Advances in lentil production through heterosis: Evaluating generations and breeding systems

Gurpreet Kaur Suri; Shivraj Braich; Dianne M. Noy; Garry M. Rosewarne; Noel O. I. Cogan; Sukhjiwan Kaur

doi:10.1371/journal.pone.0262857

Abstract

Heterosis is defined as increased performance of the F₁ hybrid relative to its parents. In the current study, a cohort of populations and parents were created to evaluate and understand heterosis across generations (i.e., F₁ to F₃) in lentil, a self-pollinated annual diploid (2n = 2× = 14) crop species. Lentil plants were evaluated for heterotic traits in terms of plant height, biomass fresh weight, seed number, yield per plant and 100 grain weight. A total of 47 selected lentil genotypes were cross hybridized to generate 72 F₁ hybrids. The F₁ hybrids from the top five crosses exhibited between 31%–62% heterosis for seed number with reference to the better parent. The five best performing heterotic crosses were selected with a negative control for evaluation at the subsequent F₂ generation and only the tails of the distribution taken forward to be assessed in the F₃ generation as a sub selection. Overall, heterosis decreases across the subsequent generations for all traits studied. However, some individual genotypes were identified at the F₂ and sub-selected F₃ generations with higher levels of heterosis than the best F₁ mean value (hybrid mimics). The phenotypic data for the selected F₂ and sub selected F₃ hybrids were analysed, and the study suggested that 100 grain weight was the biggest driver of yield followed by seed number. A genetic diversity analysis of all the F₁ parents failed to correlate genetic distance and divergence among parents with heterotic F₁’s. Therefore, genetic distance was not a key factor to determine heterosis in lentil. The study highlights the challenges associated with different breeding systems for heterosis (i.e., F₁ hybrid-based breeding systems and/or via hybrid mimics) but demonstrates the potential significant gains that could be achieved in lentil productivity.

Citation: Suri GK, Braich S, Noy DM, Rosewarne GM, Cogan NOI, Kaur S (2022) Advances in lentil production through heterosis: Evaluating generations and breeding systems. PLoS ONE 17(2): e0262857. https://doi.org/10.1371/journal.pone.0262857

Editor: Roberto Papa, Università Politecnica delle Marche, ITALY

Received: June 29, 2021; Accepted: January 6, 2022; Published: February 18, 2022

Copyright: © 2022 Suri et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: Raw data has been deposited to NCBI under accession number GSE184819.

Funding: The authors are thankful to National Lentil Breeding Programme funded by GRDC (Grains Research and Development Corporation), and Agriculture Victoria (Department of Jobs, Precincts and Regions, State Government of Victoria, Australia) for providing the facilities and supporting germplasm to conduct this research.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Lentil (Lens culinaris Medik.), a self-pollinated annual diploid (2n = 2× = 14) cool season grain legume is mostly used for human consumption due to its high level of protein, vitamins, and minerals [1]. Legumes such as lentils, have been identified as a superior and cheaper protein choice over beef, poultry or fish which possess higher amounts of saturated fat and cholesterol. Amongst plant-based foods, lentil contain high levels of folate as well as β-glucans making their glycemic index exceptionally low and suitable for wide health benefits [2]. The presence of phytochemicals such as phenolic acids, flavanols, saponins, phytic acid and condensed tannins makes lentils rich in antioxidant properties [3]. During the last decade lentil production has increased from 2.8 million tonnes to 6.3 million tonnes globally, with many world markets demanding higher quality grain [4]. Australia is one of the top ten lentil producing countries, specifically regarding the red lentil. The current yield gains for lentils from conventional plant breeding in Australia are 1.13% per annum that has led to yield of 1.5 t/ha in 2017 [5, 6].

Plant breeding has helped to increase the yield and quality of legumes as well as other grain and forage crops. Many grain legume breeding programs have achieved relatively high yield gains over time, however, created domestication bottlenecks through limited breeding and biased selections based on yield potential only [7, 8]. The lentil domestication has resulted in approximately 40% loss of genetic diversity leading to narrow gene pools within breeding programs and restricted genetic gain [9]. For instance, most of the registered lentil varieties in Canada are related to the first two cultivars that founded Canadian production: ‘Laird’ and ‘Eston’ [10, 11]. Furthermore, the narrow genetic base of lentil varieties has made them more susceptible to biotic and biotic stresses [12]. Exploration of genetic diversity and introgression of novel alleles from landraces as well as crop wild relatives is pivotal for producing high yielding, disease resistant and stress tolerant varieties. Genetic diversity can be conserved whilst accelerating genetic improvement in grain legume using innovative methods of crop breeding such as hybrid breeding [13].

Conventional plant breeding in a self-pollinated crop is based on a large number of phenotypic selections that are made from genetically diverse populations, to accumulate beneficial alleles in a stepwise manner in homozygous inbred lines [14]. These inbred lines are then passed through multiple evaluation trials for various biotic and abiotic stresses. Finally, the superior selected inbred lines are multiplied and delivered by the seed industry to growers. F₁ hybrid breeding is an alternative approach where the product delivered to growers is the first filial generation of a cross that exploits cross-breeds with improved vigor over the parental genotypes and this is referred to as hybrid vigor or heterosis [15]. To achieve this, inbred parental lines are first developed, and these are used to perform a uni-directional cross at the final stage to deliver the heterotic F₁ product. However, currently commercial exploitation of hybrid vigor is limited to a relatively small number of crops despite the evidence that heterosis above high parent exists within many species. The floral biology of the crop, the nature of pollination and natural out-crossing rate in crops plays a significant role in heterosis. Development of hybrid cultivars in some of the self-pollinated crops such as lentil has several challenges, e.g., inability to produce enough seeds per cross, possibility of outcrossing, absence of male sterile lines all of which makes it difficult to commercialise hybrids on an economic scale [16]. Benefits of heterosis can only be harnessed if F₁ seed production can be scaled up using innovative approaches. However, there are some hybrid cultivars in self-pollinating crops such as, egg plant, tomato, and pepper, with 30–60% hybrid yield advantage which is achieved due to perfect size of flower, many seeds per cross and with natural out-crossing rate close to null [17].

Heterotic F₁ hybrid can be achieved by F₁ hybrid-based breeding systems, such as artificial or chemical emasculation; cytoplasmic or nuclear male sterility [18]. In the absence of highly efficient pollination control technology, a new alternative pathway to deliver heterosis has been proposed via hybrid mimics where the first level of hybrid conventional selection from the pure breeding lines are held and selected for at every generation until the desired characteristics of F₁ hybrid are stabilised [19]. Heterosis via F₁ hybrids has been realised in both self and cross pollinated crops such as rice, wheat, Brassica napus, sorghum, sunflower, rye, onion, tomato, peppers, and barley [20–23] with up to 30% yield gains and its utilization has contributed greatly to global crop production with high level of hybrid adoption [20]. The use of hybrid maize since its inception has significantly contributed to the yield gains realised, these production gains helped expand hybrid maize acreage from less than 10% to over 90% in Iowa, United States from 1935 to 1940 [24]. Heterosis is often reported as trait-specific, for instance, tomato hybrids were found to be highly heterotic for seed number per plant, fruit number and total yield but other phenotypes such as fruit weight and seed morphology displayed no heterosis [25]. Additionally, Flint-Garcia et al. [26] identified heterosis for yield in maize whilst flowering time was additive.

To explore the underlying genetic principles of heterosis, several models have been proposed [27–30]. Heterosis through dominance occurs when the dominant alleles from one parent suppress the inferior recessive alleles from the other parent. With this model, it is theoretically possible to eliminate all deleterious alleles and/or accumulate all desired alleles to create an inbred line performing similar to the F₁ hybrid. The overdominance hypothesis details the superior performance of the hybrid due to the heterozygosity that delivers advantages over and above homozygous state. The third model is epistasis which explains the scenario where two or more non-allelic genes derived from the parental lines interact to generate heterosis. Several studies have been undertaken to understand the genetic mechanism of heterosis, however there is still no single genetic model to accurately predict the range and quantify the level of heterosis [21]. It is possible that alleles accumulate or interact with contributions from different models to explain the molecular mechanism of heterosis [20].

Understanding and quantifying heterosis in lentils will help explore the opportunities and potential gains that hybrid breeding could deliver to the agriculture industry. However, there are currently no reliable means to deliver heterotic F₁ cultivars in a commercial ready system in lentils. F₁ hybrid-based breeding systems, such as artificial or chemical emasculation; cytoplasmic or nuclear male sterility can be explored in lentils. Another potential solution could be hybrid mimics as a simpler way to deliver the benefit of heterosis without the need for complex seed multiplication and production systems. Hybrid mimics have been reported in wheat, peas, tomato, and tobacco [31–33]. Hybrid mimics have been exemplified in Arabidopsis [19] however, they have not been commercially exploited so far. Therefore, it is worthwhile evaluating the potential of hybrid mimics in lentils to explore the challenges associated with their commercial use.

