ISSR-Based Genetic Diversity Assessment of Genus Jasminum L. (Oleaceae) from Pakistan

The genus Jasminum L., of the family Oleaceae, includes many species occurring in the wild, or cultivated worldwide. A preliminary investigation based on inter-simple sequence repeats (ISSR) was performed to assess the genetic diversity among 28 accessions, representing nine species of Jasminum from various regions, representing a range of altitudes in Pakistan. A total of 21 ISSR primers were used, which produced 570 amplified bands of different sizes, with a mean polymorphic band percentage of 98.26%. The maximum resolving power, polymorphism information content, and index values of the ISSR markers recorded for primers 6, 16, and 19 were 0.40, 12.32, and 24.21, respectively. Based on the data of the ISSR markers, the resulting UPGMA dendrogram with the Jaccard coefficient divided the 28 accessions into two main clades. At the species level, the highest values for Shannon’s information index, polymorphism percentage, effective allele number, Nei’s genetic variations, and genetic unbiased diversity were found in Jasminum sambac L. and J. humile L., while the lowest were observed in J. mesnyi Hance and J. nitidum Skan. Based on Nei’s unbiased genetic identity pairwise population matrix, the maximum identity (0.804) was observed between J. elongatum Willd and J. multiflorum (Burm. f.) Andrews, and the lowest (0.566) between J. nitidum Skan. and J. azoricum L. Molecular variance analysis displayed a genetic variation of 79% among the nine populations. The study was aimed to established genetic diversity in Jasminum species using ISSR markers. With the help of this technique, we were able to establish immense intra- and interspecific diversity across the Jasminum species.


Introduction
The Oleaceae family comprises 28 genera with~900 species, wherein the genus Jasminum L. alone comprises~200 species [1]. Jasmine species grow in the wild in most of the world's tropical areas, but especially in South East Asia. Southwestern and Southeast Asia are at the center of jasmine diversity, while a few wild species have also been documented

