Pratap, Tripathi, Datta, Chatruvedi, Mallikarjuna, Anuradha, Babbar, Choudhary, Mhase, Bhardwaj, Mannur, Harer, Guo, Liang, Nadarajan and Gowda2013), while groundnut genome sequences will be available soon (http://www.peanutbioscience.com). Furthermore, resequencing of diverse germplasm collections may provide researchers opportunities to associate sequence differences with trait variations (Lai et al., Reference Lai, Li, Xu, Jin, Xu, Zhao, Xiang, Song, Ying, Zhang, Jiao, Ni, Zhang, Li, Guo, Ye, Jian, Wang, Zheng, Liang, Zhang, Wang, Chen, Li, Fu, Springer, Yang, Wang, Dai, Schnable and Wang2010; Zheng et al., Reference Zheng, Guo, He, Sun, Peng, Dong, Liu, Jiang, Ramchandran, Liu and Jing2011).

Genotyping of representative subsets has revealed abundant allelic diversity to differentiate wild relatives from cultivated types and grouped the germplasm into distinct clusters, with many of the alleles being found to be unique to particular accessions in each crop (Upadhyaya et al., Reference Upadhyaya, Dwivedi, Baun, Varshney, Udupa, Gowda, Hoisington and Singh2008a; Billot et al., Reference Billot, Ramu, Bouchet, Chantereau, Deu, Gardes, Noyer, Rami, Rivallan, Li, Lu, Wang, Folkertsma, Arnaud, Upadhyaya, Glaszmann and Hash2013; www.generation.cp.org), which may be further explored to associate allelic diversity with temporal and eco-geographical diversity.

Genome-wide association mapping in sorghum has revealed significant marker–trait associations, with many of the identified markers being co-mapped on the same linkage groups previously reported to be harboring quantitative trait loci or candidate genes associated with anthracnose, leaf rust, and grain mold resistance, tillering, and plant height and maturity (Upadhyaya et al., Reference Upadhyaya, Wang, Sharma and Singh2012c, Reference Upadhyaya, Wang, Sharma, Singh and Hasentsteind; Wang et al., Reference Wang, Bible, Longanantharaj and Upadhyaya2012; Upadhyaya et al., Reference Upadhyaya, Wang, Sharma and Sharma2013c, Reference Upadhyaya, Wang, Sharma and Sharmad).

Pre-breeding to accelerate cultivar development

Pre-breeding, the development of semi-finished products, provides a unique opportunity through introgression of desirable gene(s) from exotic germplasm into genetic backgrounds readily used by the breeders with minimum linkage drag (Sharma et al., Reference Sharma, Upadhyaya, Varshney and Gowda2013b). Resistance to pod borer has been introgressed using wild relatives from secondary and tertiary gene pools in pigeonpea, with most of these lines exhibiting variability for agronomic traits in addition to resistance to phytophthora blight, bruchid and pod fly (Mallikarjuna et al., Reference Mallikarjuna, Senapathy, Jadhav, Saxena, Sharma, Upadhyaya, Rathore and Varshney2011; Jadhav et al., Reference Jadhav, Mallikarjuna, Sharma and Saxena2012; Ramgopal et al., Reference Ramgopal, Srivastava, Pande, Rathore, Jadhav, Sharma, Gaur and Mallikarjuna2013). Wild Cicer species have been used to introgress resistance to pod borer, nematodes, phytophthora root rot, ascochyta blight and botrytis gray mold in chickpea (Ramgopal et al., Reference Ramgopal, Srivastava, Pande, Rathore, Jadhav, Sharma, Gaur and Mallikarjuna2013). Amphidiploids, originating from interspecific crosses, have been found to exhibit resistance to late leaf spot and peanut bud necrosis in groundnut (Mallikarjuna et al., Reference Mallikarjuna, Jadhav, Reddy, Husain and Das2012; Shilpa et al., Reference Shilpa, Sunkad, Kurella, Marri, Padmashree, Jadhav, Sahrawat and Mallikarjuna2012).

Chromosome segment substitution lines (CSSLs) provide additional useful genetic resources to broaden a cultigen's gene pool (Dwivedi et al., Reference Dwivedi, Upadhyaya, Stalker, Blair, Bertioli, Nielen and Ortiz2008). CSSLs have been developed in groundnut, which may be used for deciphering the molecular basis of trait variations (Fonceka et al., Reference Fonceka, Tossim, Rivallan, Vignes, Lacut, de Bellis, Faye, Ndoye, Leal-Bertioli, Valls, Bertioli, Glaszmann, Courtois and Rami2012).

Research carried out to date suggests that wild relatives have not only contributed genes for resistance to biotic stress, but also variations to yield and quality traits (Dwivedi et al., Reference Dwivedi, Upadhyaya, Stalker, Blair, Bertioli, Nielen and Ortiz2008; Imai et al., Reference Imai, Kimball, Conway, Yeater, McCouch and McClung2013). Preliminary research carried out at the ICRISAT has revealed that some cryptic introgressed lines, originating from TMV 2 and TxAG-6 cross, are phenotypically similar to TMV 2, but produce large seeds and exhibit higher pod yield (Upadhyaya et al., Reference Upadhyaya, Singh, Sharma, Varshney and Gowda2013b). TxAG-6 is a synthetic amphidiploid (Simpson et al., Reference Simpson, Starr, Nesplson, Woodard and Smith1993), while TMV 2 is a medium-maturing cultivar, adapted to peninsular India (ICRISAT, 2009).

Germplasm use and impact

To date, 75 unique germplasm accessions have been released for cultivation in 39 countries. Some of these have been widely used in breeding programmes or grown widely, e.g. an early-maturing Iniadi pearl millet landrace from West Africa (Andrews and Kumar, Reference Andrews and Kumar1996), a sorghum landrace (IS 33844) from Maharashtra, India (Reddy et al., Reference Reddy, Ramesh, Borikar and Sahib2007), early-maturing and rosette-resistant groundnut landraces ICG 12991 and ICG 12988 in Uganda (Subrahmanyam et al., Reference Subrahmanyam, van der Marwe, Reddy, Chiyembekeza, Kimmins and Naidu2000; Deom et al., Reference Deom, Kapewa, Busolo-Bulafu, Naidu, Chiyembekeza, Kimmins, Subrahmanyam and Van der Merwe2006), and a vegetable pigeonpea landrace (ICP 7035) with large seed size and resistance to sterility mosaic in many Asian countries (Shiferaw et al., Reference Shiferaw, Bantilan, Gupta and Shetty2004). The ICRISAT has notified the release of its 266 varieties/cultivars, germplasm, and elite genetic stocks with specific traits, mostly published as registration articles in Crop Science for worldwide information dissemination.

Outlook

In the context of advances in genomics, especially next-generation sequencing technologies, genomics-based germplasm science is coming up now. Although understanding diversity in germplasm collections for a few traits based on phenotyping or genotyping based on a few markers was a challenging task in the past, now it is possible to understand genome-wide diversity in germplasm collections by resequencing at least in those species where reference or draft genome sequences are available. Such datasets are also accelerating efforts made to identify marker–trait associations as well as superior lines based on genome-wide association studies. Genomics-based germplasm analysis is expected to enhance the use of germplasm and have an impact on breeding programmes in the future.