Recent Advancements in Genetic Improvement of Food Legume Crops

Abstract

Legumes are the second-largest source of food after cereals, all over the world, and an essential protein source in the vegetarian diet. These crops remain essential to subsistence production as they have the inherent capacity to survive in an adverse ecosystem and require minimum investment for their management. The increasing challenge of feeding a rapidly growing population places excessive pressure on current food production systems, which can no longer be sustained by traditional plant breeding alone. Therefore, modern breeding methods with increased genetic gains are required to meet the food demand of the increasing population. In the past few decades, the efficiency of legume crop breeding programs has increased considerably using novel and multidisciplinary approaches in breeding programs. A multidisciplinary approach combining conventional plant breeding, mutation breeding, plant biotechnology, and molecular breeding is strategically ideal for production of new and improved crop varieties. This chapter focuses on recent advancements in plant biotechnology, related molecular methods, phenomics, and their application in breeding of legume crops.

1. Introduction

Legumes are present in the diet of millions of people worldwide because these crops are associated with nutrition and health. Along with this, also have economic and environmental benefits. These are safe for consumption, relatively inexpensive, readily available, and the preferred food source after cereals. These crops have the ability of symbiotic nitrogen fixation; as a result, they help in the efficient use of fertilizers, lower emission of greenhouse gases and soil health enhancement. This ability of legumes promotes the inclusion of these crops in cultivation systems, thereby contributing to the diversified system and sustainability [1].

Food legumes are divided into two groups: 1) oil seeds and 2) pulses/ grain legumes. Oil seed legumes have high oil/fat content such as soybean and peanuts and pulses are dry seed legumes with low-fat content used as food (moong, urad, lentil, moth, etc.). Grain legumes are grown in both tropical and temperate regions of the world and used with cereals as pulse (dal). These are one of the major sources of income for smallholders who practise sustainable farming.

Today, malnutrition is more pronounced in developing countries due to the increasing population, and the most troubling one is caused by protein deficiency. Plant-based protein sources are the most desirable as they are nutritious, cheap, and easily accessible to poor people. Cereal-based diet system is deficient in protein content and essential amino acid lysine. In contrast, legumes contain protein and lysine amino acids, improving the nutritional status of diets based on cereals. Cereal diets containing legumes are considered one of the greatest therapies for protein caloric deficiency in developing nations [2]. In this way, legumes and cereals complement each other, and they must be eaten in a 35:65 ratio for nutritional balance.

In recent years, more people have substituted animal protein with vegetable protein due to increased health awareness and nutrition. With the increasing demand for vegetable protein, research on food properties and the utilization of indigenous food crops like legumes for protein-rich supplements are significantly increased [3]. These crops are also adapted to adverse climatic circumstances and are resistant to insect pests and disease, so they may be cultivated in arid climates with low or irregular rainfall.

In African and Asian countries, the primary contributors to protein and calories are legumes due to their economic and cultural reasons [3, 4]. Even though legumes crops have several benefits so far production is still less in comparison to cereals and vegetables. Cereal crops clearly overshadow these crops. It is necessary to increase awareness, spread the knowledge among the people, and encourage the farmers to grow legume crops along with cereals to increase their production.

1.1 Constraints in genetic improvement of legumes

Although legumes are a very useful protein source for humans and livestock, the research efforts to increase the productivity of legumes are lesser than the cereals. The poor yield of legumes may be due to growing these crops as subsistence in marginal lands with local varieties that do not tolerate biotic and abiotic stresses. Concerted attempts have been made during recent decades to enhance the yield potential of legumes with conventional methods, but genetic progress is poor compared to cereals [5].

The key challenges facing plant breeders in genetic improvement of legumes are discussed shortly below.

1.2 Genotype and environment interactions

Crops are largely determined by climatic conditions, and even minor changes from optimal conditions can severely affect plant growth and yield. Differential responses of improved cultivar strains are expected in different environments due to unpredictable climatic factors encountered at various sites and/ or years. G x E interactions then become a big challenge for any crop breeding program as they restrict effectiveness of breeding programs and selection responses. Legume crops show phenotypic instability due to environment (70–80%) and genotype x environment (17–27%) interactions for economically important quantitative traits, resulting in variable yield potential. The genotypic effects contribution is very less that is 1.5–7%. As a result, the environment has a crucial role in the stagnation of legume crop progress [6].

1.3 Multiple stress

Legumes are mainly cultivated in rainfed conditions on marginal lands with minimum inputs. In these risk-prone environments, legumes encounter multiple stresses such as various diseases (wilt, rust, Ascochyta blight, powdery mildew), erratic rainfall, prolonged dry spell, extreme temperatures, salinity stress, alkalinity and acidity. Most legumes are susceptible to different stresses, affecting morphological and physiological processes of plants that hamper plant growth. To ensure consistent performance of pulses in these areas, it is essential to develop multiple stress-resistant varieties.