Results
In this study, 21 ISSR primers were used, which produced a total of 570 amplified fragments of different sizes for 28 Jasminum accessions, with an average of 27.14 bands per primer. The size of the amplified fragments was 90 to 2200 bp, as shown in Table 2. The ISSR pattern achieved with primers 15, 18, and 21 is illustrated in Figure 1A-C, as an example. The maximum number of fragments (39 bands) was obtained with primer 16, whereas the minimum number was produced by primers 9 and 10 (18 bands each). Of the 570 fragments, 562 were polymorphic, with a mean polymorphic percentage of 98.26%. The number of polymorphic bands showed variation, from 16 (primer 10) to 38 (primers 16 and 20). The maximum resolving power (Rp) was 24.21, which was exhibited by primer 19, while primer 3 displayed a lower Rp value (13.00). The PIC was higher for primer 6 (0.40) and lower for primer 10 (0.28). The marker index (MI) fluctuated from 4.01 for primer 10 to 12.32 for primer 16, as shown in Table 2.
The ISSR clustering cophenetic coefficient with the Jaccard similarity matrix was 0.992. Consequently, to understand the relationships among species, a similarity matrix was constructed using the 0/1 matrix. The method of the unweighted pair group with a UPGMA arithmetic mean dendrogram based on the Jaccard coefficients separated the 28 accessions into two major clusters, as shown in Figure 2. The first cluster (C1) corresponds to Jasminum species with yellow flowers, and the second major cluster (C2) comprises Jasminum species with white flowers. C1 was further divided into two subclusters: subcluster a, containing three accessions of J. humile, and subcluster b, containing two accessions of J. mesnyi. C2, comprising white jasmines, was further divided into four subclusters (subcluster c, with nine accessions of J. sambac; subcluster d, with four accessions of J. polyanthum and J. azoricum; subcluster e, with six accessions of J. nitidum, J. multiflorum, and J. elongatum; and subcluster f, with four accessions of J. grandiflorum) with 25-27% similarity among all four subclusters of white jasmines. Among these subclusters, subcluster c was sister to subcluster d, whereas subcluster e was sister to subcluster f. Subcluster c was further clustered into three groups of J. sambac: single flower (SF), semi-double flower (SDF), and double flower (DF). The second subcluster of white jasmines, subcluster d, was divided into separate groups each for J. polyanthum and J. azoricum. The accessions of J. sambac and J. humile showed intraspecific variations and were separated by a considerable distance, whereas the accessions of all the remaining species were not detached significantly from each other within their species. These intraspecific variants showed more than 90% genetic similarity.    In addition, using Gower's coefficient, PCoA was performed based on the genetic distance matrix to better visualize the genetic diversity of the studied Jasminum species. The three principal axes showed 27.02%, 18.43%, and 16.26% of the genetic variance, with a total of 61.71%. The PCoA biplot ( Figure 3A) separated three subclusters, c, d, and e. Similarly, the PCoA biplot ( Figure 3B) produced clearly separated major clusters, comprising yellow jasmines (subclusters a and b) and a white-flowered subcluster f. Therefore, by using three coordinates in this way, we were able to place all accessions into five distinct groups: four subclusters of white jasmines and one cluster of yellow jasmines (encircled).  Population analysis showed diversity a bac. The other species have a small populati ber of species defined in this study due to ne accessions, with the exception of a few tha analysis was performed due to the wider d pulation analysis showed diversity at the species level in only J. humile and J. samother species have a small population size, and there was no increase in the numpecies defined in this study due to negligible morphological differences among the ns, with the exception of a few that had considerable variation. The population , blue), J. multiflorum ( , blue), J. mesnyi ( , orange) and J. humile ( , yellow). Population analysis showed diversity at the species level in only J. humile and J. sambac. The other species have a small population size, and there was no increase in the number of species defined in this study due to negligible morphological differences among the accessions, with the exception of a few that had considerable variation. The population analysis was performed due to the wider diversity in the floral morphologies of J. humile and J. sambac. The data regarding these species demonstrate the greatest intraspecific diversity of the studied species based on the highest values for polymorphic percentage, number of effective alleles, Nei's genetic diversity, Nei's unbiased diversity, and Shannon's information index. The lowest values for the above-described population parameters were identified for both J. mesnyi and J. nitidum (Table 3). Na (different allele number); Ne (effective allele number) = 1/(p 2 + q 2 ); I (Shannon's information index) = −1 × (p × Ln (p) + q × Ln(q)); He (heterozygosity expected) = 2pq; uHe (expected heterozygosity unbiased) = (2N/(2N − 1))He where, by assuming diploid binary data and Hardy-Weinberg equilibrium, q = (1 − Band Freq.) 0.5 and p = 1 − q; P (percentage of polymorphic loci).
The values of PhiPT were statistically significant (p = 0.001), indicating additional association based on interspecies variation. The genetic variation in the nine populations recorded was 79% (PhiPT = 0.787), as shown in Table 5. Probability, P (rand >= data), for PhiPT is on the basis of the standard permutation across the complete data set; PhiPT = AP/(WP + AP) = AP/TOT, where WP is the estimated variance within populations and AP is the estimated variance among populations.

