Development of genomic simple sequence repeat markers for Glycyrrhiza lepidota and cross-amplification of other Glycyrrhiza species

Plant material and DNA isolation

The materials used in the present study comprised 11 accessions of G. uralensis and G. glabra obtained from the Ginseng Research Division at the National Institute of Horticultural and Herbal Science (NIHHS), South Korea, and 12 accessions obtained from the United States Department of Agriculture (USDA), including G. uralensis, G. glabra, G. lepidota, G. echinata, and Glycyrrhiza spp. (Table 1). Seedlings of each licorice accession were grown in pots of sterile soil in a greenhouse with three plants per accession. Leaf samples were purposively sampled from three plants per accession from 4-week-old seedlings and ground to powder using liquid nitrogen in a pestle–mortar; their genomic DNA was extracted using the Plant gDNA Extraction Kit (GeneAll, Seoul, Korea) following the manufacturer’s protocol.

Primer design and polymerase chain reaction

In our previous study, the chloroplast genome of G. lepidota (NCBI accession no. KY038482) was obtained by the de novo assembly of the low-coverage whole-genome sequence via a bioinformatics pipeline (http://phyzen.com) (Raveendar et al., 2017). We mined SSRs based on a whole-genome sequence that was not used in the chloroplast genome analysis of G. lepidota. The whole-genome sequence of G. lepidota were searched for SSRs using the MIcroSAtellite (MISA, http://pgrc.ipk-gatersleben.de/misa/) with the search criteria for minimum number of repeats set at six for mono-nucleotide repeats, five for di-nucleotide repeats, four for tri-nucleotide repeats, and three each for tetra-, penta-, hexa-, hepta-, octa-, nona- and deca-nucleotide motifs. Next, 100 SSR primer pairs with di- and tri-nucleotide repeat motifs were randomly designed using Primer 3.0 (http://primer3.sourceforge.net/) with the following criteria: primer length (18–23 bp, with optimum value 20 bp); Tm (54 °C–56 °C, with optimum value 55 °C); GC content (40%–60%, with the optimum value 50%); maximum Tm difference between forward and reverse primer 1.5 °C and product size range (100–350 bp with optimum value 250 bp). The first polymerase chain reaction (PCR) using all the primer sets was performed for four accessions of G. lepidota (CBG20-23) to establish the PCR conditions. The primers resulting in successful amplifications were then applied to assess the genetic diversity of the 23 Glycyrrhiza spp. accessions. The PCR mixture (total volume, 40 µL) contained 20 ng genomic DNA, 10 pmol each primer, 2.5 mM MgCl₂, 0.25 mM dNTPs, and 0.5 U Taq polymerase (Inclone, Deajeon, Korea). The amplification was performed on a CFX96 PCR detection system (Bio-Rad Laboratories, Hercules, CA, USA) and included 30 cycles of pre-denaturation at 94 °C for 5 min, denaturation at 94 °C for 30 s, annealing at 55–60 °C for 45 s, and extension at 72 °C for 1 min. The PCR products were separated and visualized using a Fragment Analyzer (Agilent Technologies, Santa Clara, CA, USA).

Data analyses

The number of alleles, major allele frequency, genetic diversity, observed heterozygosity, and polymorphism information content (PIC) were analyzed using PowerMarker 3.25 (https://brcwebportal.cos.ncsu.edu/powermarker/). The rate of cross-amplification (transferability) among the 23 licorice accessions was measured using the following equation: Transferability (%) = amplicons (bands amplified by PCR) × 100/theoretical amplicons (primer number × sample size) (Lee et al., 2015; Raveendar et al., 2015). Phylogenetic analysis was performed using the Cavalli-Sforza chord distance (Cavalli-Sforza & Edwards, 1967) included in PowerMarker; the phylogenetic tree was constructed in MEGA4 (Tamura et al., 2007) using the unweighted pair group method with arithmetic mean.

Diversity of SSR markers and interspecies cross-amplification

From the whole-genome sequence of G. lepidota, 249,492 SSRs were mined. The frequency distribution of mined SSR repeat types is presented in Table S1. It was observed that mononucleotide repeats (218,320; 87.5%) were the most abundant repeat type, followed by dinucleotide repeats (11,056; 4.4%), tetra-nucleotide repeats (9,365; 3.8%), tri-nucleotide repeats (7580; 3%), penta-nucleotide repeats (3,170; 1.3%), and hexa-nucleotide repeats (1,387; 0.6%) in G. lepidota whole-genome, respectively (Table S1).

Sixty-two of the 100 selected primers amplified all the four accessions of G. lepidota (CBG20-23). Thus, these 62 primers were used to analyze the diversity of the 23 licorice accessions, which resulted in the identification of 549 alleles; the number of alleles per marker ranged from 2 (GL-gSSR-019) to 23 (GL-gSSR-028), with a mean of 9.6 alleles. The major allele frequency ranged from 0.11 (GL-gSSR-088) to 0.91 (GL-gSSR-019), with a mean of 0.349. The observed heterozygosity ranged from 0 (GL-gSSR-006, -019, -023, -068, -090, -097, and -100) to 0.70 (GL-gSSR-095), with a mean of 0.264. The maximum values of genetic diversity and PIC, indicating the genetic diversity of the markers, were 0.94 and 0.93, respectively, in GL-gSSR-028, whereas the minimum values were 0.17 and 0.15, respectively, in GL-gSSR-019. The mean values of genetic diversity and PIC were 0.760 and 0.730, respectively (Table S2). Cross-amplification analysis among licorice species revealed that G. echinata had the highest transferability (93.5%), followed by G. glabra (91.6%) and G. uralensis (91.1%); the mean value was 92.07% (Table 2).

