L. orientalis, which is a relict endemic species in our country, is one of the most important and remarkable species of Turkey in terms of biodiversity. Determining the genetic diversity of populations is very important to protect species biodiversity. As a result of the literature search, it was determined that several different methods were used to determine the genetic diversity of L. orientalis populations. On the other hand, there are no studies in the literature using ISSR markers, which also constitute the originality of this study.
When the molecular studies made with sweetgum are examined in the literature, there are some studies using isoenzymes that are biochemical markers. In addition, with the introduction of molecular markers, the use of biochemical markers has become less preferred. The reason for this is that molecular markers are more successful in detecting polymorphism than biochemical markers.
In the study carried out by Öztürk (2008), genetic diversity was investigated over 320 individuals in 14 sweetgum populations with 8 isoenzyme systems. For all populations, the average number of observed alleles (Na) was 1.84 and the average number of effective alleles (Ne) was 1.58. Although the average number of alleles from these values is lower than the value in this study, the average effective allele number is higher. The reason for this is that the lethal alleles are less in the study conducted by Öztürk (2008).
In Doğaç (2008) study, 320 individuals from 14 L. orientalis populations and 30 RAPD primers were used. As a result of the study, a total of 453 loci were identified and 6 of these loci were found to be monomorphic for all populations. According these data, it was determined that the average number of polymorphic loci detected for each primer was higher in ISSR primers (while an approximately of 15 polymorphic loci per primer was determined in RAPD primers, an approximately of 27 polymorphic loci per primer was detected with ISSR primers).
The eight populations used in this study and Doğaç's (2008) study are the same. These are Milas Selimiye, Köyceğiz Zeytinalanı, Köyceğiz Toparlar, Ula Kızılyaka, Marmaris Çetibeli, Marmaris Değirmenyanı, Fethiye Gunluklu and Fethiye Inlice populations. The observed average allele number, average effective allele number, genetic diversity in the population, polymorphic loci number and percent polymorphism values for these locations were found to be lower than the study of Doğaç (2008). The main reason for these values being higher than our study is thought to be the higher number of individuals and primers used in the study with RAPD. Although the sampling sites in both studies are the same, the sampling in this study was carried out approximately 15 years after the sampling conducted by Doğaç (2008). The loss of genetic diversity in populations during this period may also be another reason for the lower values found.
Velioglu et al. (2008) investigated the genetic diversity of L. orientalis populations using 10 RAPD primers and 25 individuals from each population in 18 populations. It was determined that all genetic diversity values obtained as a result of the study were lower than our study.
Since both ISSR and RAPD primers are dominant markers, it is very important to determine the choice of which primer to use in determining the polymorphism. When the genetic diversity values obtained in this study were compared with the values obtained by both Doğaç (2008) and Velioğlu et al. (2008), it was thought that ISSR primers were more successful in detecting polymorphism within the same population. It is thought that scanning an equal number of samples collected from the same individual at the same locations with the same number of ISSR and RAPD primers will provide more accurate data to support this idea or to prove the opposite.
When the Nei’s genetic diversity values were examined in the results of the analysis, total genetic diversity values (HT) = 0.27 ± 0.03 in based of 271 loci for all populations. It was found that 33% of the total genetic diversity in 271 loci was due to within population genetic diversity values (HS) = 0.09 ± 0.007, and 67% of genetic diversity was found to be due to among population genetic diversity with DST = 0.18 values. As a result of the molecular analysis of variance (AMOVA), it was observed that 29% of the genetic diversity was due to within population genetic diversity and 71% to among population genetic diversity. Therefore, according to the results obtained from both POPGENE ver. 1.32 and GenAlEx 6.503 software, it has been clearly revealed that the main source of genetic diversity is among the populations.
G ST is one of the values that reveal genetic differentiation. GST is observed in values ranging from 0 to 1 and expresses the level of genetic differentiation. If the GST value is 0.05 or less, genetic differentiation within the population is negligible. If it is above 0.25, the level of genetic differentiation is high. In this study, the GST value was found to be 0.63. That is, the genetic differentiation among the studied populations is quite high. In addition, NM, ie gene flow value, was calculated. It has been reported by Hamrick et al (1992) that the species with a gene flow (NM) value of 0.265 are self-pollinating and can spread their seeds and pollen to short distances such as 2–3 meters. The gene flow (NM) value of 4,750 has been reported to be the values used for species that pollinating from long distances with various seed and pollen carriers and can also spread their seeds and pollen to long distances. In addition, it has been reported that the gene flow level (NM) of 0.50 is the critical value and the values above this value are the amounts that prevent genetic drift. If the NM value is less than 1, it is considered that there is differentiation among populations (Wright 1951). In our study, the gene flow level (NM) was found to be 0.29. The fact that this value is above 0.265 indicates that the population members pollinating with the trees that are close neighbors in the distribution area, while the value below 4.750 indicates that the pollinating caused by long distances is not too much or not at all. The fact that the genetic diversity (HS) value within the population is already as low as 0.09 reveals that there is a limited gene flow (NM: 0.29). The results of the GST and NM values found and the AMOVA analysis are an indication that the studied populations may be subject to genetic drift and that individuals in the studied populations pollinating within the population and form the next generation.
