Population Genetics and Evolution Analysis Reveal Diversity and Origin of Ammopiptanthus in China.


 Background：Elucidating and revealing the population genetic structure, genetic diversity and recombination is essential for understanding the evolution and adaptation of species. Ammopiptanthus, which is an endangered survivor from the Tethys in the Tertiary Period, is the only evergreen broadleaf shrub grown in Northwest of China. However, little is known about its genetic diversity and underlying adaptation mechanisms. Results：Here, 111 Ammopiptanthus individuals collected from fifteen natural populations in estern China were analyzed by means of the specific locus amplified fragment sequencing (SLAF-seq). Based on the single nucleotide polymorphisms (SNPs) and insertions and deletions (InDels) detected by SLAF-seq, genetic diversity and markers associated with climate and geographical distribution variables were identified. The results of genetic diversity and genetic differentiation revealed that all fifteen populations showed medium genetic diversity, with PIC values ranging from 0.1648 to 0.3081. AMOVA and Fst indicated that a low genetic differentiation existed among populations. Phylogenetic analysis showed that NX-BG and NMG-DQH of fifteen populations have the highest homology，while the genetic structure analysis revealed that these Ammopiptanthus germplasm accessions were structured primarily along the basis of their geographic collection, and that an extensive admixture occurred in each group. In addition, the genome-wide linkage disequilibrium (LD) and principal component analysis showed that Ammopiptanthus nanus had a more diverse genomic background, and all genetic populations were clearly distinguished, although different degrees of introgression were detected in these groups. Conclusion：Our study could provide guidance to the future design of association studies and the systematic utilization and protection of the genetic variation characterizing the Ammopiptanthus.


Background
Throughout the world, many plant species are in need of conservation as a result of human activities such as overgrazing and deforestation. Endangered plants that occur in small and isolated populations often need to be managed to mitigate anthropogenic effects such as habitat fragmentation, habitat loss, or climate change [1]. Conservation over the past two decades has increasingly become reliant on genetic data and the insight it provides into the breeding behavior and overall genetic variation within endangered species [2,3]. Endangered wild species have many characters, such as huge ornamental and medicinal value, stress resistance [4,5]. Ammopiptanthus is an endangered species from the Tertiary Period, and it is the only evergreen broadleaf shrub grown in Northwest desert of China [6,7]. The genus Ammopiptanthus (Leguminosae) only comprises of two species, which are A. mongolicus(Maxim. ex Kom.) and A. nanus(M. Pop.). Both species are diploid with 18 chromosomes and exhibiting a wide range of adaptation [8,9].
In China, Ammopiptanthus are mainly distributed in the northern and western areas, such as Inner Mongolia, Shaanxi, Gansu, Ningxia and Xinjiang, within these areas, Ammopiptanthus is known as a drought, cold and heat tolerant shrub [10,11]. However, with the gradual warming of global climate, it remains questionable to what degree the speed of future adaptation can keep up with the pace of climate change. Therefore, a deep understanding of the genetic diversity and genetic regulation of adaptation in Ammopiptanthus is signi cant. Revealing polymorphisms and genes that determine adaptation would provide the basis for breeding genetically improved germplasms that can be applied in the breeding practice. Breeding resistant cultivars using resistant resources is an economic and effective approach to control this disease. Resistance breeding is also one of the important goals in the desert-plant breeding programme.
As known to all,the genetic diversity and relationships among resistant resources is importance for improving the tolerant varieties. Molecular markers re ect the actual level of genetic variation which exist among genotypes at the DNA level [12,13]. Hence, they provide a more accurate estimate with such variation than do either phenotypic or pedigree information. In Ammopiptanthus, the DNA markers such as inter-simple sequence repeat (ISSR), SSR and ESTs have been used for analyzing the germplasm genetic diversity, cultivar ngerprinting and identi cation of abiotic stress-responsive genes [14][15][16]. Recently, the rapid development and application of sequence-speci c markers such as SSR, EST-SSR, InDels were also reported for sesame [17]. Although various molecular markers are considered to be e cient tools for studying genetic structure, gene ow, mating systems, and estimating genetic variation within a species. Especially, microsatellite markers are used to analyze the effective pollen ow and seed dispersal among the populations and within a species [18,19]. Unfortunately, a limited number of selected markers used in these studies might provide biased estimates of genetic variability.
However, the limited number of molecular markers are too small to construct a high-density genetic map in most of plant. Fortunately, Speci c Length Ampli ed Fragment sequencing (SLAF-seq) was developed based on high-throughput sequencing technology [20]. This technology allows researchers to design the experimental system through bioinformatics and screen for fragments of a speci c length from the constructed SLAF-seq library [21]. The e ciency of SLAF-seq was tested using rice and soybean data [22][23][24], and has been used to construct the highest-density genetic map of common carp [25], without a reference genome sequence. To date, SLAF-seq has been used successfully to construct highdensity genetic maps and study the genomes of many organisms, including wax ground[26], white jute [27], crape myrtle[28], cauli ower [29], Boehmeria nivea [30] and sesame [17]. Moreover, this method has been widely used in GWAS for important traits, as well as in the development of core germplasm [31]. However, the genetic diversity and population structure of Ammopiptanthus are now have not been reported.
In this study, 111 Ammopiptanthus germplasm accessions were selected from the Northwest in China and genotyped using the SLAF-seq method to evaluate their usability as an association-mapping panel. Our objectives were as follows: (a) to assess the genetic diversity as represented by the 111 Ammopiptanthus germplasm accessions; (b) to calculate the characteristics of the population structure; and (c) to estimate the linkage disequilibrium (LD) patterns occurring in the sample of sesame germplasm. Finally, Our results not only develop a valuable resource for further genome-wide association studies in Asian Ammopiptanthus and exploiting rich allelic variation for marker-assisted breeding, but also provides guidelines for the conservation, management, and restoration of Ammopiptanthus.

