Genetic Variability Assessment of Azadirachta indica A. Juss in Eastern India: Implications for Tree Improvement ^†

Abstract

Azadirachta indica was designated the “Tree of the 21st century” by the United Nations, as it is believed to be the largest natural depository of bioactive phytochemicals. This study investigates genetic variability among 152 Candidate Plus Trees (CPTs) of A. indica selected from three agro-climatic zones (ACZs) in eastern India: the Lower Gangetic Plains (ACZ III), Middle Gangetic Plains (ACZ IV), and the Eastern Plateau and Hills region (ACZ VII). Phenotypic characters, fruit and seed morphology, kernel oil content (KOC), and Azadirachtin concentration (AC) were assessed to characterize the genetic diversity. Significant variation was observed across all parameters among individual CPTs. Girth at breast height ranged from 0.9 to 2.8 m, tree height from 6 to 16 m, and crown volume from 146.95 to 2339.86 m³. Fruit length varied from 13.55 to 21.55 mm and seed length from 9.21 to 17.37 mm. KOC ranged from 36.51 to 58.86%, with a mean of 47.22% (±0.4), while AC showed extreme variability (19.46–1823.45 μg/g seed). KOC exhibited strong positive correlations with crown diameter (R = 0.57, p ≤ 0.001) and crown volume (R = 0.45, p ≤ 0.001). Interestingly, AC did not correlate significantly with any studied parameter. Analysis of variance revealed significant differences (p < 0.05) between ACZs, but only for some traits. All of the parameters demonstrated high heritability and moderate to high genetic advance. Cluster analysis using Ward’s minimum variance criterion based on Euclidean square (D²) distances performed in RStudio grouped the CPTs into five clusters as per pooled effects of all parameters. The highest inter-cluster distance was observed between Clusters III and V (7.703), indicating a potential for heterosis in hybridization between these groups. Each cluster contained CPTs from all three ACZs, suggesting uniformly distributed variation across the study area rather than zone-specific patterns. This study provides valuable insights for improvement programs of the species and emphasizes the need for further research, including progeny trials, to comprehensively understand the genetic variability of A. indica in eastern India.

1. Introduction

Azadirachta indica A. Juss (Neem) is a member of the Meliaceae family and indigenous to the Indian subcontinent. Neem is considered a very significant multipurpose tree with great economic potential due to its high medicinal and bio-pesticidal properties [1]. The term “Neem” is derived from the Sanskrit word “Nimba”, which signifies a healthy state of physical and mental balance unaltered by diseases [2]. Neem is considered the biggest natural reservoir of bioactive phytochemicals as it contains about 300 phytochemicals [3,4]. Traditional home remedies for a wide variety of illnesses involve almost every part of this species [5]. Neem is a great alternative to dangerous chemical pesticides because it breaks down quickly, exhibits no toxicity, and does not leave behind any residue on crops [6]. The application of Neem-derived products positively influences crop yield, improves soil health, and reduces production expenses in agriculture [7]. Furthermore, the species is very climate adaptable and survives in a variety of soil types. Its root system is remarkably capable of absorbing moisture and nutrients, even in highly leached soils [8].

Due to its significant medicinal and bio-pesticidal qualities, Neem has recently gained considerable global importance [6]. The United Nations has designated Neem as the “Tree of the 21st Century”, stating that this species has the capacity to address global challenges [9]. In India and other Asian countries, Neem flourishes in a wide variety of climatic conditions. It demonstrates an extensive range of diversity in growth parameters for both juvenile and mature trees across various agro-climatic zones (ACZs) [10]. Genetic variability studies for this species are needed to fully realize the species’ economic potential and provide insights for conservation measures. This is necessary because the species has great medicinal properties and has the potential to improve the socioeconomic situation of rural indigenous people. Despite execution of two major research projects on genetic improvement of Neem through the International Neem Network (coordinated by FAO, United Nations in 1994) and the National Neem Network (coordinated by the NOVOD board, Government of India in 1999), there are significant research gaps as they both failed to tap into the genetic variation in Neem [11]. The research areas of these two projects almost excluded the eastern Indian regions. Therefore, to aid its genetic improvement, this study has been taken up to explore the existing genetic variability of Neem in Eastern India.

