Introduction Top

Genetic diversity is key to survival, adaptation, and evolution in response to abiotic and biotic stresses. Adapting to changing environmental conditions is especially important in the Mediterranean region, which is characterized by heterogeneous environmental conditions. Many species including Pinus species face a loss of adaptation ability owing to potential natural and artificial events like climate change, human activities, and fires (Richardson et al., 2007; StClair & Howe, 2011). Intraspecific variability is essential to modulate adaptive responses to changing local conditions, and forest tree species (particularly Mediterranean pines, except for P. pinea) usually harbor extremely large intraspecific genetic diversity (Heuertz et al., 2010; Fady, 2012).

Two conifers play a key role in Mediterranean coastal forests (Fady et al., 2003; Boydak, 2004). Aleppo pine (Pinus halepensis Mill.) and Brutia pine, also known as Calabrian or Brutian pine (Pinus brutia Ten.) cover together more than 8 million hectares (Daskalakou & Thanos, 2010). The natural distribution of Aleppo pine is in southern Europe (Spain, Italy, and Greece) and northern Africa (Morocco, Algeria, and Tunisia). Brutia pine is predominantly distributed in Türkiye, but some Brutia pine populations are also found in Crete, Cyprus, Syria, and Lebanon (Mauri et al., 2016). Aleppo and Brutia pines are vicariant species; the former is mainly found in the coastal areas of the western Mediterranean region while the latter is native to the eastern Mediterranean region, both of which play important roles in local economies (Fady et al., 2003; Boydak, 2004). The evolutionary and ecological characteristics of Aleppo and Brutia pines are closely related (Tozkar et al., 2009). These pines live in a Mediterranean-type climate, which is characterized by warm, dry summers and cool, mild winter conditions, and thrive in a complex range of ecological habitats from sea level up to 2,000 meters (Fady et al., 2003; Boydak, 2004; Daskalakou & Thanos, 2010). Both species are known to adapt to the Mediterranean climate. Natural and artificial hybridization between these two species has been described in many studies. Viable seeds were only obtained when Aleppo pine was used as the pollen donor and Brutia pine was used as the maternal parent (Schiller et al., 1986; Korol et al., 1995; Bucci et al., 1998). Despite their closeness, the two pines are recognized as separate species occupying different geographical ranges and bioclimates (Fady et al., 2003). Both pines can be identified using morphological, palynological, biochemical traits and DNA markers (Bucci et al., 1998; Korol et al., 2002a; Tozkar et al., 2009).

Characterizing the genetic resources of these two pines is essential to predict their adaptive potential, thereby implementing adequate strategies for the conservation, management, and improvement of these species. Aleppo and Brutia pines, separately or together were previously analyzed for their background levels of diversity, using morphological (Panetsos et al., 1997; Isik & Isik, 1999; Isik et al., 1999), biochemical traits (Korol et al., 2002a,b; Kaya et al., 2006) and DNA markers (Bucci et al., 1998; Kandemir et al., 2004; İçgen et al., 2006; Lise et al., 2007; Grivet et al., 2011; Kurt et al., 2012; Ruiz Daniels et al., 2018; Olsson et al., 2021). The results of several studies have suggested that both species are distinct and that Brutia pine is more diverse than Aleppo pine (Bucci et al., 1998; Kurt et al., 2012).

However, researchers have rarely analyzed these two pines simultaneously using the same markers. To date, only one study (Bucci et al., 1998) has used chloroplast simple sequence repeats (cpSSR) markers to analyze the genetic diversity of these two closely related species (Aleppo pine and Brutia pine). The latter study was limited by the number of populations analyzed within each species, limiting the inferences. The cpSSR primers developed by Vendramin et al. (1996) from the Pinus thunbergii Parl. chloroplast genome have been used in many pine species (Bucci et al., 1998; Robledo-Arnuncio et al., 2005; Grivet et al., 2009; Heurtz et al., 2010; Kurt et al., 2012). cpSSR markers are very useful tools for determining genetic diversity and phylogeographic analysis of closely related species due to their high polymorphism levels and easy optimization for related species (Bucci et al., 1998; Kurt et al., 2012).

Here we analyzed a comprehensive set of Aleppo and Brutia pine populations across the Mediterranean Basin using common chloroplast microsatellite markers. The main goal of this study was to characterize the genetic diversity of these two key Mediterranean pines to gain insights into the evolutionary history of the Aleppo-Brutia species complex.

Material and methods Top

Studied populations and DNA extraction

A comprehensive sampling was performed from the natural distribution of P. halepensis and P. brutia by collecting samples from 56 populations. Twenty-one populations of P. halepensis (488 individuals) and 35 populations of P. brutia var. brutia Ten. (836 individuals), and one population of P. brutia var. eldarica Medw. (Khorasan-Razavi, 20 individuals) were mostly sampled from the natural distribution range (Fig. 1 and Table 1). Few of the sampled P. brutia populations were outside the natural distribution range. Hereinafter, the P. halepensis populations and P. brutia var. brutia Ten. populations are referred to as Aleppo pine and Brutia pine, respectively, while the Khorasan-Razavi (P. brutia var. eldarica Medw.) population is referred to in the text as eldarica pine. Previously collected bulked seeds from each population were used for analyses.