Plant varieties and DNA extraction
A total of 111 Ammopiptanthus samples were evaluated in this study, these individuals were selected from the 1200 core germplasms collected from Northwest of China (Fig. 1). Geographical Coordinates of the collection location ranged from 37°N to 41°N and from 74°E to 106°E, with an altitude of 1021-2546 meters. The samples of Ammopiptanthus were stored in a -80℃ refrigerator of college of life science, Yulin University, Shaanxi province, China. Genomic DNA was extracted from Ammopiptanthus leaves according to a modi ed CTAB method [28,32]. Finally, all samples were sent to Beijing BioMarker Corporation (Beijing, China) for library construction [25,28].

Evaluation of Data Quality and Identi cation of SNP Markers
"Dual-index" software was used to analyze the original data. After removing the adapter sequences, original short reads were evaluated by GC contents and Q 30 . All SLAF reads were clustered based on sequence alignments using BLAT software [33]. Polymorphic SLAF tags were determined by comparing sequence variation between different Ammopiptanthus individuals. Meanwhile, SNP markers were identi ed for polymorphic SLAF tags using GATK and SAMtools [34,35]. Finally, SNPs that match the criteria were selected for subsequent analysis, and selecting standards of SNPs is according to previous study. [34,35] Population Structure and Linkage After SNPs pretreatment, the phylogenetic tree was constructed using the MEGA7.0 software. Meanwhile, Bayesian clustering was applied to analyze the population structure of the 111 Ammopiptanthus individuals using software STRUCTURE[36].
Linkage disequilibrium across the genome of Ammopiptanthus was calculated using TASSEL 4.0 software. LD decay for each chromosome was evaluated at a cut-off value of r 2 = 0.1. [37,38]. While the PCA was performed using Online analysis software (smart PCA, part of the EIGENSOFT v6.1).
samples. SLAF tags were mapped to the reference Ammopiptanthus genome using the BWA software, and 624,223 SLAF tags containing polymorphic SNPs were detected among the 111 Ammopiptanthus individuals ( Genetic diversity and Genetic differentiation of population Genetic diversity usually refers to the sum of genetic differentiation and variation among different individuals within a population or among different populations of a species [39,40]. Results of genetic diversity showed that the genetic diversity of Ammopiptanthus mongolicus (he = 0.32522, ho = 0.31129) was higher than that of Ammopiptanthus nanus (he = 0.26143, ho = 0.25834) ( Table 2). The population differentiation degree of Ammopiptanthus mongolicus was low (F ST =0.1248) ( indicating that Ammopiptanthus nanus population was highly differentiated and the variation mainly existed between populations. The variation of Ammopiptanthus mongolicus and Ammopiptanthus nanus exists between species, which is closely related to the geographical distance between the two species. In this study, we also assumed that cpDNA variation was in a drift-migration equilibrium [41], and considerable differentiation of the whole-genome among the Ammopiptanthus populations growing in different geographically remote areas and therefore a high degree of genetic disunity between them are evidenced by the high pairwise genetic distances(F ST , Table 3), as well as between the Ammopiptanthus mongolicus and Ammopiptanthus nanus groups (F ST = 0.5813, P < 0.0001).

