Genetic diversity of Simao pine in China revealed by SRAP markers

Introduction

Simao pine (Pinus kesiya Royle ex Gordon var. langbianensis (A. Chev.) Gaussen), a geographic variant of Pinus kesiya, is distributed naturally in the humid and sub-humid areas of the tropical and subtropical zone within Yunnan Province (Wang et al., 2012a; Wang et al., 2012b). The species is one of the most important timber and resin production trees in China (Chen, Zhao & Wang, 2002; Cai et al., 2017; Zhu et al., 2017; Wang et al., 2018), and also has ecological and social benefits, in addition to economic ones (Ou et al., 2016; Li et al., 2018a; Li et al., 2018b). There is high potential for over utilization of the tree species and Simao pine breeding programs are constrained by a narrow genetic background and overexploitation (Zhao et al., 2016). Genetic improvement will be salient to future Simao pine breeding and in order to selectively breed Simao pine in a coordinated breeding scheme, genetic diversity analysis needs to be implemented.

Sequence-related amplified polymorphism (SRAP) is a kind of molecular marker technology based on polymerase chain reaction (PCR). The method is convenient, exhibits high co-dominance, is uncomplicated in the separation of strips and sequencing, and doesn’t require knowledge of the sequence information of species in advance (Li & Quiros, 2001). The SRAP markers are simple, reliable, easily detected, genome specific, highly polymorphic and commonly used in genomic applications (Ma et al., 2015). This form of analysis is also very effective for the assessment of genetic diversity (Shaye et al., 2018). For these reasons, SRAP markers are widely used in population genetic analysis of various plant species (Liu et al., 2016; Bhatt et al., 2017; Li et al., 2018a; Li et al., 2018b).

In this study, SRAP markers were used to investigate the genetic composition of natural Simao pine populations, which are limitedly distributed in Southwest Yunnan, China. The study’s aims were to evaluate genetic diversity at population and species levels in the Simao pine; assess the distribution of the genetic variation within and among populations, and construct a dendrogram demonstrating the genetic relationships among them. The genetic information can then be used as a tool for assessing the current conservation management plan for this species and for designing conservation strategies.

Material and Methods

Plant materials

Eleven natural populations of Simao pine with a total of 290 individuals were sampled throughout the species distribution range (Fig. 1 and Table 1). Field experiments were approved by Southwest Forestry University (project number: 2013Y121). In large populations (n >  100), 30 adult individuals were selected within a distance of >  100 m. In small populations (n <30), all available adult individuals were sampled. For each sampled individual, young fresh pine needles (∼10 g) were collected, dried in silica gel, and stored at −80 °C until subsequent DNA extraction.

Table 1:

Locations and geographic characteristics of the sampled Simao pine populations.

Population name	Population abbreviation	Longitude (°E)	Latitude (°N)	Altitude (m)	Tmean (°C)	Tmax (°C)	Tmin (°C)	Pr (mm)	Prmax (mm)	Prmin (mm)	Number
Mojiang	MJ	101.62	23.40	1,578	19.0	28.2	5.9	279	1,322	15	30
Ninger	NE	100.91	23.19	1,383	20.1	29.7	6.4	283	1,342	13	30
Jingdong	JD	100.76	24.59	1,485	19.0	28.5	5.3	191	965	13	30
Jinggu	JG	100.98	24.01	1,525	19.8	29.5	5.8	285	1,342	13	30
Zhenyuan	ZY	100.58	23.23	1,964	16.8	25.9	3.3	238	1,144	14	30
Jiangcheng	JC	101.93	22.73	1,263	16.9	25.6	4.4	398	1,772	22	30
Lancang	LC	99.77	22.32	1,725	17.8	27.4	3.6	314	1,520	12	30
Menglian	ML	99.42	22.21	1,239	18.7	28.5	4.4	283	1,449	9	30
Jinghong	JH	101.09	22.41	870	20.6	30.2	7.6	290	1,434	15	20
Mengla	Mla	101.46	21.96	1,032	19.0	28.0	6.4	319	1,645	20	20
Simao	SM	100.90	22.90	1,355	18.4	27.9	4.3	324	1,491	14	10

