2015 Volume 21 Issue 3 Pages 381-389
Nucleotide sequences of the ribulose 1,5-bisphosphate carboxylase large-subunit (rbcL) in chloroplast (cp) DNA and 18S ribosomal RNA encoded in nuclear DNA (nuclear-encoded 18S-rRNA) were determined to clarify genetic diversity among cultivated teas. The nucleotide sequence alignment of rbcL in cpDNA enabled the taxonomic distinction of the section Thea. On the basis of these data, the rbcL nucleotide sequences of 195 cultivated teas in the Asian region were classified. The cultivated teas are separated into five groups: CSs type (var. sinensis), CSa type (var. assamica), type CSaa (like var. assamica), CIt type of C. irrawadiensis, and CTa type of C. taliensis. All the cultivated teas from the East Asia (China, Korea, and Japan) region were of the CSs type, whereas the cultivated teas from Bangladesh, Laos, Sri Lanka and Thailand were CSa type. The cultivated teas from India were separated into CSs and CSa types. The rbcL nucleotide sequences of the teas currently cultivated in Yunnan, China were CSa and CIt types, and the teas of Vietnam were CSa, CSs and CIt types. On the other hand, the teas cultivated in Myanmar included all types of CSs, CSa, CSaa, CIt and CTa. In addition, differences in nucleotide sequences were observed at three positions in 870 nuclear-encoded 18S-rRNA nucleotide sequences. The different nucleotide sequences of nuclear-encoded 18S-rRNA distinguished C. sinensis (var. sinensis and var. assamica) from wild tea species. Similarly, almost all cultivars tested were divided into the Cs type of C. sinensis (var. sinensis and var. assamica) and Cw type of wild tea species by alignment of nuclear-encoded 18S-rRNA nucleotide sequences, with an interspecific hybrid between C. sinensis and C. taliensis identified in a region of Myanmar. Analysis of genetic relationships indicated that many teas cultivated in the Myanmar region differ from the tea groups of East and South Asia. Our data indicated that the endemically cultivated teas in parts of Myanmar are conservationally important as new sources of desirable teas for future breeding programs and improvement of tea products.
Tea (Camellia sinensis L.) is naturally distributed in the Asian monsoon region. C. sinensis leaf tea is a major agricultural commodity produced in several Asian, African and South American countries. Due to its high economic importance, understanding of the genetic diversity and taxonomic characteristics of each individual cultivated tea is required to establish a core collection and to provide a source of desirable genes that are accessible to tea breeders.
Based on its morphological leaf and growth characteristics and geographical origin, the species C. sinensis have been classified into three taxa: C. sinensis var. sinensis (L.) O. Kuntze, C. sinensis var. assamica (Masters) Kitamura, and C. sinensis var. assamica spp. lasiocalyx Planchon ex Watt (Sealy, 1958, Wright, 1962). In addition, C. taliensis and C. irrawadiensis, which, like C. sinensis, belong to the section Thea, are naturally distributed in parts of Myanmar, Vietnam, and Yunnan, China. Some natural hybrids of C. sinensis and C. irrawadiensis are known to exist in these areas (Wright and Barua, 1957, Wood and Barua, 1958). Katoh et al., (2003) also reported that endemic cultivated teas in Myanmar and Yunnan, China are genetically similarity to C. irrawadiensis and C. taliensis. Most cultivated teas are possible hybrids resulting from the natural crossing of morphologically close taxa. Thus, it has been difficult to demonstrate the purity of cultivated tea species by morphologic and geographic characteristics. However, identification and classification at the taxonomic level is important in understanding the amount and distribution of genetic variation present in the gene pool, to establish a core collection for future improvement of tea products, and to provide a source of desirable genes.
A number of recently developed molecular techniques are efficient methods for evaluating genetic diversity in higher plants. Randomly amplified polymorphic DNAs (RAPDs) and amplified fragment length polymorphism (AFLP) have been demonstrated to be informative in the determination of the level of genetic diversity within cultivated teas (Wachira et al., 1995, 1996, 2001, Kaundun et al., 2000). In addition, the rate of chloroplast DNA (cpDNA) variation can elucidate interspecific polymorphism and times of divergence in evolution. Maternal transmission and the absence of recombination can determine phylogenetic relationships among species. Information on cpDNA variation has been widely used in pedigree analysis and population differentiation (Ecke and Michaelis, 1990, Taberlet et al., 1992, Demesure et al., 1995). Nuclear-encoded ribosomal RNA (nuclear-encoded rRNA) can be used as to trace the phylogenetic relationships of higher plants (Von de Peer et al., 1996). Moreover, the diversity of nuclear-encoded rRNA has been used extensively for the construction of genetic linkage maps (Douglas et al., 1991, Les et al., 1991).
