Wnping Fng，Lyndel W.Meinhrdt，Huwei Tn，Lin Zhou，Sue Mishke，Xinghu Wng，Dpeng Zhng，*

aCollege of Horticulture，Nanjing Agricultural University，Nanjing 210095，Jiangsu，China

bSustainable Perennial Crops Laboratory，USDA-ARS，10300 Baltimore Avenue，Bldg.001，Rm.223，BARC-W，Beltsville，MD 20705，USA

cYunnan Pu'er Tea Seeds Propagation and Extension Unit，Pu'er 665000，Yunnan，China

?

Identification of the varietal origin of processed loose-leaf tea based on analysis of a single leaf by SNP nanofluidic array

Wanping Fanga，b，Lyndel W.Meinhardtb，Huawei Tana，Lin Zhoua，Sue Mischkeb，Xinghua Wangc，Dapeng Zhangb，*

aCollege of Horticulture，Nanjing Agricultural University，Nanjing 210095，Jiangsu，China

bSustainable Perennial Crops Laboratory，USDA-ARS，10300 Baltimore Avenue，Bldg.001，Rm.223，BARC-W，Beltsville，MD 20705，USA

cYunnan Pu'er Tea Seeds Propagation and Extension Unit，Pu'er 665000，Yunnan，China

A R T I C L E I N F O

Article history:

in revised form

4 February 2016

Accepted 15 March 2016

Available online 31 March 2016

Authentication

Camellia sinensis

Conservation

Food adulteration Molecular markers

A B S T R A C T

Tea is an important cash crop，representing a$40 billion-a-year global market.Differentiation of the tea market has resulted in increasing demand for tea products that are sustainably and responsibly produced.Tea authentication is important because of growing concerns about fraud involving premium tea products.Analytical technologies are needed for protection and value enhancement of high-quality brands.For loose-leaf teas，the challenge is that the authentication needs tobeestablished on the basis ofa singleleaf，sothatthe products can be traced back to the original varieties.A new generation of molecular markers offers an ideal solution for authentication of processed agricultural products.Using a nanofluidic array to identify variant SNP sequences，we tested genetic identities using DNA extracted from single leaves of 14 processed commercial tea products.Based on the profiles of 60 SNP markers，the geneticidentityofeachteasamplewasunambiguouslyidentifiedbymultilocusmatchingand ordination analysis.Results for repeated samples of multiple tea leaves from the same products（usingthree independentDNA extractions）showed100%concordance，showingthat the nanofluidic system is a reliable platform for generating tea DNA fingerprints with high accuracy.The method worked well on green，oolong，and black teas，and can handle a large number of samples in a short period of time.It is robust and cost-effective，thus showing high potential for practical application in the value chain of the tea industry.

Production and hosting by Elsevier B.V.on behalf of Crop Science Society of China and Institute of Crop Science，CAAS.This is an open access article under the CC BY-NC-ND license（http://creativecommons.org/licenses/by-nc-nd/4.0/）.

1.Introduction

The tea plant，Camellia sinensis（L.）O.Kuntze，is a perennial woody evergreen flowering species in the family Theaceae［1，2］that has been esteemed throughout history.The present consumption of tea is billions of cups daily，making it the world's most universal beverage other than water［3］.With an annual production of approximately 4.8 million tons［4］，tearepresents a$40 billion-a-year industry［5］.Although overall consumption is expected to increase only moderately，tea drinkersarelikelytodemand，andbewillingtopayfor，ahigherquality product［6］.Commercial tea products are classified into categories based on processing techniques:degree and manner of“fermentation”（enzymatic oxidization）of the leaves and buds.Common categories include white，green，yellow，oolong，black（also known as“red”in China），and dark（genuinely fermented black）Pu'er teas.The method of processing affects the attributes of tea，including content of caffeine and polyphenols and thereby antioxidant activity and flavor［7-10］.Within each category，tea processors use leaves of various C.sinensis varieties that often differ greatly in quality.

The specialty tea market has been rapidly expanding on a global scale，resulting in higher revenues and profits for tea growers and the industry.Accurate identification of specific C.sinensis varieties is critically important for ensuring the authentication of premium tea products and maintenance of brand image.However，efficient methods for varietal authentication of specialty tea products，especially loose-leaf teas，have not yet been developed.Instrumental methods，such as near-infrared spectroscopy（NIR）have been widely applied for tea quality control［11-15］.Using NIR diffuse reflectance spectroscopy coupled with pattern recognition techniques，Tan et al.［16］were able to differentiate varieties of tea leaves from different geographical areas with a high degree of confidence（96%）.However，theanalysiswasbasedonchemicalcomponents suchaspolyphenols，theanine，caffeine，andvolatilecompounds，which are influenced by many factors including not only genetic makeup of the plant but also environmental conditions during growth，time of harvest，and postharvest factors［17］.Moreover，althoughchemicalanalysiscanreadilydifferentiateteavarieties，it is much more challenging to match a tested variety with a known onewithahighdegreeofcertainty.Positiveidentification requires more than just sensory or instrumental examination.