1.4 Limited genetic diversity

The genetic enhancement of crops largely depends on the genetic diversity available in that crop for exploitation. The variability present in legumes for selection is comparatively limited. However, India is rich in available genetic diversity for legume crops, but the production and productivity are poor compared to global production and productivity [7]. To develop new cultivars, breeders use the same germplasm repeatedly in breeding programmes, and the rate of incorporation of new germplasm is less. Extensive use of the same genotypes with common ancestry in breeding programmes is the prime reason for the narrow genetic base of developed cultivars. Thus, the developed varieties are more susceptible to insects, pests, diseases and unpredictable climatic factors.

2. Recent techniques for genetic improvement of food legumes

2.1 Genotyping-by-sequencing of food legumes

The 1st genome sequencing of Arabidopsis (Arabidopsis thaliana) was completed in 2000; after this achievement, it has been proved that information about the genome of any crop species is a major and necessary step to the advancement of that crop species (Table 1).

Crop	Genome sequence	Germplasm lines resequenced
Soybean	950 Mb of the 1115 Mb of Glycine max genome; comprised of 46,430 protein-coding genes [8]	A total of 106 soybean genomes were re-sequenced, representing wild, landrace, and elite lines of the crop. [9] 89 lines [10] 286 soybean accessions (14 wild type, 153 landrace and 119 bred accessions) were sequenced; [11]
Pigeonpea	605.78 Mb of the 833.07 Mb pigeonpea genome; total of 48680 genes [12]	Genome-wide variation in 292 Cajanus accessions, including breeding lines, landraces, and wild species. [13] 20 Cajanus spp. accessions comprising two wild; species and 18 cultivated species accessions; [14]
Chickpea	∼738-Mb whole genome sequence of Kabuli chickpea variety CDC Frontier contains 28,269 protein-coding genes [15]	35 chickpea genotypes (parental lines of 16 mapping populations); [16]) 129 chickpea varieties, comprising 88 desi and 41 kabuli [17]
Groundnut	1081 Mb (89%) of Arachis duranensis and 1371 Mb (90%) of A. ipaensis [18] 50 324 protein-coding genes; A. duranensis var PI475845; [19]	11 genotypes including synthetics and their diploid parents [19] 41 groundnut accessions and wild diploid ancestors [20]
Common bean	473 Mb of the 587-Mb genome and genetically anchored 98%; 26 279 protein-coding genes [21]
Mung bean	543 Mb (84.48%) size genome (V. radiata var. radiata VC1973A); 22 427 predicted genes [22]
Adzuki bean	75% of the 612 Mb size genome; 26857protein coding genes [23] 450 Mb (83% of the genome) sequenced; 50 accessions including 11 wild, 11 semiwild, 17 landraces, and 11 improved varieties; 34 183 predicted genes [24]
Cowpea	36 diverse cowpea accessions [25]
Pea	3920 Mb of pea cultivar ‘Caméor’, representing 88% of the estimated pea genome size.; Total number of protein-coding genes is 193,976 [26]	42 wild, landrace and cultivars [26]

Table 1.

Summary of genome sequence and resequencing efforts in important legumes.

Our knowledge of different crop plant traits, including food legumes, has expanded during the past few decades due to advances in plant biotechnology and genomic technologies [27]. Genome sequencing enables crops to be improved on the basis of genomic gains and the selection of genes that possess desirable characteristics that increase the quality and quantity of produce. This also provides detailed information on genome structure and mutagenic changes due to deletions and insertions and discloses the pathways linked to different stress responses. In legumes, Lotus japonicus (Japanese trefoil) and Medicago truncatula, these two species with small genome sizes were selected as reference genomes. The genome sequence of the majority of the legume crops is now available; for soybean, groundnut proginator, chickpea, pigeonpea, common bean, and adzuki bean. Completely sequenced legume species (with completed and annotated genomes): Cajanus cajan (833 Mb genome), Cicer arietinum (738 Mb), Glycine max (1,112 Mb), Lotus japonicus (472 Mb), Medicago truncatula (373 Mb) and Phaseolus vulgaris, respectively (588 Mb). These species have between 28,269 and 48,680 genes and 25,640 to 243,067 transcripts, respectively.

In addition to revealing the genome sequences of different legume crops, different research institutes re-sequenced legume germplasm lines because draft genomes are now available, so it is easy to deploy whole genome re-sequencing-based approaches in legumes. This approach will help to learn more about genome architecture, structural variations, genome evolution, and genome dynamics during domestication. As a result, different genotypes/lines/accessions were chosen for these legumes based on their priority in respective crop improvement programmes.

2.2 Trait mapping from diverse legumes germplasm

Several traits that are agriculturally important are complex. These traits are controlled by many genes and affected by environment and gene-environment interactions. Over the past few decades, to understand the genetics of complex traits has become a major concern. With the progress made in the area of molecular markers as well as in genomics significant number of QTLs have been found in various crops. In legumes also, several genes/ QTL controlling the target traits have been mapped (Table 2). The efficiency and accuracy of breeding practices have been improved significantly with the help of molecular marker-assisted selection of important traits. Further, the mapped gene(s) or QTLs can be introduced individually or pyramided in an improved variety.