Discussion
Given that Jasminum is a large and diverse genus in the Oleaceae family, the delimitation of its diversity and species must be carried out carefully. In addition, the usefulness of taxonomy-based classification into different sections needs to be determined. The wide geographic distribution of the genus produces continuous variations in its characteristics, which creates difficulties in the delimitation of genotypes in terms of their intraspecific and interspecific boundaries. Jasminum is divided into five sections, Jasminum, Unifoliolata, Trifoliolata, Primulina, and Alternifolia, based on flower color, leaf shape, and leaf arrangement. Representatives from each section were included in this study to identify their relationships. A diversity analysis of Jasminum was also performed on the collected variants using ISSR markers, which showed the intraspecific and interspecific variations and the relationship among the sections. Many studies have validated the use of ISSR markers for genomic characterization in the Oleaceae family [23][24][25][26] and many other plants, such as Batrachium [27], Pistacia [28], and Aerva [29].
In this study, we found higher polymorphism levels in the investigated Jasminum accessions, with values of 88.89-100% and a mean of 98.26%, as shown in Table 2. Similarly, Ghehsareh et al. [20] investigated 53 Jasminum accessions from Iran using 21 ISSR markers and evaluated their genetic diversity with high polymorphism (90.64%). Yohanan et al. [21] used 10 primers for genetic diversity evaluation within 40 accessions of 23 Jasminum species from India, with 100% polymorphism [21]. In addition, Nirmala et al. [18] assessed the genetic similarities among 48 accessions of 26 Jasminum species using 10 AFLP markers, of which four displayed 90.5% polymorphism. Qiu et al. [19] used 10 ISSR primers and reported genetic diversity among 30 accessions of J. sambac. Li and Zhang [22] used microsatellite markers to analyze J. sambac wild germplasm recourses, where six displayed polymorphisms and seven displayed fixed heterozygosity with two alleles. Phithaksilp et al. [17] validated the phylogenetic relationship between 30 genotypes of Jasminum with 10 RAPD markers with an overall polymorphism of 71.43%. These studies have shown a high degree of polymorphism with higher levels of genetic diversity in Jasminum by using different molecular markers. Furthermore, Raja [30] detected somaclonal variation in J. auriculatum using RAPD. Zietkiewicz et al. [31] established ISSR analysis to study diversity among plant species. This method overcomes some of the limitations faced when using other DNA markers. ISSR markers can provide more reliable and reproducible results for polymorphism due to their higher primer lengths and annealing temperatures [32]. The primers used in our study were previously used in other genera of the Oleaceae family, including Syringa [33] and Olea [26], for which high polymorphism percentages (92.37% and 93.42%) were found.
The values for PIC of the markers were recorded in the range of 0.31-0.40, as shown in Table 2. The PIC depends on the detectable allele number, and its frequency distribution is equal to the gene diversity. The markers that were equally distributed in the population had a higher PIC, which remained up to 0.5 in the dominant markers. The EMR was 14.22-38.00 among the 28 Jasminum accessions. Ghehsareh et al. [20] also found EMR values of 25.93-48.08 using the same primers in Jasminum accessions from Iran. The higher EMRs show the effectiveness of the marker system for diversity-focused studies [34,35]. The values for marker index (MI) (4.06-12.32) are in accordance with the outcomes of Ghehsareh et al. [20]. The MI is used to calculate the complete effectiveness of marker systems, and a higher value is always associated with better effectiveness of the procedure [34][35][36]. The Rp observed ranged between 13.00 and 24.21. A primer's ability to differentiate among higher genotype numbers can be represented by the Rp [37,38]. According to Prevost and Wilkinson [36], the Rp is a tool to measure a primer's capacity to differentiate among accessions. Primers with higher Rp values normally show improved performance in differentiating between genotypes.
The UPGMA dendrogram based on the Jaccard coefficients distributed 28 accessions into two main clusters, as shown in Figure 2. The species placed in the first cluster had yellow flowers with no or less fragrance, while the species in the second cluster had fragrant white flowers. The jasmines with yellow flowers belong to section Alternifolia, having alternate leaves, and section Primulina, having opposite leaves. The combination of both sections (Alternifolia and Primulina) showed their close relationship as compared to the white-flower sections (Unifoliolata, Trifoliolata, and Jasminum), a finding, which was also reported in the molecular phylogenetic investigations of the Oleaceae family by Wallander and Albert [39] and Dupin et al. [40]. This combination might be due to the use of a single species from each section of jasmines with yellow flowers, which may not clearly resolve their relationship. Nevertheless, it is worth mentioning that this section has less similarity (33%) with the section Alternifolia because of its opposite leaves, and is similar to jasmines with white flowers. Therefore, section Primulina could be predicted to be the point of transition from yellow to white jasmines. This state of transition is also evident from the study of Lee et al. [41], who found evidence of several gene relocation, inversion, deletion, and duplication events in Jasminum and Menodora and inferred their evolutionary relationship. Rower [42] studied the fruit development stages in J. mesnyi and Menodora and found that the discontinuity between them is not greater than that between the other sections of Jasminum. In another study, Rower [43] found more similarities in many fruit characteristics of Menodora with sections Alternifolia and Primulina compared to other sections of Jasminum. Therefore, the grouping of both yellow-flower sections in a single cluster is not unusual as they are the early descendants of this tribe compared with all white-flower jasmines, which evolved independently. Yohanan et al. [21] also proposed that the members of section Alternifolia are the ancestors of other Jasminum species.