The advantages of methods based on DNA to identify the botanical origin of food products，particularly after commercial processing，are well recognized［18，19］.Standard DNA barcodes have been used to discriminate between C.sinensis and most otherherbalteaspecies，butwerenotspecificenoughtoidentify individuals within the species［20，21］.Methods using markers basedonPCRamplificationofasequence-taggedorotherregion in a gene，and analysis of resulting restriction fragment length polymorphisms（RFLP，AFLP），have been used to identify tea varieties［22-25］.Hu et al.［26］used this method with markers from both cytoplasmic（mitochondrial and chloroplast）and nuclear tea genomes.Polymorphisms in amplification length of microsatellites or of coding and non-coding regions of specific genes have also been used for tea varietal identification［24，27-31］.However，todate，theapplication ofDNA fingerprinting has been used only to differentiate varieties，rather than confirm the genetic identity of two samples.Moreover，even with the use of microsatellite markers，resolving genotyping results from different labs has not been straightforward.It is difficult to standardize data generated on different genotyping platforms，and comparison of data is further complicated because the same alleles may be binned differently.Even on the same platform，analysis can be complicated by common PCR artifacts such as stutter due to slipped-strand mispairing，which may lead to incorrect identification of an allele，and diminished amplification of longer repeats，which may lead to scoring heterozygotes as homozygous or other spurious genotypes［32，33］.To date，none of the markers have been applied to differentially processed tea products，which are fermented，baked，or sun-dried to different extents.In processed tea，DNA is of poor quality，highly degraded，and contains PCR inhibitors that can pose problems for target amplification.Such factors that interfere with the application of simple sequence repeat（SSR）-based fingerprints for tea authentication can lead to false conclusions.

Single-nucleotide polymorphism （SNP）markers are the most abundant class of polymorphisms in plant genomes［34］. In contrast to SSR markers，accurate identification of SNPs can be performed without the requirement of DNA separation by size and can accordingly be automated in an assay array or microchip format.The biallelic nature of SNPs offers a much lower error rate in allele calling than that of SSRs，and genotyping can be multiplexed and accomplished quickly at a lower cost.Because of these advantages，SNPs have become the marker of choice for variety identification in plants［35，36］. Recently，Bazakos et al.［37］used SNP analysis to identify the varietal origin of olive oils.Development of SNP markers for the tea plant has been reported by a community of tea scientists［25，38-40］.However，application of SNP markers in genotype identification，as well as traceability and authentication of commercial tea products，has not been studied.In our previous report［41］，we demonstrated the efficacy of using a nanofluidic system to generate SNP fingerprints of the tea plant.A total of 1786 putative SNPs were identified from a tea EST database，of which 96 SNPs were evaluated in 40 fresh leaf samples of Chineseteavarieties.Theresultsshowedthateachofthetested varieties had a unique SNP profile that allowed unambiguous varietal identification.However，the efficacy of applying these SNP markers to processed，commercial tea products has yet to be systematically investigated.

For loose-leaf teas，the challenge is that DNA fingerprints must be established on the basis of a single leaf，so that the products can be traced back to the original variety.This is because，during postharvest，tea leaves from different varieties may be mixed.Adulteration can happen at any step prior to sale to traders.Moreover，the effects of different fermentation levels in various loose-leaf tea products need to be examined for this SNP genotyping system.The objective of this work was to assess the efficacy of the previously reported tea SNP panel for varietal authentication in loose-leaf tea products，using DNA extracted from a single leaf of green，black，or oolong tea.

2.Materials and methods

2.1.Sample preparation

Commercial tea products characterized in this study and their sources are listed in Table 1.A total of 14 loose-leaf tea products，including 10 green，two oolong，one black，and one raw Pu'er tea were used in this study（Table 1）.All tea products were purchased from local markets.For each product，one to three single leaves were independently and randomly sampled from the same package.

Table 1-List of 14 tea products，including 24 single leaf samples，and their origins.

2.2.Extraction and preparation of DNA from single tea leaves

DNA was extracted from each single leaf（or bud）using the DNeasy Plant Mini kit（Qiagen，Inc.，Valencia，CA），which is based on the use of silica as an affinity matrix.A single leaf was placed in a 2-mL microcentrifuge tube with one 1/4-in. ceramic sphere and a 0.15 g garnet matrix（Lysing matrix A；MP Biomedicals，Solon，OH）.The dry samples were disrupted by high-speed shaking in a TissueLyser II（Qiagen）at 30 Hz for 1 min，followed by an additional 1 min，with a 1 min rest between disruptions.Lysis solution（buffer AP1 containing 25 mg mL-1polyvinylpolypyrrolidone），along with RNase A（Ribonuclease I），was added to the powdered leaf tissue and the mixture was incubated at 65°C，as specified in the kit instructions.The remainder of the extraction method followed themanufacturer's suggestions.DNA was elutedfrom thesilica membrane with two washes with 50 μL AE buffer，which were pooled to make 100 μL of DNA solution.Using a NanoDrop spectrophotometer（Thermo Scientific，Wilmington，DE），DNA concentration was determined by absorbance at 260 nm.DNA purity was estimated by the 260/280 ratio and the 260/230 ratio. Prior to SNP genotyping，each DNAwas subjected to a multiplex Specific Target Amplification（STA）procedure using primer pools provided by Fluidigm Corp.（South San Francisco，CA）and Qiagen 2X Multiplex PCR Master Mix（PN 206143）according to the protocol recommended in the Fluidigm SNP Genotyping User Guide［42］.