Crop	QTL/Gene	Trait	Method	References
Soybean	IDC QTL; FAD2-1 and FAD2-2 genes	Iron deficiency chlorosis; Oleate biosynthesis	Association mapping; Linkage mapping	Wang et al. [28]; Bachalva et al. [29]
Pigeonpea	Hsf genes; candidate genes	Heat-response; Sterility mosaic disease (SMD), Fusarium wilt (FW) and photoperiod insensitivity	Genome-wide analysis; genome-wide association analysis	Maibam et al. [30]; [13]
Common bean	Co-1–Co-10; 10 QTLs/genes; Resistance gene analogs	Resistance to anthracnose; Resistance to anthracnose; Resistances to different pathogens	Linkage mapping; Associations mapping; Associations mapping	Kelly and Vallejo [31]; Choudhary et al. [32]
Cowpea	Candidate genes; Hbs-1–Hbs-3; Major QTL	Resistance to root-knot nematodes; Heat-induced browning of seed; drought tolerance; Resistance to root-knot nematodes	QTL mapping; QTL mapping; QTL mapping	Santos et al. [33]; Pottorff et al. [34]; Huynh et al. [35]
Pea	nod3; PsMlo; PsDREB2A	Hyper nodulation mutation; Powdery mildew resistance; Drought response	Comparative genomics; Comparative genomics; Comparative genomics	Bordat et al. [36]; Mohapatra et al. [37]; Jovanovic et al. [38]
Chickpea	Aquaporins gene family; CarERF116; Major QTLs corresponding to; flowering time genes (efl-1,; efl-3, and efl-4)	Biotic and abiotic stresses; Abiotic stress responsive; Flowering time	Comprehensive; genome-wide analysis; Genome-wide association; Analysis; QTL mapping	Deokar et al. [39]; Deokar et al. [40]; Mallikarjuna et al. [41]
Adzuki bean	VaAGL, VaPhyE, and VaAP2	Flowering time and pod maturity	QTL mapping	Li et al. [42]

Table 2.

QTL mapping in different legume crops.

There are two approaches for marker-trait associations identification in plants: (1) Linkage mapping and (2) Association or linkage disequilibrium (LD) mapping. Linkage mapping is a conventional mapping approach based on genetic recombination between two loci, whereas association mapping is a new approach and based on linkage disequilibrium.

Currently, candidate gene and whole-genome association mapping methods are used in crop plants. As a new approach to conventional linkage analysis, association mapping has the advantages of increased mapping resolution, research speed, and greater allele number. Different loci for iron deficiency chlorosis in soybeans have been effectively mapped using the candidate gene-based method [28].

Similarly, Bachalva et al. [29] mapped several candidate genes in soybean involved in oleate biosynthesis and examined their co-segregation with oleate, linolenate quantitative trait loci (QTLs). Whole-genome association mapping has been used in several legume crops; for example- Medicago truncatula, common bean, soybean, chickpea, cowpea, peanut etc. In pigeonpea, 292 accessions were characterized using genome-wide association analysis for the purpose of accelerating genetic gains and identifying associations between several candidate genes and agronomically significant traits [13]. In a diversity panel including 96 Middle American genotypes of common bean, Hoyos-Villegas, Song, and Kelly [43] studied the genetic basis of variation for drought tolerance and related traits, and the GWAS analysis enabled identification of important marker-trait associations for traits related to drought tolerance and candidate genes associated with wilting. Salinity stress, which is intensified by changing climatic conditions, has a negative impact on cowpea at the germination and seedling stages. Ravelombola et al. [44] conducted research to identify SNP markers linked to salt tolerance through association mapping.

2.3 Pan-genomes of legume crops

It is clear that a single individual’s genome does not adequately represent the diversity of genes in a species. Pangenome assemblies, which capture sequence and structural variation in a species more comprehensively, can be developed as a remedy. Pangenome includes the core complement of genes common to all individuals of the species. The variable genome contributes to species diversity and provides functions that are not essential, but which may. While the availability of reference genomes has significantly supported plant breeding and research, these reference genomes capture only a portion of the species diversity. These reference genomes provide a selective advantage under some circumstances like; biotic and abiotic stress resistance. Tettelin et al. developed the pangenome concept [45] and developed the first-ever pangenome for a bacterial species, Streptococcus agalactiae. The first legume pangenome was generated by sequencing and de novo assembly of seven phylogenetically and geographically representative accessions of the wild relatives of cultivated soybean. The soybean pangenome indicates a faster evolution and greater diversity in dispensable genes than core genes related to adaptation to environmental stresses. Recently, the Pigeonpea pangenome was developed, based on 89 accessions mostly from India and the Philippines. This reveals that in Philippine individuals, there is a substantial genetic variation that is not present in Indian individuals.