The comparison of cpDNA with maternal inheritance information has been useful for evaluating inter- and intraspecific relationships among near or distant taxa. The ribulose 1,5-bisphosphate carboxylase large-subunit (rbcL) in cpDNA is an indispensable component of photosynthetic carbon metabolism and is thus ubiquitous among green plants. The rbcL nucleotide sequence data have been accumulated for angiosperms (Les et al., 1991, Duvall et al., 1993, Hasebe et al., 1994). Furthermore, Shinozaki et al. (1983) have reported that the primary structure of rbcL in cpDNA is biochemically conserved because the rate of amino acid substitution is far below the value for hemoglobin α. The rate of nucleotide substitution is estimated to be 1.1 × 10−9 per year (Curtis and Clegg, 1984, Zurawski et al., 1984, Wolfe et al., 1987). The alignment of rbcL sequences is a straight-forward matter because its locus is highly conserved. A comparison of rbcL nucleotide sequences is often used for inference of phylogenetic relationships at higher taxonomic levels. Moreover, evidence of genetic diversity at higher taxonomic levels is especially needed in the section Thea, in which phylogenies based on morphological characters are concordant. Because small subunit ribosomal RNA (SSU-rRNA) sequences are known for many taxa, the analysis of SSU-rRNA nucleotide sequences can also be used to distinguish the phylogenetic relationships among taxa.
Based on the findings described above, the genes of rbcL in cpDNA and rRNA in nuclear DNA are the most fitting to determine the genetic relationships within domestically cultivated teas of unknown classification at the taxonomic level. Previously, the results of matK (ribosomal RNA maturase in chloroplast DNA) nucleotide sequence analysis indicated that the two taxa of C. sinensis var. sinensis and C. sinensis var. assamica are indistinguishable from each other (Katoh et al., 2003). It is thought that accurate taxonomic assessment is important for the conservation of sources of desirable genes. Therefore, we attempted to distinguish the two taxa of C. sinensis in the section Thea based on rbcL nucleotide sequence analysis. In addition, the nucleotide sequence analysis of the rbcL gene in cpDNA and 18S-rRNA in nuclear DNA (nuclear-encoded 18S-rRNA) will be useful as a tool to understand the pedigrees of domestically cultivated teas in Southeastern Asia and to conserve new sources of genes desirable for future improvement and breeding programs. In this paper, we report the identification and classification of the tea species C. sinensis, C. irrawadiensis and C. taliensis, based on nucleotide sequence comparison of rbcL in cpDNA and nuclear-encoded 18S-rRNA, and the genetic diversity within cultivated teas in the Asian tea-producing districts (India, Sri Lanka, Bangladesh, Laos, Thailand, Myanmar, Vietnam, China, Korea and Japan). In addition, we show that the endemic cultivated teas of the Myanmar region are conservationally important as new desirable sources for future improvement and breeding programs.
Plant material Teas cultivated in various Asian regions were examined (Table 1). One hundred and ninety-five cultivated teas were obtained from Bangladesh, China, India, Japan, Korea, Laos, Myanmar, Sri Lanka, Thailand and Vietnam. The samples tested in this study were obtained from traditional tea production areas or major tea estates in each country. Five samples of C. sinensis var. sinensis, C. sinensis var. assamica, C. taliensis, C. irrawadiensis, and C. chrysantha obtained from the National Institute of Vegetable and Tea Science (Makurazaki and Kanaya, Japan), and two hybrid samples (C. sinensis × C. taliensis and C. sinensis × C. chrysantha) of tea cultivars derived from seedling selection after interspecific crosses were also examined.