Phylogenetic Relationships Based on SNP Data
Phylogenetic trees could re ect the evolutionary relationships of different individuals and groups, and which close relatives tend to gather together. So, In order to ascertain the divergence of 111 of Ammopiptanthus species during evolution, we performed the phylogenetic analysis, the results showed all species are clustered into ve distinct branches (Fig. 5), and that the majority of individuals can gather in the same group. The GS-MQ population is transplanting species out of ve population (GS-MQ, GS-HSY, NX-HW, NX-BG and NMG-DQH population). The Ammopiptanthus nanus from Xinjiang province gathered together, while Ammopiptanthus nanus from Gansu province can not gather to a family group.
Meanwhile, all Ammopiptanthus mongolicus gather to a big family group. The Ammopiptanthus mongolicus GS-HSY and NX-HW are clustered on a large branch, while the remaining communities in Ningxia gather on another branch. The majority of Ammopiptanthus mongolicus gathered together (Fig. 5). Finally, all results indicated that NX-BG and NMG-DQH have the highest homology.

Population Structure and Linkage Disequilibrium Analysis
To ascertain the divergence of the NMG-DQH, NMG-ALBLG, NMG-AZQ, NMG-QLG, NMG-DKTST, NX-BG, NX-BDG, NX-BJT, NX-HW, GS-HSY, XJ-KS, XJ-AHQ, XJ -BX, XJ-WYS and GS-MQ groups during evolution, principal component, population structure analyses were performed. Additionally, the sequence diversity of the FD, SD and ND germplasms was evaluated. All of the analyses indicated that there are strong divergence between the different Ammopiptanthus groups.
A population structure analysis using the Admixture program and SNP data revealed that 111 Ammopiptanthus germplasms were mainly divided into three groups according to the cross-validation error rate (Fig. 6). Of the three groups, group 1 comprised the most germplasms with 51 followed by group 2 (41 germplasms) and group 3 (19 germplasms). The Ammopiptanthus germplasms distributed in three groups, suggesting these Ammopiptanthus species were genetically diverse.
Meanwhile, Linkage disequilibrium (LD) analysis of these fteen groups revealed that the distance of LD decay in the NX-HW and XJ-WHS group is longer than that in the other groups (Fig. 7). The results indicated that Ammopiptanthus nanus had a more diverse genomic background.
The principal component analysis  (Fig. 8), although existing some degrees of introgression in these groups. All results indicated that a strong divergence between different Ammopiptanthus groups.

