Development of EST-SSR markers based on transcriptome for genetic analysis in Radix Ardisia

Introduction

Radix Ardisia is a commonly used medicine of the Miao people in Guizhou Province. It is regarded as a good throat medicine by the Miao people and is the main component of the Miao medicine’s proven prescription (Zhao & Du, 2016; Wu, 2012). The primary source of the medicinal materials of Radix Ardisia is wild. Due to the lack of effective identification methods, the types and contents of the effective ingredients in therapeutic components from different origins are quite different (Wang et al., 2020), and the quality of the medicinal materials is difficult to guarantee. With the continuous improvement of therapeutic value and the increasing scarcity of wild resources, it is urgent to establish an effective identification method, carry out an in-depth analysis of its genetic structure, clarify the origin and genetic relationship, and carry out research on the breeding practice, which is of great significance for the protection of germplasm resources and the safety of clinical medicine.

Due to the lag of genetic research on Radix Ardisia, there are few reports on the genetic diversity analysis. Jiang et al. (2006), Yuan et al. (2024), and Tao, Liao & Luo (2011) use molecular markers to study the genetic diversity of Radix Ardisia and found that the genetic polymorphism within Radix Ardisia is high and the variation is rich. Zhang, Guo & Yang (2011), Wang (2012) and Chen et al. (2017) used sporopollen characteristics and DNA barcode technology to identify the species of Ardisia, and the results showed that the matK sequence can better identify the species of Ardisia. However, more data are needed for accurate identification and genetic research of plants within the genus. Expressed sequence tag simple sequence repeat (EST-SSR) markers have the characteristics of simple technology, good repeatability, high polymorphism and codominant inheritance in genome SSR markers but also have the advantages of cheap primer development, good versatility, clear bands and accessible statistics. In addition, as EST-SSR is a part of the coding gene, it can directly obtain the relevant information of gene expression, which may directly identify the alleles that determine critical phenotypic traits. With the development of transcriptomics technology, using its sequence to develop whole genome EST-SSR molecular markers has also become a very simple and effective method, which has been widely used in the identification of medicinal plant germplasm resources, genetic mapping analysis, polymorphism analysis, phylogenetic analysis, etc. in many plants (Zhang et al., 2023; Shi et al., 2023). Singh, Sane & Thuppil (2024) have shown that EST-SSR is an effective means for classification and identification, genetic diversity, core collection screening, etc. However, no relevant research has been carried out in the study of Radix Ardisia.

Therefore, this study intends to develop genome-wide EST-SSR primers using the transcriptome data of Ardisia crenata Sims, apply them to genetic research, analyse the genetic diversity of Ardisia species and the genetic structure of the population, to provide a reference for the identification of germplasm resources.

Materials & Methods

Results

Development and validation of the EST-SSR markers

In this study, 51,237 sequences were retrieved (Table S1), with a total length of 71.4 MB. A total of 32,827 SSR loci were detected (Table S2), averaging one SSR locus per 2.1 KB. The distribution of primers was detected. 28,322 SSR loci were unigene SSR, of which 9,455 were located in the 3′-UTR (Untranslated regions) region, accounting for 28.77%; 9,558 SSR loci were located in the CDS (Coding DNA sequence) region, accounting for 29.12%; 9,309 SSR loci were located in the 5′-UTR region, accounting for 28.36% (Table S3). 32,827 pairs of genome-wide EST-SSR markers were developed (Table S4). On average, there were 0.64 pairs of EST-SSR primers per unigene, and the sequence coverage was high. The statistical analysis results show that (3.1), six different types of nucleotide SSRs can be detected, but the number and frequency of SSRs in different primitive types vary greatly (Table 2). Mononucleotide and dinucleotide were the main repeat types, accounting for 90.51% of the total SSR. Single nucleotide SSRs were 14,692 pairs, accounting for 44.76% of the total SSRs, including 7,465 for (A), 7,043 for (T), 106 for (G), and 78 for (C), mainly of type A and T (Fig. 1); Class A accounts for 50.81% of the total number of this type, and class T accounts for 47.93%. Secondly, the frequency of dinucleotide repetition also accounts for a large proportion. The number of dinucleotide SSRs is 15,019 pairs, accounting for 45.75%. The main types of dinucleotides are AG, AT, CT, GA, TA and TC, of which (AG) has 2,628, accounting for 17.50% of the total type, (AT) has 2,282, accounting for 15.19%, (CT) has 1707, accounting for 11.37%, (GA) has 2,380, accounting for 15.85%, (TA) has 2,128, accounting for 14.17%, (TC) has 2,348, accounting for 15.63%. There are 2,337 pairs of trinucleotide SSRs, accounting for 7.12%, including 204 (AAT), 100 (GAT), 138 (TTA) and 101 (ATT). The number of tetranucleotide, pentanucleotide and hexanucleotide repeat types is relatively small, which are 544 pairs, 109 pairs and 126 pairs, respectively, accounting for 1.66%, 0.33% and 0.38% of the total SSR (Fig. 2). There are 55 main types of tetranucleotides (ATAC) and 53 tetranucleotides (AAAT). The main types of pentanucleotide are (AAAAT) and (CCAAA), with six, respectively, and the main types of hexanucleotide are (GAGAGG) with three.