| Population name | Locality | Country | Number | Variety |
|---|---|---|---|---|
| Kanaya | Shizuoka | Japan | 7 | C. sinensis var. sinensis |
| Numazu | Shizuoka | Japan | 7 | C. sinensis var. sinensis |
| Kyoto | Kyoto | Japan | 5 | C. sinensis var. sinensis |
| Kito | Tokushima | Japan | 10 | C. sinensis var. sinensis |
| Amaki | Fukuoka | Japan | 5 | C. sinensis var. sinensis |
| Seburi-yama | Saga | Japan | 1 | C. sinensis var. sinensis |
| Itsuki | Kumamoto | Japan | 5 | C. sinensis var. sinensis |
| Sunchon | Cholla-numudo | Korea | 1 | unknown |
| Shaoxing | Zhejiang | China | 10 | C. sinensis var. sinensis |
| Wuyaian | Anhui | China | 10 | C. sinensis var. sinensis |
| Qimen | Anhui | China | 10 | C. sinensis var. sinensis |
| Ninzhou | Jiangxi | China | 10 | C. sinensis var. sinensis |
| Xishuanbanna | Yunnan | China | 3 | unknown |
| Fugong | Yunnan | China | 3 | unknown |
| Luxi | Yunnan | China | 3 | unknown |
| Bangwei | Yunnan | China | 1 | unknown |
| Banshan | Yunnan | China | 1 | unknown |
| Phong Tho | Lai Chau | Vietnam | 1 | unknown |
| Suoi Giang | Yen Bai | Vietnam | 2 | unknown |
| Bac Ha | Lao Cai | Vietnam | 1 | unknown |
| Nanhsan | Shan | Myanmar | 1 | unknown |
| Lwesai | Shan | Myanmar | 1 | unknown |
| Myine | Shan | Myanmar | 5 | unknown |
| Taunggyi | Shan | Myanmar | 4 | unknown |
| Pindaya | Shan | Myanmar | 5 | unknown |
| Khamti | Sagaing | Myanmar | 5 | unknown |
| Tonhei | Sagaing | Myanmar | 2 | unknown |
| Sunparabun | Kachin | Myanmar | 11 | unknown |
| In San Yan | Kachin | Myanmar | 3 | unknown |
| Sopii | North Laos | Laos | 2 | unknown |
| Paksong | South Laos | Laos | 1 | unknown |
| Doi Saket | Chaing Mai | Thailand | 1 | unknown |
| Jorhat | Assam | India | 8 | C. sinensis var. assamica |
| Dibrugarh | Assam | India | 8 | C. sinensis var. assamica |
| Kaziranga | Assam | India | 8 | C. sinensis var. assamica |
| Shillong | Meghalaya | India | 8 | C. sinensis var. assamica |
| Darjeeling | Sikkim | India | 5 | C. sinensis var. sinensis |
| 3 | C. sinensis var. assamica | |||
| 10 | unknown | |||
| Chittagong | Chittagong | Bangladesh | 3 | C. sinensis var. assamica |
| Kandy | Central | Sri Lanka | 3 | unknown |
| Nuwara Eliya | Central | Sri Lanka | 1 | unknown |
| Hatton | Central | Sri Lanka | 1 | unknown |
DNA extraction and purification DNA was extracted from 50 mg of leaf material of all tea samples used in the experiments, and from 20 mg of anthers of C. sinensis var. sinensis, C. sinensis var. assamica, C. taliensis, C. irrawadiensis and C. chrysantha. The total DNA was purified using a DNeasy Plant Mini Kit (Qiagen Sciences., Maryland, USA). The purified DNA was dissolved in 10 mM Tris-HCl buffer (pH 8.0) containing 0.1 mM EDTA and stored at −30°C.
Amplification of ribulose 1,5-bisphosphate carboxylase large-subunit (rbcL) in cpDNA and 18S ribosomal RNA in nuclear DNA (nuclear-encoded 18S-rRNA) A partial region (1,300 bp) of rbcL (Accession number: AF380037) from the isolated DNA was amplified by PCR using an Applied Biosystems thermocycler 9700. Fifty microliters of the PCR reaction mixture contained 0.1 µg of DNA, 1 × universal buffer, 0.4 mM dNTPs, 0.2 µM of each of the forward and reverse primers, and 2 units of Taq DNA polymerase (Roche Diagnostics). The 6 primers used in the PCR reaction were universal primers to amplify the nucleotide sequences of rbcL genes in the genus Camellia: F1 (5′-ATGTCACCACAAACAGAGACTA AAGC-3′), R1 (5′-TCAACTTGGATGCCATGAG-3′), MF (5′-TA TCCCTTAGACCTCTTCGAAGAAGGTTC-3′), MR (5′-CGTTCA CCTTCTAGTTTACCTACAACAGT-3′), EF (5′-TGAAAACGTG AATTCCCAACCGTTTATGCG-3′), and ER (5′-GCAGCAGCTA GTTCCGGGCTCCA-3′), respectively. PCR conditions were as follows: initial denaturation for 2 min at 94°C, 35 cycles consisting of denaturation for 30 s at 94°C, annealing for 30 s at 48°C and extension for 1 min at 72°C, and a final extension step for 7 min at 72°C.