DOI: 10.7717/peerj.6529/table-1

Results

Cluster analysis

Based on the SRAP data, a broad range of Nei’s genetic distance was found to exist among the 11 Simao pine natural populations, varying from 0.0722 to 0.2820 (Table 6), with a mean of 0.1483. The highest genetic distance pairs were found between population SM and MLa (0.2820), and these populations may be good sources for further breeding purposes. The lowest genetic distance pairs were found between populations JD and ZY (0.0722).

Table 6:

Nei’s genetic distances and geographical distances among 11 populations.

	SM	NE	MJ	JD	JG	ZY	JC	LC	ML	JH	Mla
SM	****	32.76	92.15	187.44	50.61	123.16	107.94	133.22	171.17	57.27	118.53
NE	0.0826	****	75.92	155.03	36.23	90.48	116.85	152.635	189.02	88.41	147.32
MJ	0.1743	0.0788	****	158.33	109.97	94.52	80.35	224.71	262.28	121.92	159.41
JD	0.2025	0.1214	0.0726	****	151.42	97.75	237.96	271.14	297.85	243.03	299.15
JG	0.2135	0.1499	0.124	0.0661	****	95.92	151.44	130.17	163.98	105.53	168.09
ZY	0.2178	0.1296	0.0861	0.0722	0.0741	****	172.16	224.62	256.23	177.05	231.72
JC	0.1855	0.1135	0.0943	0.0927	0.0948	0.082	****	227.86	265.85	94.26	98.16
LC	0.2529	0.1662	0.1883	0.1969	0.1672	0.1808	0.099	****	38.31	136.48	178.92
ML	0.2433	0.1538	0.1666	0.1835	0.1694	0.1830	0.1142	0.0649	****	173.71	212.45
JH	0.2511	0.1428	0.1228	0.1605	0.1356	0.1480	0.0879	0.0747	0.0769	****	62.84
Mla	0.2820	0.1796	0.1959	0.1993	0.1899	0.2265	0.1299	0.1633	0.1924	0.1433	****

DOI: 10.7717/peerj.6529/table-6

Notes:

Above diagonal: geographical distances (km); below diagonal: genetic distances.

The applied measure of genetic similarity was used to construct UPGMA dendrograms (Fig. 2). The 11 populations were divided into four groups with population SM forming one subgroup; Mla forming another; the populations NE, MJ, JD, JG, ZY and JC forming the third group; and LC, ML and JH were included in the fourth group. The clustering results were not specified completely from location. Furthermore, the dendrogram showed that the populations were partly mixed clusters, although geographically more distant (Fig. 2). Consistent with these results, the Mantel test revealed that there was no significant correlation between geographic and genetic distance (r = 0.241, p = 0.090).

Principal coordinate analysis

Principal coordinate analysis (PCoA) was performed to determine the genetic relationships among the accessions with minimum distortion. The first two principal components displayed 37.69% and 60.12% of total variation, respectively, and the 72.74% was expressed by the first three components (Fig. 3). Corresponding with the cluster analysis, each population formed a separate plot and could be clearly distinguished from those of other populations. The first principal coordinate separated 11 populations into two groups. The first group was composed of six high latitude populations including JD, MJ, ZY, JG, NE and SM, and the remaining five southern populations were separated into the second group. Based on the results of the first principal coordinate, the second principal coordinate separated 11 populations into four groups.

Population structure analysis

The results of the Bayesian clustering analysis of genetic structure showed that the populations analyzed of Simao pine, best fit two genetic groups (K = 2, Fig. 4). The percentages of individuals in cluster I and cluster II were 61.87 and 30.38% respectively, whereas 7.75% of the individuals had mixed membership, with Q scores of 0.2 < Q < 0.8. The individuals in NE, JD, JG, ZY populations with Q values of >0.8 belonged to cluster I, and the ML, JH, Mla, SM populations with Q values of <0.2 belonged to cluster II. Individuals with admixed population assignments were from MJ, JC and LC populations.