All white-flower jasmines constituted a separate cluster that was further divided into four subclusters, as shown in Figure 2. Subcluster c placed nine accessions of J. sambac into three groups: single flower, semi-double flower, and double flower. Subcluster c belongs to section Unifoliolata, which has simple, glabrous leaves with fragrant flowers and is sister to multifoliate subcluster d species J. polyanthum (section Jasminum) and J. azoricum (section Trifoliolata) but not to Unifoliolata species of subcluster e, such as J. nitidum, J. multiflorum, and J. elongatum. These three species of subcluster e have many common characteristics, such as white, star-shaped flowers and opposite, unifoliate leaves with few differences like J. nitidum has shiny glabrous lanceolate or ovate leaves, whereas J. multiflorum and J. elongatum have pubescent ovate-lanceolate leaves with an acuminate apex and broadly ovate leaves with an acuminate apex, respectively. Here, all members of section Unifoliolata were divided into two different subclusters (c and e), with the inclusion of J. azoricum and J. polyanthum (subcluster d), showing that the species of these two Unifoliolata subclusters (c and f) are more distantly related. Such combinations have also been reported by Ghehsareh et al. [20] and Yohanan et al. [21]. Subcluster f contained all the J. grandiflorum accessions from section Jasminum, having pinnate leaves with five to seven leaflets. However, one pinnate-leaf species, J. polyanthum, was placed with the trifoliate species J. azoricum (subcluster d), confirming that the standard classification based on leaf shape and arrangement is not supported genetically [44]. This could be due to the evolution of leaf morphology (simple or compound) during multiple independent evolutionary events because their genetic makeup is distantly related, as confirmed in our study and many other molecular-based studies [14,20,21].
This study further elucidated the variations between accessions of the same species. For example, for J. sambac accessions, higher values for polymorphism percentage, Shan-non's information index, Nei's genetic diversity, effective allele number, and unbiased Nei's genetic diversity were found, due to greater variation. The nine accessions of J. sambac clearly resolved into three groups: single flower, semi-double flower, and double flower. This grouping of J. sambac is in complete agreement with the findings of Ghehsareh et al.
for Iranian jasmine species [20]. Similarly, J. humile showed the second-most genetic diversity based on greater variation among its accessions. Maximum genetic identities were displayed among J. elongatum and J. multiflorum due to similarities in most of their characteristics, for example, growth habit, pubescence ovate-lanceolate to broadly ovate leaves with an acuminate apex, and star-shaped flowers with narrowly lanceolate to elliptic corolla lobes. The accessions of other species, namely J. grandiflorum, J. multiflorum, J. elongatum, J. nitidum, J. polyanthum, and J. azoricum, did not show the considerable intraspecific variation. All these species showed higher intraspecific similarities among their accessions, which could be due to asexual propagation of cultivated species and a lower number of naturally found wild accessions in this study.
The PCoA based on genetic distance matrix differentiated the species and their variants with better accuracy. These PCoA results correspond highly with the UPGMA clustering and the first three axes explained 61.71% of the genetic variance. In previous studies, distylous morph-specific patterns were observed in J. fruticans [45,46] and J. odoratissimum [47]. The variation in floral morphology particularly lays the foundation for divergence of the species [45,47,48]. In this study, the accessions of J. humile (2, 14, and 15) were far apart from each other and collected from regions where they grow in the wild, and such morphspecific patterns were also observed in these accessions, which could be the reason for their genetic variation.
The analysis of molecular variance (AMOVA) performed for all Jasminum accessions showed statistically significant PhiPT values (p = 0.001), supporting the interspecific variation that is more likely because of the utilization of different species as populations.
Conclusively, the genetic variations within the genus Jasminum was assessed by using the ISSR molecular marker system. The results provide a preliminary indication of genetic diversity among Jasminum accessions from Pakistan. The computed results show the close identity between J. multiflorum and J. elongatum. The outcomes further present an augmented distance between the accessions of J. humile, which are characterized by distylous morph-specific pattern. Similarly, the accessions of J. sambac were transformed into three groups according to their single-, semi-double-, and double-flowering nature. On a broader spectrum, the white and yellow jasmines were clearly separated, with the indication of character evolution from yellow to white flowering, along with multiple and independent evolution of sections Unifoliolata and Jasminum. These multiple and independent evolutionary events show that the classification into diverse sections on the basis of leaf morphology, leaf arrangement, and flower color is not supported genetically, whereas these sections are paraphyletic. Therefore, our study provides preliminary information regarding the intraspecific and interspecific diversity in Jasminum and a baseline database to assist biologists and taxonomists in species delimitation. The findings will help in phylogenetic-based studies of this genus, particularly in resolving existing ambiguities in its classification.