2.3.SNP markers and genotyping

The 60 tea SNP markers used for the SNPtype genotyping panel（Fluidigm Corp）were reported in our previous publication［41］. Genotyping was performed on the high-throughput Fluidigm EP1system，usingtheFluidigmSNPtypeGenotypingReagentKit according to the manufacturer's instructions，and a nanofluidic 96.96 Dynamic Array IFC（Integrated Fluidic Circuit；Fluidigm Corp.）.This chip automatically assembles PCR reactions，enabling simultaneous testing of up to 96 samples with 96 SNP markers.Fluorescence intensities were measured with the EP1 reader and results were plotted on two axes.GenotypecallingsweremadeusingtheFluidigmSNPGenotypingAnalysis program.

2.4.Data analysis

Summary statistics for each SNP locus，including observed heterozygosity（Ho），gene diversity，and inbreeding coefficient，were computed for each locus separately as well as for all loci combined，using GenAlEx 6.5［43，44］.

To evaluate the differentiation power of SNP markers on processed tea products，multilocus matching was used to compare the samples in the data set，and the same program wasusedforgenotypematching.Samplesthatwerecompletely matched at all polymorphic loci were considered duplicates derived from the same clone.To assess the differentiation power of the SNP panel，the probability of identity among siblings（PID-sib）［45］was computed，which was defined as the probability that twosiblingindividuals drawnatrandom froma population have the same multilocus genotype.The overall PID-sib is the upper limit of the possible ranges of PID in a population，thus providing the most conservative number of loci required to resolve all individuals，including relatives［45］.

To assess the relationships among the tested tea samples，we computed the genetic distances for each possible pair of tested individuals.The matrix of genetic distances was then visualized using principal coordinates analysis（PCoA），implemented in GenAlEx 6.2［43，44］.As a complementary approach，the genetic relationship was further assessed by cluster analysis.Nei's genetic distance［46］was chosen as a distance measurement（n=20）.The computation was performed usingMicrosatellite Analyser［47］.A dendrogram was generated from the resulting distance matrix using the neighbor-joining method［48］.

3.Results

3.1.DNA extraction

DNA concentration ranged from 3.4 to 21.7 ng μL-1among the 14 tea products（represented by 24 DNA extractions），with an average of 10.02 ng μL-1per single leaf sample.The average ratio of absorbance at 260 nm and 280 nm by NanoDrop measurement was 1.62 among the 14 varieties.The lowest was found for raw Pu'er tea（1.06）and the highest for Laoshan Green（2.08）.The 260/230 absorbance ratios were 0.20 to 1.01，with an average of 0.86 among the 14 tea products（Table 2）.

3.2.Summary information of SNP fingerprints and multilocus matching of SNP fingerprints among tested samples

All 60 polymorphic SNPs were reliably scored across the 24 single leaf tea samples，unambiguously differentiating all 14 loose-leaf tea products.The reliability of the 60 SNPs was demonstratedbyrepeatedsamplingofmultipleleavesfromthe same tea products（Table 3）.Individual genotype matching（pairwise comparisons）revealed fully matched trios among these samples，with identical SNP profiles（across all 60 loci）observed for replicate samples of Long Jing 1 and Tung-Ting oolong（Table 3）.PID-sib，calculated from the 24 samples under investigation，predicted that the probability of two unrelated samples having the same genotype at all 60 SNP loci was approximately 1 in 100，000.The result also showed that the multiple leaves of these varieties had indeed been sampled from the same clone.In contrast，the other three varieties in Table 3 showed different SNP profiles among multiple leaf samples，indicating that these leaves had been harvested from different tea trees，possibly from seed progenies in which each tree was genetically different.Thiswas thecase for BiLuo Chun from China，Assam from India，and Jasmine from Vietnam.

After removal of duplicated samples，the 20 samples with uniqueSNPprofileswereusedtocomputesummaryinformation of allele frequency using GenAlex 6.5.The mean value of Shannon's information index was 0.512，ranging from 0.115 to 0.693.Themeanobservedheterozygositywas0.401，rangingfrom 0.05 to 0.95，whereas the mean expected heterozygosity was 0.341，rangingfrom0.05to0.49（Table4）.Theresultiscomparable with previously reported results［41］for these SNP markers，where high values of observed heterozygosity（Ho=0.701），gene diversity（He=0.651），and information index value（I=0.604）were found across the same 60 SNP loci in 40 Chinese tea varieties.

