1. Introduction
The lentil (
Lens culinaris Medikus ssp.
culinaris) is a self-pollinated legume species with the genome size of 4063 Mbp/1C [
1]. It is an ancient crop and its domestication dates back to the Neolithic Agricultural Revolution in the eastern Mediterranean during the 8th and 7th millennia BC [
2]. Thereafter, the crop disseminated to Central Asia, the Nile Valley and Europe during the period of Neolithic agriculture. The lentil was also part of the Harappan crop assemblage (2250 to 1750 BC) in the Indian subcontinent [
3]. The crop is cultivated for its protein-rich seeds and valuable straw in North America, South Asia, and the Mediterranean region. The global lentil production was 5.73 mtons during 2019 and Canada was the leading lentil producer followed by India [
4].
Phosphorus (P) is a non-renewable and essential nutrient with limited global reserves [
5]. The use of phosphate fertilizers has increased more than four times in the past five decades and is expected to reach 22–27 mton/year by 2050 [
6]. Being an essential element, P is involved in the synthesis of DNA, RNA and membrane proteins and lipids. It acts as a signaling molecule and is associated with cellular phosphorylation events [
7]. Plants absorb P from the soil mainly in the form of soluble inorganic P. In soil, P availability is low due to its fixation with calcium in acidic soils and iron/aluminum in alkaline soils [
8]. Therefore, to ensure high yield, farmers apply excessive P fertilizers. However, plants utilize only up to 30% of the applied P and the rest is fixed in the soil or may cause eutrophication [
9]. Improved P uptake and utilization is critical to reduce the cost of cultivation. The improvement of P use efficiency (PUE) becomes more important in legumes as legumes require more external P for growth and development as compared to other crops [
10].
Efforts have been made by breeders to improve P uptake and utilization efficiency in different crops by targeting specific traits [
11,
12]. Root traits such as root dry weight (RDW), total root length (TRL), total surface area (TSA), total root volume (TRV), total root tips (TRT) and fork number (FN) involved in P uptake are of higher significance in P-deficient soils [
13,
14,
15]. The phenotyping of black gram genotypes for root traits suggested that RDW is a potential trait for improving P uptake [
16], whereas TRL, TSA, TRV, TRT and FN and carboxylate exudation efficiency were important in green grams [
17,
18]. Shoot dry matter, total dry matter, and shoot P concentration were associated with high PUE in soybeans [
19], whereas long root hairs and a high shoot to root ratio were positively correlated to PUE in
Arabidopsis. Limited reports are available on the genotypic diversity for PUE in legumes [
20,
21]. The cultivated lentil,
Lens culinaris ssp.
Culinaris, is native to near East and Central Asia.
Lens orientalis is the probable progenitor [
22] of the lentil. Later, this genus was divided [
23] into four species including seven species/subspecies:
L. culinaris ssp.
culinaris, ssp.
orientalis (Boiss.) Ponert, ssp.
tomentosus (Ladizinsky) Fergusan, Maxted, van Slageren and Robertson, and ssp.
odemensis (Ladizinsky) Fergusan, Maxted, van Slageren and Robertson, and
L. ervoides (Brign.) Grande,
L. nigricans (M.Bieb.) Godron and
L. lamottei (Czefr). The different wild
Lens species are distributed in the Mediterranean region. The present study is the first attempt to characterize the
Lens genotypes’ germplasm under sufficient and low P conditions for the identification of PUE efficient lines and the identification of potential root and shoot traits associated with P uptake and utilization efficiency. The experiments were conducted with the objective of characterizing 85
Lens genotypes for the phenotypic variation of morphological root and shoot traits under sufficient and low P conditions, as well as to identify suitable genotypes for PupE and PUtE and phosphorus use efficient genotype(s).
3. Discussion
The present study was designed to evaluate a panel of eighty-five diverse
Lens genotypes for root and shoot traits at the seedling stage at contrasting P levels under controlled environments. The
Lens genotypes were characterized for root and shoot traits in response to different P levels for the identification of superior
Lens genotypes with significant P use. The most promising approach for developing a P use-efficient cultivar is by improving the key traits involved in the P uptake and utilization in the plant [
24,
25] Root architecture is instrumental to nutrient and water uptake, which has been poorly studied in lentils [
26]. Significant phenotypic and genetic variations, moderate to high levels of heritability, and significant correlations among different traits were recorded at different P levels (
Table 1,
Table 2 and
Table 3). This can be attributed to the diversity among and within the studied species for phosphorus uptake and utilization efficiency. These findings are in agreement with those reported in
Brassica [
27], common beans [
28], maize [
29] and green grams [
18] under nutrient stress [
18,
30,
31,
32]. Significant genotypic variation was reported for root and shoot traits in rice [
33,
34], wheat [
30], and mung beans [
18] and for P use traits in mung beans [
18]. Limited efforts have been made to study genetic diversity for phosphorus uptake and utilization efficiency in legumes.
