Variability and Association Study of  Yield and Nutritive Quality Traits in Pearl Millet (Pennisetum glaucum L. R Br)

The global health challenge of micronutrient deficiency, particularly in developing nations, has been extensively documented. Over two billion individuals worldwide, especially children and women of reproductive age, experienced the insufficient intake of iron (Fe) and zinc (Zn), from which iron deficiency being a leading cause of anemia (Boncompagni et al., 2018). This issue is most pronounced in resource-limited regions, especially among communities heavily dependent on staple cereals for their energy and nutritional requirements. Various approaches have been suggested to address micronutrient malnutrition, including medical supplementation, industrial fortification of food, dietary diversification and crop biofortification, which involves developing crop varieties with elevated micronutrient levels. Biofortification holds significant importance for populations who were reliant on agriculture, providing a cost-effective and sustainable strategy that prioritizes those most vulnerable to deficiencies (Finkelstein et al., 2017).
       
Pearl millet, a vital coarse grain cereal and forage crop grown in arid and semi-arid regions of the Indian subcontinent and Africa, plays a crucial role in ensuring dietary energy and nutritional security (Bhat et al., 2018; Srivastava et al., 2020; Aribam et al., 2024). Known for its resilience to drought conditions, pearl millet thrives in marginal areas where other major cereals often struggle to yield substantial crops (Kumawat et al., 2019; Rasitha et al., 2024). Pearl millet has a significant contribution to cereal consumption in specific regions of India, serving as an economical source of grain iron (Fe) and zinc (Zn) (Datir et al., 2018). Many of the studies have confirmed the effectiveness of iron-biofortified pearl millet in improving iron status, especially in children. Hence there is a need to alter the micronutrient composition for the benefit of end-users by exploring the presence of variation for grain Fe and Zn content among different pearl millet varieties. A wide range of genetic variability has been reported in grains specifically, Iniari germplasm accessions exhibited 51-121 mg/kg Fe and 46-87 mg/kg Zn, inbred parents and hybrids derived from diverse inbreds showed 30.3–102.0 mg/kg of Fe and 27.4-84.0 mg/kg of Zn and 25.8-80.0 mg/kg of Fe and 22.0-70 mg/kg of Zn respectively (Govindaraj et al., 2020). Selecting micronutrient-dense lines within existing breeding populations and varieties is a valuable approach for the success of crop improvement programs and also the understanding of the associations between yield attributes and the interrelation between nutritional quality traits is crucial for simultaneous crop improvement grain yield and nutritional traits or breed into on improved genetic background. The current investigation is concentrated on identifying micronutrient-dense pearl millet parental lines, marking a critical step in the development of micronutrient biofortified pearl millet hybrids or composites. Pooled analysis was applied since it has importance for a comprehensive evaluation of risk factors that are highly prevalent in the general population.

Experimental materials
 
The experimental materials consisted of 100 lines, including 42 B lines, 47 R lines and 11 land races, obtained from the Department of Millets at TNAU, Coimbatore and they were evaluated in randomized block design along with three checks, Dhanasakthi, 86M38 and COH10 Table 1. The chosen inbred lines differed exclusively for grain iron (Fe) and zinc (Zn) content. They also varied in performance for key agronomic traits, such as flowering date, plant height, panicle size, earhead weight and thousand-grain weight.


 
Field evaluation
 
A field trial was executed during three different seasons (Kharif’ 2022, Summer 2023 and Kharif 2023) at Millets Breeding Station, Department of Millets, TNAU, Coimbatore, utilizing a randomized block design with two replications. The climatic description of each environment has given Table 1. Each entry was planted in two rows, each measuring 4 meters in length, with an intra-row spacing of 15 cm and an inter-row spacing of 50 cm. The whole crop period, all required agronomic practice were followed.
 
Soil and grain micronutrient analysis
 
Soil micronutrient levels were evaluated by DTPA extractable method (Lindsay and Norwell, 1978) at the time of planting showing deficient level of Fe and Zn using Atomic Absorption Spectrometry (AAS). The scanning based method was utilized to estimate iron (Fe) and zinc (Zn) for estimation of micronutrients. The grain samples from respective entries, harvested at physiological maturity, underwent for micro nutrient analysis using Energy Dispersive X-rays Fluorescence (ED-XRF) at TRRI, Aduthurai.
 