Discussion

In this study, SRAP markers were used to evaluate levels of genetic variation within and among 11 natural populations of Simao pine in China. Previously, three natural populations of Simao pine genetic variation were investigated using nine enzyme systems (Chen, Zhao & Wang, 2002). The reported values of the proportion of polymorphic loci, mean genetic distance, and genetic differentiation were all considerably lower than the results of this study. These differences were likely due to the ability of SRAP markers to detect more loci and higher levels of polymorphism than isozymes markers. One of the benefits of SRAP over other molecular techniques was its sensitivity in stock identification without any upfront knowledge of the species genome, which provided a large number of independent markers that can be rapidly analyzed (Uzun et al., 2009; Soleimani, Talebi & Sayed-Tabatabaei, 2012; Ahamd et al., 2014; Peng et al., 2015).

The PIC and polymorphism rate (P) were used to measure genetic diversity. A high level of polymorphism was observed in numerous amplification products generated in the course of the analysis. In this study, PIC ranged from 0.43 to 0.77, with an average of 0.63, indicating that the SRAP markers could develop high loci polymorphism useful for genetic variation of accessions studied in this research (Talebi, Rahimmalek & Norouzi, 2015). The percentage of polymorphic bands (P) of Simao pine was 93.46%, revealing a considerable level of genetic diversity, similar to that of its relatives Pinus yunnanensis (P = 96.43%; Xu et al., 2015). At a species level, Nei’s genetic diversity (H) was 0.4567 and Shannon’s information indices (I) was 0.6484, indicating that Simao pine had a high basis of genetic diversity. Previous studies also showed that narrowly distributed species could have higher genetic diversity (Dolan et al., 1999; Wang et al., 2013).

In the present study, the total genetic diversity (H_t) of the species and genetic diversity within populations (H_s ) were estimated to be 0.4580 and 0.3801 respectively. Compared to previous SRAP-based studies of pine plants, Simao pine showed higher diversity than Pinus taxa (H_t = 0.2134; H_s = 0.3426; (Xie et al., 2015), but lower genetic diversity than khasi pine (H_t = 0.547; H_s = 0.285; (Rai, Ginwal & Saha, 2017). Genetic differentiation (G_st) was the ratio between the additive genetic variance among populations and the total additive genetic variance (Wright, 1965). The study revealed low genetic differentiation (G_st = 0.1701) among populations of Simao pine. This value was higher than the mean value (G_st = 0.073) recorded for 121 woody species examined using allozyme markers (Hamrick, Godt & Sherman-Broyles, 1992). The low gene differentiation among populations was mainly caused by the high level of gene flow, gene flow could give play to its homogenizing function to prevent differentiation among each population caused by genetic drift. Similar results have been reported for several woody trees that have high levels of genetic diversity, and even in trees with low or moderate levels of genetic differentiation in populations that were located several kilometers away (Lacerda et al., 2001; Arias et al., 2012). The AMOVA analysis revealed that 86% of the variation was within populations, while the remaining 14% was among populations. This was in agreement with a number of studies where it has been observed that conifers showed high levels of genetic variation within populations and relatively little differentiation among populations (Hiebert & Hamrick, 1983; Agundez et al., 1997; Xie et al., 2015).

Gene flow is the transfer of alleles from one population to another and the study of this is critical for understanding the evolution of plants and population process within and among species (Grant, 1991; Gerber et al., 2014). Gene flow among populations is mainly produced by foreign genes brought by pollens and seeds for spermatophytes (Hamrick, 1987). The N_m of Simao pine was 2.4403, showed that gene exchange among populations was moderately frequent, and weakens genetic differentiation among populations. Gene exchange in the present study was enough to rule out the possibility that some differentiation among the populations of Simao pine could be due to isolation. This result was consistent with the lack of correlation between genetic and geographic distances (r = 0.241, p = 0.090), which suggested no significant geographic restriction to gene flow among the populations. The reason for slightly higher N_m may be due to the morphological character of Simao pine, as tall plants benefit from reduced resistance to pollen movement in the air. A long-distance pollen flow also enhanced the gene recombination (Feng et al., 2009). Furthermore, a lack of effective geographic isolation may also have contributed to the improvement of gene homogenization among provenances. Simao pine forests were mainly distributed in low mountainous upland in southern China, with a closer distance between provenances.