A significant reduction in seedling growth traits (TDW, SDW, RDW) along with different root traits (PRL, TSA, ARD, TRV, TRT and RF) was recorded in response to P deficiency (
Table 1). PRL, TSA, RV and root branching are important components of root architecture and play important roles in determining the rate of nutrient uptake. Although genotype-dependent variation was evident for different traits, most of the genotypes registered a significant reduction in response to P deficiency. In mung beans, P-efficient genotypes exhibited a better RSA, TRV and carbon exudation efficiency [
17]. In rice, the RDW and RSR increased with a reduction in the SDW [
26]. Root architectural plasticity was correlated to the RDW, root length density, and lateral roots in response to low P [
35].
The Pearson correlation coefficient explained a highly significant correlation among most of the traits at the seedling stage under different P levels (
Table 3). We noted significant correlation between the TRL and TSV, TRV, TRT and RF in lentils. Similarly, the TRL was positively correlated with the TSA and TRV in maize [
36], with the TRT and RF in mung beans [
18], and the SDW, RDW, and TDW under both SP and LP conditions. In the present study, the TDW was positively correlated with the RDW and SDW and negatively correlated to the RSR under low P levels (
Table 3). The ARD showed a non-significant correlation to most of the traits in the present study as observed in chickpeas [
37]. In contrast to our findings, the ARD was highly significant and negatively correlated with the TRT and RF under LP in mung beans and can be used as an important trait to differentiate nutrient availability [
18]. A significant correlation among the root traits TRT, TRL, RSA, TRV, and ARD was reported in maize at the seedling stage [
38].
The PCA identified the most contributing traits responsible for total variation as the TRL, TSA, TRT, RF, SDW, RDW, TDW, PutiE and PupE under LP (
Figure 1). The present results are in accordance with previous studies, where P stress alters the TRL, SDW, TSA and TRT in mung beans [
18] and the TRL and TSA in common beans [
28]. We identified promising genotypes based on the TDW under P-deficient conditions (
Table 4). The selected superior genotypes belonged to the top 10% for SDW, but few differed in ranking for TRL, RDW and SDW under SP and LP, which can be due to variation in the root, shoot traits and genotype x treatment interaction [
39,
40,
41]. We observed that the SDW was positively correlated with the RDW and TDW under both P conditions. The RDW was positively correlated with the SDW under stress conditions in maize [
42], whereas the TRL and RDW were significantly correlated with root traits under P and N stress conditions in maize [
15,
43]. It was suggested that the TDW, RDW, and SDW are potential phenotypic traits for the selection of lentil genotypes under P efficiency. The evaluation of soybean germplasm revealed moderate to higher values of heritability for most of the root and shoot traits [
44,
45]. Traits such as the TRL, TSA, TRV, TRT, RDW, and SDW contributed most to the genetic diversity and significantly correlated with P accumulation under SP and LP conditions [
46,
47]. This result was similar to the results of previous studies, which revealed that the traits TRL, TSA, RDW, and SDW are the major traits for selection of genotypes in maize under different stress conditions [
36,
42].
In the present study, we observed that genotypes with a high TDW and SDW under low P were also promising for PupE and PutiE (
Table 4). PUtiE was also significantly correlated with TDW under different P regimes in rice [
48]. Though the genotypes which had higher TDW along with RDW showed better performance for physiological P use. It can be attributed to the correlation between RDW and different root traits, which resulted in efficient P uptake in these genotypes. The increased root length was responsible for higher P uptake from low P supply in barley and sugar beet roots [
49,
50] whereas the TRT, RSA, TRL and root hair length were important in P uptake in other studies [
51,
52]. In rice, the genotypic variation in P uptake was due to the RDW and RSA under P deficiency [
53]. The potential use of
L. ervoidis, L. nigricans and
L. culinaris ssp.
odemensis for drought tolerance [
54], and
L. culinaris ssp.
orientalis for cold tolerance [
55] and salinity tolerance [
56] has been investigated. The utility of wild
Lens species for phosphorus uptake and utilization requires an investigation with a large number of genotypes of each species/subspecies. In this study a limited number of genotypes of wild
Lens were investigated. The variations were recorded for different studied traits at LP and SP conditions in this study.
