Construction of fingerprinting for tea plant (Camellia sinensis) accessions using new genomic SSR markers

Transcription

1 Mol Breeding (2017) 37:93 DOI /s y Construction of fingerprinting for tea plant (Camellia sinensis) accessions using new genomic SSR markers Shengrui Liu & Hongwei Liu & Ailin Wu & Yan Hou & Yanlin An & Chaoling Wei Received: 30 November 2016 /Accepted: 20 June 2017 # Springer Science+Business Media B.V Abstract As one of the most popular non-alcoholic beverage crops, the tea plant (Camellia sinensis) plays an important role in human health and lifestyle. Genetic fingerprinting based on genomic-derived markers in tea, however, is still in the initial stages, which has limited tea germplasm resource utilization and cultivar protection. In the current study, we identified whole genome-based simple sequence repeat (SSR) loci and successfully developed 36 new genomic SSR markers, which are highly polymorphic with average allele number and polymorphic information content (PIC) of 14.9 and 0.862, respectively. A phylogenetic tree for 80 tea plant accessions was subsequently constructed based on their genotypic scores for these 36 markers. The phylogenetic relationships among the 80 accessions are highly consistent with their genetic backgrounds or original places. Noteworthy, robust fingerprinting power was performed, and the overall probability of finding two random individuals sharing identical genotypes across the 36 loci was estimated to be We subsequently identified five SSR markers as a recommended core marker set for fingerprinting the tea plant Shengrui Liu and Hongwei Liu contributed equally to this work Electronic supplementary material The online version of this article (doi: /s y) contains supplementary material, which is available to authorized users. S. Liu : H. Liu : A. Wu : Y. Hou : Y. An : C. Wei (*) State Key Laboratory of Tea Plant Biology and Utilization/Key Laboratory of Tea Biology and Processing, Ministry of Agriculture, Anhui Agricultural University, West 130 Changjiang Road, Hefei Anhui, People s Republic of China weichaoling0551@163.com cultivars or accessions. The combined PI and PIsibs of the marker set were and , respectively, which allowed us to fully discriminate all 80 tea plant accessions from one another. The SSR markers developed here will provide a valuable resource for tea plant genetics and genomic studies, as well as breeding programs. The fingerprinting profiles can serve as a database that is essential for the tea industry and commercial breeding, and for tea plant cultivar identification, utilization, and protection. Keywords Tea (Camellia sinensis). SSR markers. Fingerprinting. Phylogeny Introduction As one of the most popular non-alcoholic beverages, tea possesses numerous important properties including attractive aroma, refreshing taste, desirable physiological functions, and myriad health benefits (Sharangi 2009; Yang et al. 2009). Tea plant (Camellia sinensis) is a perennial diploid evergreen woody plant(2n = 30)that originated in Yunnan province and neighboring regions in southwestern China (Hashimoto and Takasi 1978; Chen et al. 2000). Today, the cultivated tea plant varieties are extensively grown in tropical and subtropical regions across the world, such as China, India, Kenya, Japan, Pakistan, and Argentina, where they provide crucial revenue resources and job opportunities and contribute enormously to the local economy (Yao et al. 2012). These most commercially cultivated varieties are C. sinensis

2 93 Page 2 of 14 Mol Breeding (2017) 37:93 var. sinensis and C. sinensis var. assamica of C. sinensis (L) O. Kuntze, which belongs to the section of Thea of the Camellia genus in the family Theaceae (Yang et al. 2016; Chenetal. 2000). C. sinensis var. sinensis for Chinese teas generally performed with shrub trees, small leaves, and highly cold tolerance, while C. sinensis var. assamica for Indian Assam teas characterized with semiarbor trees, large leaves, and poor cold tolerance. Several wild species and/or varieties such as C. crassicolumna, C. tachangensis F. C. Zhang, C. taliensis (W. W. Smith) Melchior, and C. gymnogyna Chang were only discovered in southwestern China. Though not cultivated commercially, these wild species represent important resources for tea germplasm conservation, utilization, and genetic improvement. Rather than being grown from seed, most cultivated tea plants are clonally propagated vegetative, because seed-grown plants exhibit genetic variation due to heterozygosity. As the most important country for tea cultivation and consumption, China designated 107 clonal tea cultivars as national tea cultivars (NTCs) and another 139 as provincial tea cultivars (PTCs) in late 2013 (Yang and Liang 2014). These outstanding tea plant cultivars are crucial for the modern tea industry because of their higher yield, better quality, and suitability for mechanized harvesting and processing (Tan et al. 2015). However, an efficient strategy for cultivar identification is urgently needed to ensure the varietal purity during propagation, commercial seedling identity in nurseries, and uniqueness of cultivars, thereby providing evidence for protecting the rights of tea breeders, farmers, and traders (Ujihara et al. 2011; Ujihara et al. 2009). There is also increasing requirement for authenticating premium tea plant cultivars in the market by identifying their cultivar information and prove of true cultivar. In recent years, it has become insufficient and impractical to identify various tea cultivars using the traditional method, in which experienced experts distinguish them based on morphological characters. Molecular markers that are based on DNA polymorphisms are appropriate and powerful tools for genomic and genetic studies including varietal genotyping and fingerprinting. Numerous molecular markers have been developed and extensively applied in genetic and genomic analysis in tea, such as restriction fragment length polymorphisms (RFLPs), random amplification of polymorphic DNAs (RAPDs), amplified fragment length polymorphisms (AFLPs), microsatellites or simple sequence repeats (SSRs), and inter-simple sequence repeats (ISSRs) (Ni et al. 2008; Mukhopadhyay et al. 2016). Of these, SSRbased markers have gained considerable importance in tea plant genetics and breeding, because of their codominant inheritance, multi-allelic nature, high abundance in the genome, and extensive genome coverage (Bali et al. 2013;Liuet al.2013). SSR markers have been frequently used in C. sinensis in genetic diversity and population structure analysis, investigations into the phylogeny and genetic origin of tea plant species, and the construction of genetic linkage maps (Fang et al. 2012; Yao et al. 2012; Taniguchi et al. 2012; Ma et al. 2014; Bali et al. 2015; Tan et al. 2016;Wambulwaetal.2016). Several studies have used SSR markers for fingerprinting of tea cultivars. A previous study showed that a combination of four SSR markers and four Indel (Insertion/ Deletion) markers were sufficient to distinguish 44 Japanese tea cultivars (Kaundun and Matsumoto 2004); Ujihara et al. (2009) reported that at least three SSR markers were required to discriminate 16 Japanese tea genotypes and one Chinese tea cultivar (Ujihara et al. 2009); Tan et al. (2016) demonstrated that eight out of 30 SSR markers as a core marker set can unambiguously fingerprinting 128 Chinese clonal tea cultivars. Nevertheless, to identify tea cultivars practically, fingerprinting data should be analyzed additionally, which has frequently been neglected. Thus far, these markers are insufficient to uniquely identify a large number of different tea cultivars, especially when new cultivars or some unknown wild species are released. The current study was endeavor toward three principal objectives: (1) to develop a set of highly polymorphic SSR markers based on tea whole genome sequence; (2) to construct a phylogenetic tree for the 80 selected tea accessions, which will be helpful to understand their relationships and genetic origins, and to validate the utility of these markers; (3) to identify a core set of markers for fingerprinting these tea accessions, which will allow for characterization of cultivars and protection of the rights of tea plant breeders or farmers. Materials and methods Plant materials and DNA extraction We sampled a total of 80 Chinese tea plant accessions, which were collected from 13 provinces that are the main tea-growing regions in China (Online Resource