The objectives of this study are to identify and quantify heterosis by phenotypic evaluation and selection for various developmental and yield related traits in lentil hybrids through multiple generations. To evaluate the range of heterosis in lentils, the percent yield gains were compared to other crops to model and inform their potential benefit for the lentil production industry. To explore if heterosis is trait specific and to detail highly heterotic traits in a more accurate way, correlations among measured traits were explored. Genetic diversity of a set of parental lines was performed to examine correlations between genetic divergence and heterosis. Furthermore, the study evaluated and considered the heterosis results observed to understand the potential of delivery via hybrid mimics. Recommendations on future potential hybrid breeding schemes in lentil breeding were made.

Material and methods

Plant material and generation advance

A total of 47 lentil genotypes (S1 Table) selected on the basis of genetic diversity, yield and its component traits, such as seed number, plant height and biomass fresh weight (S1 Table) were cross hybridized using multiple parental combinations listed in S2 Table by hand pollination, over the spring of 2016 at The Grains Innovation Park, Horsham, Victoria, Australia and progressed to develop subsequent generations in a glasshouse under controlled environmental conditions (Fig 1). A total of 72 F₁ hybrids as well as parent genotypes were initially evaluated in a randomized complete block design consisting of 4 replicates for seed number, yield, biomass fresh weight, and plant height with six crosses being taken forward to the next generation. The F₁ crosses were numbered in descending order based on levels of heterosis. The top five F₁ better parent heterotic crosses in terms of seed number were chosen and a filter on genetic diversity was also imposed to ensure that multiple genetics were selected. Along with the top five performing crosses, a negative heterosis control cross-72 was also selected. At the F₂ generation, 125–190 seeds per family were sown as separate blocks (Fig 1) and the populations were not intermixed across blocks. While designing the glasshouse trial, replicated parental controls were present in every block to evaluate environmental variance. Within the block structure, six replicates of each parent were distributed within the relevant population for all generation trials. For every F₂ family the ten best performing lines and ten worst performing lines in terms of seed number were taken forward as sub selected F₃ families (20 in total per F₁ cross) for evaluation. A total of five F₃ plants per family (total of 100 plants from 20 F₃ families that relate to each F₁ cross) were screened in a randomised block trial (Fig 1).

Download:

Fig 1. A summary of the generation advance in lentils, covering F₁ to F₃.

The plants at each generation were evaluated for seed number, seed weight, biomass fresh weight and plant height.

https://doi.org/10.1371/journal.pone.0262857.g001

Growth conditions

All plants were grown under controlled environment conditions using standard potting mix (1.3 L) in 140 mm plastic pots at 22 ± 2°C with a photoperiod of 16/8-h (light/dark). Both F₁ and F₃ generations were evaluated using a randomised complete block design and F₂ was evaluated with blocking structure around families with randomised parents in every block. A minimum of four replicates were evaluated for each parent at all three generations in every cross to evaluate environmental variation.

Phenotypic evaluation

A number of yield related traits were measured in each experiment (F₁, F₂, F₃ generational assessment) to quantify heterosis, where each individual was assessed for seed number, seed weight, biomass fresh weight and plant height at time of harvest. Each plant was harvested by taking all the above ground biomass, packing the biomass in a pre-measured paper bag, and weighed for biomass fresh weight on a balance (Thermoline Scientific Precision Balance, WLC 6/A2, d = 0.1g; New South Wales, Australia). Subsequently, these plants were oven-dried at 37°C for 48–72 h and threshed for seeds. Seeds were collected from each individual plant and the number of seeds generated was counted using an automatic seed counter (Data Count, JR; Data Technologies, Tzora, Israel) as well as generating a total yield value in terms of total seed weight per plant. In addition, a hundred seed weight per plant value was calculated by dividing the total seed weight by the number of seeds and then multiplying by a hundred. The data was evaluated for mean, and standard error using ASREML (v 4.1.0) [34].

Phenotypic data analysis

For the F₁ phenotypic data, a spatial analysis was performed, and predicted mean values were calculated for seed number, yield, biomass fresh weight, and plant height using ASREML [34]. In addition, for each trait under study one-hundred numeric values were simulated using the predicted mean values along with standard errors assuming a normal distribution.

For the F₂ and F₃ generation evaluation, the data was analysed on a single plant basis since it was not possible to have genetic replicates of the population. However, mean values of the replicated parents along with their standard errors have been considered to understand environmental variance in the glasshouse trial.

Correlation coefficients between plant height, biomass fresh weight, seed number, and yield per plant were determined using R software (v4.0.0).

The percentage of heterosis was analysed in F₁, F₂ and F₃ as better parent heterosis (BPH), which was computed BPH = (F₁-BP)/BP × 100, where BP referred to the better-parent value.

Multiple Linear Regression analysis was performed in F₂ and F₃ generations using library Tidyverse in R software (v4.0.0). A single multivariate regression model was created to predict yield based on four predictor variables, such as, biomass fresh weight, plant height, seed number and hundred grain weight.

Genotypic data of parental varieties

Leaf tissues from multiple nodes were collected from 4 weeks-old plants and were frozen immediately in liquid nitrogen and stored at −80 °C until required. Total RNA (tRNA) was extracted using RNeasy^® Plant Mini Kit (QIAGEN, Hilden, Germany) following manufacturer’s instructions. RNA-Seq libraries with an approximate insert size of 350 bp were prepared using Sure Select Strand Specific RNA library prep kit and evaluated using the Tape Station 2200 platform with HSD1000 Screen Tape System (Agilent Technologies, Santa Clara, CA, USA) according to the manufacturer’s protocols. Equal mass of each sequencing library with a unique barcode was combined to create a single pooled sample for sequencing. All reads were pair-end sequenced using the HiSeq 3000 and MiSeq platforms (Illumina Inc., San Diego, CA, USA). Raw data has been deposited to NCBI under accession number GSE184819 (Advances in lentil production through heterosis: evaluating generations and breeding systems (lentil)); Data Type: transcriptome or gene expression).

Following Fastq data generation, the raw sequence reads were filtered using a custom perl script [35] to remove adaptor sequences along with reads and bases of low quality (Q ≤ 30). Reads with three consecutive unassigned nucleotides (N) were also trimmed and finally any reads shorter than 50 bp in length were removed from the final set. The remaining high-quality trimmed sequence reads were aligned to the lentil reference transcriptome [36] using BWA-MEM [37]. The number of properly paired reads were obtained using the SAMtools flagstat option and the mapping reads were obtained [38]. Variant calling was performed using SAMtools (version-1.5) [38]. The final VCF output was then filtered using VCFtools [9] based on the following parameters: depth (DP ≥ 5), maximum allelic frequency (MAF = 0.1), maximum missing data (20%), and base quality (Q30) with a predefined SNP list [39].

Genetic diversity analysis was performed using the abovementioned filtered SNP data from all accessions. Genetic distances for each lentil accession were calculated using Nei’s method within the StAMPP package [40]. A phylogenetic tree was constructed using the unweighted neighbour-joining (NJ) method, as implemented in the DARwin-6.0.17 software [41].

Results and discussion

A total of 72 biparental crosses between varying lentil genotypes were established (S2 Table). Of the resultant F₁’s, 43 crosses exhibited positive heterosis based on seed number. Overall, the range of heterosis observed varied between -59% to 62%, with 18 out of 72 crosses exhibiting greater than 20% heterosis (S2 Table). The crosses were numbered from one to seventy-two in descending order of heterosis at the F₁ generation for the trait of seed number. The F₁ data suggested the presence of heterosis in lentils and identified some of the best performing crosses for detailed analysis of the phenomenon.

Patterns of heterosis for seed number, yield, and biomass fresh weight

Heterosis for seed number, yield, and biomass fresh weight of the five positive heterotic crosses performed in a similar way and exhibited maximal positive gains at the F₁ generation with an overall decline in the population mean in subsequent generations. However, despite the decline in heterosis, the distribution of the individuals within the generation broadened (Fig 2a–2c).

Download:

Fig 2. F₁, F₂ and F₃ hybrids from selected crosses (Cross-1, 2, 3, 8 and 10) evaluated for better parent heterosis in terms of a) seed number b) seed weight c) biomass fresh weight and d) plant height.

https://doi.org/10.1371/journal.pone.0262857.g002

Heterosis for seed number, yield, and biomass fresh weight at the F₁ generation was quantified in the range of 31%–62%, 26%–56% and 18%–50%, respectively across the five selected crosses with a global average of 31±1% (Table 1) quantifying heterosis from some of the best performing crosses for the detailed analysis of the phenomenon. Heterosis for seed number is one of the important traits affecting seed yield. Nearly 11% heterosis was observed for the number of seeds per silique in F₁ B.napus hybrids [42]. Similar to this study, 10%–30% heterosis for yield has been reported in wheat, maize, rice, pea, soybean, and B. napus in F₁ heterozygous generation [20, 43]. Dinkins et al. [44] reported 12% heterosis for biomass in F₁ hybrids of soybean and in comparison, 18%–50% heterosis for biomass was observed in the current study. Better parent heterosis for yield (25%) has also been recorded in commercial F₁ hybrid cultivars of Chinese cabbage [45].

