Cloning and Comparative Studies of Seaweed Trehalose-6-Phosphate Synthase Genes

The full-length cDNA sequence (3219 base pairs) of the trehalose-6-phosphate synthase gene of Porphyra yezoensis (PyTPS) was isolated by RACE-PCR and deposited in GenBank (NCBI) with the accession number AY729671. PyTPS encodes a protein of 908 amino acids before a stop codon, and has a calculated molecular mass of 101,591 Daltons. The PyTPS protein consists of a TPS domain in the N-terminus and a putative TPP domain at the C-terminus. Homology alignment for PyTPS and the TPS proteins from bacteria, yeast and higher plants indicated that the most closely related sequences to PyTPS were those from higher plants (OsTPS and AtTPS5), whereas the most distant sequence to PyTPS was from bacteria (EcOtsAB). Based on the identified sequence of the PyTPS gene, PCR primers were designed and used to amplify the TPS genes from nine other seaweed species. Sequences of the nine obtained TPS genes were deposited in GenBank (NCBI). All 10 TPS genes encoded peptides of 908 amino acids and the sequences were highly conserved both in nucleotide composition (>94%) and in amino acid composition (>96%). Unlike the TPS genes from some other plants, there was no intron in any of the 10 isolated seaweed TPS genes.


Introduction
Porphyra is one of the most important seaweeds. It has a global distribution and important economic value. In addition to its roles in protecting aquatic ecosystems and as sources of food, biochemicals, pharmaceuticals [1,2] and bioenergy [3,4], Porphyra is now considered the best model organism for molecular biology research [5,6] and genomic research of seaweed [7]. However, molecular biological research in seaweeds is far behind the land plants and only a few nuclear genes have been described and cloned [2].
Trehalose (α-D-glucopyranosyl-(1,1)-α-D-glucopyranoside) is a non-reducing disaccharide of two glucose units presented throughout the animal, fungal, bacterial, yeast and plant kingdom [8,9], and functions as a stress protection metabolite in the stabilization of biological structures under stress tolerance and as a storage carbohydrate in plants [10,11]. The biosynthesis of trehalose has been studied in-depth in Escherichia coli (E. coli) and Saccharomyces cerevisiae (S. cerevisiae) and involves a two-step process catalyzed by trehalose-6-phosphate synthase (TPS) and trehalose-6-phosphate phosphatase (TPP). Trehalose-6-phosphate (T6P) has a critical role in plant growth and development; it is indispensable for carbohydrate utilization and growth in Arabidopsis thaliana (A. thaliana) [11,12]. T6P is also recognized as a regulator of sugar metabolism in plants [13][14][15][16]. Recently, it was proved that T6P functions as an inhibitor of SnRK1, a central integrator of stress and metabolic signals, to promote biosynthetic reactions in growing tissues [13]. Vandesteene et al. found that Arabidopsis encodes a single trehalose-6-P synthase (TPS) next to a family of catalytically inactive TPS-like proteins that might fulfill specific regulatory functions in actively growing tissues [15]. Different aspects of plant trehalose metabolism and function have been extensively reviewed [13][14][15].
We are interested in the seaweed TPS genes for the following reasons: first, the TPS gene encodes an enzyme involved in trehalose biosynthesis, which may become a model in functional gene research in seaweed; second, some experiments have shown that TPS genes from microorganisms can be expressed in transgenic plants, and increase the drought or salt tolerance of transgenic plants [25][26][27][28][29][30]. In most plants, trehalose is present in trace amounts and does not accumulate, but their genome sequences contain trehalose biosynthesis gene families [13,15]. Considering the high-salt living conditions of seaweed, its TPS gene may confer higher resistance to environmental stress than the corresponding genes from microorganisms, and may have potential usage in crop breeding by gene transformation. Here we report the characterization and molecular cloning of the TPS gene from Porphyra yezoensis (PyTPS) by RACE (Rapid Amplification of cDNA Ends)-PCR and the comparative analysis between the PyTPS gene and the TPS genes from some other seaweed species and other organisms.