On the other hand, the nuclear encoded-18S-rRNA was amplified to evaluate the genetic diversity of samples of unknown varieties, as tea plants are outbreeders and highly heterogeneous. A forward primer F1 (5′-GTCTCAAAGATTAAGCCATGC-3′) and a reverse primer R1 (5′-CTGACTATGAAATACGAATGCC-3′) were designed for amplification of a multi-mutation region (870 bp) of nuclear-encoded 18S-rRNA (1,800 bp) (Accession number: AB120309). PCR amplifications were carried out using an Applied Biosystems thermocycler 9700 in a 50-µL reaction mixture with Taq DNA polymerase (Roche Diagnostics) according to the manufacturer's instructions. PCR conditions were: initial denaturation for 2 min at 94°C, 45 cycles consisting of denaturation for 30 sec at 94°C, annealing for 30 sec at 53°C and extension for 1 min at 72°C, with a final extension step of 7 min at 72°C.
Two microliters of PCR products were confirmed using 2% agarose gel electrophoresis (FMC BioProducts) and visualized by ethidium bromide staining. PCR products were stored at −20°C.
DNA fragment purification and sequencing PCR products were electrophoresed on 2% agarose gels in 1 × TAE buffer. Gels were stained with ethidium bromide for 30 min. DNA fragments were purified with TaKaRa Easy Trap (Takara Bio Co.). The purified DNA fragments were sequenced with a Big-Dye terminator cycle sequencing kit (Applied Biosystems) and the sequencing primers as described for the amplification of rbcL and nuclear-encoded 18S-rRNA. The fragments were analyzed on an ABI-PRISM 377 DNA sequencer (Applied Biosystems). DNA sequencing was repeated six times on the same sample, and nucleotide sequences of rbcL and nuclear-encoded 18S-rRNA were aligned using Gene Works software (Teijin Co.).
Nucleotide sequences of rbcL and nuclear-encoded 18S-rRNA in C. sinensis, C. taliensis, C. irrawadiensis and C. chrysantha The nucleotide sequences of rbcL in the cpDNA of C. sinensis var. sinensis, C. sinensis var. assamica, C. taliensis, C. irrawadiensis and C. chrysantha are shown in Table 2. The nucleotide sequences of the rbcL gene differed among the five tea taxa. Differences of nucleotide sequences were observed at positions 40, 627 and 948 in the 1,300-rbcL nucleotide sequences. In C. sinensis var. assamica, the bases detected at positions 40, 627, and 948 of the rbcL nucleotide sequences were cytosine (C), guanine (G), and adenine (A), respectively, while in C. sinensis var. sinensis, A40, G627 and G948 bases were observed. On the other hand, a different base was detected in C. irrawadiensis and C. taliensis at position 627 of rbcL nucleotide sequences. The specific base of C. irrawadiensis was thymine (T), whereas that of C. taliensis was adenine (Table 2). Overall, the nucleotide sequences of C. irrawadiensis and C. taliensis differed from those of C. sinensis. The two taxa (var. assamica and var. sinensis) of the section Thea were easily distinguished by comparison of the nucleotide sequences of rbcL in cpDNA. In addition, a base substitution was detected in C. chrysantha at position 692 of rbcL nucleotide sequences (Table 2). The characteristic base was thymine. The rbcL nucleotide sequences of the two hybrids were the same as that of C. sinensis (Table 2).