Table 2:

Distribution characteristics of repeat motif types in the whole genome.

Type	Annotation	No.	Proportion
P1	Single nucleotide repeats	14,692	44.76%
P2	Dinucleotide repeats	15,019	45.75%
P3	Trinucleotide repeats	2,337	7.12%
P4	Tetranucleotide repeats	544	1.66%
p5	Pentanucleotide repeats	109	0.33%
p6	Hexanucleotide repeat	126	0.38%
Total		32,827	100%

DOI: 10.7717/peerj.19560/table-2

Genetic diversity analysis

Sixty pairs of primers (Table S5), such as SSR1, SSR4, and SSR5, were randomly selected from the whole genome of Radix Ardisia, and PCR amplification was carried out for it. A total of 200 bands (Table S6) were amplified by 51 primers, with an average of 3.92 bands amplified by each primer, and the polymorphism rate (PPL) was 100%. The results of genetic diversity analysis showed that the genetic similarity coefficient between samples ranged from 0.6200 to 0.9350. At the species level, the number of polymorphic loci in the genus Ardisia was 200, the percentage of polymorphic loci (PPL) was 100.00%, and the total gene diversity (Ht) was 0.2181. The average number of observed alleles (Na) was 2.0000; the average number of effective alleles (Ne) was 1.3359; the average Nei’s gene diversity index (H) was 0.2181; the average Shannon’s polymorphism information index (I) was 0.3550 (Table 3). The results showed that 46 species of Ardisia had abundant genetic diversity and genetic variation among species and within species.

The principal component cluster analysis results of 46 samples of Radix Ardisia were obtained. The individuals were clustered into three dimensions by principal coordinates analysis (PCoA) (Fig. 2). PC1 represents the first principal component, PC2 represents the second principal component, and PC3 represents the third principal component. The variance contribution rates of PC1, PC2, and PC3 were 10.3%, 8.9%, and 7.07%, respectively. The population of Sapindus was split into four clusters, which corresponded approximately to the ZS group (Zhushagen, Ardisia crenata Sims), the HL group (Hongliangsan, Ardisia crenata Sims var. bicolor (Walker) C. Y. Wu et C. Chen), the BH group (Baihuazijinniu, Ardisia merrillii E. Walker), and the BL (Bailiangjin, Ardisia crispa (Thunb.) A. DC.), DF (Dongfangzijinniu, Ardisia elliptica Thunb.), GG (Jiuguanxue, Ardisia brevicaulis Diels), HS (Hushehong, Ardisia mamillata Hance), LZ (Lianzuozijinniu, Ardisia primulifolia Gardner & Champ.), XB (Xibingbailingjin, Ardisia crispa (Thunb.) A. DC. var. dielsii (Levl.) Walker), ZJ (Zijinniu, Ardisia japonica (Thunb.) Blume) group.