Seaweed materials
The filaments of Porphyra yezoensis (P. yezoensis) and Porphyra haitanensis were cultured in axenic filtered seawater for 6 weeks at 16 °C and 25 °C, respectively, before the free filaments were collected for RNA and DNA preparations. The isolated gametophytes (male and female) of Laminaria japonica (L. japonica) and Undaria pinnatifid were propagated at 7 °C for 6 weeks before RNA and DNA extraction. PESI (Provasoli′s Enriched Seawater type I) solution was used as the medium for all cultures [31]. Samples of Gracilaria lemaneiformis and Sargassum henslowianum were cultivated in a cultivation tank and harvested for RNA and DNA preparations. The seaweed materials of Monostroma angicava, Ulva pertusa, Chondrus ocellatus and Enteromorpha prolifera were collected at the intertidal areas along the Qingdao coast, China. After identification, the samples were washed and brushed several times with autoclaved seawater to eliminate the algal epiphytes. Finally, the clean seaweed materials were used for RNA and DNA extraction. The P. yezoensis cell line Qingdao-8 was used for PyTPS gene cloning.

Generation of the TPS gene from P. yezoensis
For RACE-PCR amplification, total RNA was extracted from filaments of cell line Qingdao-8 by a modified guanidine thiocyanate (GT) method [32]. In order to obtain a full-length cDNA sequence of the PyTPS gene, the SMART TM RACE cDNA Amplification Kit (Clontech) was used according to the supplier's protocol. Since a 453 base pair (bp) fragment of the PyTPS gene was already identified in our previous work [33], the gene-specific primers (GSP1 and GSP2) were designed for RACE reactions according to this sequence. Primer GSP1 (5'-CTGTTCGCCTCGTGCTCCAGGTTAAG-3') was used for generation of the 5′ end of PyTPS, while GSP2 (5'-GCATTGCCCTCAAGCTGATGGGTTTC-3') and the following designed nested PCR primers NGSP2 (5'-GGTCGTACTTGTGCAAGTTGCCATCC-3') and 2NGSP2 (5'-GACCTGTCATGGATGGAGTTGGCATTGC-3') were used for generation of the 3′ end of PyTPS. The RACE-PCR products were cloned into the pMD-18T vector (TaKaRa, Dalian, China) for sequencing.

DNA extraction
Seaweed material was ground into powder in liquid nitrogen and then DNA was extracted with a plant genomic DNA extraction kit (Tianwei Biotech, Beijing, China) as in our previous report [34].

Total RNA extraction and cDNA syntheses
To prepare cDNA template for PCR amplifications, total RNA was extracted according to a modified GT method [35]. The RNA was quantified and checked at wavelengths of 260 nm and 280 nm and by formaldehyde RNA gel electrophoresis. Five µg of total RNA was then digested with DNase I (TaKaRa, Japan) followed by first-strand cDNA synthesis using the M-MLV reverse transcriptase (Promega, USA). The first-strand cDNA was used as a template in PCR amplifications.

PCR amplification of TPS genes from other nine seaweed species
The open reading frame (ORF) sequence of the seaweed TPS gene is about 2.7 kilo bases (kb). Based on the obtained cDNA sequence of PyTPS, three primer-pairs (Table 1) were designed and used to amplify the TPS gene from cDNA and genomic DNA of the other nine seaweeds. Their amplified fragments were about 1.3, 1.1 and 0.8 kb, respectively, and overlapped. Related primer information is provided in Table 1. PCR was conducted using the LA Taq  system (TaKaRa); PCR products were confirmed by sequencing. After sequencing and assembly, the entire ORF sequences of TPS genes were identified.

Analyses and comparison of TPS genes
Analysis of the cDNA sequences was performed using the BLASTX search program (Version 2.2.21+) served by NCBI (http://www.ncbi.nlm.nih.gov/BLAST/). Multiple sequence alignments and cluster analysis of TPS genes were carried out by DNAMAN software (Version 6, Lynnon Corporation).

