Exploring genetic diversity of yellow-berried night shade (Solanum virginianum L.) using genetic divergence and molecular markers

2.1 Experimental materials

Explorations were conducted across India to obtain kantakari (Solanum virginianum L.) genetic material. In total, germplasm was collected from 54 sites across six Indian states (Fig. 1). The species identity was verified with the assistance of the Botanical Survey of India, TNAU, Coimbatore. The seeds of the collected species were deposited in the National Bureau of Plant Genetic Resources (NBPGR) Gene bank for obtaining the Indigenous Collection (IC) Number, and the seeds were available in the Ramiah gene bank, TNAU, Coimbatore. Seeds were treated with 500 ppm GA₃ for 24h to promote germination as per the trail conducted by Boomiga et al., (2021a). Treated seeds were sown in a seed pan with coir pith and vermicompost (1:1) as the potting medium. To assess and select the most appropriate genotypes, experimental fields were established at the Department of Medicinal and Aromatic Plants, TNAU Botanical Garden, Coimbatore (11°00′50.8″N 76°55′52.7″E, 411 MSL). Ten plants per accession (45-day-old seedlings) were planted in a randomized complete block design (RCBD) having three replications on October 16, 2021. The total experimental area was 0.27 acre containing 11 blocks with 100 m². Each block consist of 15 ridges of 10 m length with 15 plants planted at the spacing of 60 × 60 cm. The genotypes were cultivated as per the cultural practises proposed and standardized by NMPB (2014).

Close

Fig. 1

Map showing the locations of the kantakari accessions.

Export to PPT

Phenotypic observations were recorded from five randomly selected plants from each accession across all replications based on the National Bureau of Plant Genetic Resources minimal descriptors for Agri-horticultural crops (Part IV) (Singh et al., 2003). Morphological characters: North– South spread of the plant (cm), East– West spread of the plant (cm), number of branches plant⁻¹, diameter of the stem (mm), leaf length and width (cm), length of the petiole (cm), length of the internodes (cm), and number of thorns on the upperand lower surface of leaf were recorded upon completion of the vegetative stage of the plant. Flowering traits: days to first flowering, number of flower clusters plant⁻¹ and days taken for 50 % flowering were observed at the full bloom stage. Yield characteristics: total number of berries plant⁻¹, diameter of a berry (mm), fresh and dry weight of single berry (g/plant), fresh and dry berry yield plant⁻¹ (g) were recorded at fruit maturity. Biochemical analyses were conducted as suggested by Rakesh et al.(2021) to assess the total phenol content (mg GAE/g) by the Folin-Ciocalteu method; total protein content (mg/g) by Lowry’s method, total flavonoid content (mg QE/g) by aluminium trichloride method; and total antioxidant content using DPPH assay (mg ASAE/DW). All these observations were analyzed using Mahalanobis’s (1936) D² statistics for assessing genetic divergence between the kantakari accessions.

Variations in leaf (upper and lower side), spine, flower, and berry colour were documented using the Royal Horticultural Society (RHS) Color Chart (Fig. 2).

Close

Fig. 2

Lane 1: Variation in leaf colour (upper side); Lane 2: Leaf colour (lower side); Lane 3:Flower colour; Lane 4: Dry berry colour; Lane 5: Thorn colour.

Export to PPT

2.2 Extraction of genomic DNA

Genomic DNA analysis was performed according to the protocol outlined by Doyle and Doyle (1990). On the day of extraction, fresh young leaves of S. virginianum L. were collected from each accession, washed with tap water, and wiped with 70 % (aq) ethanol. The cleaned leaf blades were then cut into small pieces and macerated in 1000 µL of slightly modified CTAB buffer with 10 µL ofmercaptoethanol. Ground samples were incubated in a water bath at 60 °C for 45 min. Subsequently, 600 µL of a phenol, chloroform, and isoamyl alcohol solution (25:24:1) was added, and the mixture was centrifuged in a refrigerated centrifuge at 12,000 rpm for 10 min. The supernatant was collected, to which 600 µL of chloroform and isoamyl alcohol mixture (24:1) was added and centrifuged again. Because of the high phenol concentration of S. virginianum L. leaves, this procedure was performed twice to remove the phenols. Finally, 50 µL of sodium acetate (7.5 M) was added to enhance nucleic acid precipitation, and an equal amount of frigid isopropanol was added. The samples were kept overnight at 4⁰C. The DNA pellets precipitated were subsequently rinsed with 70 % ethanol on the following day after the samples had been centrifuged at 10,000 rpm for 8 min. The pellets were dried until the alcohol Odor disappeared. The pellets were then reconstituted in TE buffer and kept at −20 ⁰C for storage. A nanodrop spectrophotometer (Jenway^TM Genova) was used to evaluate the quantity and purity of DNA samples at 260 and 280 nm. Using 0.8 % (w/v) agarose gel electrophoresis and 6 X DNA loading dye (Thermo Fisher Sci, Cat. No. R06111), the DNA was examined for quality.