Download:

Table 1. Mean value of percent heterosis of selected crosses (cross-1, 2, 3, 8 and 10) along with negative control (cross-72*) for seed number (SN), yield, biomass fresh weight (BFW) and plant height (Ht) at F₁, F₂ and F₃.

The table includes Maximal (Max) and Minimal (Min) percent heterosis values at F₂ and F₃ along with parents represented as Mean±S.E.

https://doi.org/10.1371/journal.pone.0262857.t001

Heterosis was further investigated at the F₂ generation (Fig 2a–2c) with the mean heterosis values for the traits of seed number, yield and biomass fresh weight being -8% to 20%, -9% to 63%, -14% to 9%, respectively across five crosses (Table 1). The negative heterosis control cross at the F₂ stage had a migration of the mean value towards the zero heterosis level, clearly following the same trend but demonstrating a loss of negative heterosis rather than a loss of positive heterosis. However, this decrease in heterosis is in concordance with the classical theory of heterosis [46, 47]. Similarly, Sarawat et al. [48] observed up to 50% reduction in heterosis at F₂ as compared to the F₁ for grain yield and total dry matter in peas. Wang et al. [19] also noticed that yield advantage of the F₁ is lost in the F₂ and subsequent generations in Arabidopsis. Another study by Burton and Brownie [49] reported that the yield advantage declined from 16% to 5% as the hybrids progressed from F₁ to F₂ in soybean. There are other reports for significant decline in heterosis with the subsequent generations in soybean [43, 50]. Scheffler et al. [51] also reported in maize the F₂ generation averaged 32% less grain yield than the F₁ generation.

Although, the mean values of heterosis decreased at F₂ as compared to F₁, the spread of values significantly increased across the populations so there were a small proportion of individual genotypes that displayed positive heterosis. For instance, the maximal heterotic values for seed number varied from 58% to 181% across five crosses at F₂. These rare extreme genotypes delivering positive heterosis from the population at F₂ had heterosis values higher than the mean F₁ value with an exception of crosses-2 and 3 (Table 1). These rare extreme genotypes account, at the F₂ generation for 2% to 20% of individuals for the trait of seed number across all five best performing crosses (S3 Table). A similar observation has been reported by Sarawat et al. [32, 48] where some of the F₂ populations maintained the high yield levels of the corresponding F₁ hybrids in pea. Singh et al. [52] also observed the outliers at F₂ in pea.

To understand the trend of heterosis at further generations, F₂ plants that had maximal and minimal values for seed number, were evaluated as the sub selected F₃ generation (Fig 2a–2c). The mean heterosis values for seed number, yield and biomass fresh weight varied from -42% to -5%, -52% to -5% and -37% to -6% respectively across five crosses (Table 1), representing a decrease in the mean value of heterosis for all traits in comparison to both the F₁ and F₂ generations. This generation advance did not perform as expected, with the offspring from both tails of the F₂ distribution performing comparably. The sub-selected F₃ generation also was the first point where the mean values for all traits in all crosses recorded negative values. However, despite the mean negative values of heterosis at this stage, there were again a small number of individuals, a comparable proportion of the population as was identified at the F₂ generation (5% to 29%), with phenotypes at the extreme of the distribution that performed better than the F₁ generation mean value (Table 1 and S3 Table). This variation should be expected as a high number of loci at F₂ would still be segregating. The evaluation of environmental variance of all the genotypes in the trial was still not possible with segregating individuals unable to be replicated as previously discussed. However, the environmental variation as assessed by the parental genotypes provides some quantification of the variance that was experienced in the trial, and reassuringly, the extreme values from the population were still positively heterotic even considering the environmental variance as estimated through the parental replicates. Fischer and Rebetzke [53] also discussed similar issues relating to early generation selection for yield in conventional breeding systems in self-pollinated crops, where allelic segregation and recombination in early generations (F₂-F₄) leads to unreliable selection before homozygosity is attained. As a result, future studies would be suggested to be conducted from the entire F₂ derived F₃ families rather than making selections and taking the tails of the distribution. Alternatively, it might be suggested to ignore the F₂ generation evaluation entirely and simply perform a seed multiplication to advance generations and useF_2:3 families to retrospectively calculate F₂ genotype values. This would allow a form of within family replication. Alternatively modelling of rates of selection, recombination and inbreeding could indicate the most efficient strategy to advance either the entire populations or sub selected fractions to achieve advanced populations that would still have small proportions of the extreme heterotic individuals.

The identification of individuals at sequential generations that outperform the F₁ generation supports the concept that the hybrid mimic breeding strategy could be deployed for lentil varietal development. This approach utilizes recurrent selection for the critical hybrid vigor trait to advance generations while stabilising and multiplying the population to retain heterosis in pure breeding lines [15, 19]. Researchers have been able to select hybrid mimic like plants in a number of species, for instance Busch et al. [31] reported that the pure breeding F₅ lines derived from hybrid plants performed equivalent to F₁ hybrid in wheat. Similar findings have been reported in pea and tomato where researchers have been able to develop stable F₅-F₆ lines with the same characteristics as the parental F₁ hybrids [32, 33]. Sarawat et al. [32] reported equivalent performance of F₅ lines in pea as compared to F₁ hybrids (out yielded the best parent by up to 11%). Similar findings have been reported in pea, tomato, and tobacco where researchers have been able to develop stable F₅-F₆ lines with the same characteristics as the parental F₁ hybrids [32, 33, 54]. Although there are now several reports of the development of hybrid mimics, it is still challenging to extend the hybrid advantage beyond F₁ and as discussed, the method of generation advance needs careful work, consideration, and experimentation. Sarawat et al. [32] reported a similar percentage of the population (2–15%) compared to this study at F₅ performed better than the F₁ mean. The limited scale of valuable lines and attrition in the multiplication process does present additional logistical challenges on a commercial scale for hybrid mimics and necessitates the screening of a large number of lines and reliable germplasm to work with. It is likely that due to the aforementioned complexities, hybrid mimics have not been more widely commercially exploited so far.

It is understood that there are some challenges associated with the development of hybrid mimics. An alternative way to utilise heterosis is through wide hybridisation heterosis using pollination control mechanisms. Different pollination control systems have been used to breed for heterosis. In some crops such as rice, maize, wheat, B. napus, and sunflower, cytoplasmic or nuclear encoded male sterility has been explored [18]. Future research activities could focus on the identification and development of male sterile lines in lentil, through natural or induced mutations. There are reports in the literature of induced mutants in lentils for male sterility already [55] and further research on this is likely to prove fruitful.

Patterns of heterosis for plant height

In comparison to other traits in the study, the F₁ hybrids exhibited a subtly different pattern of heterosis for plant height. As compared to other traits, the level of heterosis was significantly lower for plant height except for cross-1 (Fig 2d). Heterosis for plant height varied from -3 to 32% at F₁, -10 to 1% at F₂ and -19 to 3% at sub-selected F₃ (Table 1). The populations exhibited overlapping distributions between F₁ and F₂ generations, however, a wider distribution was noticed at the sub-selected F₃. Although extreme phenotypes were observed for plant height, the extreme performance was lower than that observed for seed number, yield and/or biomass fresh weight. The highest value from the population for heterosis for plant height was 36% and 58% at the F₂ and sub selected F₃ generation respectively (Table 1). There have been previous studies, which clearly demonstrate the trait specific nature of heterosis [25, 56, 57]. In B. napus and soybean, positive heterosis was seen for yield related traits (1000 seed weight and seed yield) but not for plant height [56, 57]. van Hulten et al. [58] reported that the genetic mechanisms underlying heterosis were highly trait specific and were largely dependent on the genetic background. In agreement with the aforementioned studies, the trait specific nature of heterosis is confirmed for lentil and further, it is necessary to identify and establish the correlations between production traits on a species and potentially genotype basis. However, the exploration of heterosis in other agronomically important traits, such as flowering time, maturity time and plant architecture, could be of value to lentil breeders if it can be shown how these traits relate to yield.