Existence of a large number of repeats and several rounds of polyploidy, genome and pangenome assembly in plants is always difficult. The traditional de novo assembly and comparison approach was first used to demonstrate significant genomic differences between individuals. It has the benefit of providing the physical position of genes. During the breeding of certain crops, a decline in genetic variation has been observed, especially associated with the selection of important characteristics. This approach will help identify genes lost during breeding and selection that can be bread back into elite germplasm.

2.4 Mutmap technique in legumes

MutMap is a recently developed method based on whole-genome resequencing of pooled DNA from a segregating population of plants with a useful phenotype. The MutMap scheme to identify rice genes responsible for agronomically significant traits in a rice mutant pool that had been mutagenized by ethyl methane sulfonate (EMS). A recessive mutant from a mutant pool is backcrossed to a wild plant type in MutMap technique. The F1 plant is self-fertilized and the F2 progeny (>100) are screened for plants (>20) with a segregated mutant phenotype. In this method, only a small F2 population (>100 plants) is required for gene identification from crop plants so this is easy to maintain small population in the field. MutMap is particularly useful for identifying genes that control quantitative minor effect phenotypes, which is a challenging thing in crop improvement. This method is being used in the field of legumes to find candidate genes for leaf and plant type mutants in chickpea.

2.5 Genomic assisted breeding

The world’s population is rapidly growing and is expected to hit 9 billion by 2050. This massive pressure on population would contribute to a serious food shortage. Pulses in the vegetarian diet are essential sources of proteins, for pulses improvement, extensive breeding programmes have been done through conventional breeding and significantly developed several high-yielding varieties [46]; however, the pace of genetic improvement of pulse crops is very slow. Limited success was achieved through conventional breeding even after continuous and systematic breeding efforts due to several constraints. One of the major constraints on expression of quantitative traits is high G x E interaction leading to slow genetic gain [6]. For strengthening conventional breeding programmes, integration of novel breeding strategies and techniques is required for revolutionary changes. During the last decade, the performance of legume breeding programme has increased significantly, as a result of novel genomic tools and techniques incorporated with the conventional breeding methods. In order to incorporate genomics in breeding, genomics-assisted breeding was suggested, and it has been effective for many traits in cereals and legumes. Genomic-assisted breeding will accelerate the genetic enhancement of pulses which leads to development of cultivars with higher yield and multiple stress tolerant. The availability of molecular markers such as simple sequence repeats and SNPs has enabled the dissection of complex characteristics that limit crop production, In the case of pigeonpea and soybean, genome-wide SNPs focused on resequencing of many germplasm lines were also used to establish marker-trait associations.

2.6 Marker assisted characterization of germplasm

In the twenty-first century, food, water and land are biggest challenges for increasing population. Agricultural activities need to expand, become sustained, and be more adaptive to climate change. To improve sustainability in agricultural systems, new paradigms are required, to explore the genetic potential of the huge although unfortunately underutilized resources of genetic diversity available for different crops. For breeding of climate-resilient varieties, a better understanding of evolutionary genetic variability is essential. Genetic diversity is the precious wealth for any crop improvement programme but due to climatic changes, it is reducing continuously. In last century, 75% decline in genetic diversity was witnessed in farmers’ fields and it would further decline by about 20% by 2050.

Besides that, genetic resources can be excellent breeding material to develop superior variety in future breeding programmes. They can also be used in different breeding programs in order to increase the genetic base of cultivated crop varieties.

It has been observed that wild relatives have several desirable characteristics like resistance to biotic and abiotic streses, nutritional characteristics, cleistogamy, photo insensitivity and cytoplasmic male sterility (CMS).

In the past few decades, revolutionary approaches and systems have offered a great wealth of genetic and genomics resources that revolutionized research in both model and crop legumes.

A recent study on chickpea presents evidence of severe domestication bottleneck. Efficiency of cultivated population of chickpea is 100 times lesser than that of wild chickpea (Cicer reticulatum and C. echinospermum). In legume crops, study on landraces and wild relatives are significantly benefited by advanced technologies of genomics, phenotyping and computational biology.

The Vavilov Institute of Plant Genetic Resources (VIR), which houses a special genebank in St. Petersburg, Russia, using a mixture of genomics, computational biology and phenotyping to classify the 147 accessions of chickpea from Turkey and Ethiopia. The combination of high-density genotyping data with historical phenotypic information on these VIR landraces allowed chickpea genomes to enter ‘agro islands’ or ‘domestication islands’ that display significant associations with multiple phenotypes. These “genomic gems” have also been identified in chickpeas containing co-adapted and co-localized gene complexes. These are LG4 and LG2 in chickpea containing multiple genes/ QTLs related to drought and disease resistance, respectively. WGRS/RADSeq of 90 Cicer accessions, including cultigens, landraces and wild accessions, previously identified a wide collection of 54 genes on LG3 that could have been targeted during modern breeding efforts to manipulate salient characteristics such as flowering time.

Similarly, a genomic segment with an excess of MTAs for agronomically significant traits was observed on LG9 after re-sequencing of 292 accessions in pigeonpea [13].