Phylogenetic analysis

The 23 accessions were classified into two clusters (Fig. 1), one corresponding to the cultivated plant species (group 1) and the other to the wild species (group 2). The 15 accessions within group 1 were arranged into three subgroups: group 1–1 contained two G. glabra accessions; group 1–2 contained six G. uralensis accessions, one G. glabra accession, and one Glycyrrhiza sp. accession; and group 1-3 contained two G. glabra accessions and three Glycyrrhiza sp. accessions. The seven accessions within group 2 were arranged into two subgroups, groups 2–1 and 2–2, which contained six G. lepidota and two G. echinata accessions, respectively.

Diversity of the SSR markers

The mean PIC value of the SSR markers developed in the present study using wild (G. lepidota and G. echinata) and cultivated (G. uralensis and G. glabra) licorice accessions was 0.730, which is higher than that reported for other medicinal plants: 0.314 for Zingiber officinale (Pandotra et al., 2013), 0.57 for Codonopsis lanceolata (Kim et al., 2016), and 0.272 for Festuca arundinacea (Tehrani et al., 2009). These results showed that the SSR markers developed in the present study have a higher diversity than those developed for other species. Moreover, Um et al. (2016) reported the PIC value of 0.56 for eight SSR markers based on G. uralensis genome using 22 accessions of cultivated licorice species (G. uralensis and G. glabra). To compare the markers developed in the present study with those previously reported, the PIC values of the 11 accessions of cultivated licorice species (G. uralensis and G. glabra) were measured, and a mean value of 0.624 was obtained. Thus, the SSR markers developed in the present study had a higher mean PIC value than those previously developed for G. uralensis, despite the lower number of accessions and higher number of primers used here. This might be attributed to the fact that the primers with PIC value higher than 0.63 accounted for 66.13% of the total primers. These results suggested that the markers developed in the present study should be more efficient for molecular genetic studies than those reported in previous studies due to their higher diversity.

In the diversity analysis based on repeat motifs, the mean PIC values of dinucleotide repeat motifs (32 sets of primers) and trinucleotide repeat motifs (30 sets of primers) were 0.796 and 0.688, respectively, showing a higher value for the dinucleotide repeat motifs. The higher mean PIC value for dinucleotide repeats might be related to the fact that dinucleotide repeats are distributed throughout the genome, whereas trinucleotide repeats are mostly present in coding regions. Furthermore, trinucleotide repeats are under a relatively weaker selection pressure against mutations that alter the reading frame compared with the dinucleotide repeats, leading to a lower genetic diversity in the trinucleotide repeats (Kalia et al., 2011; Vieira et al., 2016). Considering that dinucleotide repeats had higher PIC values than trinucleotide repeats, despite the similar number of primers, the primers based on dinucleotide repeats are expected to be more useful for studies regarding diversity analysis.

Interspecies cross-amplification

The mean interspecies transferability of licorice was 92.5%. The mean interspecies transferabilities of Rubus coreanus and Allium sativum were 73.52% (Lee et al., 2015) and 58.85% (Lee et al., 2011), respectively, and that of Triticum aestivum and Secale cereale were 48.4% and 35.3%, respectively (Kuleung, Baenziger & Dweikat, 2004). According to the report of Erayman et al. (2014), the transferabilities of SSR markers developed for Medicago truncatula, Phaseolus vulgaris, and Cicer arietinum to G. glabra, G. echinata, and G. flavescens were 33% for M. truncatula, 11% for P. vulgaris, and 6% for C. arietinum. These results indicated that the markers developed for Glycyrrhiza accessions in the present study had higher transferability than those developed for other species. Interspecies crossing rates between G. lepidota and other licorice species were 51–75% (Ashurmetov, 1996), suggesting that crossing may occur under natural conditions, and genes of G. lepidota could affect other licorice species. We postulated that these are the likely reasons why interspecies transferability has become relatively high. In particular, G. lepidota had a high transferability with G. echinata, another wild species, possibly because these are more closely related to each other than to cultivated species. Such results showed that the SSR markers developed for the wild species G. lepidota can be efficiently used for other licorice species. However, as the number of resources used in the present study was smaller than that in other studies, future studies should include various species.

Phylogenetic analysis of licorice accessions

In the present study, although CBG4 and CBG6 (G. glabra accessions) clustered with CBG16 and CBG17 (also G. glabra accessions), they were separated into groups 1–1 and 1–3, respectively. Such clustering could be due to the random selection of markers. In addition, the separation of G. glabra into two subgroups might be due to their genetic differences arising from their production origins. The four Glycyrrhiza spp. accessions CGB08, CGB09, CGB10, and CGB11 were separated into two groups, suggesting that CBG09 is genetically closer to G. uralensis, whereas CBG08, CGB10, and CGB11 are genetically closer to G. glabra. Accession CBG3 (G. glabra) clustered with G. uralensis, which might be due to either incomplete identification of resources during collection or insufficient numbers of licorice accessions (Jo et al., 2018; Lee et al., 2019). However, Jo et al. (2018) estimated that the wild species G. echinata was genetically close to the cultivated species G. uralensis.