Correlation studies across generations

A correlation analysis was undertaken to investigate the consistency of heterotic patterns between the different crosses (S4 Table). Multiple linear regression analysis has also been used to explore the relationship between independent or predictor variables (S5 Table). Generally, positive correlations between yield traits are beneficial in improving productivity through conventional breeding. In the current study, yield was significantly and positively correlated with seed number, biomass fresh weight, plant height and 100 grain weight in all six crosses across the F₁, F₂ and sub selected F₃ populations (r = 0.23–0.99) except for cross-2 where yield was not significantly correlated to 100 grain weight at F₁. Also, cross-1 showed no significant correlation between yield and plant height at F₂. The degree of correlation was evaluated across the generations in all the crosses under investigation and it was observed that yield correlated to plant height (r = 0.23–0.99), biomass fresh weight (r = 0.37–0.98), seed number (r = 0.32–0.98), and 100 grain weight (r = 0.26–0.66). Multiple regression analysis also confirmed that yield per plant at F₁, F₂ and sub selected F₃ generation could be predicted by hundred grain weight (coefficient = 0.061–2.10) and seed number (coefficient = 0.01–0.06) within each generation in all the populations under study (multiple R-squared values > 0.91, S5 Table). Broadly, the correlation study suggested that plants with greater biomass fresh weight, yielded a greater number of seeds with higher yield per plant. Parallel to the current study where yield correlated with seed number and biomass fresh weight in all the six crosses under study, similar observations have been reported in B. napus where seed yield correlated significantly with nine yield related traits (including seed number, biomass yield, pod number, and plant height) for both trait performance and mid parent heterosis [59]. Significant correlations between yield related traits have been observed in a number of other crops such as tomato, wheat, maize, and sunflower [25, 60–63]. There are reports where high heritability values have been recorded in lentil for grain weight (h² = 0.87) but moderate heritability has been observed for yield (h² = 0.5; [64, 65]. This indicates that grain weight experiences less genotype x environment interactions, whilst yield can be heavily affected by the environment. Therefore, it is important to understand the correlations between genetically controlled traits and traits impacted by environmental variance. Further, correlation studies will be useful for the development of hybrid mimics and evaluation of germplasm for generation advance. Efforts are also required to extend the work from this controlled environment study to a field setting.

Genetic diversity analysis

In an attempt to evaluate and predict heterosis and heterotic combinations, a genetic diversity analysis was performed, and the full set of parents used to create the 72 initial F₁ crosses were evaluated for their genetic distances and relationships (Fig 3). The genetic distance matrix along with a phylogenetic tree revealed that genetic distance between all the parents of F₁ hybrids was in the range of 0.049–0.68. Broadly, looking at the top ten heterotic crosses in the phylogenetic tree, they are not most genetically consistent and do not provide a clear picture of heterosis, however, they sit on different clades along with some non-heterotic crosses (S1 Fig). Narrowing the results to the selected six crosses, the average genetic distance between parent 1 and parent 2 of a particular F₁ hybrid was 0.45 and across other crosses was not less than 0.23. This gives an initial set of boundary parameters for further evaluation to test genotypes with a degree of diversity for heterosis. There was a commonality of genetics between parents of cross-2 and cross-72, apart from being genetically distant (genetic distance between parents of cross-2 and cross-72 was 0.36, 0.51, respectively) they did not exhibit heterosis (Table 2; Fig 4). From this analysis it is evident that genetic distance alone does not predict heterosis, although all the crosses had distance values greater than 0.23, which is an indicator that can be derived from this approach (Table 2). The maximal heterotic crosses also clearly failed to define specific groupings or pools. More crosses are needed to dissect the relationships and provide finer resolution over the nature of heterosis in lentils. There have been reports for both positive (within a range of divergent parental lines) and negative (when the genetic distance was extremely high) correlation between genetic distance and heterosis [66]. A positive correlation between genetic distance and heterosis has been discussed by Birchler et al. [27]. However, there have been several studies where genetic distance and heterosis are not always positively correlated [67–69]. Van Hulten et al. [58] observed no correlation between heterosis levels and the genetic distance between parental lines. Fujimoto et al. [20] suggested that F₁ hybrids between genetically distant parents would not necessarily produce heterotic F₁ hybrids, only specific combinations of parental lines exhibit heterosis relative to the parental lines which correlates well to the current study. Maximal gains in performance are most likely to be realised if heterotic loci can be identified and selected for genomically from specific lentil plants.

Download:

Fig 3. Correlation between genetic distance (nei’s genetic distance) of the selected parents and percent heterosis for seed number at F₁.

https://doi.org/10.1371/journal.pone.0262857.g003

Download:

Fig 4. Phylogenetic tree of lentil parents used for creating crosses under heterosis study.

The parents of the selected crosses used in the study have been colour coded and abbreviated for ease of visualisation, e.g C1P1 = Cross-1; Parent-1.

https://doi.org/10.1371/journal.pone.0262857.g004

Download:

Table 2. Genetic distance (Nei’s genetic distance) between the parents of the heterotic crosses under study.

https://doi.org/10.1371/journal.pone.0262857.t002

Conclusion

Heterosis does exist in lentils and up to 62% heterosis being identified at F₁ in a controlled environment. However, the yield advantage diminishes with the successive generations, but despite of the reduction, the range of heterosis spreads broadly and some individual high performing genotypes were identified at F₂ and sub selected F₃ generations. Heterosis was evaluated for yield related agronomically important traits and it was noticed that yield correlated with seed number, biomass fresh weight, plant height and 100 grain weight. Further, to explore the reasons underlying heterosis, genetic diversity analysis of all the F₁ parents was performed and the study failed to identify a correlation between genetic distance and divergent parents with heterotic F₁’s. Thus, the study can be further utilised to explore ways to achieve potential yield gains from heterosis either by F₁ hybrid-based breeding systems and/or via modelling through computational simulation of different breeding programmes and recombination patterns to inform the best way to develop hybrid mimics. Despite the potential challenges, the potential yield gains in lentil from heterosis has the ability to transform the industry and warrants further efforts to realise the benefits.

Supporting information

S1 Fig. Phylogenetic tree describing the position of top ten heterotic crosses (highlighted in coloured boxes) along with other less heterotic and non-heterotic crosses.

https://doi.org/10.1371/journal.pone.0262857.s001

(TIF)

S1 Table. Mean value for plant height, seed number, biomass fresh weight and yield of 47 selected genotypes used for producing heterotic F1’s.

https://doi.org/10.1371/journal.pone.0262857.s002

(XLSX)

S2 Table. Percent heterosis of F1 hybrids for seed number, yield, biomass fresh weight, and plant height.

https://doi.org/10.1371/journal.pone.0262857.s003

(XLSX)

S3 Table. Percentage of individuals performing better at F1, F2 and F3.

https://doi.org/10.1371/journal.pone.0262857.s004

(XLSX)

S4 Table. Correlation coefficients between growth and yield related parameters in different populations evaluated at F2 and F3 stage (*P≤0.5).

https://doi.org/10.1371/journal.pone.0262857.s005

(XLSX)

S5 Table. Multiple linear regression analysis across generations in test populations.

https://doi.org/10.1371/journal.pone.0262857.s006

(XLSX)

