Morphological method and molecular marker determine genetic diversity and population structure in Allochrusa

The Caryophyllaceae family is complex. Several attempts have been carried out in the past to study Caryophyllaceae members. This study mainly focused on Allochrusa Bunge to determine its genetic structure and used ISSR markers, ITS, and rps16 data to classify and differentiate Allochrusa species. We collected 122 Allochrusa specimens. Our analysis included morphological and molecular method approaches. Morphometry analysis indicated that floral characters could assist in the identification of Allochrusa species. A. persica (Boiss.) Boiss. and A. versicolor Fisch. & C.A.Mey. showed affinity to each other. A. bungei Boiss. formed a separate group. Analysis of molecular variance showed significant genetic differentiation in Allochrusa (p= 0.001). The majority of genetic variation was among the Allochrusa population. We recorded minimum gene flow (Nm=0.176) between Allochrusa species. Besides this, isolation by distance occurs in Allochrusa members, as shown in the Mantel test result (r = 0.01, p = 0.0002). STRUCTURE analysis revealed three genetic groups. It is evident that A. persica, A. versicolor, and A. bungei differ genetically from each other. Our current findings have implications in plant systematics and biodiversity management.


INTRODUCTION
Caryophyllaceae contains 88 genera and 2,200 species. The Caryophyllaceae family is subdivided into three subfamilies, ie. Caryophylloideae, Alsinoideae, and Paronychioideae (Greenberg and Donoguhe 2011;Pirani et al. 2014;Hernandez-Ledesma et al. 2015). The Caryophyllaceae has a worldwide distribution, and this family is diverse. The Mediterranean region is consid-ered a hot spot or center of diversity for Caryophyllaceae (Harbaugh et al. 2010;Greenberg and Donoguhe 2011).
Acanthophyllum Meyer plant species are shrubs and perennial. The majority of Acanthophyllum occurs in Iran and Central Asian countries (Ghaffari 2004;Pirani et al. 2014;Mahmoudi Shamsabad et al. 2020). The Caryophyllaceae family is a complex taxonomical family. Therefore given the taxonomic complexity in Caryophyllaceae, some studies were conducted to resolve taxonomical and classification issues. For instance, phylogenetic data on Acanthophyllum supports the notion of inclusion of Allochrusa within Acanthophyllum ). However, traditional taxonomical and morphological characters are dissimilar between Acanthophyllum and Allochrusa. Henceforth, Allochrusa is classified as a separate genus .
According to Madhani et al. (2018) the Acanthophyllum clade includes Allochrusa, Gypsopgila herniarioides, and Allochrusa species. They revealed that both markers (ITS) and the chloroplast gene rps16 does not allow Allochrusa to differentiate from Acanthophyllum. The species of the genus Allochrusa were considered once as members of Acanthophyllum subgenus. Allochrusa (Schischkin 1936) and molecular phylogenetic studies by Madhani et al. (2018) corroborate the taxonomic treatment performed by Pirani et al. (2014) and contradict the treatment by Hernandez-Ledesma et al. (2015), where it was recognized provisionally at the generic level. According to this concept, it is necessary to resurrect the generic name Acanthophyllum for some taxa treated as Allochrusa in recent taxonomic surveys .
Advent in molecular biology has paved our understanding to characterize genetic diversity and population structure in plant species (Shakoor et al. 2021). Molecular markers played a vital role in conservation biology and plant genetic resources (Erbano et al. 2015;Esfandani-Bozchaloyi and Sheidai 2018 ). Molecular markers, including Inter Simple Sequence Repeats (ISSR) and ITS phylogenetic studies on the Caryophyllaceae family, showed the significance of molecular methods to resolve the genetic and evolutionary relationship within the members of Caryophyllaceae (Greenberg and Donoguhe 2011;Korkmaz and Yildirim 2015).
Allochrusa lutea is restricted to the Zanjan province, while its closest relative species (A. Persica) occurs in NW Iran. The altitudinal range is 1300-1600 m a.s.l. A. lutea grows on low montane steppe life zone in open, disturbed, and dry areas with a high percentage of Scree on the ground (Mahmoodi and Falatoury 2016 A. bungei, and A. persica are found in Iran. These species have almost similar morphological features. It is difficult to identify and separate these species on the basis of traditional taxonomy and morphology. Therefore, due to complexity in identification, we only used ISSR markers to identify/ separate these species. The phylogenetic approach has been used on other accessions, and no unedited sequences were produced. Our approach integrated morphological and molecular methods to analyze Allochrusa species.