3 Mol Breeding (2017) 37:93 Page 3 of ). Detailed information of the samples including accession name, subspecies, germplasm type, and cultivation region is available (Table 1). Of the 80 accessions, 24, 22, and 34 accessions belonged to the NTC, PTC, and LTC groups, respectively. Young leaves were collected from all accessions, immediately frozen in liquid nitrogen, and then stored at 70 C until utilization. Total genomic DNA was extracted using the EZgene CP Plant Miniprep Kit (Biomiga, San Diego, USA) according to the manufacturer s protocol. The quality and quantity of the DNA samples were detected using 1% agarose gels and the NanoDrop 2000 UV-Vis spectrophotometer, respectively. The concentration of each DNA sample was adjusted to approximate 400 ng/μl, and a small amount of each sample was diluted to ng/μl for use in consequent PCR amplifications. SSR marker development and genotyping To develop new markers, we screened the genome sequence of C. sinensis (unpublished data) for SSR loci using the MISA program ( de/misa/). To avoid previously developed, we aligned the sequences containing SSRs with the DNA sequences that were used for developing EST and genomic SSR markers previously. A total of 180 SSR loci were selected, which contained five types of repeat motifs (di-nucleotide, tri-nucleotide, tetra-nucleotide, penta-nucleotide, and hexa-nucleotide) and are distributed on different genomic scaffolds. PCR primers were designed based on the sequences flanking the SSR loci using the Primer 5.0 program, with expected amplicon sizes were between 100 and 300 base pairs (bp). To validate the primers, six tea accessions including BTieguanyin,^ BShuchazao,^ BLongjingchangye,^ BYungui,^ BYunkang 10,^ and BYunmei^ were selected for PCR amplification. Accessions BTieguanyin,^ BShuchazao,^ and BLongjingchangye^ belong to C. sinensis var. sinensis and originated in Fujian, Anhui, and Zhejiang provinces, respectively; whereas BYungui,^ BYunkang 10,^ and BYunmei^ belong to C. sinensis var. assamica and originated in Yunnan but have different genetic backgrounds. Initially, 1.5% agarose gels were used for screening of primer pairs; subsequently, primers that produced unambiguous bands were further screened against the six tea accessions using the Fragment Analyzer 96 (Advanced Analytical Technologies, Inc., Ames, IA) (data not shown). The resulting primers were then used for identification of the 80 tea accessions. PCR amplifications were performed in 20 μl reaction mixtures, containing 2 μl oftemplate DNA (20 50 ng/μl), 2 μl of 10 PCR buffer, 0.2 μl of dntps (10 mm), 0.5 μlofeachprimer(10μm), and 0.5 unit of Taq polymerase (Takara, Dalian, China). Thermocycling conditions were as follows: denaturation at 94 C for 4 min, 35 cycles of 40 s at 94 C, 30 s at C (depending on the primer pair), and 45 s at 72 C with a final extension at 72 C for 10 min (S1000 Thermal Cycler, Bio-Rad). The amplification products were initially examined by electrophoresis on 1.5% agarose gels for identification of primers. Amplified SSR-containing fragments were separated on a 96-capillary automated DNA fragment analyzer (Fragment Analyzer 96, Advanced Analytical Technologies, Inc., Ames, IA). All required reagents were from the double-stranded DNA kit DNF-900 (Advanced Analytical Technologies, Inc.) including FA dsdna gel, dsdna inlet buffer, dilution buffer 1 TE, intercalating dye, markers, DNA ladder (including 35, 75, 100, 150, 200, 250, 300, 400, and 500 bp), 5 capillary conditioning solution, capillary storage solution, and mineral oil. PCR products were diluted ten folds using 1 TE dilution buffer depending on the DNA concentration, dilution, and injection voltage which were adjusted to prevent overloading of the PCR products. Mixtures consisting of 2 μl PCR products and 18 μl of 1 TE were added to the plate wells, and 24 μl DNA ladder was included in the H12 position of the 96-well plate. This DNA fragment analyzer separates amplicons ranged from 35 to 500 bp, resolving 1- bp differences between distinct alleles. The data was normalized to the 35-bp lower marker and 500-bp upper markers and calibrated to the 75 to 400 bp range using PROSize 2.0 software (Advanced Analytical Technologies, Inc.). The PROSize 2.0 software included in the advanced Fragment Analyzer 96 system was used initially to visually select strong, clear polymorphic DNA bands for scoring. Data analysis Genetic statistics for the genomic SSR markers, including the number of alleles (N A ), major allele frequency (MAF), observed heterozygosity (H O ), polymorphic information content (PIC), and Nei s genetic distance (Nei et al. 1983), were calculated using PowerMarker (Liu