References

1. Adsule R, Lawande K, Kadam S. Pea. Handbook of world food legumes; nutritional chemistry, processing technology and utilization CRC Press, Inc, Boca Raton. 1989:215–51.
2. Kim H, Behall K, Conway J, editors. Short-term satiety and glycemic response after consumption of whole grains varying in the amount of β-glucan in overweight adults. North American Association for the Study of Obesity 2005 Annual Meeting Obesity Research; 2005.
3. Durazzo A, Turfani V, Azzini E, Maiani G, Carcea M. Phenols, lignans and antioxidant properties of legume and sweet chestnut flours. Food chemistry. 2013;140(4):666–71. pmid:23692751
- View Article
- PubMed/NCBI
- Google Scholar
4. FAOSTAT. Crops. 2020.
5. Corporation GRD. DAV00154—Pulse Breeding Australia: Lentil Breeding 2021 [https://grdc.com.au/research/projects/project?id=4297.
6. Corporation GRD. Still loving those legumes—more adventures with lentils 2021 [https://grdc.com.au/resources-and-publications/grdc-update-papers/tab-content/grdc-update-papers/2018/02/still-loving-those-legumes.
7. Foyer CH, Lam H-M, Nguyen HT, Siddique KH, Varshney RK, Colmer TD, et al. Neglecting legumes has compromised human health and sustainable food production. Nature plants. 2016;2(8):1–10. pmid:28221372
- View Article
- PubMed/NCBI
- Google Scholar
8. Lombardi M, Materne M, Cogan NO, Rodda M, Daetwyler HD, Slater AT, et al. Assessment of genetic variation within a global collection of lentil (Lens culinaris Medik.) cultivars and landraces using SNP markers. BMC genetics. 2014;15(1):1–10. pmid:25540077
- View Article
- PubMed/NCBI
- Google Scholar
9. Danecek P, Auton A, Abecasis G, Albers CA, Banks E, DePristo MA, et al. 1000 Genomes Project Analysis Group. The variant call format and vcftools. Bioinformatics. 2011;27(15):2156–8.
- View Article
- Google Scholar
10. Khazaei H, Caron CT, Fedoruk M, Diapari M, Vandenberg A, Coyne CJ, et al. Genetic diversity of cultivated lentil (Lens culinaris Medik.) and its relation to the World’s agro-ecological zones. Frontiers in Plant Science. 2016;7:1093. pmid:27507980
- View Article
- PubMed/NCBI
- Google Scholar
11. Wright DM, Neupane S, Heidecker T, Haile TA, Chan C, Coyne CJ, et al. Understanding photothermal interactions will help expand production range and increase genetic diversity of lentil (Lens culinaris Medik.). Plants, People, Planet. 2021;3(2):171–81.
- View Article
- Google Scholar
12. Singh M, Sharma SK, Singh B, Malhotra N, Chandora R, Sarker A, et al. Widening the genetic base of cultivated gene pool following introgression from wild Lens taxa. Plant Breeding. 2018;137(4):470–85.
- View Article
- Google Scholar
13. Varshney RK, Kudapa H, Pazhamala L, Chitikineni A, Thudi M, Bohra A, et al. Translational genomics in agriculture: some examples in grain legumes. Critical Reviews in Plant Sciences. 2015;34(1–3):169–94.
- View Article
- Google Scholar
14. Borlaug NE. Contributions of conventional plant breeding to food production. Science. 1983;219(4585):689–93. pmid:17814030
- View Article
- PubMed/NCBI
- Google Scholar
15. Wang L, Wu LM, Greaves IK, Zhu A, Dennis ES, Peacock WJ. PIF4-controlled auxin pathway contributes to hybrid vigor in Arabidopsis thaliana. Proceedings of the National Academy of Sciences. 2017;114(17):E3555–E62. pmid:28396418
- View Article
- PubMed/NCBI
- Google Scholar
16. Tewodros M, Mohammed A. Heterotic response in major cereals and vegetable crops. International Journal of Plant Breeding and Genetics. 2016;10(2):69–78.
- View Article
- Google Scholar
17. Nienhuis J, Sills G. The potential of hybrid varieties in self-pollinating vegetables. Reproductive biology and plant breeding: Springer; 1992. p. 387–96.
18. Fu D, Xiao M, Hayward A, Fu Y, Liu G, Jiang G, et al. Utilization of crop heterosis: a review. Euphytica. 2014;197(2):161–73.
- View Article
- Google Scholar
19. Wang L, Greaves IK, Groszmann M, Wu LM, Dennis ES, Peacock WJ. Hybrid mimics and hybrid vigor in Arabidopsis. Proceedings of the National Academy of Sciences. 2015;112(35):E4959–E67. pmid:26283378
- View Article
- PubMed/NCBI
- Google Scholar
20. Fujimoto R, Uezono K, Ishikura S, Osabe K, Peacock WJ, Dennis ES. Recent research on the mechanism of heterosis is important for crop and vegetable breeding systems. Breeding science. 2018:17155. pmid:29875598
- View Article
- PubMed/NCBI
- Google Scholar
21. Vasseur F, Fouqueau L, De Vienne D, Nidelet T, Violle C, Weigel D. Nonlinear phenotypic variation uncovers the emergence of heterosis in Arabidopsis thaliana. PLoS biology. 2019;17(4):e3000214. pmid:31017902
- View Article
- PubMed/NCBI
- Google Scholar
22. Vacher M, Small I. Simulation of heterosis in a genome-scale metabolic network provides mechanistic explanations for increased biomass production rates in hybrid plants. NPJ systems biology and applications. 2019;5(1):1–10.
- View Article
- Google Scholar
23. Ahmar S, Gill RA, Jung K-H, Faheem A, Qasim MU, Mubeen M, et al. Conventional and molecular techniques from simple breeding to speed breeding in crop plants: Recent advances and future outlook. International journal of molecular sciences. 2020;21(7):2590. pmid:32276445
- View Article
- PubMed/NCBI
- Google Scholar
24. Crow JF. 90 years ago: the beginning of hybrid maize. Genetics. 1998;148(3):923–8. pmid:9539413
- View Article
- PubMed/NCBI
- Google Scholar
25. Semel Y, Nissenbaum J, Menda N, Zinder M, Krieger U, Issman N, et al. Overdominant quantitative trait loci for yield and fitness in tomato. Proceedings of the National Academy of Sciences. 2006;103(35):12981–6. pmid:16938842
- View Article
- PubMed/NCBI
- Google Scholar
26. Flint-Garcia SA, Buckler ES, Tiffin P, Ersoz E, Springer NM. Heterosis is prevalent for multiple traits in diverse maize germplasm. PloS one. 2009;4(10):e7433. pmid:19823591
- View Article
- PubMed/NCBI
- Google Scholar
27. Birchler JA, Yao H, Chudalayandi S. Unraveling the genetic basis of hybrid vigor. Proceedings of the National Academy of Sciences. 2006;103(35):12957–8. pmid:16938847
- View Article
- PubMed/NCBI
- Google Scholar
28. Lippman ZB, Zamir D. Heterosis: revisiting the magic. Trends in genetics. 2007;23(2):60–6. pmid:17188398
- View Article
- PubMed/NCBI
- Google Scholar
29. Charlesworth D, Willis JH. The genetics of inbreeding depression. Nature reviews genetics. 2009;10(11):783–96. pmid:19834483
- View Article
- PubMed/NCBI
- Google Scholar
30. Seymour DK, Chae E, Grimm DG, Pizarro CM, Habring-Müller A, Vasseur F, et al. Genetic architecture of nonadditive inheritance in Arabidopsis thaliana hybrids. Proceedings of the National Academy of Sciences. 2016;113(46):E7317–E26. pmid:27803326
- View Article
- PubMed/NCBI
- Google Scholar
31. Busch R, Lucken K, Frohberg R. F1 Hybrids Versus Random F5 Line Performance and Estimates of Genetic Effects in Spring Wheat 1. Crop science. 1971;11(3):357–61.
- View Article
- Google Scholar
32. Sarawat P, Stoddard F, Marshall D. Derivation of superior F 5 lines from heterotic hybrids in pea. Euphytica. 1993;73(3):265–72.
- View Article
- Google Scholar
33. Williams W. The isolation of ‘pure lines’ from F 1 hybrids of tomato, and the problem of heterosis in inbreeding crop species. The Journal of Agricultural Science. 1959;53(3):347–53.
- View Article
- Google Scholar
34. Gilmour A, Gogel B, Cullis B, Thompson R. ASReml user guide. Release; 2009.
35. Braich S, Baillie RC, Spangenberg GC, Cogan NO. A New and Improved Genome Sequence of Cannabis sativa. bioRxiv. 2020.
- View Article
- Google Scholar
36. Sudheesh S, Verma P, Forster JW, Cogan NO, Kaur S. Generation and characterisation of a reference transcriptome for lentil (Lens culinaris Medik.). International journal of molecular sciences. 2016;17(11):1887.
- View Article
- Google Scholar
37. Li H. Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM. arXiv preprint arXiv:13033997. 2013.
38. Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, et al. The sequence alignment/map format and SAMtools. Bioinformatics. 2009;25(16):2078–9. pmid:19505943
- View Article
- PubMed/NCBI
- Google Scholar
39. Malmberg MM, Pembleton LW, Baillie RC, Drayton MC, Sudheesh S, Kaur S, et al. Genotyping-by-sequencing through transcriptomics: implementation in a range of crop species with varying reproductive habits and ploidy levels. Plant biotechnology journal. 2018;16(4):877–89. pmid:28913899
- View Article
- PubMed/NCBI
- Google Scholar
40. Pembleton LW, Cogan NO, Forster JW. St AMPP: An R package for calculation of genetic differentiation and structure of mixed-ploidy level populations. Molecular ecology resources. 2013;13(5):946–52. pmid:23738873
- View Article
- PubMed/NCBI
- Google Scholar
41. Perrier X. DARwin software. http://darwinciradfr/darwin. 2006.
42. Radoev M, Becker HC, Ecke W. Genetic analysis of heterosis for yield and yield components in rapeseed (Brassica napus L.) by quantitative trait locus mapping. Genetics. 2008;179(3):1547–58. pmid:18562665
- View Article
- PubMed/NCBI
- Google Scholar
43. Brim CA, Cockerham CC. Inheritance of quantitative characters in soybeans. Crop Science. 1961;1:189–90.
- View Article
- Google Scholar
44. Dinkins R, Keim K, Farno L, Edwards L. Expression of the narrow leaflet gene for yield and agronomic traits in soybean. Journal of Heredity. 2002;93(5):346–51.
- View Article
- Google Scholar
45. Saeki N, Kawanabe T, Ying H, Shimizu M, Kojima M, Abe H, et al. Molecular and cellular characteristics of hybrid vigour in a commercial hybrid of Chinese cabbage. BMC plant biology. 2016;16(1):1–15.
- View Article
- Google Scholar
46. Fisher RA. Theory of inbreeding: Oliver and Boyd, London; 1949.
47. Kearsey M, Pooni H. Genetical analysis of quantitative traits: Garland Science; 2020.
48. Sarawat P, Stoddard F, Marshall D, Ali S. Heterosis for yield and related characters in pea. Euphytica. 1994;80(1):39–48.
- View Article
- Google Scholar
49. Burton JW, Brownie C. Heterosis and inbreeding depression in two soybean single crosses. Crop science. 2006;46(6):2643–8.
- View Article
- Google Scholar
50. Lewers K, St. Martin S, Hedges B, Palmer R. Testcross evaluation of soybean germplasm. Crop science. 1998;38(5):1143–9.
- View Article
- Google Scholar
51. Scheffler T, Hallauer AR, Lamkey KR, White PR. Estimates of heterosis and inbreeding depression for crosses of lowa maize populations. Maydica. 2008;53(3):189.
- View Article
- Google Scholar
52. Singh S, Singh H, Singh N, Srivastava J. Heterosis in pea. INDIAN JOURNAL OF AGRICULTURAL SCIENCES. 1978;48(12):705–10.
- View Article
- Google Scholar
53. Fischer R, Rebetzke G. Indirect selection for potential yield in early-generation, spaced plantings of wheat and other small-grain cereals: a review. Crop and Pasture Science. 2018;69(5):439–59.
- View Article
- Google Scholar
54. Smith H, Gowen J. Heterosis. Iowa State College Press, Ames. 1952.
55. Toker C, Yadav SS, Solanki I. Mutation breeding. Lentil: Springer; 2007. p. 209–24.
56. Diondra W, Ivey S, Washington E, Woods S, Walker J, Krueger N, et al. Is there a correlation between plant height and yield in soybean. Reviews Biol Biotechnol. 2008;7(2):70–6.
- View Article
- Google Scholar
57. Wolko J, Dobrzycka A, Bocianowski J, Bartkowiak-Broda I. Estimation of heterosis for yield-related traits for single cross and three-way cross hybrids of oilseed rape (Brassica napus L.). Euphytica. 2019;215(10):1–17.
- View Article
- Google Scholar
58. van Hulten MH, Paulo M-J, Kruijer W, Blankestijn-De Vries H, Kemperman B, Becker FF, et al. Assessment of heterosis in two Arabidopsis thaliana common-reference mapping populations. PloS one. 2018;13(10):e0205564. pmid:30312352
- View Article
- PubMed/NCBI
- Google Scholar
59. Shi J, Li R, Zou J, Long Y, Meng J. A dynamic and complex network regulates the heterosis of yield-correlated traits in rapeseed (Brassica napus L.). PLoS One. 2011;6(7):e21645. pmid:21747942
- View Article
- PubMed/NCBI
- Google Scholar
60. Leilah A, Al-Khateeb S. Statistical analysis of wheat yield under drought conditions. Journal of Arid environments. 2005;61(3):483–96.
- View Article
- Google Scholar
61. Saed-Moucheshi A, Pessarakli M, Heidari B. Comparing relationships among yield and its related traits in mycorrhizal and nonmycorrhizal inoculated wheat cultivars under different water regimes using multivariate statistics. International Journal of Agronomy. 2013;2013.
- View Article
- Google Scholar
62. Wadan A, Abd E-A, Mohsen E. Analysing and modeling the relationship between yield and yield components in sunflower under different planting dates. World Journal of Agricultural Research and Food Safety. 2013;1:46–55.
- View Article
- Google Scholar
63. Tang L, Ma W, Noor MA, Li L, Hou H, Zhang X, et al. Density resistance evaluation of maize varieties through new “Density–Yield Model” and quantification of varietal response to gradual planting density pressure. Scientific reports. 2018;8(1):1–16.
- View Article
- Google Scholar
64. Tambal H, Erskine W, Baalbaki R, Zaiter H. Relationship of flower and pod numbers per inflorescence with seed yield in lentil. Experimental Agriculture. 2000;36(3):369–78.
- View Article
- Google Scholar
65. Biçer B, ŞAKAR D. Heritability and path analysis of some economical characteristics in lentil. Journal of Central European Agriculture. 2008;9(1):175–80.
- View Article
- Google Scholar
66. Moll R, Lonnquist J, Fortuno JV, Johnson E. The relationship of heterosis and genetic divergence in maize. Genetics. 1965;52(1):139. pmid:17248265
- View Article
- PubMed/NCBI
- Google Scholar
67. Barth S, Busimi A, Utz HF, Melchinger A. Heterosis for biomass yield and related traits in five hybrids of Arabidopsis thaliana L. Heynh. Heredity. 2003;91(1):36–42. pmid:12815451
- View Article
- PubMed/NCBI
- Google Scholar
68. Girke A, Schierholt A, Becker HC. Extending the rapeseed gene pool with resynthesized Brassica napus II: Heterosis. Theoretical and applied genetics. 2012;124(6):1017–26. pmid:22159759
- View Article
- PubMed/NCBI
- Google Scholar
69. Kawamura K, Kawanabe T, Shimizu M, Nagano AJ, Saeki N, Okazaki K, et al. Genetic distance of inbred lines of Chinese cabbage and its relationship to heterosis. Plant Gene. 2016;5:1–7.
- View Article
- Google Scholar