2.3 RAPD amplification

Ten decamer oligonucleotides derived from Operon series (OP) RAPD primers (Eurofins Advinus Agrosciences Services Pvt. Ltd., Bengaluru) (Table 1) were used to evaluate genetic polymorphisms in S. virginianum L. Amplification was performed in a 10 µL reaction volume, with 2 µL of 100 ng DNA, 2 µL of primer and 6 µL of 2X mastermix (smartPRIME). In a Proflex Thermal Cycler, PCR reactions were performed with an initial denaturation at 94 °C for 5 min, 45 cycles at 94 °C for 1 min, 37 °C for 1 min, and 72 °C for 1 min, and a final extension at 72 °C for 5 min. The amplified products were separated using a 1.5 % agarose gel in 0.5X TBE buffer and then electrophoresed at 90 V for 3h. The Gel Documentation System was used to visualize the results using EtBr staining. Using GeneDirex Cat. No. DM0150 as a molecular standard, the 1-kb marker RTU was used to quantifythe amplified product size.

After gel electrophoresis, the amplified fragments were scored in a binary fashion, and a phylogenetic tree was created using the neighbour joining approach with 100 bootstraps in the DARWIN computer program (Perrier and Jacquemoud-Collet, 2006). The PIC values were calculated using the formula, $$\mathit{PIC}=1-\sum {\mathrm{f}\mathrm{i}}^2$$ where, “f” − represents the “ith” allele frequency (Botstein et al. 1980).

The formula for calculating resolving power was also as follows: $$MarkerIndex\left(MI\right)=PIC\times EMR$$ where, EMR is the multiple of number of polymorphic loci and fraction of polymorphic loci (Milbourne et al. 1997) $$\mathit{Rp}=\sum Ib(1-(2\times |0.5-\mathrm{p}|)$$ where, Ib (band informativeness) and 'p' is the number of individuals containing the band (Prevost and Wilkinson, 1999). Using AMOVA, which is a feature of the GenAlEx 6.5 Excel software suite, molecular variation within and across populations was analyzed (Peakall and Smouse, 2012).

3.1 Variation in phenotypic traits

Table 2 represents the assessment of variance results unequivocally demonstrated significant accession-level differences for each of the twenty-four characteristics.

The observation of qualitative traits also revealed variation within the accessions. Three leaf-colour classes were observed: 16 accessions showed dark green and strong yellowish green leaf colour whereas 43 showed dark yellowish green and moderate yellowish green colour and only one accession was categorized in the greyish olive green and moderate yellowish green group (Fig. 2; Lanes 1 and 2). In case of thorn colour, there were two types (Fig. 2; Lane 5). Three groups were observed in flower as well as berry colour (Fig. 2; Lanes 3 and 4).

3.2 Genetic divergence

Genetic diversity is one of the most significant features of plant breeding. Measures of genetic divergence can be used to optimise accession selection for breeding programmes by providing information on both the type and amount of diversity present within plant populations (Smith et al., 2015). The significance of parental accessions selected based on a set of critical qualities may outweigh those selected based on a single characteristic. In order to develop novel variability and suitable recombinants with improved yield and quality, genetically diverse parents must be included. Therefore, multivariate analysis using Mahalanobis D² statistics has been employed to quantitatively measure phenotypic divergence (Kovacic, 1994). Table 3 revealed that these 54 kantakari accessions clustered into six groups with appreciable diversity. Fresh single berry weight, days to 50 % flowering, berries plant⁻¹, flower clusters plant⁻¹ and dry berry yield plant⁻¹ were the most significant contributors to overall differentiation (Fig. 3).

Close

Fig. 3

Percent contribution of characters towards genetic divergence in kantakari. * FSBW- Fresh weight of a berry, PSEW −Plant spread (E-W), IL −Internode length, NTUL- Number of thorns on upper leaf surface, DRBW- Dry weight of single berry, SG- Stem girth, PSNS- Plant spread (N-S),NBr- Branches per plant, NYLL- Number of thorns on lower leaf surface, FSBY- Berry yield (fresh), LW- Leaf width, LL- Length of the leaf, BD-Diameter of the berry PL- Petiole length, DFF- First flowering (days),DRBY- Yield of berry (Dry), NFCB- Flower cluster per branch, NB- number of berries and DF- days to 50 % flowering.

Export to PPT

The inter-cluster values of D² analysis ranged from 1183.81 (between Cluster-V and VI) down to 674.46 (between Cluster-III and IV) whereas, the highest intra- cluster distance was observed in Cluster-IV (D²= 275.67) with nine accessions (Table 4). Considering that the greatest phenotypic distance between clusters VI and V, the accessions selected from these clusters for hybridization should yield broad variation in segregating generations, perhaps including transgressive segregates. Similarly, Boomiga et al. (2021b) classified Solanum surattense accessions into six groups based on quantitative traits. These findings are consistent with studies on other Solanaceous crops, including tomato (Meena and Bahadur, 2015), eggplant (Begum et al., 2013), and chili (Rahevar et al., 2021).