4 93 Page 4 of 14 Mol Breeding (2017) 37:93 Table 1 Detailed information for the 80 tea plant accessions used in this study Sample Accession name Species Germplasm Cultivation region 1 BXianyuzao^ C. sinensis var. sinensis PTC Anhui, China 2 BNongkangzao^ C. sinensis var. sinensis LTC Anhui, China 3 BFuzao 2^ C. sinensis var. sinensis NTC Anhui, China 4 BXianghongdian 1^ C. sinensis var. sinensis PTC Anhui, China 5 BDuokangxiang^ C. sinensis var. sinensis LTC Anhui, China 6 BAnhui 3^ C. sinensis var. sinensis NTC Anhui, China 7 BHuangshantezaoya^ C. sinensis var. sinensis PTC Anhui, China 8 BHuoshanjinjizhong^ C. sinensis var. sinensis LTC Anhui, China 9 BHuangshanyecha 18^ C. sinensis var. sinensis LTC Anhui, China 10 BHuangshanyecha 27^ C. sinensis var. sinensis LTC Anhui, China 11 BShidacha 1^ C. sinensis var. sinensis LTC Anhui, China 12 BShidacha 2^ C. sinensis var. sinensis LTC Anhui, China 13 BShidacha 6^ C. sinensis var. sinensis LTC Anhui, China 14 BShidacha benzhong^ C. sinensis var. sinensis LTC Anhui, China 15 BShidacha huangzhong^ C. sinensis var. sinensis LTC Anhui, China 16 BHuangkui^ C. sinensis var. sinensis PTC Anhui, China 17 BHuoshanhuangya^ C. sinensis var. sinensis NTC Anhui, China 18 BShuchazao^ C. sinensis var. sinensis NTC Anhui, China 19 BChunyu 2^ C. sinensis var. sinensis NTC Zhejiang, China 20 BLongjingchangye^ C. sinensis var. sinensis NTC Zhejiang, China 21 BJuhuachun^ C. sinensis var. sinensis LTC Zhejiang, China 22 BDayeyunfeng^ C. sinensis var. sinensis LTC Zhejiang, China 23 BJingfeng^ C. sinensis var. sinensis NTC Zhejiang, China 24 BTiantaihuangcha^ C. sinensis var. sinensis PTC Zhejiang, China 25 BLiuyezao^ C. sinensis var. sinensis PTC Hubei, China 26 BYichangdaye^ C. sinensis var. sinensis NTC Hubei, China 27 BEcha 1^ C. sinensis var. sinensis PTC Hubei, China 28 BJunshanyinzhen 1^ C. sinensis var. sinensis NTC Hunan, China 29 BHuangjincha 2^ C. sinensis var. sinensis PTC Hunan, China 30 BAnmingzao^ C. sinensis var. sinensis PTC Hunan, China 31 BTaoyuandaye^ C. sinensis var. sinensis PTC Hunan, China 32 BZhuyeqi^ C. sinensis var. sinensis NTC Hunan, China 33 BXiangbolv^ C. sinensis var. sinensis PTC Hunan, China 34 BGaoqiaozao 51^ C. sinensis var. sinensis PTC Hunan, China 35 BDongtingchun^ C. sinensis var. sinensis PTC Jiangsu, China 36 BSuyuhuang^ C. sinensis var. sinensis PTC Jiangsu, China 37 BShangmeizhou^ C. sinensis var. sinensis NTC Jiangxi, China 38 BNingzhou 2^ C. sinensis var. sinensis NTC Jiangxi, China 39 BGancha 1^ C. sinensis var. sinensis PTC Jiangxi, China 40 BFoshou^ C. sinensis var. sinensis NTC Fujian, China 41 BManqilan^ C. sinensis var. sinensis LTC Fujian, China 42 BBenshan^ C. sinensis var. sinensis NTC Fujian, China 43 BTieguanyin^ C. sinensis var. sinensis NTC Fujian, China 44 BHuangdan^ C. sinensis var. sinensis NTC Fujian, China