Table 2-Concentration and quality of DNA samples extracted from a single leaf/bud in 14 loose-leaf tea products.

3.3.Genetic relationship among tested tea samples

The genetic distances among the tea samples are presented in Table 5.The smallest pairwise distance（D=20）was between two samples from the Chinese green tea Bi Luo Chun，whereas the largest genetic distance（D=64）was between the Chinese green tea Dong Ping Gaoshan and Assam B from India.The genetic relationships among the tested samples were shown by PCoA（Fig.1）.The first three main PCO axes accounted for 26.1%，17.4%，and 13.4%of total variation，respectively.In the plane of coordinate 1 vs.2（Fig.1），there is an apparent pattern of clustering among the 20 samples.It appeared that the tea products from India and Vietnam，including the green tea Jasmine and the black tea Assam，were at a distance from the tea products from mainland China，Taiwan of China，and Japan.The Japanese tea Genmaicha was closely affiliated with the tea products from Zhejiang and Anhui，eastern China.The two oolong teas from Taiwan of China（Tung-Ting oolong and Da Yu Ling oolong）were very similar to each other，but showed substantial differences from other Chinese teas.The raw Pu'er and the green tea Bi Luo Chun（B）fell between the India/Vietnam and the China/Japan clusters.

A neighbor-joining dendrogram based on Nei's distance provided a complementary supporting view of the 20 samples，revealing a pattern of relationships consistent with those revealed in the PCoA（Fig.1）.The NJ tree showed that the 20 samples can be grouped into two main clusters（Fig.2）.The first cluster comprised all three Jasmine tea samples from Vietnam and all three Assam tea samples.The second cluster included all the green，raw Pu'er，and oolong teas from mainland China，Japan，and Taiwan of China.The two Long Jing tea products（Long Jing 1 and Long Jing 2）were grouped in different subclusters，indicating that different varieties were used to produce the same brand of products.The same case was observed in green tea Bi Luo Chun，where the three individual leaf samples were grouped in three different subclusters，revealing a mixture of different varieties in the tea package.

4.Discussion

Loose-leaf tea comprises the bulk of the specialty tea market. Green tea alone accounts for approximately one million metrictons of global tea production［49］.To date，it has not been possible to discern the unambiguous genetic identity of a tea variety by morphological and biochemical characteristics，especially for processed tea.Owing to insufficient throughput，accuracy，and data standardization，existingmolecular markerbased technologies such as SSR marker fingerprinting，are of limited use.Furthermore，processed tea leaf in commercial products usually contain high levels of polyphenolic and other PCR-inhibitory compounds and there can also be residue from microorganisms resulting from the fermentation and drying processes.Because of these problems，a robust analytical system is needed for genotyping tea DNA.

Table 3 - Examples of SNP fingerprints based on single leavs for five green， black， and oolong tea products. The tabl shows only 22 of the full array of 60 SNPs.

In the present study，we demonstrated a DNA fingerprinting method that uses a small set of SNP markers to verify the genetic identity of a processed single tea leaf.Our results showed that a nanofluidic array of SNP markers is particularly suitable for this purpose.The specific target amplification protocol［50］efficiently addressed potential problems of quality and/or quantity of DNA extracted from a single processed tea leaf.This protocol，performed before genotyping，is a multiplex PCR reaction that uses primers for all loci of interest，but without targeting the specific alleles，thus proportionally increasing the amplified copies of these loci.This procedure solved our problem of recovery of low DNA concentration from processed commercial tea leaves.Results for repeated samples of multiple tea leaves from the same products（using three independent DNA extractions）showed 100%concordance，suggesting that the nanofluidic system is a reliable platform for generating tea DNA fingerprints with high accuracy.The method worked well on green tea，which is not fermented，for moderately fermented oolong tea，and for deeply fermented blacktea.Thismethod canhandlealarge numberofsamplesin a short period of time and the results are highly robust and repeatable.