| Sample Name | Base and position of substitution | |||||||
|---|---|---|---|---|---|---|---|---|
| rbcL | 18S-rRNA | |||||||
| 40 | 281 | 627 | 692 | 948 | 15 | 390 | 481 | |
| Kanaya/Japan (C. sinensis var. sinensis) | A | T | G | G | G | T | A | C |
| Chittagong/Bangladesh (C. sinensis var. assamica) | C | T | G | G | A | T | A | C |
| Makurazaki/Japan (C. irrawadiensis) | C | T | T | G | A | C | G | T |
| Makurazaki/Japan (C. taliensis) | C | T | A | G | A | C | G | T |
| Makurazaki /Japan (C. chrysantha) | C | T | G | T | A | C | A | T |
| Makurazaki /Japan (C. sinensis × C. taliensis) | A | T | G | G | G | C | G | T |
| Makurazaki /Japan (C. sinensis × C. chrysantha) | A | T | G | G | G | C | A | T |
| Taunggyi /Myanmar-(unknown) | C | C | G | G | A | T | A | C |
| Lwesai /Myanmar-(unknown) | C | T | A | G | A | T | A | C |
The nucleotide sequences of the nuclear-encoded 18S-rRNA of C. sinensis var. sinensis, C. sinensis var. assamica, C. taliensis, C. irrawadiensis, and C. chrysantha were aligned. The nucleotide sequences of nuclear-encoded 18S-rRNA of C. sinensis (var. sinensis and var. assamica) and the wild tea species (C. taliensis and C. irrawadiensis) differed (Table 2). As shown in Table 2, differences in nucleotide sequences were observed at positions 15, 390, and 481 in the 500 nuclear-encoded 18S-rRNA nucleotide sequences. In C. sinensis (var. sinensis and var. assamica), the bases detected at positions 15, 390, and 481 of the nuclear-encoded 18S-rRNA nucleotide sequences were T, A, and C, respectively, while C15, G390, and T481 bases were observed in the wild species (C. taliensis and C. irrawadiensis). The bases detected at positions 15, 390, and 481 of the nuclear-encoded 18S-rRNA nucleotide sequences in C. chrysantha were C, A, and T, respectively. Additionally, the nucleotide sequences of the nuclear-encoded 18S-rRNA of the two hybrids were determined. As shown in Table 2, these nuclear-encoded 18S-rRNA nucleotide sequences corresponded with that of C. taliensis or C. chrysantha.
Genetic differences of rbcL DNA among cultivated teas The rbcL nucleotide sequences of 195 cultivated teas were examined. From the nucleotide sequence alignments, base differences were detected at positions 40, 281, 627, and 948 in the rbcL DNA (Table 3). The rbcL nucleotide sequences of the cultivated teas were defined as five types (CSs, CSa, CSaa, CIt, and CTa).
| Type | Base and position of substitution | |||
|---|---|---|---|---|
| 40 | 281 | 627 | 948 | |
| CSs | A | T | G | G |
| CSa | C | T | G | A |
| CSaa | C | C | G | A |
| CIt | C | T | T | A |
| CTa | C | T | A | A |
As compared with the rbcL sequence data of Table 2, the five types of CSs, CSa, CSaa, CIt, and CTa were shown as follows: the CSs type had the same rbcL nucleotide sequences as C. sinensis var. sinensis. The CSa type was the same as C. sinensis var. assamica. The CSaa type was placed in the fifth group as unknown. The rbcL nucleotide sequences of CSaa type were the same as that of CSa except for a cytosine base at position 281. The groups of C. irrawadiensis belonged to type CIt, whereas type CTa was revealed to belong to C. taliensis.
The variations of rbcL of individual tea are shown in Table 4. Eighty-one samples from the eastern part of China, Korea and Japan were all identified as CSs type, whereas 12 cultivars of Bangladesh, Laos, Sri Lanka and Thailand were identified as CSa type. The 50 cultivars of Assam and Meghalaya in India were separated into 13 CSs and 37 CSa types. On the other hand, samples of Yunnan (China), Myanmar, and Vietnam were revealed to comprise 5 types (CSs, CSa, CSaa, CIt, and CTa) (Table 4). The 11 cultivated teas of Yunnan (China) were divided into 9 CSa, 1 CSs, and 1 CIt. The samples from Vietnam were revealed to be 1 CSa, 2 CIt, and 1 CSs. Additionally, the cultivated teas of Myanmar were revealed to be 3 CSs, 21 CSa, 7 CSaa, 2 CIt, and 4 CTa types. The CSaa type was observed in the tea samples from the Myanmar regions of Taunggyi, Nanhsan, and Pindaya.