5 Mol Breeding (2017) 37:93 Page 5 of Table 1 (continued) Sample Accession name Species Germplasm Cultivation region 45 BYuanxiaolv^ C. sinensis var. sinensis PTC Fujian, China 46 BZhenghedabaicha^ C. sinensis var. sinensis NTC Fujian, China 47 BFuandabaicha^ C. sinensis var. sinensis NTC Fujian, China 48 BDahongpao^ C. sinensis var. sinensis NTC Fujian, China 49 BShuijingui^ C. sinensis var. sinensis LTC Fujian, China 50 BYinghong 9^ C. sinensis var. assamica PTC Guangdong, China 51 BYinghong 1^ C. sinensis var. assamica NTC Guangdong, China 52 BGuilv 1^ C. sinensis var. sinensis NTC Guangxi, China 53 BGuixiang 18^ C. sinensis var. sinensis NTC Guangxi, China 54 BQianfu 4^ C. sinensis var. assamica LTC Guizhou, China 55 BMingshanbaihao 131^ C. sinensis var. assamica NTC Sichuan, China 56 BChuanhong 1^ C. sinensis var. assamica LTC Sichuan, China 57 BJinxuan^ C. sinensis var. sinensis NTC Taiwan, China 58 BWuheichangye^ C. sinensis var. assamica LTC Yunnan, China 59 BChangyebaihao^ C. sinensis var. assamica PTC Yunnan, China 60 BYuangui^ C. sinensis var. assamica PTC Yunnan, China 61 BYunkang 10^ C. sinensis var. assamica PTC Yunnan, China 62 BYunkang 27^ C. sinensis var. assamica PTC Yunnan, China 63 BYunmei^ C. sinensis var. assamica PTC Yunnan, China 64 BY1^ C. sinensis var. assamica LTC Yunnan, China 65 BY2^ C. sinensis var. assamica LTC Yunnan, China 66 BY3^ C. sinensis var. assamica LTC Yunnan, China 67 BY4^ C. sinensis var. assamica LTC Yunnan, China 68 BY5^ C. sinensis var. assamica LTC Yunnan, China 69 BY6^ C. sinensis var. assamica LTC Yunnan, China 70 BY7^ C. sinensis var. assamica LTC Yunnan, China 71 BY8^ C. sinensis var. assamica LTC Yunnan, China 72 BY9^ C. sinensis var. assamica LTC Yunnan, China 73 BY10^ C. sinensis var. assamica LTC Yunnan, China 74 BY11^ C. sinensis var. assamica LTC Yunnan, China 75 BY12^ C. sinensis var. assamica LTC Yunnan, China 76 BY13^ C. sinensis var. assamica LTC Yunnan, China 77 BY14^ C. sinensis var. assamica LTC Yunnan, China 78 BY15^ C. sinensis var. assamica LTC Yunnan, China 79 BY16^ C. sinensis var. assamica LTC Yunnan, China 80 BY17^ C. sinensis var. assamica LTC Yunnan, China NTC national tea cultivar, PTC provincial tea cultivar, LTC local tea cultivar and Musel 2005). A phylogenetic tree was constructed basedonnei s genetic distance and the neighborjoining method as implemented in MEGA 4.0 (Tamura et al. 2007). To assess the fingerprinting power of these markers, we calculated the probability of identity (PI) for each marker and their combinations (PIs) using Genalex 6.5 (Peakall and Smouse 2012). PI is the average probability of two random individuals that shared the same genotype, and the calculation formula of PI is the following: PI = 2( p i 2 ) 2 p i 4, where p i represents the frequency of the ith allele at

6 93 Page 6 of 14 Mol Breeding (2017) 37:93 a locus. For multiple loci combination, PI was calculated as the product of individual locus PIs while assuming that all loci segregate independently. The PIsibs, corresponding to the overall probability of finding two full-sib individuals from a population that possesses the same genotype by chance, was also calculated. The formula for calculating PIsibs is the following: PIsibs = (0.5 p i 2 ) + [0.5( p i 2 ) 2 ] (0.25 p i 4 ), where p i represents the frequency of the ith allele at a locus. Core SSR marker set selection To obtain a reliable core SSR marker set, several criteria were applied (Patzak et al. 2007; Tan et al. 2015) including (1) the amplified products are unambiguous without non-specific amplification; (2) the marker shows high degree of polymorphism with a PIC > 0.5; (3) the fragment size differences between two adjacent alleles should be >3 bp to allow resolution by gel electrophoresis; and (4) the PI < 0.2, so that a combination of some markers can provide sufficient discriminatory power. Results Development of genomic SSR markers Based on the tea whole genome sequence, we identified a huge number of SSR loci using the MISA program ( Only SSRs of di-, tri-, tetra-, penta-, and hexa-nucleotide motifs with a number of uninterrupted repeat units greater than 6, 5, 4, 4, and 4, respectively, were included. A total of 180 SSR loci that included five SSR types and that were distributed on different genomic scaffolds were selected for developing new markers. A group of six tea accessions were selected for the initial screening of the primer pairs on 1.5% agarose gels, and those primer pairs that gave unambiguous amplification bands were further screened using the DNA Fragment Analyzer 96. As a consequence, a total of 36 pair primers, giving reproducible amplification of polymorphic fragments in the initial group of tea accessions, were identified, and we regarded them as new informative SSR markers. The other primer pairs were discarded from further analyses based on non-specific amplification, total amplification failure, amplification of ambiguous products, and fragments size >500 bp, possibly due to the presence of introns. The amplification results using the 36 primer pairs and the primer sequences were given in Online Resources 2 and 3, respectively. Using the 36 newly developed SSR markers, we identified 80 Chinese tea plant accessions using the Fragment Analyzer 96. Subsequently, genetic properties for these markers were calculated including NA, MAF, HO, and PIC (Table 2). A total of 535 alleles were detected, which varied from 7 (G- SSR2) to 20 (G-SSR34) per SSR marker with an average of The MAF for each marker ranged from (G-SSR33) to (G-SSR29) with an average of The HO varied from (G- SSR23) to (G-SSR25 and G-SSR26) with an average of 0.568, and the PIC ranged from (G-SSR29) to (G-SSR2) with an average of These results demonstrated that the new genomic SSR markers are highly polymorphic, providing valuable resource for genetic and genomic analyses in Camellia sinensis. Phylogenetic analysis A phylogenetic tree was constructed based on the Nei s genetic distance and neighbor-joining method in MEGA 4.0 (Fig. 1). All of the 80 tea accessions clustered into four main groups, which is highly consistent with their genetic backgrounds or original places. The four groups (A to D) contained 28, 15, 9, and 28 members, respectively. Group A consisted of accessions from Anhui (17), Jiangsu (2), Zhejiang (4), Hubei (2), Hunan (2), and Jiangxi (1) province, representing the central part of the tea-growing region in China. Group B contained all the accessions that originated from Fujian, and two members from Guangxi, one from Hunan and one from Jiangxi. Group C consisted of nine members: four from Hunan and one each from Anhui, Zhejiang, Hubei, and Jiangxi province. All accessions from Yunnan clustered exclusively in group D. Remarkably, accessions BYinghong 1^ and BYinghong 9,^ which are considered to be landraces from Yunnan, clustered into group D, and BQianfu 4^ from Guizhou, BMingshanbaihao 131,^ and BChuanhong 1^ from Sichuan province also clustered into this group, suggesting that accessions from the same or neighboring geographical regions generally clustered together. This group (D) can be further divided into two subgroups (D1 and

