Chinese traditional herbal medicine, Danshen, is the radix of perennial herbs of Salvia miltiorrhiza Bunge of the family Labiatae. It has been traditionally and extensively used in clinical practice to treat various ailments such as cardiovascular, cerebrovascular, hyperlipidemia, and acute ischemic stroke diseases [1] [2] [3] [4] [5].

Danshen owns abundant germplasm resources and many cultivated populations in China. In recent years, with the increase in market demand, the often-chaotic introduction of varieties in field cultivation and the nonstandard field management cause some confusion and result in poor quality of the herbal medicine. Extensive researches have been conducted on its cultivation, germplasm resources protection, and molecular identification.

However, research progress in the functional genes for pharmacologically active constituents has been slow. There are two major classes of pharmacologically effective components in Danshen, the hydrophilic salvianolic acids, and the lipophilic tanshinones [6]. So far, the key genes for the biosynthesis of effective components in S. miltiorrhiza, of 4-hydroxycinnamate coenzyme A ligase [6], 4-hydroxyphenylpyruvate reductase [7], cinnamic acid 4-hydroxylase [8], tyrosine aminotransferase [9], 3-hydroxy-3-methylglutaryl coenzyme A reductase [10], and phenylalanine ammonia-lyase genes [11], have been characterized and studied respectively.

The biosynthetic pathway of rosmarinic acid consists of two parallel phenylalanine and tyrosine branches [12]. Tyrosine aminotransferase (TAT) (EC 2.6.1.5) is the rate-limiting step in the tyrosine branch, which catalyzes the formation of 4-hydroxyphenylpyruvate from tyrosine and ultimately the biosynthesis of salvianolic acids. Various other structurally diverse natural compounds are also derived from the tyrosine metabolic pathway, among which tocopherols, plastoquinone, and ubiquinone are essential to plant survival [13].

So far, our knowledge of the plant tyrosine metabolism pathway remains rudimentary, and genes encoding the pathway enzymes have not been fully defined, despite that the tyrosine aminotransferase genes have been cloned in a few other plants such as Coleus blumei (AJ458993), Arabidopsis thaliana [14], Glycine max (AAY21813), Medicago truncatula (DQ006809) as well as S. miltiorrhiza, and functionally studied by overexpression of single gene and coexpression of several in S. miltiorrhiza hairy root cultures [15]. The structure and the genetic diversity of tyrosine aminotransferase genes among the various cultivated populations of S. miltiorrhiza are still unknown.

In this research, the genomic tyrosine aminotransferase genes of the 38 cultivated populations of S. miltiorrhiza from the major cultivation regions of China, were for the first time cloned by walking technology, and its sequences were analyzed bioinformatically to understand the genomic structure, genetic diversity, and phylogeny of the cultivated S. miltiorrhiza populations.

Plant materials

Seeds of 38 cultivated Danshen populations were collected from three major seed industries representing more than 30 regions of China (Table 1); uniform seeds preliminarily selected according to size, color and shape were used for sowing in Southwest University Agricultural Station, Chongqing; and morphologically representative single plants of each population were used for extraction of genomic DNAs.

Primer design

Two pairs of primers for cloning of the genomic SmTAT genes of cultivated

Notes: HDSI—Hengda Seed Industry, China; TDSI—Tongda Seed Industry, China; FHSI—Fenghong Seed Industry, China; SC—Self-Collected.

S. miltiorrhiza populations were designed based on the reference accession (EF192320.1) with Primer 5 (Table 2).

Genomic DNA extraction

Total leaf genomic DNAs were extracted with Qiagen DNeasy Plant Mini Kit (Multi Sciences, Hangzhou, China) according to the manufacturer’s instructions. The purity was assessed by agarose gel electrophoresis followed by Goldview staining, and the quantity was determined spectrophotometrically by Shimadzu UV mini-1240. Purified genomic DNAs were dissolved in 10 mmol/L Tris-HCl buffer and stored at −70˚C.