Plants collection
122 plant samples were collected. Overall, seven natural populations were sampled. Five to eight specimens from each plant population were recorded. Further details about the plant location are provided (Table 1, Figure 1). We carefully identified the plant species, i.e., Allochrusa versicolor, A. bungei, and A. persica according to previous identification protocols (Boissier 1867;Schischkin 1936;Cullen 1967;Schiman-Czeika 1988). Dr. Shahram Mehri helped in plant collections. Plant samples were deposited in the Islamic Azad University herbarium. We examined 38 morphological characters (10 qualitative, 28 quantitative). The details of morphological characters are provided (Table 2).

Plant morphology analysis
Before morphometric analysis, we transformed data. Mean and variance was coded as 0 and 1. To measure the similarity among plant individuals, we followed Euclidean distance (Podani 2000). Multidimensional scaling (MDS) and Unweighted Pair-Group Method with Arithmetic Mean (UPGMA) method to group the plant species (Podani 2000). Principal component analysis (PCA) to find the variation in the morphological characters of Allochrusa plant species. These analyses were done in the PAST software, version 2.17. (Hammer et al. 2001).

Phylogenetic reconstruction
Two different nuclear and chloroplastidial DNA markers (ITS an rps16 respectively) were prelimi-  (Swofford 2002). The heuristic search option was used for each of the two single region datasets, using tree bisectionreconnection (TBR) branch swapping, with 1,000 replicates of the random addition sequence. Uninformative characters were excluded from the analysis. Branch support values were calculated using a full heuristic search with 1,000 bootstrap replicates (Felsenstein 2005), each with a simple addition sequence. The Combinability of these two datasets was assessed by use of the partition homogeneity test (the incongruence length difference test (ILD) of Farris et al. (1995) as implemented in PAUP (Swofford 2002). The test was conducted with invariant characters excluded (Felsenstein 2005), using the heuristic search option involving 100 replicates of the random addition sequence and TBR branch swapping with 1,000 homogeneity replicates. The maximum number of trees was set to 500. The model of sequence evolution for each dataset was selected by use of the software MrModeltest v. 2.3 (Kumar et al. 2016) as implemented in MrMTgui based on the Akaike information criterion (AIC) (Edgar 2004). All datasets were analyzed as a single partition with the Kimura 2-parameters + G model by Bayesian inference (BI) using the software MrBayes version 3.12 (Ronquist and Huelsenbeck 2003) . Posteriors on the model parameters were estimated from the data using the default priors. The analysis was performed with 4 million generations, using Markov chain Monte Carlo search. MrBayes performed two simultaneous analyses starting from different random trees (Nruns = 2) each with four Markov Chains trees sampled every 100 generations. No new sequences were produced. We downloaded the ITS and rps16 data on Allochrusa species from National Center for Biotechnology Information. Accession numbers obtained from NCBI are provided in Appendix. Acanthophyllum mucronatum and Acanthophyllum cerastioides (D.Don) Madhani & Zarre were used as outgroup taxa.