7 Mol Breeding (2017) 37:93 Page 7 of Table 2 Key genetic statistics of the 36 SSR markers Marker ID N A MAF H O PIC PI PIsibs G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR G-SSR Mean NA number of alleles, MAF major allele frequencies, Ho observed heterozygosity, PIC polymorphic information content, PI probability of identity, PIsibs the overall probability of finding two full-sib individuals from a population that possesses the same genotype by chance D2), in which, all the accessions from Yunnan are mixed together and indistinguishable by their geographical location, giving rise of complicated phylogenetic relationships. Power of fingerprinting Based on the genotyping data, we performed a multilocus match analysis in GenAlex 6.5 and no match was

8 93 Page 8 of 14 Mol Breeding (2017) 37:93 Fig. 1 Unrooted neighbor-joining phylogenetic tree based on the SSR marker genotypes for the 80 tea plant accessions using Nei s genetic distances. There are four main groups A, B, C, and D, and group D is further divided into two subgroups D1 and D2 detected, indicating that each of the 80 tea accessions had a unique SSR fingerprint profile. To evaluate the fingerprinting power of the 36 markers, we calculated the two pivotal statistics of PI and PIsibs (Table 2). For each SSR marker, the PI ranged from (G-SSR2 and G-SSR33) to (G-SSR15), and the average value was Assuming that all SSR marker loci segregate independently, the probability of identifying two random individuals, sharing the same genotypes at all the 36 loci, was estimated to be At the upper limit of PI, the PIsibs varied from (G-SSR2 and G-SSR33) to (G-SSR15) with an average of 0.326, and the combined PIsibs was Both PI and PIsibs had extremely low values, indicating the elegant fingerprinting power of these newly developed SSR markers. Based on our results, SSR marker combinations were evaluated for their ability to discriminate each tea plant accession. As a result, we identified five selected SSR markers that gave unambiguous fragment amplification

9 Mol Breeding (2017) 37:93 Page 9 of and had high PIC values as a core marker set for fingerprinting the 80 tea accessions (Fig. 2a). Ultimately, markers G-SSR2, G-SSR7, G-SSR19, G-SSR20, and G-SSR22 were chosen for the core marker set, and their allele sizes and frequencies are also summarized (Fig. 2b). The combined PI and PIsibs of the core marker set were and ,respectively.Remarkably, G-SSR2 had the strongest fingerprinting power with the lowest PI value, and 30 of the 80 accessions can be distinguished fully by this marker. The G-SSR19 Fig. 2 Evaluating the fingerprinting power of SSR markers in tea plant accessions. a Evaluation of the probability of identification of SSR markers. b Allele sizes and frequencies for the five core SSR markers based on the genotyping data from 80 tea plant accessions