According to the classification approach reported earlier [
33,
57], lentil genotypes were classified into four categories based on P efficiency and responsiveness: ER, ENR, IR, and INR. Under both P circumstances, the most efficient genotypes, EC714238 and EC718339, were classified as ER. Under HP conditions, the genotypes EC 718,330 and EC 7188276, which were in the ER category under LP, were moved to the IR category. This emphasizes the significance of categorization at both high and low P levels. The genotypes classified as ER were well suited to soils with variable levels of P. The genotypes in the ENR group, on the other hand, could be successfully cultivated in P-depleted soils. The IR genotypes could be employed in a crossbreeding program to incorporate P-responsive characteristics. The genotypes of the INR category, on the other hand, have no part in the PUE enhancement program [
58]. This method of categorization allows for the identification of genotypes suitable for a variety of growing conditions and P levels [
59]. This strategy, however, is primarily dependent on the population mean. As a result, the difference between efficient and inefficient, responsive and nonresponsive kinds is relatively thin [
58]. The genotypes EC 718,276 and PL 97 in the LP condition and NDL-1 and EC 718,287 in the HP condition, for example, were at the borderline between efficient and inefficient groups. As a result, genotypes with a minor deviation from the population mean are difficult to categorize as efficient or inefficient, or responsive or nonresponsive. As a result, this technique is ineffective for investigating and categorizing genotypes on a wide scale [
57].
The genotypes were classified into nine groups (LDM-HP, LDM-MP, LDM-LP, MDM-HP, MDM-MP, MDM-LP, HDM-HP, HDM-MP, and HDM-LP) using a graph with the TDM and TPU on the x and y axes, respectively, in both HP and LP conditions [
60]. By generating nine groups, this categorization method may discern tiny changes between genotypes [
59]. This method, on the other hand, is more suited to categorizing genotypes at low P levels [
57]. The efficient genotype EC714238 with high TPU and TDM was clustered in HDM-MP under both HP and LP conditions in this investigation. HDM-HP genotypes are efficient in P uptake, and their use for biomass production suggests the genotype’s ability to produce greater biomass under a variety of P regimes [
30]. L 4717 was classified as MDM-MP in both LP and HP situations, whereas L 4650 was classified as LDM-MP in LP and MDM-MP in HP. Both P absorption and use in biomass production are inefficient in LDM-LP genotypes. The genotypes in the LDM-MP group are good at absorbing P but not so good at being used for biomass production.
As a result, the three-way categorization of genotypes (low, medium, and high) allows for the discovery of significant differences between the high and low groups while also providing the most room for medium type genotypes. Such differences explain genotype adaptation across a wide range of P regimes and provide the genetic foundation for PUE enhancement in breeding strategies. The categorization of genotypes using stress tolerance indices calculated from genotype dry mass under control and stress situations was reported earlier [
32,
61]. The P deficiency tolerance indices of all genotypes were computed using TDM under both HP and LP conditions in the current investigation. The susceptibility indices SSI, TI, and SI have a negative association with yield/biomass and are used to distinguish between stress-tolerant and susceptible genotypes [
62], whereas the tolerance indices MPI, GMPI, and STI have a positive connection with yield/biomass and can be used to select genotypes with high average yield/biomass and stress tolerance [
63].
The studied
Lens genotypes exhibited significant genetic variability and heritability for different root and shoot traits at two contrasting levels of phosphorus. Most of the recorded parameters revealed a remarkable reduction in P deficient conditions. Wild
Lens species exhibited low PupE and PutiE in comparison to cultivated species (
Figure 2L,M). Genotypes EC718309, EC718348, EC718332 and PL06 were better in P uptake as well as in P utilization compared to other genotypes. The identified genotypes can be utilized for the development of mapping population for the identification of QTLs responsible for P uptake and utilization efficiency. The characterization of the studied genotypes using different techniques identified the genotype EC714238 as having the highest STS score, indicating that it was highly phosphorus use efficient under LP conditions. The genotype can be utilized in breeding programs and in genetic studies.