2. Materials and Methods

2.1. Study Area

A total of 152 A. indica Candidate Plus Trees (CPTs) situated across three ACZs, viz., the Lower Gangetic Plains zone (ACZ III), Middle Gangetic Plains zone (ACZ IV) and Eastern Plateau and Hills region (ACZ VII) were selected for the study (Figure 1). Geographically, the study area consists of four states of eastern India, viz., West Bengal, Jharkhand, Bihar, and eastern Uttar Pradesh. The trees with better phenotypic characters, growth, and stoutness than the population mean were selected as CPTs. During the selection of CPTs, certain screenings and checks were implemented, such as the trees having a minimum girth at breast height (GBH) of 80 cm, crown diameter of 7 m, being disease, insect, and deformity free, and having a minimum age of 8 years. The agro-meteorological parameters and the number of CPTs chosen from each ACZ are listed in Table S1.

2.2. Data Collection of Phenotypic Traits of the CPTs

Several phenotypical data of the selected 152 CPTs were recorded in field, like tree height, crown length, crown percentage, crown diameter, crown volume, and GBH (Figure S1). Estimation of crown volume was carried out using the UrbanCrowns software [12]. Mature fruits were collected from the 152 CPTs. Fruit length, fruit width, and test weight of fruit (100 fruit weights) were recorded. The collected fruits were depulped and washed thoroughly to extract seeds. The seeds were sun-dried properly and then the seed length, seed width, and test weight of seed (100 seed weights) were recorded.

2.3. Estimation of Kernel Oil and Azadirachtin Content

From the collected Neem seeds, kernel oil extraction was performed using the modified Soxhlet Extraction Method [13,14,15]. Dried seeds were crushed to remove the husk and extract kernel. The kernel was pressed and crushed with a mortar–pestle to powder form. For each sample, 10 g of kernel powder was used to extract oil in a Soxhlet extractor with Finar Finsolv^® AR solvent for 6–8 h at 50 °C (Figure 2). After extraction, the kernel oil content was determined as

$$\mathrm{Kernel} \mathrm{oil} \mathrm{content} (\mathrm{Wt}\%)={\displaystyle \frac{\mathrm{W}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t} \mathrm{o}\mathrm{f} \mathrm{e}\mathrm{x}\mathrm{t}\mathrm{r}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{e}\mathrm{d} \mathrm{o}\mathrm{i}\mathrm{l}}{\mathrm{W}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t} \mathrm{o}\mathrm{f} \mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l} \mathrm{s}\mathrm{a}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{e}}}\times 100$$

(1)

Azadirachtin concentration (µg/g seed) of the collected Neem seeds was estimated using a Waters LC Module I quaternary automated liquid chromatograph equipped with an autoinjector, high-sensitivity tunable UV and photodiode array detectors, and a Novapak RP-18 column [10] from Amity University, Noida, India.

2.4. Statistical Analysis of Genetic Variability

Biometrical techniques were used to estimate the presence of variability among the 152 selected CPTs. Descriptive statistical analysis was performed upon studied characters to explore the maximum value, minimum value, mean, standard error (SE), standard deviation (SD), co-efficient of variation (CV), and inter-quartile range (IQR) for each character in IBM SPSS 25.0. Pearson’s correlation analysis and parallel plotting were carried out in Origin Pro v2022 to understand the underlying correlations and interactions among the studied characters. Analysis of variance (ANOVA) was performed to explore any significant difference in the variability of the characters across the three ACZs. Phenotypic and genotypic coefficients of variability (PCV and GCV), broad sense heritability (h²), and genetic advance (GA) at 5% selection intensity were analyzed in a General R-shiny based Analysis Platform Empowered by Statistics (GRAPES) v.1.0 [16]. To further understand the genetic divergence among the CPTs, agglomerative hierarchical clustering (AHC) was performed using Ward’s minimum variance criterion [17] in RStudio v.2023.06.0–421 based on Euclidean square (D²) distances [18] as per the pooled effects of all the parameters studied. Cluster means and average intra- and inter-cluster distances were also calculated while performing the analysis.