Based on the EST-SSR primers developed, the population structure was analysed through the Structure software. In contrast to the PCoA results, when K = 2, the error rate of cross-validation is the minimum, so the optimal number of clusters is 2. 46 germplasm resources are divided into two subgroups, which shows a large genetic difference between the species, forming a significant genetic differentiation of the population (Fig. 3). When K was equal to 3–6, individuals of subgroup 2 were consistently divided into several subgroups.

Table 3:

Genetic diversity analysis of Radix Ardisia.

Locus	Sample size	na	ne	h	I	Ht
Mean	46	2	1.3359	0.2181	0.355	0.2181
St. Dev		0	0.2995	0.1516	0.1968	0.023

DOI: 10.7717/peerj.19560/table-3

The cluster analysis (Table S7) showed that Ardisia’s genetic distance ranged from 0.106 to 1.242. The results showed that EST-SSR could accurately distinguish the species of Ardisia and its relatives (Fig. 4); Zhushagen (Ardisia crenata Sims) and Hongliangsan (Ardisia crenata Sims var. bicolor (Walker) C. Y. Wu et C. Chen) are gathered into one branch, and Bailiangjin, Xibingbailiangjin, Dayebailiangjin, Zijinniu (Ardisia japonica (Thunb.) Blume), Baihua Zijinniu (Ardisia merrillii E. Walker), Dongfang Zijinniu (Ardisia elliptica Thunb.), Jiuguanxue (Ardisia brevicaulis Diels), and Hushehong (Ardisia mamillata Hance) are gathered into one branch respectively. Bailiangjin, Zijinniu, Baihua Zijinniu and Jiuguanxue can be clearly distinguished, and they are far away from Zhushagen and Hongliangsan, indicating that they are closely related, and their differentiation has prominent regional characteristics. At the same time, the genetic distance between Zhushagen, Hongliangsan and other species is far, and the species can be clearly distinguished. This shows that the EST-SSR used in the research can distinguish the Radix Ardisia and its easily mixed species of Ardisia, which can be applied to the identification. However, other data must be explored to distinguish Zhushagen and Hongliangsan.

Discussion

As a natural attribute formed in organisms’ long-term evolution and development, genetic diversity is reflected not only among populations and within populations but also among individuals, including population and individual genetic variation (Wang et al., 2023). The genetic diversity of a species is not only the result of its long-term survival but also the result of evolution and adaptation (Parveen et al., 2022). Generally, the average value of the nucleotide diversity index of inbred species is 0.51, and that of outcrossing species is less than 0.1 (Wen, Han & Wu, 2010). In this study, the nucleotide diversity index of Ardisia is less than 0.1, which indicates that the cinnabar root is mainly cross-pollinated, and most of the genetic variation exists between populations.

The Radix Ardisia is the dried root and rhizome of Zhushagen, Hongliangsan and Bailiangjin. In the new edition of Flora of China, Zhushagen and Hongliangsan are combined into one species. In terms of appearance, the two are different in red or purple red. However, in the process of market use, the root is usually used, but the root cannot be identified through its appearance. Therefore, it needs to be accurately identified by microscopy (Chen, Tong & Zhong, 2020), molecular identification (Zhang, Xing & Xue, 2021; Wei et al., 2020) and other technologies; in this experiment, the genetic diversity of the three original plants of the plant were studied by EST-SSR markers. The intraspecific genetic distance of the three original plants was less than the interspecific genetic distance of the plant of the same genus, which showed significant differences between the medicinal material of the plant of the same genus at the molecular level. It could be well separated by K2P genetic distance. At the same time, the evolution tree constructed by the neighbour-joining method showed that the medicinal materials of Radix Ardisia could be effectively identified from other species of the same genus. However, Zhushagen and Hongliangsan can not be well distinguished, possibly because Hongliangsan is a variety of Zhushagen. There is a crossover phenomenon in genetic evolution.