10 93 Page 10 of 14 Mol Breeding (2017) 37:93 marker can discriminate 21 accessions out of the remaining 50 tea accessions, followed by G-SSR7, which can distinguish 13 out of 29 tea accessions. The last 16 tea accessions can be distinguished by the last two markers, G-SSR20 and G-SSR22. Fingerprinting of tea plant accessions To distinguish the 80 tea plant accessions from one another, we analyzed the DNA fragments amplified by five core markers against the 80 accessions. Each tea plant accession was given an identifying fingerprinting profile based on the order of markers and the amplified allele sizes (Online Resource 4). The fingerprinting codes can be further converted into two-dimension barcode using a website ( For example, as one of the most important national tea cultivars, C. sinensis var. BShuchazao^ had five unique amplified fragments from the five core markers using Fragment Analyzer 96. The fragments sized by PROSize 2.0 for G-SSR2, G-SSR7, G-SSR19, G-SSR20, and G-SSR22 were 247, 263 and 272, 267, 179, and 264 and 275 bp, respectively, which displayed a unique fingerprinting profile based on marker order and fragment sizes (Fig. 3a). Furthermore, the two-dimension barcode of C. sinensis var. BShuchazao^ was established, and the information including accession name, type of germplasm, systematics, cultivation region, and fingerprinting code can be scanned by computers or a mobile service (Fig. 3b). Therefore, the core marker set containing five SSR markers is suitable for constructing fingerprinting of a diverse group of tea cultivars/accessions. Discussion As one of the most popular genetic markers, SSRs play a significant role in plant genetics and breeding because of their multiple-allelic nature, stability, co-dominant inheritance, and relative abundance in the genome (Tautz and Renz 1984; Liu et al. 2013;Sardaretal.2016). However, the effectiveness and successes in using SSRs rely considerably upon the quality of the markers, the accuracy of genotyping data, and the selected plant materials. In this study, we paid great attention to the process of marker selection and selected 36 markers out of 180 SSR loci for further analysis. Stable and unambiguous amplification were demonstrated for the 36 markers against the 80 tea accessions. It was shown that these 36 unique SSR markers were highly polymorphic in the 80 tea accessions with an average of 14.9 alleles for each marker and an average PIC of The PIC value was highly consistent with the result reported in a previous study, which demonstrated an average PIC value of 0.86 for 13 genomic SSRs (Freeman et al. 2004). Remarkably, the average allele number and PIC values were far greater than those reported in several previous studies for EST-SSR markers (Ma et al. 2010; Fangetal.2012; Yao et al. 2012; Balietal. 2013; Tan et al. 2015). It is possible that some SSR motifs, located in the non-transcribed regions, have relatively lower levels of sequence conservation or higher mutation rates compared with SSRs located in gene exons. Another alternative explanation is the higher resolution of the capillary DNA Fragment Analyzer 96, which contributed tremendously to a higher polymorphism for scoring of alleles. Compared with polyacrylamide gel electrophoresis, the DNA Fragment Analyzer presented overwhelming advantages including high throughput fragment separation, time and labor saving, and higher accuracy in allele discrimination. It is noteworthy that the Fragment Analyzer 96 used in this study has lower reagent costs, higher sensitivity, and higher efficiency compared with the ABI 3730 DNA Analyzer (Nimmakayala et al. 2014; Singh et al. 2015). Nevertheless, the differences in genetic diversity reported for the different studies are determined by various factors including sampling schemes, the number of SSR markers, sizes and types of SSR repeats, and the location of the SSR motifs in the genome (Yao et al. 2012). Collectively, the PICs calculated for all these markers were >0.70, with the highest maximum of 0.919, indicating that the new SSR markers are suitable for phylogenetic and genetic diversity analysis, and fingerprinting of tea plant cultivars and accessions. Phylogenetic trees can reflect both the pedigree relationships and the geographic distances among plant species and varieties. The neighbor-joining tree for the 80 tea accessions grouped them into four major clusters on the base of Nei s genetic distance. Apparently, the tea accessions that originated from the southwestern China were found exclusively in group D, which was the longest branch of the phylogenetic tree, similar to the results of several previous studies (Fang et al. 2012; Tan et al. 2015; Wambulwa et al. 2016). It was not unexpected that accessions BYinghong 1^ and BYinghong 9^ collected from Guangdong province were clustered in this group because they are actually clonal cultivars from C. sinensis var. assamica populations from Yunnan province. The BMingshanbaihao 131^ and BChuanhong 1^ accessions were selected from Sichuan province and BQianfu 4^

11 Mol Breeding (2017) 37:93 Page 11 of Fig. 3 Evaluating the fingerprinting power of SSR markers in tea plant accessions. a Evaluation of the probability of identity of SSR markers. b Allele sizes and frequencies for the five core SSR markers based on the genotyping data from 80 tea plant accessions originated from Guizhou province. Both of the two provinces are located in southwestern China and near to Yunnan. Group D was further divided into two subgroups, suggesting that there is a higher relative level of genetic variation among the accessions between the two subgroups. The accessions in the other three groups (A, B, and C) had relatively close relationships with each other compared with those in group D, supporting the notion that these cultivars or accessions from central China have retained distinct phenotypes compared with those originated from C. sinensis var. assamica populations. Furthermore, these accessions in groups A, B, and C can be further distinguished based on their geographical regions. Interestingly, BLongjingchangye^ from Zhejiang province clustered close to the cultivars or landraces from Anhui province in group A. The result is consistent with the previous study, revealed by Tan et al. (2015), who speculated that the national tea plant cultivar BShuchazao^ is probably derived from BLongjing 43^ (propagated from the BLongjingchangye^ cultivar population) based on their