Plant Identification and Herbarium Preparation
Plants were primarily identified by botanical experts, and their herbarium specimens were prepared. These specimens were submitted to the Pakistan Museum of Natural History (PMNH), Islamabad, Pakistan, as herbarium specimen vouchers (Table 1).

Total Genomic DNA Extraction
Dried leaf samples were taken from the herbarium specimens and washed with distilled water, followed by 70% ethanol, and then dried with silica gel. Genomic DNA was extracted by grinding the dried leaf samples of 28 accessions into a fine powder. The DNeasy Plant Mini Kit (Qiagen) was used for DNA extraction following the prescribed protocols.

Quantification of Genomic DNA
After extraction, the quality and quantity of the extracted DNA were evaluated using an ND 2000 spectrometer (NanoDrop Technologies, Wilmington, USA) by assessing the absorbance at A260 and A280 [49]. The extracted genomic DNA was diluted to a concentration of 15 ng/µL for the ISSR reaction using standard methods [20].

ISSR Primers and Their PCR Amplification
In this study, 21 ISSR primers ( Table 2) were selected that had shown successful results in previous studies on three Oleaceae genera: Syringa [33], Jasminum [20,21], and Olea [25]. The inter-repeat region of the Jasminum accession genome was amplified with ISSR primers following standard PCR procedures, as described by Ghehsareh et al. [20].
Briefly, the PCR profile for the amplification of genomic regions using an ISSR marker is given in Table 2. Each 25 µL of the PCR reaction mixture contained 0.5 µL of dinucleotide triphosphate (2.5 mM each; total 10 mM), 0.7 µL of MgCl 2 (50 mM), 17.55 µL of doubledistilled water, 2.5 µL of 10× PCR buffer, 0.25 µL (5 U/µL) of Taq-polymerase (Qiagen, USA), 1.5 µL (50 µM) of primer, and 2 µL of template DNA (15 ng/µL). A sample with no DNA was taken as a negative control. Gel electrophoresis was carried out to visualize the amplified genomic region in a gel using DNA standard size markers of 1 kb (N-3232L, Biolabs Company), as described in Ghehsareh et al. [20].

Data Analysis
All the amplified fragments were deliberated as a dominant locus-specific allele. The bands obtained from each locus were noted as absent (0) or present (1), and the ISSR loci data matrix was gathered for more scrutiny. The UPGMA dendrogram based on the arithmetic mean unweighted pair group and the similarity matrices were produced with the Jaccard coefficient. The coefficient of cophenetic correlation was calculated to assess the goodness of fit for the dendrogram and the similarity matrices with a test, as described by Mantel [50]. Principal coordinate analysis (PCoA) was performed with Gower's coefficient using Paleontological Statistics (PAST) software, version 3 [51]. To distinguish among accessions, the capacity of each primer studied was evaluated on the basis of the marker index (MI) [34], polymorphic information content (PIC) [52], and resolving power (Rp) [38]. The PIC was estimated by the following formula: PIC = 1 − pi 2 − qi 2 (1) where q is the null allele frequency and p is the visual allele frequency [53]. In addition, MI = PIC × polymorphic loci number and Rp = ΣIb, where b is the formativeness band, taking values 1 − (2 × (0.5 − p)), and p is the amount of each genotype encompassing a band.
To calculate the polymorphic loci percentage, GenAlEx 6.503 was used [54]. The MI, PIC, Rp, and effective multiplex ratio (EMR) were premeditated with Microsoft Excel, version 16.9. Diversity was determined based on calculation of Nei's index and Shannon's information index (I) [55]. AMOVA was used to estimate the variance between populations. Wright's F ST analogue is considered a PhiPT value, and assessment of PhiPT with AMOVA provides species variation information. GenAlEx 6.503 was used to quantify the pairwise genetic distance with Nei's unbiased genetic distance among populations, Shannon's information index (I), genetic diversity index (He), different allele number (Na), polymorphic loci percentage, effective allele number (Ne), and ANOVA.