The effectiveness of individual identification via SNP fingerprints depends on the number of loci used for genotyping.An important statistical parameter for determining the number of loci required to identify all distinct individuals with the needed confidence level is the probability of identity（PID）.Multilocus PID values can be obtained by multiplying together single-locus PID values，assuming independence of loci.A stringent PID value is needed for domesticated crop species because they often share similar ancestors.Thus，a PID calculated for sibs would provide a highly conservative threshold for a domesticated crop species.The present results show that using the 60 SNP loci，the chance of sampling identical genotypes from a random mating population would be 1 of 100，000.It thus predicts the high statistical power of using this set of SNPs for tea genotype verification.Given that tea is an outcrossing species，each tea tree derived from seed is expected to have a unique genotype.The present result shows that the multiple samples of Bi Luo Chun，Jasmine，and Assam had different genotypes，suggesting that these tea leaves were sampled from a population of trees propagated by seed.For such tea plants，varietal authentication would need to be performed at the population level and a tested tea sample would need to be comparedwiththeSNPprofileofthereferenceseedpopulation. Statistical approaches such as assignment test，which assigns an unknown sample to a given population based on multilocus DNA marker profiles，may need to be employed.Using thisapproach，F(xiàn)ang et al.［51］were able to test the varietal authenticity of fine-flavored cocoa beans.The same principle can be applied to tea authentication.In recent years，clone propagation has been widely promoted as a means of improving the consistency of tea quality［52］.Both Long Jing 1 from Zhejiang and Tung-Ting oolong from Taiwan are well-known premium loose teas that command a high price in the specialty tea market，and the results of this study confirm that they were sampled from clonal trees.For clonally propagated varieties，authentication through SNP fingerprinting is straightforward. Varietal authentication can be achieved by comparing any processed tea samples labeled as Long Jing with the reference Long Jing clone.

Table 4-Shannon's information index，heterozygosity，and inbreeding coefficient of the 60 SNP loci，scored on 14 loose-leaf tea products.

The result of multivariate analysis by PCoA revealed diversity within tea types and a significant difference between Assam-type and China-type teas.This result is consistent with previously reported studies based on AFLP［23］，SSR［27，53，54］，and CAP markers［26］.It thus further supports the classification of C.sinensis into different varieties including C.sinensis var.assamica（“Assam”）and C.sinensis var.sinensis［2，55-57］.The Vietnamese Jasmine tea was foundtobe grouped together with the Assam type.This finding is consistent with Vo's finding［58］that some promising tea clones in Vietnam（suchasclonePH1）wereselectedfromtheAssamgermplasm.It also supports the finding of Wachira et al.［59］that grouped tea varieties from the Indochina peninsula together with those from India.The close proximity of Japanese Genmaicha and Sencha teas with the Chinese varieties supports the notion that Japanese tea was introduced from China.The Taiwanese Tung-Ting oolong tea occupies a distinct position in the PCoA plot，suggesting that its genetic background is substantially different from those from Zhejiang and Jiangsu in mainland China.In our previous study we showed that the tea landraces from Zhejiang and Jiangsu had different genetic backgrounds from those from Fujian Province in China.However，given that no Fujian tea was included in the present study，we cannot determine whether the Tung-Ting oolong was from native Taiwanese tea or was based on introduced Fujian tea germplasm.The present study serves as a proof of principle that a robust SNP profile can be established based on a single leaf（or bud）of commercial tea products.However，for practical application of this protocol in the value chain of the tea industry，a much larger number of samples will need to be analyzed for each tea product or variety，in order to address the issueofintra-varietalvariation.Moreloose-leafteaproductsare being tested in our laboratory using this protocol.Our goal is to establish a comprehensive database so that reference SNP profiles can be found for each tea variety.

Table 5-Matrix of genetic distances among 20 single-leaf tea samples（representing 14 tea products），based on the profiles with SNP markers.

In conclusion，we conducted a pilot study of varietal authentication for loose tea.Our main objective was to show thatvarietalauthenticationofteacanbeachievedbygenotyping a single leaf of a processed tea product，irrespective of fermentation method.We showed that use of a nanofluidic array with the small set of tea SNP markers was efficient andreliable for varietal verification.This technology enabled us to generate high-quality SNP profiles basedon DNAextracted from processed tea products，including green，oolong，and black teas. To our knowledge，this is the first authentication study of commercial tea products using SNP molecular markers.The approach is sufficiently robust for verification of authenticity of specialty tea varieties and thus has high potential for practical application in the tea value chain.

Fig.1-PCoA plot of 14 loose-leaf tea products represented by 20 single leaf samples.Identification of accessions corresponds to samples listed in Table 1.Only one sample each from Long Jing 1 and Tung-Ting oolong was retained in the clustering analysis，after verification of sample homogeneity.The plane of the first three main PCO axes accounted for 56.9%of total variation.The first axis represents 26.1%of total information，the second 17.4%，and the third 13.4%.

Fig.2-Neighbor-joining dendrogram depicting the relationships among 14 tea products represented by 20 single leaf samples. Identification of accessions corresponds to samples listed in Table 1.Only one sample each from Long Jing 1 and Tung-Ting oolong was retained in the clustering analysis，after verification of sample homogeneity.

Acknowledgements

We thank Stephen Pinney of USDA-ARS，Sustainable Perennial Crops Laboratory，for technical support in SNP genotyping. References by the USDA to a company and/or product are only for the purposes of information and do not imply approval or recommendation of the product to the exclusion of others that may also be suitable.

R E F E R E N C E S

［1］P.K.Barua，Classification of the Tea Plant，Two and a Bud，101963 3-11.

［2］T.L.Ming，B.Bartholomew，Theaceae，in:Z.Y.Wu，P.H.Raven，D.Y.Hong（Eds.），F(xiàn)lora of China，Science Press&Missouri Botanical Garden Press，Beijing&St，Louis 2007，pp.367-412.