Figure 1. Distribution map of 56 studied populations and their haplotypes in different colors (red dot: Aleppo pine; blue square: Brutia pine (see Table 1 for more information about populations). The haplotypes of 14 western populations (left) and 7 eastern populations (center) of Aleppo pine are presented above. The Brutian pine populations’ haplotypes are presented in the right part of the figure. Ind: ındividuals. Ind1-4: total haplotypes of individuals (from 1 to 4 individuals). The green and yellow areas indicate the natural distribution range of Aleppo and Brutia pine, respectively.

Randomly selected 30-35 seeds from each population were germinated for DNA extractions. DNA was extracted according to Doyle & Doyle (1990) protocol. The average number of analyzed samples from each population was 24, ranging from 19 to 28 individuals per population (Table 1).

Chloroplast microsatellites and PCR products

For a preliminary analysis, ten primer pairs were selected from Vendramin et al. (1996) and tested on 16 individuals from eight geographically distant populations. Based on the preliminary analysis results, only five primer pairs (Pt15169, Pt30204, Pt41093, Pt87268, and Pt110048) were used to analyze all populations, while the others were discarded as they did not give any product in PCR amplification or were not polymorphic. PCR amplifications were performed in a total volume of 10 μL containing 5 ng of template DNA, 2.5 mM MgCl2, 0.3 mM dNTP mix, 0.5 U Taq polymerase, 1X reaction buffer (Thermo Fisher Scientific), and different amounts of forward and reverse primers for studied primer pairs (1 μM for Pt15169, Pt41093 and Pt87268; 2 μM for Pt110048; and 2.5 μM for Pt30204). PCR reactions were carried out on an Applied Biosystems 9700 thermal cycler according to the following conditions: initial denaturation at 95°C for 5 min, followed by 25 cycles of 1 min at 94°C, 45 s at 55°C, and 1 min at 72°C, and a final extension step of 8 min at 72°C. The amplified fragments were resolved on an ABI 310 genetic analyzer (Applied Biosystems). The size of fragments was determined by GeneScan® Analysis Software 3.1 (Thermo Fisher Scientific) concerning the GeneScan-500 LIZ size standard (Applied Biosystems).

Data analysis

The combination of size variants across five chloroplast microsatellite regions was defined as haplotypes. The chloroplast haplotype variation parameters including the total number of haplotypes within populations, effective number of haplotypes, number of private haplotypes, haplotypic richness, and haplotypic Nei’s genetic diversity (Nei, 1987) were calculated using HaplotypeAnalysis 1.05 (Eliades & Eliades, 2009). A median-joining network including potential median vectors was constructed using Network 10 with cpSSRs data (Bandelt et al., 1999). The average genetic distance among individuals within populations (Goldstein et al., 1995) was estimated according to the stepwise mutation model, as defined by Vendramin et al. (1998). Analysis of molecular variance (AMOVA) was performed using Arlequin 3.1 to estimate the neutral genetic differentiation of species and populations within species (Excoffier et al., 2005). The unweighted pair group method with arithmetic mean (UPGMA) dendrogram of population pairs was obtained from Nei’s (1987) genetic distance values, which were obtained from haplotype data using FigTree 1.4.4 Software (Rambaut, 2018).

The spatial and non-spatial genetic mixture analyses were applied to the haplotypes of populations (Corander et al., 2003, 2008). The program Structure 2.3.4 (Pritchard et al., 2000) was used to evaluate the admixture structure patterns of the populations within each species. The analysis was performed according to the MCMC algorithm with 50000 Burn-in periods, 500,000 replications, and 10 iterations. Results of the structure analysis were evaluated using the web-based program Structure Harvester (Earl & von Holdt, 2012) and the number of clusters (K) was obtained by employing Evanno et al. (2005) and Jakobsson & Rosenberg (2007) computations in Structure Harvester.

ResultsTop

Chloroplast microsatellite variation

All analyzed primers were polymorphic for Aleppo, Brutia, and eldarica pines. Overall, 51 size variants were detected from the five primers in the studied species. Almost half of the size variants (24 alleles) were species-specific (private alleles) and each species had private alleles. The total number of size variants for Aleppo pine and Brutia pine were 33 and 42, respectively (Fig. S1 [suppl]). Most private alleles were found in Brutia pine populations (Table S1 [suppl]). The common alleles of primers for each species had distinct size variants. eldarica pine size variants were found to be closer to the Brutia pine than the Aleppo pine alleles. All the studied primers showed mutational steps among alleles from 1-bp to 5-bp in the Brutia pine populations. Mutational steps of the size variants of alleles were observed in three primers (Pt15169, Pt41093, and Pt87268) for Aleppo pine (Fig. S1).