In a recent study, to understand the genetic relationships between various lentil species/subspecies, a lentil collection comprised of 467 wild and cultivated genotypes originating from 10 different geographical regions was evaluated. A total of 422, 101 high-confidence SNP markers were identified against the reference lentil genome (cv. CDC Redberry). Phylogenetic analysis clustered the germplasm collection into four groups, namely, Lens culinaris/Lens orientalis, Lens lamottei/Lens odemensis, Lens ervoides, and Lens nigricans. Results of this study indicated that L. nigricans is most distantly related to L. culinaris and major differences were observed in six genomic regions with the largest being on Chromosome 1 (c. 1 Mbp) and further additional structural variants are likely to be identified from genome sequencing studies. In order to improve germplasm and for introgression of novel genes, this will provide further insights into the evolutionary relationship between cultivated and wild lentil germplasm.

Guar (Cyamopsis tetragonoloba (L.) Taub.) is primarily grown as an industrial crop due to its high-quality galactomannan gum used as a thickener, flocculant, emulsion stabiliser and gelling agent. Therefore, the novel set of molecular markers (nSSR) could be adopted as a useful tool to characterize the guar accessions for future breeding programmes.

2.7 Marker assisted backcrossing (MABC)

Research on legumes has greatly benefited from different available molecular markers in crop plants. Association between molecular markers and plant traits in these crops has introduced a novel approach to breeding that is based on the crossing of selected genotypes and selection of suitable progenies based on associated markers/QTL(s) rather than depending solely on phenotypes. Over the past three decades, the advancement and development of molecular marker technologies have been steady, such as low-throughput restriction fragment length polymorphisms (RFLPs) in the 1980s, high-throughput array-based markers in the 2000s and next-generation ultrahigh-throughput sequence-based marker systems in the 2010s. RFLP, RAPD, AFLP and SSR markers are low-throughput marker systems and are also considered past molecular markers. Besides these, next-generation sequencing (NGS) and genotyping by sequencing (GBS) are high and ultrahigh-throughput marker systems. These are based on low-cost and high-throughput sequencing technologies and are considered as present marker systems.

In cereals, so many outstanding achievements of marker-assisted breeding are available, but in legumes, negligence and lack of genomic resources adversely affected their initial establishment in the field of molecular breeding. Now recent advances in pulse genomics have led to the launch of several marker-assisted breeding projects.

RAPD and RFLP markers were used in five wild lentil taxonomic groups to understand their genetic makeup [47]. A genetic linkage map was also constructed in lentils with RAPD, AFLP and RFLP markers [48].

For shielding the varieties against various biotic and abiotic stresses and for ensuring crop productivity; gene mapping, tagging and marker-assisted selection have vital importance. Identifying and deploying molecular markers/QTLs in a desired background would be a priority. Marker-aided selection (MAS) greatly reduce the time and effort required to recover high levels of resistance from the donor and simultaneously recover the genomes of the recurrent parent. It has become more easier to transfer desirable genes/QTLs from wild relatives to existing cultivars due to MAS and transgenics.

Fusarium wilt (FW) and Ascochyta blight (AB) are two major constraints in chickpea (Cicer arietinum L.) production. The most affordable approach for long-term control of ascochyta blight and fusarium wilt in chickpeas is known to be the use of varieties with high resistance levels.

The availability of molecular markers associated with QTL for ascochyta blight resistance provided an opportunity to introgress the traits into adapted chickpea cultivars. backcrossing between moderately resistant donors (CDC Frontier and CDC 425-14) and the adapted varieties (CDC Xena, CDC Leader and FLIP98-135C) resulted in a variety with improved resistance to ascochyta blight [49].

More recently, five resistant lines representing foc2 gene introgressed into the background of Pusa 256 were reported with the help of foreground selection aided by two SSR markers (TA 37 and TA110). Cultivar Vijay was used as a donor of foc2 gene [50]. Annigeri 1 and JG 74 are elite high-yielding desi cultivars of chickpea, in Karnataka and Madhya Pradesh. in recent years, have become susceptible to race 4 of Fusarium wilt (FW).

A widely grown cowpea variety in Africa was improved by adding drought tolerance, striga and root knot nematode resistance QTLs using SNP markers. The major QTL region on LG 8 was introgressed from cultivar V-16 into the bacterial leaf blight susceptible variety C-152 through marker-assisted backcrossing (MABC) [51]. Similarly, By backcrossing resistance to CpMv gene was transferred into variety C-152. Cowpea mosaic virus (CpMV) was responsible for 80–100% yield loss in cowpea. SSR markers were used for linkage map construction and indicated that two markers MA15 and MA 80 were linked to CpMV resistance.

At ICRISAT in hybrid pigeonpea programmes, markers associated with fertility restoration and CMS are being used. This improved the selection efficiency of hybrid breeding and accelerated the breeding work [52]. In addition, a range of markers, including SSRs and SNPs are now available to enable genetic purity testing of pigeonpea hybrids and their parents. Recently, ICRISAT has launched a collaborative effort with ICAR-IIPR and other NARS institutions/universities to accelerate and target the improvement of ruling mega varieties of pigeonpea in India.