The genus Liquidambar is represented by four different species on Earth. There are some studies in the literature to determine genetic diversity by using ISSR markers in L. formasana, one of these species. Bi et al. (2010) found the GST value of the L. formasana populations to be 0.185 and the NM value to be 2.194. In addition, as a result of molecular variance analysis, it was determined that the genetic variation originates from among populations with a rate of 14.51% and within the population at a rate of 85.49%. Rongxi et al. (2016), according to the results of AMOVA, genetic variation in L. formasana populations originates from within the population with a rate of 94.02% and among populations with a rate of 5.98%. Rongxi et al. (2016) and Bi et al. (2010) revealed that genetic diversity is mostly caused by the population, with AMOVA, GST and NM values. These values were reported by Rongxi et al. (2016) and Bi et al. (2010) show that the populations studied do not have a risk of genetic drift, unlike the L. orientalis species we studied. Rongxi et al. (2016) and Bi et al. (2010), the most important reason for the higher genetic diversity values and the absence of genetic drift risk for the species is that the L. formasana species has populations that spread over very large and undivided areas. For this reason, it is very important to protect the L. orientalis populations whose spread areas are decreasing day by day.
L. orientalis species is considered by some researchers to have two varieties, L. orientalis Mill. var. orientalis and L. orientalis Mill. var. integriloba Fiori, in terms of morphological character. It can be said that L. orientalis Mill. var. orientalis and L. orientalis Mill. var. integriloba Fiori varieties do not differ from each other within the determined 271 loci. Because when the genetic distance values given in Fig. 2 are examined, the second location pair with the least genetic distance from each other is Fethiye İnlice (L. orientalis Mill. var. orientalis) and Ula Kızılyaka 1 (L. orientalis Mill. var. integriloba Fiori) populations. In other words, L. orientalis species, which is morphologically divided into two varieties, could not be genetically separated into two different varieties after this study.
When Nei's genetic distance values were examined, the two populations with the least genetic distance to each other were found with a value of 0.161, Kızılyaka 1 and Kızılyaka 2. Kızılyaka 1 and Kızılyaka 2 populations were separated from each other due to the tree cuttings and roads made by humans. For this reason, it is an expected result that the genetic distance between them is the least compared to other populations. Among the studied populations, the locations with the furthest distance are the Selimiye population in Milas district and the İnlice and Gunluklu populations in Fethiye district. Therefore, it was expected that the genetic distance from each other would be the highest between these populations in the analysis results. However, in the analysis results, the populations with the highest genetic distance from each other were determined as Marmaris Değirmenyanı and Kızılyaka 2 populations with a value of 0.334. As seen in Fig. 1, although they are located in neighboring districts, these two populations are quite far from each other. In addition, the reason for the highest genetic distance value may be the geographical isolation between the two populations created by the mountains and the sea.
The UPGMA method divided the populations into three groups. The first of these groups consisted of the Marmaris Cetibeli, Ula Kızılyaka 1, Ula Kızılyaka 2, Milas Selimiye and Koycegiz Zeytinalanı populations, the second from the Fethiye Gunluklu, Fethiye Inlice and Marmaris Degirmenyanı populations, and the third from the populations of Marmaris National Park and Koycegiz Toparlar. When the geographical locations and distances of the populations given in Fig. 1 are examined, it is clear that the closest locations to each other are the Kızılyaka 1 - Kızılyaka 2 population pairs and the Fethiye Inlice - Fethiye Gunluklu population pairs. Therefore, these two population pairs were expected to be in the same tree branches in the UPGMA dendrogram. The obtained UPGMA dendrogram is given in Fig. 3 and supports the expected result.
Species must be protected in order to preserve genetic diversity and transfer it to the next generations. As a result of this study, the three populations with the highest parameters (average number of alleles, number of effective alleles, Nei’s genetic diversity, Shannon's constant, number of polymorphic loci and polymorphic loci) revealing genetic diversity were determined as Fethiye Gunluklu, Fethiye İnlice and Kızılyaka 1. In the studies conducted by Doğaç (2008) and Özdilek (2007), it was determined that the Gunluklu and Kızılyaka populations were among the populations with the highest genetic diversity values. In this context, it is suggested that priority should be given to these three populations in a conservation program to be initiated for L. orientalis species.