［3］A.MacFarlane，I.MacFarlane，The Empire of Tea:The Remarkable History of the Plant that Took Over the World，The Overlook Press，New York，2004.

［4］FAO，F(xiàn)AOSTAT Database.Food and Agriculture Organization of the United Nations.（http://faostat.fao.org/site/567/ DesktopDefault.aspx？PageID=567#ancor，on 2015-2-12）

［5］World Tea News Market Report.Issue 4.http://www. worldteanews.com/wp-content/uploads/Final-October-Market-Report.pdf，（2015-02-09）.

［6］J.Clay，World Agricultureand the Environment:A Commodityby-Commodity Guide to Impacts and Practices，Island Press，Washington，DC，2004.

［7］P.Carloni，L.Tiano，L.Padella，T.Bacchetti，C.Customu，A. Kay，E.Damiani，Antioxidant activity of white，green and black tea obtained from the same tea cultivar，F(xiàn)ood Res.Int. 53（2013）900-908.

［8］M.J.Kim，K.M.Maria John，J.N.Choi，S.Lee，A.J.Kim，Y.M.Kim，C.H.Lee，Changes in secondary metabolites of green tea during fermentation by Aspergillus oryzae and its effect on antioxidant potential，F(xiàn)ood Res.Int.53（2013）670-677.

［9］A.B.Sharangi，Medicinal and therapeutic potentialities of tea（Camellia sinensis L.）-a review，F(xiàn)ood Res.Int.42（2009）529-535.

［10］L.Zhang，Z.Z.Zhang，Y.B.Zhou，T.J.Ling，X.C.Wan，Chinese dark teas:postfermentation，chemistry and biological activities，F(xiàn)ood Res.Int.53（2013）600-607.

［11］Q.S.Chen，J.W.Zhao，X.Y.Huang，H.D.Zhang，M.H.Liu，Simultaneous determination of total polyphenols and caffeine contents of green tea by near-infrared reflectance spectroscopy，Microchem.J.83（2006）42-47.

［12］Q.S.Chen，J.W.Zhao，M.H.Liu，J.R.Cai，J.H.Liu，Determination of total polyphenols content in green tea using FT-NIR spectroscopy and different PLS algorithms，J.Pharm.Biomed. Anal.46（2008）568-573.

［13］Y.He，X.L.Li，X.F.Deng，Discrimination of varieties of tea using near infrared spectroscopy by principal component analysis and BP model，J.Food Eng.79（2007）1238-1242.

［14］J.Luypaert，M.H.Zhang，D.L.Massart，F(xiàn)easibility study for the use of near infrared spectroscopy in the qualitative and quantitative analysis of green tea，Camellia sinensis（L.），Anal. Chim.Acta 478（2003）303-312.

［15］H.Schulz，U.H.Engelhardt，A.Wegent，H.H.Drews，S. Lapczynski，Application of near-infrared reflectance spectroscopy to the simultaneous prediction of alkaloids and phenolic substances in green tea leaves，J.Agric.Food Chem. 47（1999）5064-5067.

［16］S.M.Tan，R.M.Luo，Y.P.Zhou，H.Gong，Z.Tan，Rapid and non-destructive discrimination of tea varieties by near infrared diffuse reflection spectroscopy coupled with classification and regression trees，Afr.J.Biotechnol.11（2012）2303-2312.

［17］J.E.Lee，B.J.Lee，J.O.Chung，J.A.Hwant，S.J.Lee，C.H.Lee，Y.S. Hong，Geographical and climatic dependencies of green tea（Camellia sinensis）metabolites:a1H NMR-based metabolomics study，J.Agric.Food Chem.58（2010）10582-10589.

［18］P.Madesis，I.Ganopoulos，I.Sakaridis，A.Argiriou，A. Tsaftaris，Advances of DNA-based methods for tracing the botanical origin of food products，F(xiàn)ood Res.Int.60（2014）163-172.

［19］S.Ni，M.Z.Yao，L.Chen，L.P.Zhao，X.C.Wang，Germplasm and breeding research of tea plant based on DNA marker approaches，F(xiàn)ront.Agric.Chin.2（2008）200-207.

［20］S.Longhi，L.Giongo，M.Buti，N.Surbanovski，R.Viola，R. Velasco，J.A.Ward，D.J.Sargent，Molecular genetics and genomics of the Rosoideae:state of the art and future perspectives，Hortic.Res.1（2014）1-18.

［21］M.Y.Stoeckle，C.C.Gamble，R.Kirpekar，G.Young，S.Ahmed，D.P.Little，Commercial teas highlight plant DNA barcode identification successes and obstacles，Sci.Rep.1（2011）42-49.