| Population name | Locality | Country | Number | rbcL | 18S-rRNA |
|---|---|---|---|---|---|
| Kanaya | Shizuoka | Japan | 7 | CSs | Cs |
| Numazu | Shizuoka | Japan | 7 | CSs | Cs |
| Kyoto | Kyoto | Japan | 5 | CSs | Cs |
| Kito | Tokushima | Japan | 10 | CSs | Cs |
| Amaki | Fukuoka | Japan | 5 | CSs | Cs |
| Seburi-yama | Saga | Japan | 1 | CSs | Cs |
| Itsuki | Kumamoto | Japan | 5 | CSs | Cs |
| Sunchon | Cholla-numudo | Korea | 1 | CSs | Cs |
| Shaoxing | Zhejiang | China | 10 | CSs | Cs |
| Wuyaian | Anhui | China | 10 | CSs | Cs |
| Qimen | Anhui | China | 10 | CSs | Cs |
| Ninzhou | Jiangxi | China | 10 | CSs | Cs |
| Xishuanbanna | Yunnan | China | 2 | CSa | Cs |
| 1 | CIt | Cw | |||
| Fugong | Yunnan | China | 2 | CSa | Cs |
| 1 | CSs | Cs | |||
| Luxi | Yunnan | China | 3 | CSa | Cs |
| Bangwei | Yunnan | China | 1 | CSa | Cs |
| Banshan | Yunnan | China | 1 | CSa | Cs |
| Phong Tho | Lai Chau | Vietnam | 1 | CSa | Cs |
| Suoi Giang | Yen Bai | Vietnam | 2 | CIt | Cw |
| Bac Ha | Lao Cai | Vietnam | 1 | CSs | Cs |
| Nanhsan | Shan | Myanmar | 1 | CSaa | Cs |
| Lwesai | Shan | Myanmar | 1 | CTa | Cs |
| Myine | Shan | Myanmar | 3 | CTa | Cw |
| 2 | CIt | Cw | |||
| Taunggyi | Shan | Myanmar | 4 | CSaa | Cs |
| Pindaya | Shan | Myanmar | 2 | CSaa | Cs |
| 3 | CSa | Cs | |||
| Khamti | Sagaing | Myanmar | 5 | CSa | Cs |
| Tonhei | Sagaing | Myanmar | 2 | CSa | Cs |
| Sunparabun | Kachin | Myanmar | 11 | CSa | Cs |
| In San Yan | Kachin | Myanmar | 3 | CSs | Cs |
| Sopii | North Laos | Laos | 2 | CSa | Cs |
| Paksong | South Laos | Laos | 1 | CSa | Cs |
| Doi Saket | Chaing Mai | Thailand | 1 | CSa | Cs |
| Jorhat | Assam | India | 8 | CSa | Cs |
| Dibrugarh | Assam | India | 8 | CSa | Cs |
| Kaziranga | Assam | India | 8 | CSa | Cs |
| Shillong | Meghalaya | India | 8 | CSa | Cs |
| Darjeeling | Sikkim | India | 13 | CSs | Cs |
| 5 | CSa | Cs | |||
| Chittagong | Chittagong | Bangladesh | 3 | CSa | Cs |
| Kandy | Central | Sri Lanka | 3 | CSa | Cs |
| Nuwara Eliya | Central | Sri Lanka | 1 | CSa | Cs |
| Hatton | Central | Sri Lanka | 1 | CSa | Cs |
Symbols (CSs, CSa, CSaa, CIt, CTa) of rbcL are the same as in Table 3. Cs and Cw of 18S-rRNA represent Camellia sinensis and wild species (C. irrawadiensis and C. taliensis), respectively. Comparison of matK nucleotide sequences resulted in the following classification: CSs was placed in the CJ and AA types, CSa in types AA, AB, AE and AD, CIt in types AC, IC and IM, and CTa in types TM and TV (Katoh et al., 2003).
Genetic differences in nuclear-encoded 18S-rRNA among cultivated teas The nuclear-encoded 18S-rRNA nucleotide sequences of the cultivated teas were divided into two types. One hundred and forty-three samples from eastern China, Korea, Japan, Bangladesh, Laos, Sri Lanka, and India were all identified as types of C. sinensis (var. sinensis and var. assamica). On the other hand, samples from Yunnan (China), Myanmar, and Vietnam were revealed to be of the following two types. The 52 teas currently cultivated in Yunnan (China), Myanmar, and Vietnam were identified as types of C. sinensis (var. sinensis and var. assamica). The 9 endemic teas classified as types CIt or CTa based on the analysis of rbcL nucleotide sequences were placed in the group of wild tea species (C. irrawadiensis and C. taliensis). However, 1 of the CTa samples in Myanmar had the same nucleotide sequence of nuclear-encoded 18S-rRNA as C. sinensis. The Lwesai sample from Myanmar was grouped with C. sinensis according to the nuclear-encoded 18S-rRNA nucleotide sequence alignment (Table 2). The variations in nuclear-encoded 18S-rRNA of individual teas are shown in Table 4.