Observations recorded
 
Data recorded for thirteen morphometric traits from five plants which were selected randomly at all the experimental material. Morphometric traits viz. days to 50% flowering (DFF),  days to maturity (DTM), plant height (PH), leaf sheath length (LSL), length of leaf (LL), leaf breadth (LB),  number of economic tillers per plant (NET), panicle length (PL), panicle girth (PG), thousand grain weight (TGW),  grain Fe content (Fe), grain Zn content (Zn) and grain yield/single plant (SPY) were noted.
 
Statistical analysis
 
Ahead of combined analysis, Bartlett ‘t’ test was performed to test the error variance for its  homogeneity  across the seasons. Analysis of variance across the seasons and the association of traits over the seasons was done by WINDOSTAT statistical software version 7.1. The genetic parameters such as PCV and GCV, heritability and GAM were computed by using following formula.

 

The phenotypic coefficient of variation facilitates the comparison of total variation within the experimental material across different characteristics and was computed using the following formula:
                                        

       

       
The genotypic coefficient of variation aids in assessing the extent of genotypic variation within the experimental material across different characteristics and was computed using the following formula:
 

  

       
Heritability is the inheritable portion of the phenotypic variance and is estimated by dividing the genotypic variance by the phenotypic variance (Lush, 1949), expressed as a percentage.
 

   

       
Genetic advance mean signifies the percentage improvement in the genotypic mean value of the selected progenies compared to their parental population after the process of selection. The calculation and classification were carried out following the method outlined by Johnson et al. (1955).
 

A significant portion of the universal population experiences the shortage of micronutrients such as with iron affecting 60-80% and zinc affecting thirty per cent of the population being particularly prevalent. These deficiencies have severe social impacts, leading to conditions such as anemia (due to iron deficiency) and stunted growth (due to zinc deficiency), which have devastating effects on countries (Yadav et al., 2023). Ensuring these essential micronutrients are available through staple diets is a reliable way to enhance global human health. To meet out this micronutrient needs, evolving nutrient-rich and agronomically superior cultivars is a crucial goal in plant breeding. The biofortification breeding focuses on creating lines that produce hybrids and commercial varieties with higher iron and zinc content than currently available. To meet these ambitious breeding targets, leveraging that genetic diversity found in available material is a rapid method for identifying mineral-rich accessions. In this regard, one hundred and three bajra genotypes evaluated for their variability for yield and nutritive quality traits and the results described below.
       
Combined analysis of variance Table 2 disclosed that existence of variation among 103 genotypes for studied morphometric and nutritive quality traits under except leaf breadth for environmental variance (Anuradha et al., 2020) over the seasons. The examination of mean values and various heritable variability parameters Table 3 indicated the significant variations for most of the studied traits and it could be utilized to develop hybrid varieties with simultaneous improvement in yield and former yield traits (Rajpoot et al., 2023).

​

​
 
Variability studies for yield and nutritive quality traits
 
Higher phenotypic coefficient of variation (PCV) was observed Table 3 and it indicating that the impact of environment on the observed traits. Notably, the yield traits including test weight and leaf sheath length exhibited higher PCV and GCV, suggesting significant variability amid the genotypes for these traits (Rajpoot et al., 2023). These high variability estimates indicated that presence of ample variation for the studied traits and it could be used for further selection and crop improvement. In the sense of yield higher PCV and moderate GCV observed across the environments and also individual environments Supplementary Table 1. In contrast to this, Anuradha et al., (2018) and Kumawat et al., (2019) reported higher variability. Considering nutritive traits higher variability observed for Fe content whereas moderate variability observed for Zn content.


       
Heritability of traits provides insights on the effectiveness of selection concerning inheritance. However, combining heritability with genetic advance (GAM) offers a more dependable estimate of selection. Most of the studied traits observed with high heritability (broad sense) and genetic advance as % of mean. These findings suggested that these characters governed by additive genes and direct selection strategies could effectively enhance these specific traits including grain micronutrients viz., Fe and Zn (Govindaraj et al., 2020).
       
The genotypes viz., PT 6476, PT 6583, PTB 7079, PTB 7095, PT 6679, PT 7054, PTB 7082, PT 7062 and ICMB 98222 and PMC 23B, PT 5315, PTB 7086, DMR 3/1, PT 6476, PTB 7098, PT 5188, ICMB 99222 and PT 6676 were exhibited the higher reports of >80 mg/kg and >50 mg/kg Fe and Zn contents respectively Table 4 with considerable agronomic superiority across the three seasons. Hence these lines serves as key genotypes and it can be utilized for the bio-fortification programs.