3.3 Quantitative trait cluster mean performance

In represent to Table 5 plant spread (N-S) showed the widest plants in cluster II (mean = 106.10 cm) whereas cluster VI (at 105.86 cm) recorded the widest plants for plant spread (E-W). In addition, cluster VI produced the most branches per plant (6.14). The highest number of flower clusters per branch was observed in clusters I and VI (15.20). The fewest thorns on the upper and lower sides of the leaf were also noted in cluster VI (11.90 and 12.20 respectively). Cluster II recorded the highest mean number of berries (85.52). The yield attributes: fresh (3.26 g) and dry (0.75 g) single berry weight, and fresh (238.09 g) and dry (74.07 g) berry yield per plant was highest in cluster V. Hence, the selection of parental lines for hybridization from diverse clusters results in greater variability. Dutta et al. (2009) have similarly observed similar findings in aubergine. Hybrids with desired segregants could potentially result from accessions originating from the most divergent clusters. These results are in line with the kantakari divergence studies of Boomiga et al. (2021) and research demonstrated substantial variation in the yield of dried berries/plant among the 34 traits analyzed. and that 49 accessions could be grouped into six groups using cluster analysis.

3.4 AMOVA

Data from Table 6, revealed that just 3 % of variation was identified among populations, whereas 97 % of the total geneticdifference was found within populations. This significant level of within-population variation may be indicative of elevated gene flow or the presence of allelic variation that was not resolved in the observed banding patterns. High levels of within-population variation allow populations to better adapt to changing environments (Danusevičius et al., 2024). Future studies should investigate these RAPD banding patterns to determine whether single bands are truly mono-allelic, and, if so, whether those alleles are identical by descent. The findings of this study echo earlier research conducted by Akaffou et al. (2020), wherein molecular variance analysis revealed 14 % of molecular variance among groups, with the remaining 86 % attributed to within-group diversity in brinjal. Additionally, Shimira et al. (2021) reported that within-population variation in scarlet eggplant accounted for 81 %, while among-population variation constituted 19 %. In the case of potato, analysis of molecular variance demonstrated 3 % variation among locations, with 97 % stemming from diversity within locations, and further among the elevations, 2 % variability coupled with 98 % attributable to diversity within elevation (Anoumaa et al., 2017).

3.5 Molecular characterization by RAPD primers

Ten RAPD primers (Table 7), producing repeatable, polymorphic banding patterns (Fig. 4), were used to characterize the 54 kantakari accessions. Out of total 106 scorable bands, 79 were polymorphic, average of 7.9 polymorphic bands per primer. Among the ten primers, six showed polymorphism levels above 80 %, whereas primers OPP 09 and OPR 02 were fully polymorphic. The marker index was determined for all markers, as it quantifies their discriminatory strength. A marker index value of 2.46 was determined for this set of RAPD primers, a value resulting from a higher multiplex ratio component (7.49), results consistent with those of previous authors (Ansari and Singh, 2013; Tiwari et al., 2009; Verma et al., 2012,).(See Fig. 5).

Close

Fig. 4

Dendrogram using neighbour-joining method, utilizing kantakari populations screened with RAPD markers. (The number following accession names denote individual genotypes with boot-trap values, expressed as a percentage of 100 replicates, indicated at the nodes.).

Export to PPT

Close

Fig. 5

Gel image of OPT −17 RAPD primer showing amplified bands of 54 kantakari accessions.

Export to PPT

PIC of a primer is a crucial component that shows how well it can distinguish between different species. An average PIC of 0.30 was observed for the ten RAPD primers, whereas the highest PIC was observed for primer OPR 02 (0.37) (Table 6). In addition, the resolving power of each marker was assessed, indicating the primer's ability to effectively differentiate between genotypes or individuals. Across the primers, the mean resolving power was 4.76 with the highest value (7.74) detected for primer OPT 17. A neighbour-joining algorithm was used to generate a cluster analysis for the RAPD markers (Fig. 4). Larger differences in bootstrap values, greater divergence in genotypes and greater stability within clusters. The tree produced three major groupings (but with very low bootstrap values 18 accessions form Cluster I with two sub-clusters (shown in pink and blue); 31 accessions form Cluster II also with two sub-clusters (in black and green), and 5 accessions form Cluster III (in red). In an earlier study involving 24 accessions of S. surattense from Yadav et al. (2013), accessions were divided into 4 groups using UPGMA dendrogram with a 40 % resemblance for RAPD and ISSR markers. In this study, it is noteworthy that the dendrogram did not unveil a discernible pattern of geographic grouping. Similar agro-climatic conditions and widespread gene flow may be the cause of these associations between accessions from various locales (Saengprajak and Saensouk, 2012). Parallel research outcomes were documented by prior investigations conducted by Mastan et al. (2012) for Jatropa and Sarwat et al. (2011) for Terminalia. Semagn (2002) explained that the low relationship between molecular markers and phenotypical traits is indeed relevant to our study. Molecular markers, such as RAPD, encompass a broader portion of the genome, spanning both coding and non-coding regions, unlike morphology. Furthermore, molecular markers are not subject to artificial selection as morphology is. Consequently, phenotypic and molecular data approaches might not always be tightly linked. Enhanced association between them might be achieved with the inclusion of additional morphological markers in the analysis or by utilizing a wider array of RAPD primer combinations.