The detected 51 alleles were combined into 204 different haplotypes. The frequency of the fourteen (H1-H14) haplotypes was greater than 1%, and the total frequency of these haplotypes corresponded to 65.8%. The distribution of haplotypes in the species was 63 and 150 haplotypes in Aleppo and Brutia pines, respectively. The eldarica pine population had four private haplotypes. The most common haplotypes of each species were found at much lower frequencies in the other species. Ten haplotypes were shared between Aleppo and Brutia pine populations. The Aleppo pine populations of the eastern and western Mediterranean Basin had different haplotypes, except for two common haplotypes (H2 and H3) (Figs. 1 and 2). The number of haplotypes ranged from 4 to 18, and the effective haplotype number (Ne) was between 1.52 and 12.52 in Aleppo pine populations. In Brutia pine populations, the number (A) of haplotypes ranged from 5 to 23, and Ne was between 1.45 and 23 (Table 1). If all populations were analyzed together, the mean A and the mean Ne were 9.79 and 6.02, respectively (Table 1). Haplotypic richness values ranged from 2.05 (Tivissa) to 10.18 (Kassadra) in Aleppo pine populations. In the Brutia pine populations, the haplotypic richness values ranged from 2.96 (Isparta-Sutculer) to 15.51 (Amasya-Bafra). Nei’s gene diversity values (h) showed similarity with the haplotypic richness value of populations. The mean genetic distance among individuals within populations for each species was 2.29 and 3.33 for Aleppo pine and Brutia pine, respectively. If the eastern and western populations of Aleppo pine were considered separately (except for the Tabarka population from Tunisia, which was grouped with Brutia pine populations), the mean genetic distance among individuals within populations changed dramatically (4.24 vs 0.89) (Table 1). The values of genetic diversity parameters were more similar between the Aleppo pine populations of the eastern Mediterranean Basin and Brutia pine populations than the western Aleppo pine populations (Table 1). The Aleppo pine populations from the eastern range (Türkiye, Greece, and Italy) were more than two-fold more diverse than the 13 populations from the western range (Spain and Morocco). When the level of genetic diversity between the eastern range Aleppo pine and Brutia pine populations were compared, the eastern range Aleppo pine populations had slightly higher genetic diversity than Turkish red pine populations (Table 1).

A median-joining network was constructed to understand the relationships between the 37 most common cpSSR haplotypes detected in Aleppo pine and Brutia pine (Fig. 2). The haplotype network showed a minimum number of evolutionary events that separated each haplotype. The network indicated that H1 was centered in Brutia pine haplotypes, while H2 and H3 were centered in the Aleppo pine haplotypes. Although the divergent center is not clear for Brutia pine, H2 might be considered the divergent center of Aleppo pine.

Figure 2. Median-joining network for the 37 most frequent (>5 frequency) cpSSR haplotypes of Aleppo and Brutian pines.

Phylogeographic comparison of species and populations

The molecular variance analysis based on the stepwise mutation model showed that between species and among populations within species variation based on assuming no population structure were 51.98 and 11.87% of the total variation, respectively (Table 2).

AMOVA results showed that haplotypic differentiation among species was statistically significant. However, the genetic diversity of Brutia pine populations was higher than among Aleppo pine populations (Table 3). The dendrogram indicated that the genetic distance between the Aleppo and Brutia pine populations was generally compatible with geographic separation at the species level, except for a few populations. The Amasya-Amasya, Amasya-Bafra, K.Maras-K.Maras and Balikesir-Ayvacik populations of Brutia pine clustered with the Tunisian population (Tabarka) of Aleppo pine. The eldarica pine population (Khorasan-Razavi) clustered with the Bursa-MKP population of Brutia pine (Fig. 3). Brutia pine populations did not present any clear geographic structure within their natural distribution range. However, the Aleppo pine populations showed clear east-west differentiation across the Mediterranean Basin. The two populations of Aleppo pine from Türkiye were in the same group as those from Greece and Italy. The Aleppo pine populations from the western Mediterranean clustered together, except for the Tivissa population from Spain. The Tivissa population also clustered with the Greek and Turkish populations in the eastern groups (Fig. 3 and Table 1). The eastern Mediterranean group of Aleppo pine is composed of Turkish, Greek, and Italian populations while the western Mediterranean group is composed of Spanish and Moroccan populations.

Figure 3. UPGMA dendrogram of studied populations based on Nei’s genetic distance

In the structural analysis, ΔK showed clear peaks for K=2 in both Aleppo and Brutia pine populations, and most of the populations were admixtures of two clusters, while a few populations had specific population structures according to K=2. For Brutia pine, Cluster 1 was reported to be predominantly in Adapazari-Geyve, Amasya-Amasya, Amasya-Bafra, Ankara-Nallıhan, Antalya-Antalya, Antalya-Gündoğmuş, Isparta-Sütçüler, İstanbul-Kanlıca, İzmir-İzmir, Kastamonu-Durağan, Kmaraş-Antakya, Kmaraş-Kmaraş, Mersin-Bozyazi, Mersin-Tarsus and Kibris1 (>90%); Cluster 2 was mainly detected in Balıkesir-Ayvacık, Bolu-Göynük, Denizli-Acipayam, Muğla-Marmaris and S.Urfa-Adiyaman (>90%) (Fig. S2). For Aleppo pine, Cluster 1 was observed to be predominantly in western Mediterranean populations (>90%), while Cluster 2 was generally detected in eastern Mediterranean populations (>90%) (Fig. S3).