 
Association studies for yield and nutritive quality traits
 
Selecting superior genotypes by yield as such will not be effective (Bikash et al., 2013). Correlation of distinct traits provides the idea on being inherited together from generations. It helps in rambling selection for the multifaceted trait like grain yield by selection through other biometrical traits which are closely and positively associated. This association is by cause of pleiotropic gene action or linkage or more likely both (Dapke et al., 2014).
       
Correlations between yield characters and quality components were examined and the resultant correlation coefficients are presented in Table 5. The correlation coefficient due to genotypic was more than that phenotypic for the studied characters which denoted that existence of considerable inherent association of the traits. Yield per plant expressed the significant positive correlations with the traits panicle girth, thousand grain weight, leaf breadth, leaf length, leaf sheath length, length of single panicle and grain Fe. These results inferred that choice of these traits would lead to simultaneous improvement in yield/plant. Grain Fe exhibited an optimistic relation with yield, suggesting ample progress of both yield and micronutrient (Fe) content simultaneously. This study concludes that selecting for high yield and Fe will be achieved without adversely affecting grain yield whereas Chakraborti et al., (2010) reported a noteworthy negative relationship among Fe content and grain yield in maize. No correlation was exhibited among kernel iron and test weight, in contrast Pujar et al., 2020 found significant associations for both micronutrients.

​
       
The current investigation emphasizes a substantial positive correlation of micronutrients across the seasons and as well in individual seasons, indicating concurrent selection of these interrelated traits to enhance nutritional quality. Earlier reports in pearl millet by Govindaraj et al., (2013) and Rai et al., (2014) have consistently revealed the strong association between grain nutritive contents and these likely due to shared physiological mechanism or localized quantitative trait loci (QTLs) for iron and zinc accumulation, as reported in pearl millet (Kumar, 2011). This suggests that selecting for both micronutrients simultaneously could be effective. The consistent association across diverse environments allows to rank the genotypes for micronutrients which are expected to be less affected by environmental factors. Since both micronutrients are additively inhibited, it is crucial to incorporate these qualities into parents to produce hybrids with high Fe/Zn content and high yields.
       
Although negative correlation was identified between zinc and yield and Pujar et al., (2020) reported the same relationships in pearl millet. Comparable negative correlations between kernel zinc and yield have been observed in sorghum (Ashok Kumar  et al., 2009), wheat (Morgounov et al., 2007) and Maize (Chakraborti et al., 2010).  These negative linkages can be overcome through intentional assortment in the populations drawn from crosses between high-Fe and high-yielding parents.
       
Estimation of associations of traits alone not be sufficient because of reciprocated cancellation of related traits, hence there is a call for to estimate the path co-efficients, which takes into account, the cause of relationship in addition to the degree of relationship. The trait test weight (TGW) and grain Fe showed moderate optimistic direct effect on grain yield whereas grain Zn exhibited negative effect on yield Fig 1 and these traits were important as component traits for form selection indices in future (Kalagare et al., 2022).

The study showed considerable variability in critical nutrients and varied agronomic traits. Recovering different micronutrients and traits simultaneously is achievable if they are not negatively related with each other or with morphometric traits in tested lines, which are more probable to be use as future parents. The high estimates of hereditary parameters indicated that the minor influence of genotype-environment interactions, suggesting that simple selection-based breeding will be effective for genetic improvement. A strong correlation among grain micronutrient contents suggested that the opportunity of parallel improvement by conventional breeding. The positive relation among kernel iron and yield indicates that biofortification of Fe and Zn with competitive yields is feasible. The B lines such as PTB 7079, PTB 7095, PTB 7082, ICMB 98222, PMC 23B, PTB 7086, PTB 7098, ICMB 99222 along with the R lines such as PT 6476, PT 6583, PT 6679, PT 7054, PT 7062, PT 5315, DMR 3/1, PT 5188 and PT 6676 were identified with high level of micronutrients and the accessibility and use of these elite nutrient-dense breeding lines by breeding organizations could facilitate the introgression and mainstreaming of nutrition traits in commercial hybrid breeding in the near future.

All the authors declared that they have no conflicts of interest regarding the publication of this article. No funding or sponsorship influenced the design of the study, data collection, analysis, decision to publish or preparation of the manuscript.