Phenotypic assessment of the populations
Ear placement is a quantitative traits which mean that expression is caused, not only by genetic factors, but also influenced by environmental and genotype × environment interaction effects (Shahrokhi et al. 2013). Analysis of variances for all the observed traits in six generations i.e., P1, P2, F1, F2, BC1.1 and BC1.2 indicated that there are significant differences among generations (Table 1). Therefore, it was possible to conduct generation mean analysis.
Higher mean value of F1 generation than both the parents for plant height, ear height, number of leaves, leaf width, lower intermodal length, upper intermodal length and 100- grain weight while higher than the mid-parent value for the remaining traits under investigation illustrated the presence of partial dominance or over-dominance for all the traits under study. Likewise, higher mean value of F2 generation for plant height than both the parents indicating the presence of heterosis for plant height. However, mean value of F2 generation for ear height was less than parent AKON40A having high ear placement.
The results obtained revealed that the hybridization based breeding methods will help in obtaining improved plant and ear height, based on the parents employed for the cross. As, it was noticed that in other traits, the heterosis was even less than 1% indicating that these traits cannot be manipulated to much extent through breeding programs involving hybridization. Similar results were obtained by Saleem et al. (2002), Ji et al. (2006), Shahrokhi et al. (2013), Haq et al. (2013), Dorri et al. (2014) and Hussein et al. (2017) on heterosis studies for ear height and other related traits in maize.
Scaling test, Genetic components and Components of variance
The values were significant for scaling tests in all the traits under investigation, except upper intermodal length which illustrated that non-allelic interactions were present for all these traits. In case plant height and ear height, it was revealed from significant values of A, B and C scaling test that there is presence of additive × additive, additive × dominance and dominance × dominance interactions. As there were significant values of scaling tests for most of traits, hence types of epistasis and the nature of gene action was estimated on the basis of six parameter model to understand the expression of all the traits under study. Most of the agro-morphological traits are highly influenced by environment and it is tough to believe that no epistasis gene interaction is present for these traits (Sofi et al. 2006) which were in accordance with our results. Presence of epistasis and its significant role in the expression of heterosis has been well documented by several researchers (Lukens and Doebley 1999; Hinze et al. 2003).
Significant dominance gene effect [h], additive gene effect [d] and additive × additive interactions [i] has been observed for plant height, indicating that both additive and dominant gene action played an important in the inheritance of this trait. In ear height, additive gene effect [d], dominance gene effect [h], additive × additive interaction [i] and additive × dominance interactions [j] were significant revealing that both dominance and additive effects were important in the inheritance of the trait. Previously studies concluded that most of the variance in ear height was due to additive and dominance gene effects and epistasis was of little significance (Harville et al. 1978). Russell (1976) observed that additive gene effect is of great significance for ear height whereas; Guo et al. (1986) emphasized the presence of dominance gene effects for ear height. The significant epistatic effect contributes positively for the expression of heterosis. Recent studies suggest that dispersion of favourable desirable alleles coupled with complementary epistatic effects is the main components of heterosis in maize (Sofi et al. 2006).
Additive variance (D) was found to be higher than dominance variance (H) for most of the traits. For many of traits except plant height, ear height and 75% husk brown, F value was positive which revealed that dominant alleles were more in number as compared to recessive alleles in the parents, therefore indicated the prevalence of partial dominance gene action in the inheritance of the traits investigated. Environment variance has been observed for plant height and ear height which means that these traits were affected by environmental factors to some extent. The average degree of dominance was less than unity for most of the traits under investigation indicated that these traits were influenced by additive effects of some genes which are controlling the traits under study, which is also shown by low broad-sense heritability. The degree of dominance was highest for number of leaves which revealed that the trait is governed by highly dominant genes. Similar results were obtained by Dorri et al. (2014) who also illustrated the importance of additive gene action for plant height and ear height in maize.
Heritability and Gene number
Highest broad-sense heritability for 75% husk brown (0.64), while lowest broad sense heritability for number of leaves (0.09) have been observed which revealed the preponderance of dominance variance in governing the 75% husk brown. The narrow-sense heritability estimates for all the traits were high which indicated the high importance of additive effects rather than dominance effects in controlling the traits studied in present investigation. Previously, several studies observed higher broad sense heritability for ear height and plant height (Ghimire and Timsina 2015; Bartula et al. 2019) which indicates less influence of environment and high correlation of phenotype and breeding values. Thus, high scope of genetic improvement of these traits through selection could be possible.
For most of the traits, more than one gene was responsible for governing the trait which suggested the polygenic inheritance of these traits in maize. However, Milus and Line (1986) argued that the genes which control quantitative traits are more likely to be linked and therefore, can segregate as a group or effective factors. Hence, if this is occurring in present study, the formulae would underestimate the number of effective factors controlling the traits. Thus, the number of individual genes would have been higher.