PCR amplification

About 1.0μg genomic DNA templates were amplified with primer pairs TAT-FP1/RP1 and TAT-FP2/RP2 respectively in a reaction mixture of 50 μL: 1.1 × T3 Super PCR Mix 36.0 - 44 μL, 10 μmol/L primers each 2.0 μL (final concentration 0.4 μM) in Biometra TGRADIENT thermocycler (Biometra GmbH, Germany) with the programme: initial-denaturation at 98˚C for 3 min followed by 35 cycles of denaturation at 98˚C for 10 s, annealing at 55˚C/60˚C for 10 s and elongation at 72˚C for 5 - 15 s, and a final extension at 72˚C for 2 min.

Amplified products were electrophoresed in 1% agarose gel and visualized with Goldview stain. And after recovery and purification, they were bidirectionally sequenced by dideoxy chain termination with ABI Prism 310 Genetic Analyzer (Applied Biosystems, Foster City, USA) and manually spliced and checked.

Sequence data processing

The BLAST confirmed two segments of the SmTAT gene sequences of the 38 cultivated populations of S. miltiorrhiza were spliced with Vector NTI Advance11. The spliced whole sequences were manually checked to ensure the quality of sequences and BLAST analyzed to confirm the gene of interest. The spliced whole SmTAT gene sequences were deposited in GenBank. Sequences were aligned with Vector NTI Advance11 to identify the nucleotide variation sites. Phylogenetic trees were constructed with MEGA X based on Neighbor-Joining (NJ) with a bootstrap value of 1000.

Structural features of the SmTAT genes of the cultivated S. miltiorrhiza populations

The SmTAT genes of the 38 cultivated populations were successfully amplified by the walking primers designed (TAT-FP1/RP1 and TAT-FP2/RP2). Agarose gel electrophoresis showed distinct single bands of about 700 bp and 2000 bp respectively with the two primer pairs.

All the obtained SmTAT gene sequences of the 38 cultivated populations of S. miltiorrhiza were BLAST confirmed and submitted to GenBank (Accession numbers shown in TableA1).

The sequence of accession EF192320.1 was used as the reference to demarcate the SmTAT gene sequences. Results showed that the full-length of genomic SmTAT gene was 2296 - 2444 bp, consisting of 6 exons with a total length of 1236 bp among all the tested populations, and 5 introns with a total length of 1060 bp for 37 populations except for W-SCHY-W-1, whose total intron length increased to 1208 bp, most of the increased 148bp distributed in introns 1 - 4; the corresponding exons or introns are equal in size for the majority (37 populations) except for W-SCHY-W-1. The total length of the SmTAT gene for the majority (37 populations) was 2296 bp, while for W-SCHY-W-1, 2444 bp (Table 3).

Nucleotide variations in the introns of the SmTAT genes

The 5 intron sequences of SmTAT of the 38 cultivated populations were aligned and compared. Results showed that there were 290 nucleotide variations with a rate of 24%, and all occurred in the population W-SCHY-W-1(Table 4). Statistics showed that there were 110 variations (11conversions, 21 transversions, 4 deletions and 74 insertions) in intron 1, 13 variations (4 transversions and 9 insertions) in intron 2, 120 variations (19 conversions, 21 transversions, 18 deletions and 62 insertions) in intron 3, 34 variations (2 conversions, 2 transversions, 30 insertions) in intron 4, and 13 variations (2 transversions, 2 transversions, 7 deletions and 2 insertions) in intron 5, accounting for variation rates of 26.3%, 15.7%, 30.9%, 45.3% and 13.5% respectively (Table 4).

Nucleotide variation in exons of the SmTAT genes

Alignment of the spliced 6 exon sequences of all the 38 cultivated S. miltiorrhiza populations showed that there were 23 nucleotide variation sites, of a variation rate of 1.86%, among which 12 conversions, 11 transversions. Most variations (22) occurred in population W-SCHY-W-1 and were distributed mainly in exons 2-4 (Table 5).

Amino acid variations in the deduced amino acid sequences of SmTAT

The spliced exon sequences of the SmTAT gene were 1236 bp in length with a complete reading frame of 1233 bp encoding 411 amino acid residues. The deduced amino acid sequences are highly conserved. All showed the aminotransferases family-I pyridoxal-phosphate attachment site (SLSKRWLVPGWRLG)

Note: E and I for exons and introns respectively.