[ref1] 1. Adsule R, Lawande K, Kadam S. Pea. Handbook of world food legumes; nutritional chemistry, processing technology and utilization CRC Press, Inc, Boca Raton. 1989:215–51.

[ref2] 2. Kim H, Behall K, Conway J, editors. Short-term satiety and glycemic response after consumption of whole grains varying in the amount of β-glucan in overweight adults. North American Association for the Study of Obesity 2005 Annual Meeting Obesity Research; 2005.

[ref3] 3. Durazzo A, Turfani V, Azzini E, Maiani G, Carcea M. Phenols, lignans and antioxidant properties of legume and sweet chestnut flours. Food chemistry. 2013;140(4):666–71. pmid:23692751
View Article
PubMed/NCBI
Google Scholar

[4] View Article

[5] PubMed/NCBI

[6] Google Scholar

[ref4] 4. FAOSTAT. Crops. 2020.

[ref5] 5. Corporation GRD. DAV00154—Pulse Breeding Australia: Lentil Breeding 2021 [https://grdc.com.au/research/projects/project?id=4297.

[ref6] 6. Corporation GRD. Still loving those legumes—more adventures with lentils 2021 [https://grdc.com.au/resources-and-publications/grdc-update-papers/tab-content/grdc-update-papers/2018/02/still-loving-those-legumes.

[ref7] 7. Foyer CH, Lam H-M, Nguyen HT, Siddique KH, Varshney RK, Colmer TD, et al. Neglecting legumes has compromised human health and sustainable food production. Nature plants. 2016;2(8):1–10. pmid:28221372
View Article
PubMed/NCBI
Google Scholar

[11] View Article

[12] PubMed/NCBI

[13] Google Scholar

[ref8] 8. Lombardi M, Materne M, Cogan NO, Rodda M, Daetwyler HD, Slater AT, et al. Assessment of genetic variation within a global collection of lentil (Lens culinaris Medik.) cultivars and landraces using SNP markers. BMC genetics. 2014;15(1):1–10. pmid:25540077
View Article
PubMed/NCBI
Google Scholar

[15] View Article

[16] PubMed/NCBI

[17] Google Scholar

[ref9] 9. Danecek P, Auton A, Abecasis G, Albers CA, Banks E, DePristo MA, et al. 1000 Genomes Project Analysis Group. The variant call format and vcftools. Bioinformatics. 2011;27(15):2156–8.
View Article
Google Scholar

[19] View Article

[20] Google Scholar

[ref10] 10. Khazaei H, Caron CT, Fedoruk M, Diapari M, Vandenberg A, Coyne CJ, et al. Genetic diversity of cultivated lentil (Lens culinaris Medik.) and its relation to the World’s agro-ecological zones. Frontiers in Plant Science. 2016;7:1093. pmid:27507980
View Article
PubMed/NCBI
Google Scholar

[22] View Article

[23] PubMed/NCBI

[24] Google Scholar

[ref11] 11. Wright DM, Neupane S, Heidecker T, Haile TA, Chan C, Coyne CJ, et al. Understanding photothermal interactions will help expand production range and increase genetic diversity of lentil (Lens culinaris Medik.). Plants, People, Planet. 2021;3(2):171–81.
View Article
Google Scholar

[26] View Article

[27] Google Scholar

[ref12] 12. Singh M, Sharma SK, Singh B, Malhotra N, Chandora R, Sarker A, et al. Widening the genetic base of cultivated gene pool following introgression from wild Lens taxa. Plant Breeding. 2018;137(4):470–85.
View Article
Google Scholar