The genetic relationship among the 11 analyzed populations of Simao pine was represented in a dendrogram showing a clear genetic differentiation of the SM population, which was genetically distinct and could be considered to be more genetically diverse compared to the other ten populations. Principal coordinate analysis (PCoA) was performed to provide spatial representation of the relative genetic distances among populations defined by cluster analysis. The results of UPGMA and PCoA also showed that the assignment of populations was not completely consistent with their geographic origin, a result that has been observed in previous studies (Liu et al., 2013; Cheng, Zheng & Sun, 2015). This finding indicated that geographical distance had no obvious effect on genetic differentiation of Simao pine populations.

A Bayesian cluster analysis performed with STRUCTURE, showed that the most probable number of genetic groups in the data was two (K = 2) for Simao pine. Cluster I was composed of the NE, JD, and ZY populations from the northwest of the Simao pine natural range, collected from regions that had higher altitude (mostly above 1,400 m) and latitude. The ML, JH, Mla, and SM populations which belonged to cluster II were collected from southern regions that had lower altitude (mostly below 1,300 m) and latitude. These results possibly reflected an adaptational difference. It implied that geographical and environmental factors together created stronger and more discrete genetic differentiation than isolation by distance alone.

In the study, there were no significant correlations between the ecological factors and genetic diversity parameters according to Spearman nonparametric correlation analysis. It was suggested that the ecological factors had less effect on genetic diversity. This may be due to the limited distribution, and the small geographical and ecological differences of the studied populations. The high level of genetic diversity in Simao pine is mainly caused by the traits of long-life, generation overlap, wind pollination, and wind-seed dispersal, which lay a broad genetic base for the species (Chen, Zhao & Wang, 2002).

Retaining genetic diversity was the key to multi-generation improvement of forest trees and fine provenances should have an extensive genetic base and significant genetic gain (Feng et al., 2009). Thorough understanding of the extent and patterns of genetic diversity in Simao pine was essential for its conservation and utilization. The results reported here revealed high genetic diversity at the population level and low genetic diversity at the species level in Simao pine. Neither the percentage of polymorphic loci nor diversity index displayed significant difference among the 11 provenances. Therefore, selective breeding of Simao pine forests can disregard the limitation of geographic location and should directly select individuals with advanced growth traits. The low genetic differentiation among populations implied that conservation efforts should aim to preserve all existing populations of this species. Based on this, the in situ conservation method was proposed as it is of paramount urgency that sufficient natural population numbers and sizes were conserved to prevent a reduction in genetic diversity. Moreover, in order to achieve effective conservation of germplasm resources, efforts were required to carefully plan and construct pollen and gene banks for Simao pine. In the meantime, natural protection areas should be established to conserve and restore the habitat and populations. In this study, population JC displayed relatively high genetic diversity, and should therefore be a priority site for in situ conservation. In addition, it was important to develop a core collection of Simao pine, which would not only mitigated the pressure of excessive exploitation of wild resources, but can also helped in achieving more effective management and utilization of germplasm.

Conclusions

The present study is understood to be the first genetic investigation of Simao pine using SRAP markers to understand distribution and genetic variation. The results indicated that SRAP markers can be efficiently used in the study of genetic diversity and genetic variability of Simao pine. The study revealed high levels of genetic diversity and low or moderate levels of genetic differentiation. Slightly frequent gene flow existed within Simao pine populations, weakening genetic differentiation among these populations. Based on these results, conservation strategies for Simao pine have been formulated.

Supplemental Information