Due to the economic and agricultural value of tea, attempts have been made to estimate the level of genetic diversity in cultivated teas to aid in breeding and conservation. The classification of C. sinensis has been reported. Paul et al. (1997) classified the cultivated teas into China (var. sinensis), Assam (var. assamica), and Cambod types (assamica ssp. lasiocalyx) using AFLP markers. Magoma et al. (2000) demonstrated that it was possible to separate C. sinensis var. assamica, C. sinensis var. sinensis, and C. sinensis var. assamica ssp. lasiocalyx using catechins as biochemical markers. However, biological descriptors are considered to be less useful in analyzing the genetic relationships among closed taxa because they are greatly affected by environmental conditions. Additionally, RFLP and AFLP require large amounts or higher grades of DNA, and RAPD lacks reproducibility (Chen and Yamaguchi, 2002). Nevertheless, the groupings in C. sinensis and its wild Camellia relatives were assessed using RAPD and AFLP markers. Wachira et al. (2001) reported that two of the groups corresponded to varieties of assamica and sinensis, while the third group consisted of a heterogeneous mix of teas. Chen and Yamaguchi (2002) re-constructed the phylogenetic relationships of the section Thea in the genus Camellia using RAPD markers. It is possible to separate C. sinensis var. sinensis, C. sinensis var. assamica, and C. sinensis var. assamica ssp. lasiocalyx by biological descriptors and RFLP, RAPD, and AFLP markers. In this study, we demonstrated that the rbcL and 18S-rRNA nucleotide sequence analysis could distinguish tea taxa at the species level or below, as with the method using RFLP, RAPD, and AFLP ((Jones et al., 1997, Chan et al., 2012, Hayashi and Takeda, 2012).
In the section Thea, C. sinensis var. sinensis, C. sinensis var. assamica, C. irrawadiensis, and C. taliensis were classified into five groups by five base differences in the rbcL nucleotide sequences in cpDNA. The nucleotide sequence data comparison supports the view that the tea varieties and wild species have high genetic similarity to each other (Katoh et al., 2003). Nevertheless, the rbcL locus is taxonomically conservative because the rate of nucleotide substitution is estimated to be about 1.1 × 10−9 per year per site (Curtis and Clegg, 1984, Zurawski et al., 1984, Wolfe et al., 1987). This finding indicates that a single base change in rbcL nucleotide sequences is a significantly distinctive characteristic for discrimination of the genetic relationships among taxa. Consequently, the difference of bases at positions 40, 627, and 948 in the rbcL nucleotide sequences in this study allowed complete separation of two tea varieties (C. sinensis var. sinensis and C. sinensis var. assamica) and two wild tea species (C. irrawadiensis and C. taliensis).
Further, the genetic diversity within the cultivated teas was assessed. Comparison of rbcL nucleotide sequences classified the samples into five types: CSs, CSa, CSaa, CIt, and CTa. The CSs type comprises the cultivated teas of eastern China excluding Yunnan, Japan and Korea. In addition, tea samples from India, Bangladesh, Laos, Sri Lanka, and Thailand were placed into types CSs and/or CSa. All commercial cultivars from these countries were types CSs or CSa. The comparison of nucleotide sequences shows that type CSs has the same rbcL as C. sinensis var. sinensis. The CSa type belongs to C. sinensis var. assamica. On the other hand, all the types CSa, CSs, CSaa, CIt, and CTa were found in endemic and currently cultivated teas in parts of Myanmar. The samples from Yunnan (China) were CSa and CIt types, and the teas of Vietnam were CSa, CSs, and CIt types. The analysis of nucleotide sequences indicates that type CTa was a member of C. taliensis, while the CIt type was a member of C. irrawadiensis. It has been reported that several cultivated teas in the southeastern Asia region differ from the tea group of East Asia. In regions of Myanmar, some hybrids of C. sinensis and C. irrawadiensis are already known to exist among the cultivated teas (Wood and Barua, 1958). Wachira et al. (1995) has shown the existence of a commercial tea cultivar related to a lineage of C. irrawadiensis in the southeastern Asia region. Katoh et al. (2003) also showed that several endemic teas were clustered with two species of C. irrawadiensis and C. taliensis. The members of the southeastern Asia region were revealed to be types CJ, AA, AB, AC, AE, AD, IC, IM, TM, and