Discussion
High-density genetic maps play an essential role in facilitating discovery of functional genes and comparative analysis of genome structure [42]. However, most current genetic maps contain only about hundreds of markers, due to discovery technologies and genotyping costs [43,44]. Advances in genome sequencing technologies have paved the way for signi cant improvements in the rapid detection of genetic variation as well as the throughput and wealth of the information obtained. To date, a number of reduced representation sequencing methods have been developed, such as genotyping by sequencing (GBS), type IIB restriction site associated DNA (2b-RAD) [45], and SLAF-sEq. We chose the latter for this study because of its advantages, such as lower sequencing costs, higher genotyping accuracy, and e cient detection system.
Genetic diversity is the amount of variation observed between DNA sequences from distinct individuals of a given species [46]. This pivotal concept of population genetics has implications for species health, domestication, management and conservation. As Known to all, population diversity and structure is of fundamental importance for crop breeding programs. Normal passport (geography and/or pedigree) and phenotype data, traditionally used for the assessment of genetic architecture of the population, has been recently paralleled by the use of molecular markers. This is because the molecular markers allowed researchers to distinguish closely related samples and give more precise variation information among genotypes. SLAF-seq methods are currently used in a wide range of applications. These applications mainly include constructing high-density linkage maps [20,47], identifying candidate genes and mutant [48,49], drafting the genome [50] and constructing the core set for rare species [51]. However, the application of population genetics in endangered plant Ammopiptanthus has not been reported. In this study, a high quality of 111,735,304 80-bp long paired-end reads, 467,735 SLAFs and 1,261,501 SNPs were generated using high-throughput SLAF-sEq. Then we analyzed the genetic diversity(both morphology and molecular)and population structure of 111 kinds of Ammopiptanthus that grow in different locations of China in order to begin to lay a foundation for future yield-improvement endangered plant breeding programs.
Species that are considered threatened or endangered often exhibit low genetic diversity, especially when populations are small or disconnected [52]. Higher levels of genetic diversity allow species to adapt to environmental changes and reduces their susceptibility to catastrophic extinction events [53]. In this study, the 111 Ammopiptanthus lines were used as materials, by comparing the genetic diversity of Ammopiptanthus mongolicus and Ammopiptanthus nanus, most groups can be divided into the independent group, may be experienced the founder effect or the result of the bottleneck effect, GS-MQ group as compared to other Ammopiptanthus nanus on the related closer to the Ammopiptanthus mongolicus group, showing the geographical position close group is affected by the same climate, tending to be more consistent in evolutionary history.
The species evolution is largely related to the genetic diversity. Generally, species with high genetic diversity are better able to adapt to changing ecological environment, while species with low genetic diversity are less adaptable to the environment and are passive in the permanent evolution. In this study, the genetic diversity and genetic structure of different geographic populations in Ammopiptanthus mongolicus and Ammopiptanthus nanus was explored by SLAF for the rst time. Previous studies have shown that the heterozygosity is an important measure of overall genetic diversity. Our results showed that the genetic diversity of Ammopiptanthus mongolicus (he = 0.32522, ho = 0.31129) was higher than that of Ammopiptanthus nanus (he = 0.26143, ho = 0.25834). The population differentiation degree of Ammopiptanthus mongolicus was low (F ST =0.1248), indicating that the population differentiattion was moderate and the variation mainly existed in the population. While the differentiation degree of Ammopiptanthus nanus population in Xinjiang was high (F ST = 0.6989), indicating that Ammopiptanthus nanus population was highly differentiated and the variation mainly existed between populations. The variation of Ammopiptanthus mongolicus and Ammopiptanthus nanus exists between species, which is closely related to the geographical distance between the two species.
Gene ow are integral to interpretation of microevolutionary patterns and geographic structure. Through gene ow, we strive to gain insights into evelutionary independence and potential for population diversi cation, differentiation and ultimately speciation. Gene ow can halt or reduce genetic divergence that accumulates due to population isolation. In this study, the gene ow value (Nm) among Ammopiptanthus mongolicus population was 3.506, it indicates that genetic differentiation between Ammopiptanthus mongolicus populations was small, while the gene ow value (Nm) among Ammopiptanthus nanus population was 0.431, its genetic differentiation is higher than Ammopiptanthus mongolicus. Overall, the average of gene ow (Nm) was 0.360, indicating that the gene ow level of the whole population was low and the genetic differentiation was high. The main reason may be related to geographical distribution and interspeci c isolation. Our analysis results are consistent with the previous studies.
Recently, Ding [54] found that the mountain and monsoon are mainly reasons of terrestrial biodiversity, and by the early to middle Miocene, accelerated diversi cation and colonization of adjacent regions were likely driven jointly by mountain building and intensi cation of the monsoon. Similarly, based on our research, we speculate on a possible migration model for Ammopiptanthus species in China. The possible migration route may be from Xinjiang to Gansu, then to Ningxia and Inner Mongolia. It is why we can nd some Ammopiptanthus nanus species in Mingqin county of Gansu province. In addition, introduction and cultivation are also the reasons of terrestrial biodiversity, this accuracy of this model needs to be veri ed in future experiments.
In summary, a SLAF-seq approach for Ammopiptanthus was developed to reveal the genetic diversity, genetic structure and relationship between Ammopiptanthus mongolicus and Ammopiptanthus nanus. Meanwhile, SNP including hundreds of genes between Ammopiptanthus mongolicus and Ammopiptanthus nanus were identifed further. Consequently, our study not only provides a cost-effective approach for Ammopiptanthus genome-wide screening, but also contribute to understanding the diversity and origin of Ammopiptanthus in future research.

Conclusions
The aim of this study was to reveal the genetic diversity and underlying adaptation mechanisms in Ammopiptanthus. Overall, our results indicated that all fteen populations showing medium genetic diversity, with PIC values ranging from 0.1648 to 0.3081. AMOVA and Fst indicated that a low genetic differentiation existed among populations. Phylogenetic analysis showed that NX-BG and NMG-DQH of fteen populations have the highest homology, while the genetic structure analysis revealed that these Ammopiptanthus germplasm accessions were structured primarily along the basis of their geographic collection, and that an extensive admixture occurred in each group. In addition, the genome-wide linkage disequilibrium (LD) and principal component analysis showed that Ammopiptanthus nanus had a more diverse genomic background, and all genetic populations were clearly distinguished, although different degrees of introgression were detected in these groups.

Declarations
Authors' contributions YD designed the study. GC, YD, PJ, ZD, FK, JS and YH performed experiments. GC and YD analyzed the SLAF-seq data and drafted the manuscript. GC, YD and ZD revised the manuscript. All authors contributed to the article and approved the submitted version.

Availability of data and materials
All the data pertaining to the present study has been included in table and/or gure form in the present manuscript and authors are pleased to share analyzed/raw data and plant materials upon reasonable request.
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.