[29] View Article

[30] Google Scholar

[ref13] 13. Varshney RK, Kudapa H, Pazhamala L, Chitikineni A, Thudi M, Bohra A, et al. Translational genomics in agriculture: some examples in grain legumes. Critical Reviews in Plant Sciences. 2015;34(1–3):169–94.
View Article
Google Scholar

[32] View Article

[33] Google Scholar

[ref14] 14. Borlaug NE. Contributions of conventional plant breeding to food production. Science. 1983;219(4585):689–93. pmid:17814030
View Article
PubMed/NCBI
Google Scholar

[35] View Article

[36] PubMed/NCBI

[37] Google Scholar

[ref15] 15. Wang L, Wu LM, Greaves IK, Zhu A, Dennis ES, Peacock WJ. PIF4-controlled auxin pathway contributes to hybrid vigor in Arabidopsis thaliana. Proceedings of the National Academy of Sciences. 2017;114(17):E3555–E62. pmid:28396418
View Article
PubMed/NCBI
Google Scholar

[39] View Article

[40] PubMed/NCBI

[41] Google Scholar

[ref16] 16. Tewodros M, Mohammed A. Heterotic response in major cereals and vegetable crops. International Journal of Plant Breeding and Genetics. 2016;10(2):69–78.
View Article
Google Scholar

[43] View Article

[44] Google Scholar

[ref17] 17. Nienhuis J, Sills G. The potential of hybrid varieties in self-pollinating vegetables. Reproductive biology and plant breeding: Springer; 1992. p. 387–96.

[ref18] 18. Fu D, Xiao M, Hayward A, Fu Y, Liu G, Jiang G, et al. Utilization of crop heterosis: a review. Euphytica. 2014;197(2):161–73.
View Article
Google Scholar

[47] View Article

[48] Google Scholar

[ref19] 19. Wang L, Greaves IK, Groszmann M, Wu LM, Dennis ES, Peacock WJ. Hybrid mimics and hybrid vigor in Arabidopsis. Proceedings of the National Academy of Sciences. 2015;112(35):E4959–E67. pmid:26283378
View Article
PubMed/NCBI
Google Scholar

[50] View Article

[51] PubMed/NCBI

[52] Google Scholar

[ref20] 20. Fujimoto R, Uezono K, Ishikura S, Osabe K, Peacock WJ, Dennis ES. Recent research on the mechanism of heterosis is important for crop and vegetable breeding systems. Breeding science. 2018:17155. pmid:29875598
View Article
PubMed/NCBI
Google Scholar

[54] View Article

[55] PubMed/NCBI

[56] Google Scholar

[ref21] 21. Vasseur F, Fouqueau L, De Vienne D, Nidelet T, Violle C, Weigel D. Nonlinear phenotypic variation uncovers the emergence of heterosis in Arabidopsis thaliana. PLoS biology. 2019;17(4):e3000214. pmid:31017902
View Article
PubMed/NCBI
Google Scholar

[58] View Article

[59] PubMed/NCBI

[60] Google Scholar

[ref22] 22. Vacher M, Small I. Simulation of heterosis in a genome-scale metabolic network provides mechanistic explanations for increased biomass production rates in hybrid plants. NPJ systems biology and applications. 2019;5(1):1–10.
View Article
Google Scholar

[62] View Article

[63] Google Scholar

[ref23] 23. Ahmar S, Gill RA, Jung K-H, Faheem A, Qasim MU, Mubeen M, et al. Conventional and molecular techniques from simple breeding to speed breeding in crop plants: Recent advances and future outlook. International journal of molecular sciences. 2020;21(7):2590. pmid:32276445
View Article
PubMed/NCBI
Google Scholar

[65] View Article

[66] PubMed/NCBI

[67] Google Scholar

[ref24] 24. Crow JF. 90 years ago: the beginning of hybrid maize. Genetics. 1998;148(3):923–8. pmid:9539413
View Article
PubMed/NCBI
Google Scholar

[69] View Article

[70] PubMed/NCBI

[71] Google Scholar

[ref25] 25. Semel Y, Nissenbaum J, Menda N, Zinder M, Krieger U, Issman N, et al. Overdominant quantitative trait loci for yield and fitness in tomato. Proceedings of the National Academy of Sciences. 2006;103(35):12981–6. pmid:16938842
View Article
PubMed/NCBI
Google Scholar

[73] View Article

[74] PubMed/NCBI

[75] Google Scholar

[ref26] 26. Flint-Garcia SA, Buckler ES, Tiffin P, Ersoz E, Springer NM. Heterosis is prevalent for multiple traits in diverse maize germplasm. PloS one. 2009;4(10):e7433. pmid:19823591
View Article
PubMed/NCBI
Google Scholar

[77] View Article

[78] PubMed/NCBI

[79] Google Scholar

[ref27] 27. Birchler JA, Yao H, Chudalayandi S. Unraveling the genetic basis of hybrid vigor. Proceedings of the National Academy of Sciences. 2006;103(35):12957–8. pmid:16938847
View Article
PubMed/NCBI
Google Scholar

[81] View Article

[82] PubMed/NCBI

[83] Google Scholar

[ref28] 28. Lippman ZB, Zamir D. Heterosis: revisiting the magic. Trends in genetics. 2007;23(2):60–6. pmid:17188398
View Article
PubMed/NCBI
Google Scholar

[85] View Article

[86] PubMed/NCBI

[87] Google Scholar

[ref29] 29. Charlesworth D, Willis JH. The genetics of inbreeding depression. Nature reviews genetics. 2009;10(11):783–96. pmid:19834483
View Article
PubMed/NCBI
Google Scholar

[89] View Article

[90] PubMed/NCBI

[91] Google Scholar

[ref30] 30. Seymour DK, Chae E, Grimm DG, Pizarro CM, Habring-Müller A, Vasseur F, et al. Genetic architecture of nonadditive inheritance in Arabidopsis thaliana hybrids. Proceedings of the National Academy of Sciences. 2016;113(46):E7317–E26. pmid:27803326
View Article
PubMed/NCBI
Google Scholar

[93] View Article

[94] PubMed/NCBI

[95] Google Scholar

[ref31] 31. Busch R, Lucken K, Frohberg R. F1 Hybrids Versus Random F5 Line Performance and Estimates of Genetic Effects in Spring Wheat 1. Crop science. 1971;11(3):357–61.
View Article
Google Scholar

[97] View Article

[98] Google Scholar

[ref32] 32. Sarawat P, Stoddard F, Marshall D. Derivation of superior F 5 lines from heterotic hybrids in pea. Euphytica. 1993;73(3):265–72.
View Article
Google Scholar

[100] View Article

[101] Google Scholar

[ref33] 33. Williams W. The isolation of ‘pure lines’ from F 1 hybrids of tomato, and the problem of heterosis in inbreeding crop species. The Journal of Agricultural Science. 1959;53(3):347–53.
View Article
Google Scholar

[103] View Article

[104] Google Scholar

[ref34] 34. Gilmour A, Gogel B, Cullis B, Thompson R. ASReml user guide. Release; 2009.

[ref35] 35. Braich S, Baillie RC, Spangenberg GC, Cogan NO. A New and Improved Genome Sequence of Cannabis sativa. bioRxiv. 2020.
View Article
Google Scholar

[107] View Article

[108] Google Scholar

[ref36] 36. Sudheesh S, Verma P, Forster JW, Cogan NO, Kaur S. Generation and characterisation of a reference transcriptome for lentil (Lens culinaris Medik.). International journal of molecular sciences. 2016;17(11):1887.
View Article
Google Scholar

[110] View Article

[111] Google Scholar

[ref37] 37. Li H. Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM. arXiv preprint arXiv:13033997. 2013.

[ref38] 38. Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, et al. The sequence alignment/map format and SAMtools. Bioinformatics. 2009;25(16):2078–9. pmid:19505943
View Article
PubMed/NCBI
Google Scholar

[114] View Article

[115] PubMed/NCBI

[116] Google Scholar

[ref39] 39. Malmberg MM, Pembleton LW, Baillie RC, Drayton MC, Sudheesh S, Kaur S, et al. Genotyping-by-sequencing through transcriptomics: implementation in a range of crop species with varying reproductive habits and ploidy levels. Plant biotechnology journal. 2018;16(4):877–89. pmid:28913899
View Article
PubMed/NCBI
Google Scholar

[118] View Article

[119] PubMed/NCBI

[120] Google Scholar

[ref40] 40. Pembleton LW, Cogan NO, Forster JW. St AMPP: An R package for calculation of genetic differentiation and structure of mixed-ploidy level populations. Molecular ecology resources. 2013;13(5):946–52. pmid:23738873
View Article
PubMed/NCBI
Google Scholar

[122] View Article

[123] PubMed/NCBI

[124] Google Scholar

[ref41] 41. Perrier X. DARwin software. http://darwinciradfr/darwin. 2006.