3. Results

3.1. Variability in Phenotypic and Biochemical Parameters

Large variation in terms of phenotypic traits studied among the 152 CPTs were observed across the three ACZs (Figure 3). Descriptive statistical analysis revealed a better insight into this variability (Table S2). An accession from Siwan, Bihar (AI-4-79), exhibited the highest GBH, whereas the maximum height and crown length were recorded in an accession from Ramgarh, Jharkhand (AI-7-9). However, the largest crown diameter was observed in an accession from Saraikela, Jharkhand (AI-7-14). An accession from Ghazipur, Uttar Pradesh (AI-4-108), had the highest crown percentage and fruit length. Similarly, fruit and seed characters along with biochemical parameters exhibited wide variation (Figure 4). The widest fruits were observed from an accession of Gumla, Jharkhand (AI-7-49). The highest fruit test weight and longest seeds were also recorded in Gumla (AI-7-4). Interestingly, the maximum seed width and test weight were found in accessions from Bihar (AI-4-82 and AI-4-90, respectively). Maximum oil content (58.86%) was yielded from the accession of Paschim Medinipur, West Bengal (AI-3-87). Nonetheless, the highest Azadirachtin content (1823.45 µg/g seed) was observed in the accession from Bokaro, Jharkhand (AI-7-38).

3.2. Correlation Analysis and ANOVA

Pearson’s correlation analysis (Figure 5) revealed that kernel oil content exhibited a strong positive correlation with crown diameter (R = 0.57, p ≤ 0.001) and crown volume (R = 0.45, p ≤ 0.001). In addition, GBH was weakly, yet positively, correlated with kernel oil content (R = 0.31, p ≤ 0.001). ANOVA performed for all the studied parameters revealed significant differences across the three ACZs only in GBH, crown percentage, all fruit parameters, and Azadirachtin content (Table S3). Other parameters varied widely among the individuals, though the variation was not significant (p > 0.05) among the ACZs.

3.3. Analysis of Variance Components

Different variance components like PCV, GCV, h², and GA for fruit and seed morphological characters along with biochemical traits of all 152 CPTs were calculated (

Table 1. Average intra- and inter-cluster distances (D² values) derived from AHC.

Parameters	PCV (%)	GCV (%)	Heritability	Genetic Advance (%)
Fruit length	9.703	9.395	0.937	18.739
Fruit width	11.664	11.401	0.955	22.955
Fruit test weight	25.579	25.467	0.991	52.235
Seed length	10.108	9.83	0.946	19.692
Seed width	11.384	11.118	0.954	22.366
Seed test weight	21.857	21.726	0.988	44.487
Kernel oil content	10.751	10.456	0.946	20.947
Azadirachtin concentration	72.046	71.986	0.998	148.17

). A high assessment of the GCV and PCV (>20%) were recorded for the test weight of fruits, seeds, and Azadirachtin concentration. On the other hand, kernel oil percentage, seed length along with width of fruits, and seeds demonstrated moderate variations (10–20%). However, only fruit length exhibited a low estimate of variation (<10%). Very high broad sense heritability (>60%) was exhibited by all the parameters. Except fruit and seed length, all other parameters exhibited high GA at 5% selection intensity (>20%).

3.4. Agglomerative Hierarchical Clustering (AHC)

The dendrogram based on Euclidean square (D²) distances [18] grouped 152 Neem CPTs into five clusters as per the pooled effects of all parameters studied (Figure 6). Average intra- and inter-cluster distances in five clusters of Neem CPTs indicated significant variation among the clusters obtained (

Table 2. Average intra- and inter-cluster distances (D² values) derived from AHC.

Clusters	I	II	III	IV	V
I	4.263	4.481	5.425	6.084	5.652
II		3.721	4.697	5.394	6.055
III			4.520	6.647	7.703
IV				4.488	5.619
V					4.586

). The highest intra-cluster distance was observed in Cluster V (4.586), whereas the highest inter-cluster distance was recorded between Cluster III and Cluster V (7.703). The lowest intra-cluster distance was observed in Cluster I (4.263), whereas the lowest inter-cluster distance was recorded between Cluster I and Cluster II (4.481). Upon further inspection of the clusters obtained, it was revealed that every cluster contained accession (CPTs) from all three ACZs. The mean cluster values of all the parameters are depicted in Table S4.