［22］S.S.Kaundun，S.Matsumoto，Identification of processed Japanese green tea based on polymorphisms generated by STS-RFLP analysis，J.Agric.Food Chem.51（2003）1765-1770.［23］R.K.Sharma，M.S.Negi，S.Sharma，P.Bhardwaj，R.Kumar，E. Bhattachrya，S.B.Tripathi，D.Vijayan，A.R.Baruah，S.C.Das，B. Bera，R.Rajkumar，J.Thomas，R.K.Sud，N.Muraleedharan，M. Hazarika，M.Lakshmikumaran，S.N.Raina，P.S.Ahuja，AFLP-based genetic diversity assessment of commercially important tea germplasm in India，Biochem.Genet.48（2010）549-564.

［24］T.Ujihara，S.Matsumoto，N.Hayashi，K.Kohata，Cultivar identification and analysis of the blended ratio of green tea production on the market using DNA markers，F(xiàn)ood Sci. Technol.Res.11（2005）43-45.

［25］T.Ujihara，F(xiàn).Taniguchi，J.Tanaka，N.Hayashi，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（2011）1557-1564.

［26］C.Y.Hu，Y.Z.Tsai，S.F.Lin，Development of STS and CAPS markers for variety identification and genetic diversity analysis of tea germplasm in Taiwan，Bot.Stud.55（2014）12-27.

［27］W.P.Fang，H.Cheng，Y.Duan，X.Jiang，X.Li，Genetic diversity and relationship of clonal tea（Camellia sinensis）cultivars in China as revealed by SSR markers，Plant Syst.Evol.298（2012）469-483.

［28］S.S.Kaundun，S.Matsumoto，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（2004）895-902.

［29］F.Taniguchi，K.Kimura，T.Saba，A.Ogino，S.Yamaguchi，J. Tanaka，Worldwide core collections of tea（Camellia sinensis）based on SSR markers，Tree Genet.Genomes 10（2014）1555-1565.

［30］R.K.Sharma，P.Bhardwaj，R.Negi，T.Mohapatra，P.S.Ahuja，Identification，characterization and utilization of unigene derived microsatellite markers in tea（Camellia sinensis L.），BMC Plant Biol.9（2009）53-77.

［31］T.Ujihara，R.Ohta，N.Hayashi，K.Kohata，J.Tanaka，Identification of Japanese and Chinese green tea cultivars by using simple sequence repeat markers to encourage proper labeling，Biosci.Biotechnol.Biochem.73（2009）15-20.

［32］L.Flores-Rentaria，A.Krohn，Scoring Microsatellite Loci，in: S.K.Kantartzi（Ed.），Microsatellites:Methods and Protocols，Series of Methods in Molecular Biology，vol.1006，Humana Press，New York 2013，pp.319-336.

［33］A.C.Kelly，N.E.Mateus-Pinilla，M.Douglas，M.Douglas，P. Shelton，J.Novakofski，Microsatellites behavingbadly:empirical evaluation of genotyping errors and subsequent impacts on population studies，Genet.Mol.Res.10（2011）2534-2553.

［34］A.Rafalski，Applications of single nucleotide polymorphisms in crop genetics，Curr.Opin.Plant Biol.5（2002）94-100.

［35］N.K.Korir，J.Han，L.F.Shangguan，C.Wang，E.Kayesh，Y.Y. Zhang，J.G.Fang，Plant variety and cultivar identification: advances and prospects，Crit.Rev.Biotechnol.33（2013）111-125.

［36］K.Weising，H.Nybom，M.Pfenninger，K.Wolff，G.Kahl，DNA FingerprintinginPlants:Principals，Methods，andApplications，second ed.CRC Press，Boca Raton，F(xiàn)lorida，2005.

［37］C.Bazakos，A.O.Dulger，A.T.Uncu，S.Spaniolas，T.Spano，P. Kalaitzis，A SNP-based PCR-RFLP capillary electrophoresis analysis for the identification of the varietal origin of olive oils，F(xiàn)ood Chem.134（2012）2411-2418.

［38］J.Q.Jin，L.Chen，M.Z.Yao，X.C.Wang，C.L.Ma，Simplification of EcoTILLING technique for tea plant，J.Tea Sci.30（2010）19-26（in Chinese with English abstract）.

［39］J.B.Yang，S.X.Yang，H.T.Li，J.Yang，D.Z.Li，Comparative chloroplast genomes of Camellia species，PLoS One 8（2013），e73053，http://dx.doi.org/10.1371/journal.pone.0073053.

［40］C.C.Zhang，L.Y.Wang，K.Wei，H.Cheng，Development and characterization of single nucleotide polymorphism markers in Camelliasinensis（Theaceae），Genet.Mol.Res.13（2014）5822-5831.

［41］W.P.Fang，L.W.Meinhardt，H.W.Tan，L.Zhou，S.Mischke，D. Zhang，Varietal identification of tea（Camellia sinensis）using nanofluidic array of single nucleotide polymorphism（SNP）markers，Hortic.Res.1（2014）1-8.