Cloning the PyTPS gene into vector pET22b
First, total RNA was used as a template to synthesize first-strand cDNA. The entire PyTPS gene was then generated from cDNA by RT-PCR using primers TPS-R1 (a 5′ primer incorporating an Nde1 site overlapping the PyTPS initiator ATG codon) and TPS-b2 (a 3′ primer with a HindIII site incorporated downstream of the TPS translation stop codon) ( Table 1). PCR was conducted using the LA Taq  system (TaKaRa) to generate a ~2.7 kb fragment (Nde1-HindIII). After the amplified fragment was gel purified and digested with restriction enzymes Nde1 and HindIII (New England BioLabs, Inc.), the fragment was ligated into the pET22b vector (Novagen) using the Nde1 and HindIII sites to yield plasmid pET22b/PyTPS.
The plasmid pET22b/PyTPS was transformed into E. coli strain BL21(DE3) [36] for PyTPS overexpression. The transformants were grown in LB medium with ampicillin (50 µg/mL) at 37 °C to mid-logarithmic phase. PyTPS expression was induced by addition of 1mM IPTG (isopropylthio-β-Dgalactoside) and growth was continued for 4 h at 37°C. An aliquot of 1 mL cells was harvested and resuspended in 150 µL TE and separated by SDS-polyacrylamide gel electrophoresis (PAGE; 7.5%).

Generation of full-length PyTPS cDNA
Three successive rounds of RACE-PCR were performed to reach the 3′ end of the PyTPS gene, while only one round of RACE-PCR was performed to reach the 5′ end of the PyTPS gene. The RACE-PCR products were sequenced, analyzed and assembled by BLAST.
After assembly, a 3219 bp full-length cDNA of the PyTPS gene was obtained, and then it was deposited in GenBank (NCBI) with the accession numbers AY729671 (mRNA) and AAW27916 (protein). AY729671 contains a 216 bp 5′-leader sequence upstream of the ATG initiation codon, 276 bp of 3′ UTR (untranslated region) downstream of the termination codon (TAG), and an ORF (2727 bp) coding the TPS protein of 908 amino acids and a stop codon with a calculated molecular mass of 101,591 Daltons. The nucleotide sequence of the coding region and the deduced amino acid sequence of the PyTPS gene are shown in Figure 1.

Expression of the PyTPS gene in E. coli
The plasmid pET22b/PyTPS was constructed and transformed into E. coli strain BL21(DE3). Electrophoresis results showed that a strong PyTPS protein band was observed in the sample carrying pET22b/PyTPS (Figure 2, lane 2), but that no band was found in the sample carrying pET22b ( Figure 2, lane 1). The result proved that the PyTPS gene was highly expressed in the E. coli strain.

Domain analysis of the PyTPS protein
The BLAST results showed that, similar to the TPS proteins from other higher origins, the deduced PyTPS protein consists of a TPS domain at the N-terminus and a putative TPP domain at the C-terminus (Figure 3). The PyTPP domain has two typical sequences (LFDYDGTLT and GDDRTDEDMF) at amino acid positions 603-611 and 795-804 of the TPS protein that are conserved regions in the phosphatase family [37,38].

Comparison of PyTPS with TPS genes from other nine seaweeds
In addition to PyTPS, the TPS genes were PCR amplified from nine other seaweed species. Three of them (Porphyra haitanensis, Gracilaria lemaneiformis and Chondrus ocellatus) are Rhodophyta; three (Monostroma angicava, Ulva prolifera and Enteromorpha prolifera) are Chlorophyta, and three (Laminaria japonica, Undaria pinnatifida and Sargassum henslowianum) are Phaeophyta. The nine TPS genes were successfully PCR amplified from cDNA and genomic DNA. The PCR products were sequenced and the identified TPS genes were deposited in GenBank (NCBI); their accession numbers are listed in Table 2.
Comparison of nucleotide sequences of the TPS genes from the nine seaweed species with PyTPS indicated that all of these TPS genes contained an ORF with the same size of 2727 nucleotides. The identity of the 10 nucleotide sequences is higher than 94% (Table 3 and Figure 5).  Table 3. Variation of nucleotide and deduced amino acid sequences between the 10 seaweed TPS genes. (See Table 2 for the nomenclature of the TPS genes). The homologies of nucleotide sequences and their deduced amino acid sequences were calculated by comparing the differences in sequences between the indicated TPS and PyTPS or TPS and PyTPS, respectively.

