The importance of LT in plant adaptation and crop production has been well documented. Due to the limited dimensions of LT, previous genetic studies on this characteristic in cereals has all been based on indirect estimations. In the study reported here, we demonstrated for the first time that targeting LT directly in the genetic studies is now feasible. By assessing the RIL population consisting of 201 lines, we did not only detect QTL for LT in barley but also showed that QTL detected for FLT are larger and more stable compared with those for other leaf characteristics including length, width, and area. Although with reduced magnitudes, QTL for LT with similar locations were also detected from measuring 2LL after anthesis as well as from measuring the 3rd leaves of developing plants. These results indicate that the thicknesses of different leaves in a plant are correlated, and it likely has a more simpler inherence than other leaf characteristics. The importance of FLT and 2LLT is shown by its strong correlation with HD, FTN, SRT, KL and KNPS. Taken advantage of the high-quality genome assemblies for both parents of the mapping population used in this study, we also identified candidate genes underlying the most significant QTL on chromosome 6H based on the orthologous analysis.
In addition to the major locus on chromosome 6H, several other loci detected from fully grown plants were also detected from measuring 3L of seedlings especially with the use of the BLUP values. However, the magnitudes of the loci detected from seedlings were all significantly smaller. Importantly, the strong correlations between LT and yield-related traits obtained from measuring leaves of fully-grown plants were not detected from measuring seedlings. One of the possible reasons for these differences could be caused by the likelihood that HD could have a larger effect on the 3L in developing seedlings compared with that on leaves of fully grown plants. However, it is difficult to understand why the magnitude and consistency of the loci detected from the two different leaves of fully-grown plants also differ so much. Our results show that, when possible, data from FL should be collected for LT.
In mapping loci for traits related to seedling vigour, Capo-chichi et al. (2021) detected multiple QTL for SLA on each of the seven chromosomes in barley. Of them, six were on chromosome 6H. It is likely that one of these six loci shares a similar location with the one on 6HL detected in this study. However, none of the six loci reported earlier comes close to the latter in regarding to either the magnitude or stability. Loci for SLA have been reported previously based on assessing either plants after anthesis (Yin et al. 1999a, b) or young seedlings (Elberse et al. 2004; Poorter et al. 2005). However, loci on chromosome 6H was not detected in any of these studies. The different results between the study reported here and those earlier ones could be due to direct vs indirect measurements as found in the study on desert-adapted tomato (Coneva et al. 2017). As only loci segregating in a population can be detected, another likely reason for the different results is due to the different materials used among these studies.
High quality genome assemblies are available for both parental genotypes of the mapping population used in this study (Liu et al. 2020), which made it easier to identify candidate genes targeting a given region based on orthologous analysis (Zhou et al. 2021). Based on such an analysis, three candidate genes were detected for the major locus on chromosome arm 6HL. One of these genes, HORVU6Hr1G057630, is orthologous to OSPRR1 in rice which is involved in tiller bud outgrowth (Strable 2020). The orthologs of this gene are involved in photoperiodic flowering response in barley and Arabidopsis (Matsushika et al. 2000; Pruneda-Paz et al. 2009; He et al. 2019). The second gene HORVU6Hr1G060990 is homologous with OsVPE3 in rice. It has been reported that suppression of this gene could decrease the leaf width and guard cell length (Lu et al. 2016). The ortholog for the third gene HORVU6Hr1G068370 is OsGRF4 in rice and it is a positive regulator of genes that promote cell proliferation (Hu et al. 2015; Sun et al. 2016) and activates transcription of expansin promoters in protoplasts leading to a potential function in cell expansion (Liebsch and Palatnik 2020). Orthologs of this gene have also been found to be involved in multiple development processes in various species (Liebsch and Palatnik 2020). All three genes contain non-synonymous variations in their exons between the two parental genotypes which lead to amino acid substitutions. They form the primary targets to identify the gene(s) underlying this major locus for LT.
Correlations among various traits are common in any plant species thus it is not surprising that strong correlations between LT and several other traits were detected. Covariance analysis has been widely used to estimate the effects of such interaction between traits, but the power of such statistical analysis can be limited. One of the examples is the locus conferring plant height on chromosome 3H in barley. This locus was also shown to confer crown rot resistance and the effect of the locus on crown rot was still highly significant when the effect of plant height was removed by a covariance analysis (Li et al. 2009). However, further analyses based on near isogenic lines (NILs) for various plant height genes showed that all of height genes affect CR severity significantly (Liu et al. 2010) and the attempt to exploiting the 3H locus in generating barley breeding lines suggest that the observed CR resistance at this locus was a by-product of plant height (Unpublished). NILs have been effectively used to study the effect of a given locus for different traits in various plant species (Liu et al. 2010; Yan et al. 2011; Ma et al. 2012; Habib et al. 2016; Gao et al. 2019; Chen et al. 2021). With the adoption of techniques in rapidly generating materials with high-level of homozygosity (Zheng et al. 2013; Liu et al. 2016; Yan et al., 2017; Wanga et al. 2021), generating NILs for a given locus in many plant species is not a time-consuming process anymore. The size and stability of the loci detected for LT in this study suggest that developing NILs for some of these loci can be straightforward. As only two isolines need to be compared, effects of a given LT locus in multiple genetic backgrounds can be conveniently and accurately assessed in different environments once a few sets of NILs become available.