12 93 Page 12 of 14 Mol Breeding (2017) 37:93 phylogenetic tree. Our results showed that BLongjingchangye^ was closely related to the cultivars or landraces from Anhui, although their detailed pedigree relationship is still unclear. Group B contained 15 tea accessions, most of which were collected from Fujian province, with the exception of two from Guangxi province, one from Hunan province and one from Jiangxi province. It is expected that BGuilv 1^ would cluster in this group because it is a clonal cultivar selected from Zhejiang. However, it was unexpected that the landraces of BNingzhou 2^ from Jiangxi, BGaoqiaozao^ from Hunan, and BGuixiang 18^ from Guangxi were clustered into this group, which should be further clarified. Though group C consisted of the smallest number of tea accessions, they covered a wide geographic region. Of the nine in group C, four were from Hunan, two from Zhejiang, and one each from Jiangxi, Anhui, and Hubei provinces, respectively. In summary, almost all the tea accessions were grouped with respect to their known genetic relationships and geographic origins, indicating that the phylogenetic tree reflects the relationships among the 80 accessions based on the new SSR marker data. The two parameters of PI and PIsibs have been extensively used for identifying the fingerprinting power of molecular markers (Waits et al. 2001; Tan et al. 2015). It is worth noting that the combined PI for the 36 SSR markers was ,implying that it is almost impossible to find two distinct tea accessions with the same SSR fingerprinting profile. In fact, it has been suggested that the theoretical PI can be overestimated due to population substructure and the assumption of independent segregation among loci is not authentic (Waits et al. 2001). Therefore, tea accessions with relatively large genetic variation were preferentially selected to minimize the impact of population substructure. On the other hand, the probability of random two markers is linked together in the large tea genome (2.2 4 Gb) (Tanaka and Taniguchi 2006; Huang et al. 2013), which is extremely low, thereby, the impact of nonauthentic assumption to theoretical PI can be neglected. The PIsibs, which considered as a conservative upper boundary for PI, was calculated and itsvaluewasalsoverylow( ). Waits et al. (2001) suggested that a PI of and is sufficiently low for most applications in natural populations. On the base of our results, the combined PI and PIsibs were sufficiently low for fingerprinting of all samples when five random markers were selected as a core marker set. The combined PI and PIsibs of the selected five SSR markers were and , respectively, thus, the core marker set should be adequate to discriminate all tea accessions used in this study. Ultimately, the established fingerprinting profiles can serve as a database for the elite tea plant cultivars or landraces and can provide outstanding tools for commercial tea plant variety certification and protection. Our future efforts will focus mainly on four objectives, which are also proposed for tea-breeding projects: (1) to collect worldwide tea plant resources, especially some unknown and wild species; (2) to develop alternative reliable and more efficient molecular markers, such as SNP and Indel markers (Celik et al. 2016; Zhang et al. 2016), and to combine those molecular markers that have been developed previously and here; (3) to provide a platform which can be used to define the true origin of tea plants that have been cultivated for centuries; and (4) to provide a better classification of more tea accessions, or even different tea plant species such as C. crassicolumna, C. tachangensis, C. taliensis, and C. gymnogyna, by combining physiological and morphological characters with molecular evidences. Conclusions SSR markers are powerful tools that have been extensively used for genetic diversity and phylogenetic analysis, fingerprinting, germplasm identification and conservation, and linkage map construction in C. sinensis. Based on the selected 180 SSR loci, we developed 36 new genomic SSR markers that are highly polymorphic and reliable for scoring of genotypes. Phylogenetic relationships of 80 tea plant accessions were successfully constructed, providing valuable information for understanding phylogeny and genetic origin of these accessions. Five SSR markers were recommended as a core of marker set for fingerprinting of the tea plants, thereby a unique fingerprint profile of these 80 tea plant accessions was established. Overall, the SSR markers, phylogenetic relationships, and constructed fingerprint profiles of tea accessions are valuable for breeding programs, various genetic and genomic studies, and true cultivar/accession confirmation. Acknowledgments This work was supported by the National Natural Science Foundation of China ( ), the Special

13 Mol Breeding (2017) 37:93 Page 13 of Innovative Province Construction in Anhui Province (15czs08032), the Central Guiding the Science and Technology Development of the Local ( B024), and the Anhui Natural Science Foundation ( J08, QC57). We appreciate the anonymous reviewers for constructive comments on this manuscript. Author contributions SRL performed collections of tea plant materials, data analysis, and manuscript drafting. HWL involved in DNA isolation, designing genomic SSRs primers, SSR genotyping, PCR amplifications, and testing amplified fragments. ALW and YH coordinated collections of plant materials and DNA isolation. YLA involved in PCR amplifications and testing amplified fragments. CLW involved in experimental design, data analysis, and manuscript preparation. Compliance with ethical standards Competing interests competing interests. References The authors declare that they have no Bali S, Mamgain A, Raina SN, Yadava SK, Bhat V, Das S, Pradhan AK, Goel S (2015) Construction of a genetic linkage map and mapping of drought tolerance trait in Indian beveragial tea. Mol Breed 35:112 Bali S, Raina SN, Bhat V, Aggarwal RK, Goel S (2013) Development of a set of genomic microsatellite markers in tea (Camellia L.) (Camelliaceae). Mol Breed 32: Celik I, Bodur S, Frary A, Doganlar S (2016) Genome-wide SNP discovery and genetic linkage map construction in sunflower (Helianthus annuus L.) using a genotyping by sequencing (GBS) approach. Mol Breed 36:133 Chen L, Yu FL, Tong Q, Q. (2000) Discussions on phylogenetic classification and evolution of section Thea. J Tea Sci 20:89 94 Fang W, Li X, Cheng H, Duan Y, Jiang X (2012) Genetic diversity and relationship of clonal tea (Camellia sinensis) cultivars in China as revealed by SSR markers. Plant Syst Evol 298: Freeman S, West J, James C, Lea V, Mayes S (2004) Isolation and characterization of highly polymorphic microsatellites in tea (Camellia sinensis). Mol Ecol Notes 4: Hashimoto M, Takasi S (1978) Morphological studies on the origin of the tea plant V, a proposal of one place of origin by cluster analysis. Jpn J Crop Agr 21: Huang H, Tong Y, Zhang QJ, Gao LZ (2013) Genome size variation among and within Camellia species by using flow cytometric analysis. PLoS One 8:e64981 Kaundun SS, Matsumoto S (2004) PCR-based amplicon length polymorphisms (ALPs) at microsatellite loci and indels from non-coding DNA regions of cloned genes as a means of authenticating commercial Japanese green teas. J Sci Food Agric 84: Liu K, Musel SV (2005) PowerMarker: an integrated analysis environment for genetic marker analysis. Bioinformatics 21: Liu SR, Li WY, Long D, Hu CG, Zhang JZ (2013) Development and characterization of genomic and expressed SSRs in citrus by genome-wide analysis. PLoS One 8:e75149 Ma JQ, Yao MZ, Ma CL, Wang XC, Jin JQ, Wang XM, Chen L (2014) Construction of a SSR-based genetic map and identification of QTLs for catechins content in tea plant (Camellia sinensis). PLoS One 9:e93131 Ma JQ, Zhou YH, Ma CL, Yao MZ, Jin JQ, Wang XC, Chen L (2010) Identification and characterization of 74 novel polymorphic EST-SSR markers in the tea plant, Camellia sinensis (Theaceae). Am J Bot 97:e153 e156 Mukhopadhyay M, Mondal TK, Chand PK (2016) Biotechnological advances in tea (Camellia sinensis [L.] O. Kuntze): a review. Plant Cell Rep 35: Nei M, Tajima F, Tateno Y (1983) Accuracy of estimated phylogenetic trees from molecular data. J Mol Evol 19: Ni S, Yao MZ, Chen L, Zhao LP, Wang XC (2008) Germplasm and breeding research of tea plant based on DNA marker approaches. Front Agric China 2: Nimmakayala P, Abburi VL, Abburi L, Alaparthi SB, Cantrell R, Park M, Choi D, Hankins G, Malkaram S, Reddy UK (2014) Linkage disequilibrium and population-structure analysis among Capsicum annuum L. cultivars for use in association mapping. Mol Gen Genomics 289: Patzak J, Vrba L, Matousek J (2007) New STS molecular markers for assessment of genetic diversity and DNA fingerprinting in hop (Humulus lupulus L.) Genome 50:15 25 Peakall R, Smouse PE (2012) GenAlEx 6.5: genetic analysis in Excel. Population genetic software for teaching and research an update. Bioinformatics 28: Sardar SS, Pradhan K, Shukla RP, Saraswat R, Srivastava A, Jena SN, Das AB (2016) In silico mining of EST-SSRs in Arachis hypogaea L. and their utilization for genetic structure and diversity analysis in cultivars/breeding lines in Odisha, India. Mol Breed 36:49 Sharangi AB (2009) Medicinal and therapeutic potentialities of tea (Camellia sinensis L.) a review. Food Res Int 42: Singh NV, Abburi VL, Ramajayam D, Kumar R, Chandra R, Sharma KK, Sharma J, Babu KD, Pal RK, Mundewadikar DM, Saminathan T, Cantrell R, Nimmakayala P, Reddy UK (2015) Genetic diversity and association mapping of bacterial blight and other horticulturally important traits with microsatellite markers in pomegranate from India. Mol Gen Genomics 290: Tamura K, Dudley J, Nei M, Kumar S (2007) MEGA4: molecular evolutionary genetics analysis (MEGA) software version 4.0. MolBiolEvol24: Tan LQ, Peng M, Xu LY, Wang LY, Chen SX, Zou Y, Qi GN, Cheng H (2015) Fingerprinting 128 Chinese clonal tea cultivars using SSR markers provides new insights into their pedigree relationships. Tree Genet Genomes 11:90 Tan LQ, Wang LY, Xu LY, Wu LY, Peng M, Zhang CC, Wei K, Bai PX, Li HL, Cheng H, Qi GN (2016) SSR-based genetic mapping and QTL analysis for timing of spring bud flush, young shoot color, and mature leaf size in tea plant (Camellia sinensis). Tree Genet Genomes 12:52 Tanaka J, Taniguchi F (2006) Estimation of the genome size of tea (Camellia sinensis), Camellia (C. japonica), and their