Molecular marker assay (ISSR)
We extracted DNA from the fresh leaves of plants. Plant DNA was extracted according to a previous proto- col (Esfandani-Bozchaloyi et al. 2019). The plant leaves samples were dried with the aid of silica gel. Twenty-two ISSR primers from the University of British Columbia were initially chosen for the ISSR assay. However, we selected 10 primers that could amplify the DNA and yielded clear bands (Table 3). The ISSR marker had a 16-18 bp nucleotide repeat sequence. DNA amplification was done through PCR. A 25μl volume containing 10 mMTris-HCl buffer at pH 8; 50 mMKCl; 1.5 mM MgCl2; 0.2 mM of each dNTP (Bioron, Germany); 0.2 μM of single primer, 20 ng of genomic DNA; and 3 U of Taq DNA polymerase were subjected to PCR reactions (Bioron, Germany). The PCR was carried out in Techne thermocycler (Germany). The initial denaturation stage of 5 minutes is 94°C. The initial denaturation step was followed by 36 cycles of 1 minute at 95°C, 1 minute at 50-52°C and 1 minute at 72°C. The final extension stage of 5-10 min at 72°C completed the reaction. The quality of the amplified product was checked on 1% agarose gel. Ethidium bromide was used to dye the gel. We used a 100 bp molecular size ladder to compare the fragment size of the PCR product.
We conducted genetic diversity, gene diversity (H), Shannon information index (I), number of effective alleles, and percentage of polymorphism analysis while following previous protocols (Weising et al. 2005;Freeland et al. 2011). Neighbor-joining (NJ) algorithm (Saitou and Nei 1987) was used to detect the evolutionary relationship between plant populations. We also performed network computation, i.e., TCS (Clement et al. 2002), to construct the Allochrusa plant population network. TCS analysis was done in the PopART (Population Analysis with Reticulate Trees) (Clement et al. 2002). The Mantel test was performed in the PAST program ( Hammer et al. 2001) to know the correlation between geographical and genetic distances between Allochrusa plant population. We investigated the genetic differentiation between plant populations through the AMOVA test (Analysis of molecular variance) in GenAlex 6.4 (Peakall and Smouse 2006). The data was iterated1000 times to infer the statistical significance. To unveil Allochrusa plant population genetic structure, we did genetic structure analysis through a Bayesian-based model in STRUCTURE software (Pritchard et al. 2000). Under the correlated allele frequency model, we used the admixture ancestry model. We ran twenty times Markov chain Monte Carlo simulation to get the reliable results of K. Besides this, the Evanno test (Evanno et al. 2005) was done to discern correct values of K. Since our prime aim was to describe the genetic structure of the Allochrusa plant population. Therefore, gene flow analysis was carried out in PopGene version 1.32 (Yeh et al. 1999).

Morphometry
Our clustering analysis showed the same results. UPGMA cluster results were generated based on morphological characters (Figure 2). Morphological characters failed to separate A. versicolor (2) and A. persica (3). The principal component results explained the morphological variation within species. Overall first three Table 3. Details about the banding pattern revealed by ISSR primers.

Primers
Primers sequence (5'-3') GTGTGTGTGTGTGTGTYG ISSR-10 CACACACACACACACARG components explained the majority of variation (74%) in Allochrusa species. Among three components, the first component described 55% of the total variation. Floral characters such as calyx teeth, petals, and limb shape showed a positive correlation (> 0.70). The second PCA component explained the variation in ovary shape, seed morphology. A. bungei, A. persica, and A. versicolor had morphological differences.

Phylogenetic tree
The reconstructed phylogenetic tree is shown ( Figure  3). Acanthophyllum mucronatum and Acanthophyllum cerastioides constituted in a single clade, while other species were in two separate clades. ITS and rps16 data set supported separation of A. versicolor, A. Persica and, A. Bungei with high bootstrap value (> 0.98) (Figure 3). The results show that Allochrusa species are monophyletic.