In groundnut breeding, the use of molecular markers in backcross breeding programme accelerated selection of recombinant progenies bearing nematode resistance and high oleic acid. Selection for high oleic acid content in groundnut was facilitated by one CAPS marker along with gel-free SNP assay using HybProbe design for the selection of nematode resistance SCAR, SSR and CAPS marker were used.

Recently in peanuts, two ahfad2 alleles from SunOleic 95R were introgressed into ICGV 05141 using marker-assisted selection. Marker-assisted breeding effectively increased oleic acid and oleic to linoleic acid ratio in recombinant lines up to 44% and 30%, respectively as compared to ICGV 05141. Subsequently, In the marker-assisted backcrossing-introgression lines, a 97% increase in oleic acid, and a 92% reduction in linoleic acid content were observed in comparison to the recurrent parent [53].

As opposed to traditional breeding, gene stacking or pyramiding is a useful strategy for transferring multiple desired genes or QTLs from various parents into a single genotype in the shortest possible time (two to three generations). Molecular markers that may be beneficial for marker-assisted selection and gene pyramiding have been identified through genetic linkage analyses and QTL mapping. The most effective and inexpensive means of combating plant diseases is the use of genetic resistance. Gene pyramiding is thus a sensible approach to creating multiple and enduring resistance. Most successful approach in common bean for wide spectrum control of common mosaic virus is to combine I, bc-u, bc-12, bc-22, and bc-3 genes. SCAR marker was used for MAS. In lentils, molecular marker-assisted gene pyramiding was used for resistance to ascochyta blight and anthracnose. In this research, two genes for resistance to ascochyta blight and the gene for anthracnose resistance in lentil breeding lines were pyramided using linked RAPD marker [49].

2.8 Genome editing

Crop plant genome editing is a faster-growing technique for inserting specified changes into the genome precisely and with great accuracy. Genome editing has emerged as an alternative approach to conventional plant breeding, and transgenic (GMO) approaches to improve food legumes and their sustainable production. Instead of spontaneous non-specific changes caused by radiation or chemical mutagenesis, crop researchers have long required mutations at specific sites in the genome. This approach allows for site-specific DNA insertion, deletion, modification, or replacement in a living organism’s genome. The plant research community has not been widely involved with earlier SSN-specific (sequence-specific) genome-based editing technologies, because of the complex design and labour-intensive assembly of particular DNA binding protein for each target gene. A relatively new and comparatively easier technique for genome editing is CRISPR (clustered regularly interspaced short palindromic repeats) technique which is based on a simplified version of the bacterial CRISPR-Cas9 antiviral defence system. CRISPR genome editing technique is based on Cas9 protein which is an endonuclease. This endonuclease induces double-strand breaks using a guide RNA that is complementary to a target gene [54]. In order to create mutants for inaccessible genes, CRISPR-Cas9 would be a very useful technique. It can mutate multiple loci and make large deletions, thereby speeding the plant breeding without directly adding any transgene. The sequence-specific nucleases-based plant genome editing has a great potential to develop modified crops which can address the increased global food requirements and sustainable agriculture production. CRISPR/Cas9 was applied first in model legume plants to induce targeted mutagenesis.

A web tool was designed to identify potential CRISPR/Cas9 target sites and also a soybean codon-optimized CRISPR/Cas9 platform to induce mutation at target sites in somatic cells of Glycine max and Medicago truncatula [55]. In a recent study, an efficient CRISPR/Cas9 system was developed for targeted gene mutations in the model legume M. truncatula. A specific sgRNA was designed that targeted medicago phytoene desaturase (MtPDS) gene involved in the carotenoid biosynthesis. Very recently in Cowpea, the representative SNF gene has been effectively disrupted with an efficient CRISPR/Cas9-mediated genome editing. Guide RNAs (gRNAs) for the symbiosis of receptor-like kinase (SYMRK), reached ~67% mutagenesis efficiency in plants with hairy roots, and the formation of nodules in both mutants was totally prevented [56]. Conventional breeding is based on natural genetic variation and rigorous back-crossing systems are needed to incorporate the selected traits into an elite genotype. Unlike conventional breeding techniques, the present diversity does not limit CRISPR because it can directly integrate new mutations. This approach will benefit particularly those crops which have narrow genetic diversity and low variability for traits. Therefore, genome editing can speed up plant breeding programmes by inserting correct and predictable modifications directly in desirable backgrounds. The CRISPR/Cas9 system is especially beneficial because multiple traits can be modified simultaneously.