4. Discussion

The variability in phenotypic traits observed in this study across the three ACZs agrees with the findings of many previous reports where the researchers studied phenotypic variation among Neem trees across several provenances in India [10,19,20,21,22]. The variation recorded in the fruit and seed parameters also attested the findings of Kumaran et al. [23], where the authors reported significant variation between populations of Neem. They also reported seed length to be highly heritable and the test weight of seeds to be a robust selection index of high vigor of the tree.

In the correlogram, positive associations were noted among kernel oil content, crown diameter, crown volume, and GBH, implying that those characteristics may be regulated by closely associated genes. Our analysis demonstrated an absence of any correlation between Azadirachtin concentration and the examined parameters, suggesting that it may be governed by distinct genetic processes [24]. All parameters exhibited high heritability values. The test weight of the fruit, seed, and Azadirachtin content interestingly showed excellent heritability and high GA, while the length of the fruits, the seeds, and kernel oil content exhibited high heritability and moderate GA.

It is worth mentioning that as Azadirachtin concentration varied extremely in our population, it exhibited an astonishingly higher GA value. Therefore, long-term monitoring of this parameter with year-wise assessment is very much necessary for truly understanding its potential to be used as a selection criterion of superior genotypes. Sidhu et al. [1] also denoted that evaluating only Azadirachtin content is controversial for assessing genetic variability in Neem. A parallel plotting of Azadirachtin and kernel oil content of the selected 152 CPTs also revealed no significant correlation (Figure S2). Individuals with very high and very low Azadirachtin content were observed within a single agroclimatic zone, indicating it is not only highly affected by environmental factors but is also dependent on individual genetic makeup. The result aligns with the findings of Sidhu et al. [1] and Shirin et al. [24] implying the need to consider G × E interaction while studying variability in Neem.

The results of ANOVA are in corroboration with Kaushik et al. [10] and Tomar et al. [25], where the authors were unable to find significant variation in a number of parameters between different agro-ecological zones (AERs) of Gujarat despite the presence of large genetic variation in all parameters studied among individuals. It is important to note that the highest values of each studied parameter are not confined to any of the particular ACZs. They were recorded from accessions of different regions of varied states. Similar trends have been observed in a recent study by Garg et al. [26], where the authors found ample genetic variability among the Neem accessions from different AERs of southern and western India.

Furthermore, it can be implied from the results of AHC that hybridization between CPTs of Clusters III and V is expected to exhibit good heterosis. In addition, CPTs belonging to Clusters I and II are probably genetically closer, which can be determined through further progeny trials. AHC also validated the results of ANOVA carried out among the ACZs. It also shows that there was no directional selection pressure that may have caused any discriminate clustering [8,27]. In our study, parameters exhibiting high heritability and GA highlights the possibility of selective breeding for the improvement of Neem [26]. According to Sidhu et al. [1], such an assessment of genetic divergence should further be validated by performing a half-sib progeny trial of mother trees with desired traits.

5. Conclusions

Despite the execution of two major projects, a considerable research gap is still present in the field of research on genetic variability of A. indica. Our thorough study of A. indica across eastern India has uncovered valuable insights into its genetic landscape. Our analysis of 152 CPTs revealed remarkable genetic variability in their phenotypic characters, fruit and seed parameters, kernel oil content, and Azadirachtin content. We found it intriguing that significant variation between different ACZs was not observed for all the traits. It indicates a homogeneous distribution of CPTs amongst all ACZs and also that the CPTs are well adapted to different climatic conditions. We observed that trees with higher stoutness and spread (crown diameter, crown volume, and GBH) will usually yield higher kernel oil content. However, the Azadirachtin content was found highly variable and requires careful assessment due to possibilities of high G × E interaction. Looking ahead, we suggest further validation of these results by means of recurrent assessments and half-sib progeny trials of a chosen tree, preferably for Plus Trees (PTs). Additionally, the development of reliable biochemical and molecular markers to investigate genetic diversity in naturally occurring populations of Neem is essential for implying tree improvement methodologies. This would assist filling in any remaining knowledge gaps and help to enhance the future tree improvement programs of A. indica by fully exploring the genetic variability present among the CPTs in eastern India.