The results obtained in present investigation were in agreement with earlier studies. Dorri et al. (2014) also revealed that dominance variance (H) was higher than additive variance (D) in ear height and plant height. He documented that over dominance and dominance effect had an important role in controlling no. of leaves, leaf area, 100 grain weight and days to anthesis in maize. Shahrokhi et al. (2013) revealed that epistasis have an important role in controlling ear height, days to silking, days to anthesis, 100 kernel weight as well as physiological maturity in maize. He suggested the utilization of hybrid breeding is effective as the traits are mainly controlled by additive and additive × additive gene action and dominance gene effect together with higher broad-sense heritability of the agronomic traits. In case of plant height, the presence of additive and dominance gene action was in accordance with those of Haq et al. (2010), Kumar et al. (2005) and Singh & Roy (2007). Haq et al. (2010) also reported that both additive and dominant gene effects are effective in ear height. However, over dominance observed for ear height, on the basis of (H/D)1/2 ratio, was not in agreement with present study which may be due to variation in the genetic materials and environmental conditions to which genotypes were subjected. Paul and Debanth (1999) also suggested the presence of both additive and non-additive gene action for plant height, ear height and days to silking.
Construction of linkage map and gene/QTL analysis
Nine primers were used for constructing linkage map in F2 population. Three linkage groups were formed in which LG1 spanned 26.9 cM on chromosome 3 while, LG2 and LG3 spanned 42.1 and 6.7 cM on chromosome 6 and 7, respectively. Gene controlling the trait ear height was mapped between the SSR primers umc1979 and umc2317 on chromosome 6 having LOD value of ~3 and the phenotypic variance (R2) of 8.5. The results obtained were in agreement with Yang et al. (2008) who constructed linkage map using 150 SSR markers for the maize hybrid of cross between Ye478 and Dan340. They observed that mean contribution of 10.2% and 22.8 % of ear position at the interval phi029-umc1102 on chromosome 3 and phi109188-bnlg1118 on chromosome 5. The main QTL for plant height and ear height were both found at origin of bin 5.05-5.07 on chromosome 5. Sibov et al. (2003) used F2:3 population (with 400 families) and detected five QTL for ear height which described 20.91% of the phenotypic variance. These six QTL are present on Chr.1 (bin 1.05), Chr.2 (bin 2.09), Chr.3 (bin 3.07), Chr.5 (bin 5.05) Chr.6 (bin 6.04), and Chr.8 (bin 8.06) (Li et al. 2014). Lima et al. (2006) used maize inbred lines which were L- 02-03D and L-20-01F as parental lines and found 9 QTL for ear height; these were present on chromosome numbers 2, 3, 4, 7, 9 and 10. Li et al. (2019) found ZmRPH1, a novel MAP, which is responsible for regulating the cell elongation in maize. They found that ZmRPH1 when overexpresses lowered plant and ear height by reduction of the length of internodes and thus increased the lodging resistance, also without reduction of the maize yield. Zhang et al. (2006) found 14 QTLs using CIM method: five for ear height and nine for plant height. They revealed that the QTLs identified showed dominance or partial dominance gene action and were identified to chromosome 2,3,4,8 and 9. Lu et al. (2020) used SNP markers to identify the QTLs. They identified that out of 25 SNPs used for the mapping 15 SNPs were closely linked to the height and these were present on chromosome numbers 1, 2, 4, 6, 7, 8, 9, and 10, which explained 30% phenotypic variability for the trait. Veldboom et al. (1994) found five genomic regions significantly related to ear height i.e., 1L, 3S, 5L, 6L, 7L and collectively these explained 61.8% of the phenotypic variation.
However, the results obtained were not in agreement with some of the earlier studies which may be due to genotypic and environmental differences. Li et al. (2012) found fifty-five QTLs with five QTLs identified on chromosomes 1 and 8. Li et al. (2014) detected one QTL for ear height on chromosome I (Qeh) which was flanked by phi308707 and umc1358. The QTL explained the phenotypic variance of 9.55%. They also observed that due to the additive effect, the QTL was able to increase ear placement by 4.86 cm. On the basis of phenotypic values of the parental lines as well as the additive effect values of the QTL obtained, they determined that the QTL which was found in their study was from Mo17 (lower ear height parent). Choi et al. (2013) found one QTL related to EH (qEH10) which was identified on chromosome 10 and explained 18.09% of the phenotypic variance.
More than 283 and 46 QTLs related to plant height and ear height, respectively has been published till now in MaizeGDB (http://www.maize gdb.org/) database but Most of the QTLs were not up to the marks in terms of preciseness and accuracy (Fei et al. 2022). Previously, 30 QTLs for ear height and 25 QTLs for ear height were mapped using doubled haploid mapping population (Ertiro et al. 2020). Likewise, Wang et al. (2018) developed RIL populations to identify QTLs for plant height and ear height which were distributed on chromosomes 1, 3, 4, 5, 7, 8, and 10. Earlier, researchers concluded the correlation between plant height and ear height (Li et al. 2013) which is in the accordance with our results. Several genes governing the plant height and ear height in maize have been cloned previously, such as dwaft3 on chromosome 9 related to gibberellin synthesis (Teng et al. 2013), brachytic2 on chromosome 1 responsible for polar auxin transport (Xing et al. 2015), ZmGA3ox2 on chromosome 3 involved in gibberellin signal transduction pathways (Lawit et al. 2010). More than 40 genes involving in different biosynthetic pathways have been reported (Peiffer et al. 2014), but still the underlying mechanism of the relevant trait remains unclear. Molecular mapping of major genes/QTLs might help to better understand the mechanism of plant height and ear height development in maize.