Numbers of amino acid substitutions
Homology of nucleotide sequences (%)   Table 2 for nomenclature of the TPS genes in detail.

Comparison of the TPS gene sequences from cDNA and genomic DNA
Nucleotide sequence comparison between cDNA and genomic DNA of the TPS genes from 10 different seaweed species indicated that the sequences from cDNA and from genomic DNA were identical, confirming that no intron existed in all of the 10 TPS genes investigated.

Comparison of the TPS proteins from the 10 seaweed species
Comparison of the amino acid sequences of TPS proteins from 10 different seaweed species indicated that they were identical in size (908 amino acids); and that their sequences had an identity higher than 96% (Table 3, Figures 6 and 7).

Discussion
Trehalose might interfere with the sugar sensing mechanisms and other signal transduction pathways [39,40]. In Selaginella lepidophylla, trehalose forms glasses (vitrification) in the dry state for the stabilization of macromolecules [23]. The trehalose pathway is now known to be ubiquitous in plants [41]. The reported results proved that in Arabidopsis it is indispensable for carbohydrate utilization during plant growth and development [11,16].
Most plant TPS genes have introns. In the A. thaliana genome there are 11 TPS homologs: AtTPS1~4 contain 16 introns and AtTPS5~11 contain two or three introns [24,41]. In cultivated cotton (Gossypium hirsutum L.) the TPS gene was separated by two introns [38]. Sequence analysis indicated that, unlike the situation in TPS genes of higher plants, which have introns and exons in their genomic DNA sequences, there is no intron in any of the 10 seaweed TPS genes investigated in this study, which included species from red algae, brown algae and green algae. The E. coli otsA gene and yeast TPS genes are also without introns. This may reflect that seaweed belongs to lower plants in evolutionary taxonomy, and is very close to the prokaryote E. coli and the lower eukaryote yeast. Furthermore, the 10 TPS genes show highly conserved DNA sequences; their nucleotide sequence identity is higher than 94% ( Figure 5); however the identity between seaweed and other organisms is much lower (Figure 4).    It has been reported that two plant TPS genes, AtTPS1 and SlTPS cloned from A. thaliana and Selaginella lepidophylla, could partially complement an S. cerevisiae tps1∆ mutant, but most plant TPS genes failed to complement the S. cerevisiae tps1∆ mutant [21,23,42]. In addition, AtTPPA and AtTPPB were able to complement the yeast tps2 mutant [22]. In our experiments the cloned PyTPS gene failed to complement the tps1∆ and tps2 mutant (data not shown). We think the reason may be due to the structure of PyTPS gene itself. By applying BLAST analysis to compare the protein sequences of PyTPS and the 11 TPS proteins from Arabidopsis, the results indicated that the highest identity was found between PyTPS and AtTPS7 (identity = 37.7%) and the lowest identity was found between PyTPS and AtTPS1 (identity = 27.7%). Vogel et al. [40] had reported that the AtTPS7 and AtTPS8, although expressed, appeared to lack both TPS and TPP activity in yeast transformants. We think the similarity between PyTPS and AtTPS7 may make PyTPS more like AtTPS7 in lacking TPS and TPP activity in yeast transformants.
In recent years, trehalose metabolism has been implicated with stress tolerance and the control of yeast glycolysis [41]. Some experiments have indicated that transgenic plants expressing TPS genes from microorganisms exhibited increased stress tolerance. Seaweed is a kind of lower plant and belongs to algae, which can synthesis and accumulate trehalose [15]. So far no report has characterized the seaweed TPS gene. In this study, firstly we cloned TPS gene from the seaweed P. yezoensis, and it was studied in comparison with subsequently isolated TPS genes from other nine seaweed species. The results reported here will be helpful for the continued study of the function of seaweed TPS gene in stress tolerance, and for exploring its possible application in stress tolerance breeding of grain plants by gene transformation.
Recently, the PyTPS gene has been transformed into cultivated rice by agro-bacterium mediated transformation in our laboratory and some transgenic lines show increased salt/drought tolerance [43]. This will have potential applications in crop breeding in the future.