[ref42] 42. Radoev M, Becker HC, Ecke W. Genetic analysis of heterosis for yield and yield components in rapeseed (Brassica napus L.) by quantitative trait locus mapping. Genetics. 2008;179(3):1547–58. pmid:18562665
View Article
PubMed/NCBI
Google Scholar

[127] View Article

[128] PubMed/NCBI

[129] Google Scholar

[ref43] 43. Brim CA, Cockerham CC. Inheritance of quantitative characters in soybeans. Crop Science. 1961;1:189–90.
View Article
Google Scholar

[131] View Article

[132] Google Scholar

[ref44] 44. Dinkins R, Keim K, Farno L, Edwards L. Expression of the narrow leaflet gene for yield and agronomic traits in soybean. Journal of Heredity. 2002;93(5):346–51.
View Article
Google Scholar

[134] View Article

[135] Google Scholar

[ref45] 45. Saeki N, Kawanabe T, Ying H, Shimizu M, Kojima M, Abe H, et al. Molecular and cellular characteristics of hybrid vigour in a commercial hybrid of Chinese cabbage. BMC plant biology. 2016;16(1):1–15.
View Article
Google Scholar

[137] View Article

[138] Google Scholar

[ref46] 46. Fisher RA. Theory of inbreeding: Oliver and Boyd, London; 1949.

[ref47] 47. Kearsey M, Pooni H. Genetical analysis of quantitative traits: Garland Science; 2020.

[ref48] 48. Sarawat P, Stoddard F, Marshall D, Ali S. Heterosis for yield and related characters in pea. Euphytica. 1994;80(1):39–48.
View Article
Google Scholar

[142] View Article

[143] Google Scholar

[ref49] 49. Burton JW, Brownie C. Heterosis and inbreeding depression in two soybean single crosses. Crop science. 2006;46(6):2643–8.
View Article
Google Scholar

[145] View Article

[146] Google Scholar

[ref50] 50. Lewers K, St. Martin S, Hedges B, Palmer R. Testcross evaluation of soybean germplasm. Crop science. 1998;38(5):1143–9.
View Article
Google Scholar

[148] View Article

[149] Google Scholar

[ref51] 51. Scheffler T, Hallauer AR, Lamkey KR, White PR. Estimates of heterosis and inbreeding depression for crosses of lowa maize populations. Maydica. 2008;53(3):189.
View Article
Google Scholar

[151] View Article

[152] Google Scholar

[ref52] 52. Singh S, Singh H, Singh N, Srivastava J. Heterosis in pea. INDIAN JOURNAL OF AGRICULTURAL SCIENCES. 1978;48(12):705–10.
View Article
Google Scholar

[154] View Article

[155] Google Scholar

[ref53] 53. Fischer R, Rebetzke G. Indirect selection for potential yield in early-generation, spaced plantings of wheat and other small-grain cereals: a review. Crop and Pasture Science. 2018;69(5):439–59.
View Article
Google Scholar

[157] View Article

[158] Google Scholar

[ref54] 54. Smith H, Gowen J. Heterosis. Iowa State College Press, Ames. 1952.

[ref55] 55. Toker C, Yadav SS, Solanki I. Mutation breeding. Lentil: Springer; 2007. p. 209–24.

[ref56] 56. Diondra W, Ivey S, Washington E, Woods S, Walker J, Krueger N, et al. Is there a correlation between plant height and yield in soybean. Reviews Biol Biotechnol. 2008;7(2):70–6.
View Article
Google Scholar

[162] View Article

[163] Google Scholar

[ref57] 57. Wolko J, Dobrzycka A, Bocianowski J, Bartkowiak-Broda I. Estimation of heterosis for yield-related traits for single cross and three-way cross hybrids of oilseed rape (Brassica napus L.). Euphytica. 2019;215(10):1–17.
View Article
Google Scholar

[165] View Article

[166] Google Scholar

[ref58] 58. van Hulten MH, Paulo M-J, Kruijer W, Blankestijn-De Vries H, Kemperman B, Becker FF, et al. Assessment of heterosis in two Arabidopsis thaliana common-reference mapping populations. PloS one. 2018;13(10):e0205564. pmid:30312352
View Article
PubMed/NCBI
Google Scholar

[168] View Article

[169] PubMed/NCBI

[170] Google Scholar

[ref59] 59. Shi J, Li R, Zou J, Long Y, Meng J. A dynamic and complex network regulates the heterosis of yield-correlated traits in rapeseed (Brassica napus L.). PLoS One. 2011;6(7):e21645. pmid:21747942
View Article
PubMed/NCBI
Google Scholar

[172] View Article

[173] PubMed/NCBI

[174] Google Scholar

[ref60] 60. Leilah A, Al-Khateeb S. Statistical analysis of wheat yield under drought conditions. Journal of Arid environments. 2005;61(3):483–96.
View Article
Google Scholar

[176] View Article

[177] Google Scholar

[ref61] 61. Saed-Moucheshi A, Pessarakli M, Heidari B. Comparing relationships among yield and its related traits in mycorrhizal and nonmycorrhizal inoculated wheat cultivars under different water regimes using multivariate statistics. International Journal of Agronomy. 2013;2013.
View Article
Google Scholar

[179] View Article

[180] Google Scholar

[ref62] 62. Wadan A, Abd E-A, Mohsen E. Analysing and modeling the relationship between yield and yield components in sunflower under different planting dates. World Journal of Agricultural Research and Food Safety. 2013;1:46–55.
View Article
Google Scholar

[182] View Article

[183] Google Scholar

[ref63] 63. Tang L, Ma W, Noor MA, Li L, Hou H, Zhang X, et al. Density resistance evaluation of maize varieties through new “Density–Yield Model” and quantification of varietal response to gradual planting density pressure. Scientific reports. 2018;8(1):1–16.
View Article
Google Scholar

[185] View Article

[186] Google Scholar

[ref64] 64. Tambal H, Erskine W, Baalbaki R, Zaiter H. Relationship of flower and pod numbers per inflorescence with seed yield in lentil. Experimental Agriculture. 2000;36(3):369–78.
View Article
Google Scholar

[188] View Article

[189] Google Scholar

[ref65] 65. Biçer B, ŞAKAR D. Heritability and path analysis of some economical characteristics in lentil. Journal of Central European Agriculture. 2008;9(1):175–80.
View Article
Google Scholar

[191] View Article

[192] Google Scholar

[ref66] 66. Moll R, Lonnquist J, Fortuno JV, Johnson E. The relationship of heterosis and genetic divergence in maize. Genetics. 1965;52(1):139. pmid:17248265
View Article
PubMed/NCBI
Google Scholar

[194] View Article

[195] PubMed/NCBI

[196] Google Scholar

[ref67] 67. Barth S, Busimi A, Utz HF, Melchinger A. Heterosis for biomass yield and related traits in five hybrids of Arabidopsis thaliana L. Heynh. Heredity. 2003;91(1):36–42. pmid:12815451
View Article
PubMed/NCBI
Google Scholar

[198] View Article

[199] PubMed/NCBI

[200] Google Scholar

[ref68] 68. Girke A, Schierholt A, Becker HC. Extending the rapeseed gene pool with resynthesized Brassica napus II: Heterosis. Theoretical and applied genetics. 2012;124(6):1017–26. pmid:22159759
View Article
PubMed/NCBI
Google Scholar

[202] View Article

[203] PubMed/NCBI

[204] Google Scholar

[ref69] 69. Kawamura K, Kawanabe T, Shimizu M, Nagano AJ, Saeki N, Okazaki K, et al. Genetic distance of inbred lines of Chinese cabbage and its relationship to heterosis. Plant Gene. 2016;5:1–7.
View Article
Google Scholar

[206] View Article

[207] Google Scholar

Figures

Abstract

Introduction

Material and methods

Plant material and generation advance

Growth conditions

Phenotypic evaluation

Phenotypic data analysis

Genotypic data of parental varieties

Results and discussion

Patterns of heterosis for seed number, yield, and biomass fresh weight

Patterns of heterosis for plant height

Correlation studies across generations

Genetic diversity analysis

Conclusion

Supporting information

S1 Fig. Phylogenetic tree describing the position of top ten heterotic crosses (highlighted in coloured boxes) along with other less heterotic and non-heterotic crosses.

S1 Table. Mean value for plant height, seed number, biomass fresh weight and yield of 47 selected genotypes used for producing heterotic F1’s.

S2 Table. Percent heterosis of F1 hybrids for seed number, yield, biomass fresh weight, and plant height.

S3 Table. Percentage of individuals performing better at F1, F2 and F3.

S4 Table. Correlation coefficients between growth and yield related parameters in different populations evaluated at F2 and F3 stage (*P≤0.5).

S5 Table. Multiple linear regression analysis across generations in test populations.

References