2.9 Mutation breeding in legumes improvement

The basis for any crop improvement programme is the variations present in the concerned crop. For generation of new variations, mutation is a prerequisite. These mutations are caused by various factors and are broadly divided into two major categories: spontaneous and induced mutations. Natural causes like as ultraviolet (UV) irradiation, reactive oxygen species, and transposable elements may generate spontaneous mutations in nature. On the other hand, physical and chemical mutagens cause artificial mutations. Different mutagenesis techniques have been successfully utilized in molecular plant breeding to study gene functions. The alterations induced can be random or particular to the target. Chemicals and physical mutagens cause random mutations. Unfortunately, random mutagenesis is costly, time-consuming and also difficult to screen desirable mutants from a large, mutated population. In addition to conventional plant breeding and GMO techniques, targeted mutagenesis has arisen as an alternative for improving crop plants. This approach relies on the use of nucleases that allow for precise double-stranded breaks to occur at certain sites within the genome. The specific methodologies for targeted mutagenesis include PCR-based techniques for in vitro mutation generation and analysis, transposon mutagenesis, RNA interference (RNAi), TILLING (Targeting Induced Local Lesion IN Genomes), and programmed meganucleases [also called homing endonucleases, site-directed nucleases (SDNs) or site-specific nucleases (SSNs)]. TALENs, ZFNs, and CRISPR/Cas9 are frequently used meganucleases.

2.9.1 Tilling

Identifying a mutation in a particular gene and relating this mutation to the phenotypic alteration in the mutant organism is one of the most straightforward ways of determining gene function. TILLING (Targeting induced local lesions in genomes) is a non-transgenic, high throughput, general reverse-genetic strategy which aims to identify SNPs (single nucleotide polymorphisms) and/or INDELS (insertions/deletions) in a gene/gene of interest from a mutagenized population. TILLING has developed a few decades ago as an alternative to insertional mutagenesis in Arabidopsis thaliana. High-throughput TILLING provides a quick and cheap diagnosis method of induced mutations in artificially mutagenized populations. The important feature of TILLING is that it can be applied to any species, regardless of its genome size and ploidy level.

2.9.2 Eco tilling

EcoTILLING (Ecotype Targeting Induced Local Lesions IN Genomes) is the modification of TILLING, which identifies natural genetic variations in populations in contrast to induced mutations in TILLING. This has been successfully used in animals and plants to discover SNPs and small INDELs. The classical method of Eco-TILLING is based on the enzyme endonuclease (Cel1, Endo1), which cleaves at the point of mutation by detecting mismatches in double-strand DNA. EcoTILLING is convenient for those plant species in detection of natural mutations where chemical mutagenesis is not suitable.

So far, TILLING and EcoTILLING have been implemented in many legume crops. In soybean, Tilling was used to screen more than 40,000 mutant lines and to create novel mutant alleles [57]. In chickpea, TILLING was also used to diagnose mutations in the M2 generation. Recently, in mungbean, five exon residing mutations were identified by TILLING and confirmed the potential role of each mutation in altering mungbean plant architecture to develop an ideal plant type [58].

2.10 Transgenic approaches/genetically modified legumes

Traditional breeding is tedious and success rate of obtaining desirable gene/genes or gene combinations from a large number of crosses is very less. These limitations hamper the desirable changes in crop plants. Therefore, biotechnological approaches are complementary to traditional breeding methods for addressing global food demands. Today we have access to vast gene pools due to new biotechnological approaches, which can be utilized in food crops to add favourable features. In this way, Genetically-modified (GM) crops can contribute to satisfying the food demand by developing varieties which are high yielding, good in quality, nutrition-rich and different kinds of stress-tolerant. Genetically modified crops are plants in which one or more genes have been introduced using genetic engineering techniques to produce desirable traits for agricultural purposes. Genetic engineering facilitates the direct gene transfer not only within the species and between the different plant species but also from unrelated organisms as well as also resolves the problem of linkage drag. Soybean was the first grain legume for which transgenic plants were developed [59]. Glyphosate-resistant soybean was developed by transferring gene derived from Agrobacterium sp. strain CP4, which encodes a glyphosate-tolerant enzyme EPSP synthase. Genetically transformed other legume species have successfully developed glyphosate-resistant lines for example- narrow-leaf lupin (Lupinus angustifolius L.) [60]. This is an easy way of weed control, reduces the cost of production and has a positive impact on the environment. Water stress causes significant yield losses in soybean crops; to resolve this problem, transgenic soybean was developed by transferring a gene encoding an osmotin-like protein extracted from Solanum nigrum var. americanum [61].

Helicoverpa armigera, a food legume insect, causes significant yield losses in pigeonpea. To minimize the losses caused by Helicoverpa armigera; transgenic pigeonpea was developed by transferring two synthetic Bacillus thuringiensis insecticidal crystal protein genes, cry1Ac and Cry2Aa. The transgenic pigeonpea expressed Cry1Ac and Cry2Aa proteins exhibited 80–100% mortality of insect [62]. Chickpea crop often encounter terminal drought stress that affects its production. Desi chickpea variety C235 that has 120 days of crop cycle, a transcription factor DREB1A was transferred and observed better root and shoot partitioning as well as higher transpiration efficiency in transgenic chickpea under drought stress [63]. In storage, cowpea seeds are severely damaged by storage pests (Callosobruchus maculatus and C. chinensis). Introduction of the bean (Phaseolus vulgaris) α-amylase inhibitor-1 (αAI-1) gene into a commercial Indian cowpea cultivar (Pusa Komal) strongly inhibited the development of these insects [64].