TV by the comparison of maturase K (matK) nucleotide sequences in cpDNA (Katoh et al. 2003). The matK sequences were fragmented into smaller clusters as compared to the rbcL sequences. Based on the comparison of rbcL nucleotide sequences, these matK types were classified as follows: the CJ type was placed in the CSs group; the AB, AE, and AD types were placed in the CSa group; and then the AC, IC, and IM types were placed in the CIt group, and the TM and TV types in the CTa group. The AA type of matK had a strong affinity to both CSs and CSa groups. Additionally, this AA type is closely related to the CSaa type of rbcL sequences of teas from the regions of Taunggyi, Nanhsan, and Pindaya in Myanmar. More specifically, the rbcL amino acid sequences of the CSaa type were the same as that of C. sinensis var. assamica, because the characteristic base T281 of CSaa is a silent mutation that induced synonymous substitutions in the rbcL gene. Considering that the type related to C. chrysantha was not detected, the CSaa type may be related to C. sinensis var. assamica ssp. lasiocalyx, which belongs to C. sinensis var. assamica. Moreover, the CSaa type of teas from the regions of Taunggyi, Nanhsan, and Pindaya in Myanmar is conservationally important as a new source of desirable genes.
Chloroplasts in C. sinensis are genetic evidence of maternal inheritance (Carriveau and Coleman, 1988). The analysis of rbcL in cpDNA will be useful in clarifying the origin of one side of the genome of the cultivars. Tea plants are outbreeders and highly heterogeneous (Ackerman, 1973, Takeda, 1990). Owing to extensive internal hybridization between taxa, several intergrades and introgressants have been found (Wood and Barua, 1958). Classification of these hybrids is mainly based on morphological characteristics, which are greatly influenced by environmental factors. Therefore, it is difficult to identify the hybrids by their specific morphological similarity. Assessment of hybridization at a taxonomic level is important for individual clone identification used to establish a core collection. As a result of 18S-rRNA nucleotide sequence analysis of cultivated teas, type Cs was put into C. sinensis and type Cw belongs to C. taliensis or C. irrawadiensis. The 18S-rRNA nucleotide sequence of the sample Lwesai in Myanmar, classified as type CTa according to rbcL analysis, is the same as that of C. sinensis. Wright and Barua (1957), and Wood and Barua (1958) previously reported that an interspecific hybrid exists in regions of Myanmar. An endemic cultivated tea, Lwesai likely resulted from seedling selection after open pollination. We thus emphasize that nucleotide sequence analysis of nuclear-encoded 18S-rRNA is a potential tool in discriminating the origin of genomes in cultivated teas (Zao et al., 2014).
Overall, discrimination using cpDNA will enable understanding of the pedigrees of cultivated teas. The nucleotide sequence of rbcL in cpDNA is a significantly distinctive characteristic that can discriminate genetic relationships among taxa. In addition, Katoh et al. (2003) have reported that the cultivated teas from India, Bangladesh, Myanmar, Thailand, Laos, Vietnam, China, and Japan shared 10 different types of variations in the matK nucleotide sequences of cpDNA. Thereby, the nucleotide sequence alignments of rbcL in the cpDNA ensure the taxonomic classification into four groups (C. sinensis var. sinensis, C. sinensis var. assamica, C. irrawadiensis, and C. taliensis). The cultivated teas can then be fragmented into smaller population clusters, i.e., 10 types of CJ, AA, AB, AC, AD, AE, IC, IM, TM, and TV by the variation in the matK nucleotide sequences. From comparison analysis of rbcL nucleotide sequences, these 10 types of matK were classified into CSs, CSa, CSaa, CIt, and CTa. The nuclear-encoded 18S-rRNA will be significant for understanding the amount and distribution of hybrids present in tea cultivars. The nucleotide sequence comparison of matK and rbcL in cpDNA and nuclear-encoded 18S-rRNA will highlight the necessity for conserving sources of desirable genes within native tea cultivars. In addition, we emphasize that endemic cultivated teas in regions of Myanmar are conservationally important as new sources for the qualitative and quantitative improvement of tea products.
Acknowledgements This work was supported by grants (No. 14580141) from the Society of the Promotion of Science, and from the Ministry of Education, Culture, Sports, Science and Technology.