ISSR and genetic diversity
We conducted detailed genetic diversity and other genetic parameters on the ISSR generated data (Table 4). A. versicolor showed high polymorphism (57.53%), gene diversity (0.33), and Shannon information index (0.30). A. persica plant population had low polymorphism and Shannon information index (0.15). Analysis of molecular variance showed population differentiation in Allochrusa (p= 0.001). Seventy-three percentage of genetic variation was among the Allochrusa population. Comparative less genetic variation, i.e., 27%, was reported within the population (Table 5). FsT pairwise analysis showed that Allochrusa members are genetically dissimilar. Minimum gene flow occurs (Nm=0.176) between Allochrusa species. A. versicolor and A. persica were genetically related (0.88). These species are more closely related to each other. A. versicolor and A. persica can exchange genetic material and hybrid with each other.
The Mantel test result indicated a positive correlation (r = 0.01, p = 0.0002) between genetic and geographical distances among Allochrusa taxa.
TCS network analysis and clustering results showed a similar clustering pattern (Figure 4 A, B). ISSR molecular primer demonstrated its utility to divide Allochrous species into different groups or clades, as evident in the WARD tree (Figure 4 A). It is evident that A. persica, A. versicolor, and A. bungei differ genetically from each other (different segment colors) ( Figure 5). STRUCTURE analysis revealed three genetic groups. Yellow and blue segments indicated individuals of A. bungei and A. versicolor. On the other hand, the green color segment highlighted A. persica specimens ( Figure 5).
Allochrous show genetic variability within taxa due to introgression (hybridization) between different spe-   cies. Henceforth, we performed Horizontal gene transfer (HGT) analysis on ISSR and ITS data of the studied Allochrous species (Figure 6). We obtained two introgression events between A. persica and A. versicolor, and the same events happened between A. persica and A. bungei.
Allochrous species revealed 0.2-0.3 observed heterozygosity (Ho) value. In addition to this, inbreeding depression showed high values (F IS = 0.3-0.7). DISCUSSION We used traditional taxonomical and molecular methods to understand genetic and population structure in Allochrusa. The current climate change scenario and biodiversity threats have emphasized the need to conduct genetic diversity studies. Given the progress in molecular tools, several investigations have been done to analyze population structure in plants Erbano et al. 2015;Esfandani-Bozchaloyi et al. 2017;Esfandani-Bozchaloyi et al. 2018;Shakoor et al. 2021).
Current morphological findings showed the importance of floral characters to explain the variation and difference among Allochrusa species. PCA analysis highlighted the significance of calyx teeth, petals, and limbs to identify the Allochrusa species. In Iran, a new Allochrusa was reported based on floral characters (Mahmoodi and Falatoury 2016). Past and current ecological and taxonomical investigations have successfully implemented morphological characteristics to study plant species (Neal et al. 1998;Borba et al. 2002;Mahmoodi and Falatoury 2016;Chen et al. 2020). However, the rationale for choosing molecular tools to study Allochrusa was the overlapping of morphological characters in Allochrusa. Besides using ISSR markers, we also assessed the evolutionary relationship among Allochrusa members. Our results revealed genetic differentiation among studied species. A. persica and A. versicolor had a close genetic affinity between them. Genetic association and relationship studies were conducted in Caryophyllaceae ( Fior et al. 2006;Pirani et al. 2014;Madhani et al. 2018). These studies recommended the use of ITS, cpDNA, and matk to classify Caryophyllaceae plant individuals. Genetic diversity is a central theme in plant adaptability to cope with changing environments (Tomasello et al. 2015). Our analysis showed genetic diversity was low within the same individuals; however, comparative high genetic differentiation existed between different plant specimens of Allochrusa. Previous scientific data suggests that genetic diversity is linked with plant ability to endure against perturbation in the environment (Booy et al. 2000). A. persica showed a low level of genetic diversity in our analysis. The reason for such finding could be the small number of populations. Common logic suggests that population size correlates with genetic diversity (Leimu-Brown et al. 2006). Present results (Mantel test)   about genetic and geographical distances indicated the distance isolation occurs in Allochrusa species.
We detected high inbreeding depression showed high values in the Allochrusa population. High inbreeding depression reduces plant ability to survive against biotic and abiotic stress (Ramsey and Vaughton 1998). Inbreeding depression occurs due to reduced population size (Lonn and Prentice 2002). Inbreeding depression analysis is critical in the biodiversity management sector (Neaves et al. 2015). Molecular markers provide in-depth analysis and several genetic diversity parameters to describe inbreeding depression in plant species (Glemin et al. 2006).
Current results showed limited gene flow in the Allochrusa population. Indeed, a low level of gene flow hinders the exchange of genetic material between species. It may pose survival threats to a small-sized plant population (Booy et al. 2000).
Neighbor-joining and STRUCTURE indicated three groups of Allochrusa. Genetic variation among the three groups had the same pattern. Two hypotheses have been proposed in the past to explain the genetic variation pattern. Genetic diversity is maintained through gene flow; another explanation is connectivity among plant populations (Dostalek et al. 2010).
A. bungei and A. versicolor had similar macro and micromorphological similarities. Nonetheless, they are recognized as a separate taxon. The main differences noted in stem and calyx indumentum, pedicle size, the calyx teeth, petal apex, and limb shape were significant to separating the taxon. These findings are in accordance with Mahmoodi and Falatoury (2016). They also showed that A. lutea is close to A. persica morphologically. A. persica and A. lutea are similar in habit, leaves shape. A. bungei is a subshrub, covered with glandular hairs. A. persica is perennial herbs with thick woody caudex, without distinctive glandular hairs, petals white with purple striate on the claw (Schischkin 1936;Schiman-Czeika 1988).
Our findings suggest the use of plant morphology features and molecular data to identify Allochrusa species. Species identification and differentiation is an essential task for systematic and evolutionary studies. We showed that molecular markers have resolving power to solve the plant systematics complex questions. Present results have applications in biodiversity and conservation management.