Genetic engineering, however, has excellent potential to maximize crop performance coupled with conventional methods, even though there is somewhat risk related to the effects of transgenic crops on the environment and human health. To overcome these risks, each product should be critically examined. Appropriate bio-safety and food safety measures should be strictly followed.

2.11 Phenomics

A better understanding of the biological processes is required to increase yield potential and multiple stress tolerance. Any crop for its improvement majorly depends on favourable genetic changes in the crop genome, but the current pace of crop improvement is incapable of meeting future food demands. Therefore, crop improvement requires introducing new approaches for genetic changes in crop plants and their breeding. Marker-assisted breeding/ molecular breeding gives more importance to genotypic information of a crop, but phenotypic information is also equally important. Plant phenotyping is now a bottleneck in advancing crop yield. To enhance the selection efficiency of crop plants, phenotyping is also important, along with genotyping. The rapid and accurate evaluation of the phenotype of breeding lines and different crop populations is required for new variety development.

Phenomics is the investigation of phenomes, which are the collection of phenotypes (physical and biochemical traits) that a given organism may generate during development and in response to environmental effects. Crop phenomics is a multidisciplinary approach which integrates agronomy, life sciences, information science, math and engineering sciences and combines high-performance computing and artificial intelligence technology. This technique provides non-destructive and non-invasive ways of imaging, including colour imaging, near-infrared imaging, far-infrared imaging and fluorescence imaging for different phenomena like; plant structure, biomass, leaf health, for measuring soil and tissue water content, canopy/leaf temperature measurement etc. High-throughput phenotyping has been widely used, offering automated digital analyses of large data samples. The main benefit of high throughput phenomics approaches is the speed at which data can be collected: field data that could take several days to collect using conventional methods can be collected in a matter of hours using several sensors installed on a phenotyping platform. This saves time and allows several observations of a given plant/plot in a single day. These phenomics tools and techniques are making way for crop plant genetic improvement by using the potentiality of genomic resources.

2.12 Rapid generation advancement approaches in legumes

The biggest challenge for breeding higher-yielding and more resilient crops is the inability to complete more generations in lesser time. Generally, legume crops complete one or two breeding cycles in a year, so developing a new variety is time-consuming. Speed breeding is a rapidly emerging method among plant breeders to develop new varieties in a short period of time. This technique greatly enhances breeding and research speed by reducing generation time. Plants are grown in controlled growth chambers or greenhouses with optimal light intensity and quality, as well as specific day length and temperature (22 h light, 22 °C day/17 °C night, and high light intensity), to speed up different physiological functions; especially photosynthesis and flowering, and thus reduce generation time. Under normal glasshouse conditions, 2–3 generations can be produced per year, while speed breeding can produce up to 4–6 generations per year. Chickpea was induced to flower early by Gaur et al. [65] using 24-hour photoperiod, which, with the aid of offseason nurseries, allowed the production of three generations per year. Similarly, early and late flowering genotypes of pea, chickpea, faba bean, lentil and lupin were grown by Croser et al. [66] in controlled environments under different light spectra (blue and far red-enriched LED lights and metal halide). The time it took for the first seed to germinate was reduced significantly in half, and pollen viability was enhanced. In addition, costs for speed breeding are also reduced by combining this with genomics-based breeding and high-throughput phenotyping.

3. Conclusion and future perspectives

Legumes have a lot of potential to offset the effects of climate change by contributing to sustainable cultivation and expanding the crop spectrum, which is largely controlled by a few major cereal crops. Furthermore, in legumes, considerable advancement has been made in identifying novel genes for useful traits; however, the full potential of legume crops is still unknown. A range of tools and techniques have been provided by advancements in biotechnology, molecular breeding and genomics that can significantly enhance the hereditary gains in legume breeding programs. Genome sequencing allows the improvement of legume crops based on desirable gene selection and now the genome sequence of most legume crops has been sequenced. Marker-assisted breeding also significantly improved the accuracy and efficiency of crop breeding practices. Different loci have been mapped in several legume crops. The high-throughput genotyping platform would undoubtedly allow for low-cost, large-scale screening of segregating individuals to select suitable genotypes. In the future, it will increase the utility of MAS breeding for legume crops, being productive, and cost-effective. Marker-assisted breeding/ molecular breeding gives more importance to genotypic information of a crop, but phenotypic information is also equally important. This gives the information about phenome of an organism eventually increasing the selection efficiency and reducing the time required for evaluation. CRISPRs allow the development of novel cultivars containing multiple genes in just one generation. Besides these, the availability of the reference genome, combined with high-density genotyping and sequencing assays, opens new possibilities for harnessing genetic variations for climate-resilient traits. These modern techniques are significantly accelerating the pace of legume crop development, ensuring overall food security.