［42］Fluidigm SNP Genotyping User Guide.Rev.H1，PN 68000098，South San Francisco，CA:Fluidigm Corporation.Rev.，（http:// www.mscience.com.au/upload/pages/fluidigmtech/ fluidigm-snp-genotyping-user-guide-151112.pdf，2015-2-10）

［43］R.Peakall，P.E.Smouse，Genalex 6:genetic analysis in Excel. Population genetic software for teaching and research，Mol. Ecol.Notes 6（2006）288-295.

［44］R.Peakall，P.E.Smouse，GenAlEx 6.5:genetic analysis in Excel. Population genetic software for teaching and research-an update，Bioinformatics 28（2012）2537-2539.

［45］L.P.Waits，G.Luikart，P.Taberlet，Estimating the probability of identity among genotypes in natural populations:cautions and guidelines，Mol.Ecol.10（2001）249-256.

［46］M.Nei，F(xiàn).Tajima，Y.Tateno，Accuracyofestimatedphylogenetic trees from molecular data，Anal.Chim.Acta 19（1983）153-170.

［47］D.Dieringer，C.Schl?tterer，Microsatellite Analyser（MSA）:a platform independent analysis tool for large microsatellite data sets，Mol.Ecol.Notes 3（2003）167-169.

［48］N.Saitou，M.Nei，The neighbor-joining method:a new method for reconstructing phylogenetic trees，Mol.Biol.Evol. 4（1987）406-425.

［49］Food and Agriculture Organization.Committee on Commodity Problems，Intergovernmental Group on Tea，18th Session，Hangzhou，China.（ftp://ftp.fao.org/docrep/fao/ Meeting/013/K2054e.pdf，2015-2-12）.

［50］J.Wang，M.Lin，A.Crenshaw，A.Hutchinson，B.Hicks，M. Yeager，S.Berndt，W.Y.Huang，B.Hayes，S.J.Chanock，R.C. Jones，R.Ramakrishnan，High-throughput single nucleotide polymorphism genotyping using nanofluidic dynamic arrays，BMC Genomics 10（2009）561，http://dx.doi.org/10.1186/ 1471-2164-10-561.

［51］W.P.Fang，L.W.Meinhardt，S.Mischke，C.M.Bellato，L. Motilal，D.Zhang，Accurate determination of genetic identity for a single cacao bean，using molecular markers with a nanofluidic system，ensures cocoa authentication，J.Agric. Food Chem.62（2014）481-487.

［52］L.Chen，Z.Apostolides，Z.M.Chen，Global Tea Breeding: Achievements，Challenges and Perspectives，Zhejiang University Press&Springer-Verlag，Zhejiang，China&Berlin，Germany，2012.

［53］P.Bhardwaj，R.K.Sharma，R.Kumar，H.Sharma，P.S.Ahuja，SSR marker based DNA fingerprinting and diversity assessment in superior tea germplasm cultivated in Western Himalaya，Proc.Indian Natl.Sci.Acad.80（2014）157-162.

［54］M.Z.Yao，C.L.Ma，T.T.Qiao，J.Q.Jin，L.Chen，Diversity distribution and population structure of tea germplasms in China revealed by EST-SSR markers，Tree Genet.Genomes 8（2012）205-220.

［55］L.Chen，Z.X.Zhou，Y.J.Yang，Genetic improvement and breeding of tea plant（Camellia sinensis）in China:from individual selection to hybridization and molecular breeding，Euphytica 154（2007）239-248.

［56］T.L.Ming，Monograph of the Genus Camellia，Yunnan Science and Technology Press，Kunming，China，2000.

［57］J.R.Sealy，ARevisionoftheGenusCamellia，RoyalHorticultural Society，London，1958.

［58］T.D.Vo，Assessing Genetic Diversity in Vietnam Tea［Camellia sinensis（L.）O.Kuntze］Using Morphology，Inter-Simple Sequence Repeat（lSSR）and Microsatellite（SSR）Markers（Dissertation for the Doctoral Degree）Georg-August G?ttingen University，G?ttingen，Germany，2006.

［59］F.N.Wachira，R.Waugh，W.Powell，C.A.Hackett，Detection of genetic diversity in tea（Camellia sinensis）using RAPD markers，Genome 38（1995）201-210.

26 October 2015

Abbreviations:IFC，integrated fluidic circuit；PCoA，principal coordinates analysis；PIC，polymorphism information content；PID，probability of identity；PID-sib，probability of identity among siblings；STA，specific target amplification.

*Corresponding author.Tel.:+1 301 504 7477；fax:+1 301 504 1998.

E-mail address:Dapeng.Zhang@ars.usda.gov（D.Zhang）.

Peer review under responsibility of Crop Science Society of China and Institute of Crop Science，CAAS.

http://dx.doi.org/10.1016/j.cj.2016.02.001

2214-5141/Production and hosting by Elsevier B.V.on behalf of Crop Science Society of China and Institute of Crop Science，CAAS.This is an open access article under the CC BY-NC-ND license（http://creativecommons.org/licenses/by-nc-nd/4.0/）.