In this study, transcriptome data used to develop EST-SSR molecular markers for genetic research can be better used in population structure analysis and identification research. EST-SSR is a powerful tool for genetic analysis. In addition to the advantages of high polymorphism, codominant inheritance, simple technology and good repeatability in gSSR markers, EST-SSR markers also have many other advantages. For example, EST-SSR markers can reduce the cost of primer development and significantly improve the utilization efficiency of existing sequencing data, making marker development more economical and practical. Expressed sequence tag (EST) is a partial sequence of cDNA, representing a partial sequence of a complete gene. Therefore, SSR markers contained in EST can be used as direct markers of specific traits or genes. At the same time, EST-SSR flanking sequences are highly conservative and have good universality among different species of the same genus. Therefore, EST-SSR markers can be developed for those species with less research, less EST sequence information, or no EST data from their relatives. Developing molecular markers using EST sequences is a better way to develop and utilize EST databases. With the continuous enrichment of EST databases, developing SSR markers using EST sequences has become a simple and effective method. The development and research technology of EST-SSR marker is relatively mature and widely used in plants.

The EST-SSR sequence is a part of the coding gene, and its flanking sequence is highly conservative, so the primers designed can often be used for other species of the same genus or even family, which has been reported in many Rosaceae plants (Preethi et al., 2024; Lu et al., 2021; Liu et al., 2022). With 60 pairs of primers, 51 pairs of primers were found to have a good polymorphism in the amplification of 46 germplasm materials of Ardisia, indicating that the SSR flanking sequence in the target gene (target band) amplified by the 51 pairs of primers was highly conservative, which could not only be used for the study of the related molecular of Radix Ardisia but also be used for the further study of germplasm identification and genetic diversity analysis of other Ardisia plants. In addition, although the target fragment (main band) amplified by 51 pairs of primers is consistent with the expected product size, it may also contain false positive or amplification fragments without SSR repeat sequence, which must be sequenced.

Based on the extensive capacity transcriptome library of Zhuashagen that has been constructed, 32,827 SSR loci were developed for the first time through bioinformatics methods, with an average of one SSR locus every 2.1 Kb. This indicates that the EST sequence in Radix Ardisia contains rich repeats like other conventional crops, which provides an essential basis for identifying SSR in traditional Chinese medicine crops; it provides a basis for enriching the application of SSR in the role of traditional Chinese medicine. The dinucleotide content is the highest among the SSR sites developed in the selected dinucleotide repeats. The number of AG/GA/TC is the largest in different repeat units, consistent with the results of most species studies. This study also analysed the distribution characteristics of EST-SSR, which showed that 28.77% of EST-SSR was located in 3′-UTR, 28.36% of SSR was located in 5′-UTR and 29.12% of SSR was located in the CDS region, which was consistent with the distribution characteristics of EST-SSR in gene positions found in the study of dicotyledons by Kumpatla and Mukhopadhyay. It can be seen that EST-SSR has a strong conservatism in biological evolution.

Due to the lack of effective protection measures for the germplasm resources of Ardisia in China for a long time, there is still no unified standard for the effective division of varieties and germplasm resources in China. This is precisely why many cultivated varieties or wild germplasm resources containing excellent genetic information in the production of Ardisia are gradually lost and not used. It is difficult to distinguish or identify many varieties using traditional morphological identification methods. The existing markers, such as RAPD and ITS2, have been identified as effective in Rehmannia glutinosa. However, because it is difficult to distinguish the effective heterozygotes of germplasm, and these markers have complex steps and relatively high requirements, their promotion and utilisation in growth practice are affected. Therefore, there is an urgent need for a modern molecular marker method to carry out the most fundamental differentiation and identification from its “internal” way. Among the molecular markers used, only SSR is the most simple and economical way. In Zijinniu, due to the fuzzy genomic information and the few available genetic resources, the application of SSR in variety breeding is minimal; this research is the first large-scale development of practical EST-SSR markers with polymorphism of Ardisia based on the transcriptome. At the same time, this research is also the first to carry out preliminary systematic genetic clustering of existing germplasm resources, which will play an important role in the subsequent cultivar cultivation and identification of germplasm resources.

Conclusions

The results of systematic clustering showed that the EST-SSR used could well distinguish Radix Ardisia and its easily mixed species; it can be applied to the identification of Radix Ardisia, as well as the molecular identification between the original Bailiangjin, Zhushagen and Hongliangsan. This study can provide a reference for the genetic analysis of the Radix Ardisia.

Supplemental Information