14 93 Page 14 of 14 Mol Breeding (2017) 37:93 interspecific hybrids by flow cytometry. J Remote Sens Soc Jpn 101:1 7 Taniguchi F, Furukawa K, Metoku SO, Yamaguchi N, Ujihara T, Kono I, Fukuoka H, Tanaka J (2012) Construction of a highdensity reference linkage map of tea (Camellia sinensis). Breed Sci 62: Tautz D, Renz M (1984) Simple sequences are ubiquitous repetitive components of eukaryotic genomes. Nucleic Acids Res 12: Ujihara T, Ohta R, Hayashi N, Kohata K, Tanaka J (2009) Identification of Japanese and Chinese green tea cultivars by using simple sequence repeat markers to encourage proper labeling. Biosci Biotechnol Biochem 73:15 20 Ujihara T, Taniguchi F, Tanaka J, Hayashi N (2011) Development of expressed sequence tag (EST)-based cleaved amplified polymorphic sequence (CAPS) markers of tea plant and their application to cultivar identification. J Agric Food Chem 59: Waits LP, Luikart G, Taberlet P (2001) Estimating the probability of identity among genotypes in natural populations: cautions and guidelines. Mol Ecol 10: Wambulwa MC, Meegahakumbura MK, Chalo R, Kamunya S, Muchugi A, Xu JC, Liu J, Li DZ, Gao LM (2016) Nuclear microsatellites reveal the genetic architecture and breeding history of tea germplasm of East Africa. Tree Genet Genomes 12:11 Yang CS, Wang X, Lu G, Picinich SC (2009) Cancer prevention by tea: animal studies, molecular mechanisms and human relevance. Nat Rev Cancer 9: Yang H, Wei CL, Liu HW, Wu JL, Li ZG, Zhang L, Jian JB, Li YY, Tai YL, Zhang J, Zhang ZZ, Jiang CJ, Xia T, Wan XC (2016) Genetic divergence between Camellia sinensis and its wild relatives revealed via genome-wide SNPs from RAD sequencing. PLoS One 11:e Yang YJ, Liang YR (2014) Clonal tea cultivars in China. Shanghai Scientific and Technical Publishers, Shanghai Yao MZ, Ma CL, Qiao TT, Jin JQ, Chen L (2012) Diversity distribution and population structure of tea germplasms in China revealed by EST-SSR markers. Tree Genet Genomes 8: Zhang JZ, Liu SR, Hu CG (2016) Identifying the genome-wide genetic variation between precocious trifoliate orange and its wild type and developing new markers for genetics research. DNA Res 23: