Phylogenetic relationships of sucrose transporters (SUTs) in plants and genome-wide characterization of SUT genes in Orchidaceae reveal roles in floral organ development

Identification and characterization of SUT proteins in Orchidaceae

The genome, gene and corresponding protein sequences of three sequenced Orchidaceae species, D. officinale (Zhang et al., 2016; Yan et al., 2015), P. equestris (Cai et al., 2014), and A. shenzhenica (Zhang et al., 2017), were downloaded from NCBI (https://www.ncbi.nlm.nih.gov/assembly/GCF_001605985.2/) and OrchidBase (http://orchidbase.itps.ncku.edu.tw/est/home2012.aspx). All members of the SUT family contain the GPH_sucrose (TIGR01301) domain, the seed sequence of which was downloaded from the TIGRFAMS database (http://tigrfams.jcvi.org/cgi-bin/index.cgi). ClustalW (Thompson, Gibson & Higgins, 2003) was used for sequence alignment, and a hidden Markov model (HMM) (Eddy, 1998) was constructed for SUT proteins. The HMMER program was used to search for SUT proteins among all D. officinale, P. equestris, and A. shenzhenica proteins, with a cutoff E-value of 1e ⁻⁴, using the HMM as a query. If the location of two SUT genes in the genome was less than 10 kb part, they were considered homologous genes generated by fragment duplication; if not, they were considered homologous genes generated by genome-wide duplication. After a comprehensive check, the candidate proteins that contained only fragmented SUT domains were eliminated. The ProtParam (http://web.expasy.org/protparam/) website was used to determine the molecular weight of each gene, and the theoretical isoelectric point (pI) of each protein was also predicted.

Phylogenetic analysis of SUT proteins

The amino acid sequences of SUT proteins identified in three Orchidaceae species (A. shenzhenica, D. officinale, P. equestris) and 21 other species were used in a phylogenetic analysis that included algae, moss, lycophytes, and angiosperms: Chlamydomonas reinhardtii (Cre), Volvox carteri (Vca), Physcomitrella patens (Ppa), Selaginella moellendorffii (Smo), Aquilegia coerulea (Aco), Picea abies (Pab), Brachypodium distachyon (Bdi), Oryza sativa (Osa), Zea mays (Zma), Vitis vinifera (Vvi), Eucalyptus grandis (Egr), Malus domestica (Mdo), Carica papaya (Cpa), Cucumis sativa (Csa), Daucus carota (Dca), Solanum lycopersicum (Sly), Asparagus officinalis (Aof), Populus trichocarpa (Ptr), Arabidopsis thaliana (AT), Glycine max (Gma), and Theobroma cacao (Tca). The protein sequences were downloaded from the Pfam database (https://phytozome.jgi.doe.gov/) and Phytozome database (https://phytozome.jgi. doe.gov/). MEGA 6 (V6.0, Tokyo Metropolitan University, Tokyo, Japan) was used to systematically analyze the protein sequences of the SUTs. First, CLUSX2 in MEGA 6 was used for multiple sequence alignment, and then the maximum likelihood (ML) method with the Jones-Taylor-Thornton (JTT) model was used to construct a phylogenetic tree. Moreover, 1,000 bootstrap replicates and a partial deletion with a site coverage cutoff of 70% were used for gap treatment. The phylogenetic trees were visualized using FigTree v1.4.2 (http://tree.bio.ed. ac.uk/software/figtree/).

Gene structure and motif analyses

The Gene Structure Display Server tool (http://gsds.cbi.pku.edu.cn/; v2.0) was used to analyze the gene structure of all the SUTs identified in D. officinale, P. equestris, and A. shenzhenica. MEME software (http://meme.nbcr.net/meme; v4.11.0) was then used to search for motifs in SUT proteins, with motif window lengths from 10 to 100 bp; the maximum number of motifs was set at 20, and motifs present in at least three SUT proteins were identified as true motifs.

Analysis of SUT gene expression in different tissues of D. officinale

First, we performed RNA-seq on different tissues of D. officinale. Three-year-old D. officinale plants were grown in glasshouses at the Mulberry Field Station of Zhejiang Academy of Agriculture Science (Hangzhou, China). Four different tissues—roots, stems, leaves and flowers—were collected, frozen in liquid nitrogen, and then stored at −80 °C until use. Each tissue was sampled three independent times. Total RNA was extracted using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions. The library preparations were sequenced on an Illumina HiSeq 2000 platform (Illumina, Inc; San Diego, CA, U.S.), and 150 bp paired-end reads were generated for the 12 samples. Then, the expression profiles of all the Dendrobium genes were obtained via fragments per kilobase of exon per million fragments mapped (FPKM) values using Cufflinks software (http://cole-trapnell-lab.github.io/cufflink; v2.2.1) under the guidance of annotated gene models with a GFF file. The SUT gene expression profile from each sample was analyzed using the HemI program (http://hemi.biocuckoo.org/) with the average hierarchical clustering method.

Determination of the total water-soluble polysaccharide content

Three-year-old D. officinale plants were grown in glasshouses at the Mulberry Field Station of Zhejiang Academy of Agriculture Science (Hangzhou, China). Four D. officinale tissues—roots, stems, leaves and flowers (three replicates for each tissue)—were collected and dried in an oven at 105 °C until a constant weight was achieved. The 12 samples were independently ground into fine powders by a mixing mill (MM 400, Retsch). Total polysaccharides were extracted using the water extraction and alcohol precipitation methods, and the content of total polysaccharides was measured using the phenol-sulfuric acid method.

Total polysaccharide extraction: Approximately 0.05 g of each sample was weighed, added to one mL of water, and fully homogenized. Each sample was then extracted in a water bath at 100 °C for 2 h and subsequently centrifuged at 10000 × g for 10 min after cooling, and the supernatant was removed. Then, 0.2 mL of the supernatant was collected, and 0.8 mL of anhydrous ethanol was slowly added. After mixing, the mixture was stored overnight at 4 °C. After centrifugation at 10,000 g for 10 min, the supernatant was discarded, and one mL of water was added to the precipitate, after which the mixture was thoroughly mixed and dissolved.

To calculate the total polysaccharide content, the microplate reader was preheated for more than 30 min, and the wavelength was adjusted to 490 nm. Two hundred microliters of the supernatant was extracted, and 100 µL of the reagent and 0.5 mL of concentrated sulfuric acid were added. After the contents of the wells were mixed together, the mixtures were incubated in a 90 °C water for 20 min. A 200 µL mixture was extracted and added to an enzyme-labeled plate, and the absorbance value (A) was determined at 490 nm. Glucose was used as a reference. The regression equation under standard conditions was y = 7.981x−0.0037, R2 = 0.9973; were, x represents the glucose content (mg/mL), and y represents the absorbance value. The total polysaccharide content (µg/g dry weight) was calculated as (A+0.0037) ÷7. 981 ×V1 ÷V 2 ×V3 ÷W ×1000 = 626.49 ×(A + 0.0037) ÷W. Here, V1 is the redissolved volume after alcohol precipitation (one mL); V2 is the volume of alcohol precipitation (0.2 mL); V3 is the volume of water added during extraction (one mL); W is the sample weight (g); and 1000 is the conversion coefficient for milligrams to micrograms.

Determination of the sucrose content

After drying, the 12 samples were ground into a fine powder independently with a mixing mill (MM 400; Retsch). Twenty milligrams of powder was diluted in 500 µL of a methanol:isopropanol:water (3:3:2 v/v/v) solution. The extract was centrifuged at 14,000 rpm at 4 °C for 3 min. Fifty microliters of the supernatant and an internal standard (Shanghai ZZBIO Co., Ltd.) were subsequently mixed together, evaporated under a stream of nitrogen gas, and then transferred to a lyophilizer for freeze drying. The residue was subjected to further derivatization. A sample of small-molecule carbohydrates and a 100 µL solution of methoxyamine hydrochloride in pyridine (15 mg/mL) were then mixed together. The mixture was incubated at 37 °C for 2 h. Then, 100 µL of BSTFA was added to the mixture and incubated at 37 ° C for 30 min after vortexing. The mixture was subsequently diluted and analyzed via GC-MS/MS according to the methods of Gómez-González et al. (2010) and Sun et al. (2016), with modifications. An Agilent 7890B gas chromatograph coupled to a 7000D mass spectrometer equipped with a DB-5 MS column (30 m length ×0.25 mm i.d. ×0.25 µm film thickness, J&W Scientific, USA) was used for GC-MS/MS analysis of the sugars. Helium was used as the carrier gas at a flow rate of one mL/min. The injections were made in split mode at a ratio of 3:1, and the injection volume was 3 µL. The oven temperature was set at 170 °C for 2 min, raised to 240 °C at 10 ° C/min, raised to 280 °C at 5 °C/min, raised to 310 °C at 25 °C/min and then held for 4 min. All the samples were analyzed in selective ion monitoring mode. The injector inlet and transfer line temperatures were 250 °C and 240 °C, respectively.

RNA extraction and qRT-PCR analyses

Total RNA was extracted from three D. officinale tissues, flowers, stems and leaves, using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions. DNase I was used to purify potential contaminated genomic DNA. The quality of total RNA was checked with 1% denaturing agarose gels and a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Beijing, China). First-strand cDNA synthesis was performed with PrimeScript reverse transcriptase (TaKaRa Biotechnology, Dalian, China), with RNA used as the template. Gene-specific primers were designed with the Primer Premier 5.0 program (Table S3). The DnActin (comp205612_c0) gene was used as an internal standard for normalizing the gene expression data (Chen et al., 2017). The expression levels of DoSUTs were analyzed via a qRT-PCR assay, which was completed with a SYBR Green qPCR kit (TaKaRa Biotechnology, Dalian, China) and a Stratagene Mx3000P thermocycler (Agilent, Santa Clara, CA, USA). The PCR program was as follows: 95 °C for 10 min, followed by 40 cycles of 95 °C for 15 s and 60 °C for 60 s. The relative SUT gene expression levels were calculated with the 2^−ΔΔCt method (Livak & Schmittgen, 2001). The analysis included three biological replicates, each with three technical replicates (Table S4). The expression levels in the different tissues were visualized with a histogram using the average values.

Statistical analysis

Statistical analysis was performed to calculate the average values and standard errors of three replicates. SPSS software (v. 16.0) was used to determine the significant differences in sugar content among the different tissues using one-way ANOVA and post hoc analysis. P value = 0.05 indicates a significant difference and is represented by an asterisk (*) in the figures; p value = 0.01 indicates a very significant difference and is represented by two asterisks (**).

Genome-wide identification of SUT genes in Orchidaceae species

To understand the potential roles of SUTs in orchids, three sequenced Orchidaceae species, D. officinale, P. equestris, and A. shenzhenica, were used for genome-wide identification and characterization of SUT genes. The HHM profile of the SUT proteins was used as a query to perform an HMMER search against the genome assemblies of the three species. Bioinformatics analysis identified a total of 22 SUT s with different serial numbers from the three species, which were designated ‘DoSUT’ for D. officinale, ‘PeqSUT’ for P. equestris, and ‘AsSUT’ for A. shenzhenica (Table 1, Table S1). Among them, D. officinale had eight genes (DoSUT01-08), P. equestris had eight genes (PeqSUT01-08) and A. shenzhenica had six genes (AsSUT01-06). These results agree with those of previous reports that plant sucrose transporters are encoded by relatively small gene families.

According to the phylogenetic tree, the 22 SUT genes from the three orchids could be classified into four subgroups: subgroups A, B2.1, B2.2 and C (Fig. 1, Table 1). Subgroup A included three genes: DoSUT01, PeqSUT01 and AsSUT01. There were four genes in subgroup C (DoSUT03, DoSUT04, PeqSUT08 and AsSUT02) and three genes in subgroup B2.2 (DoSUT02, PeqSUT03 and AsSUT03). However, the SUT genes had significantly expanded in the monocot-specific subgroup B2.1, which comprised 12 genes. Phylogenetically, the sucrose transporters in D. officinale were more closely related to those in P. equestris than to those in A. shenzhenica.

The molecular weights of the SUTs ranged from 18.81 to 106.90 kDa, with pI values ranging from 4.95 to 10.12. Most of these genes were ∼500 aa or ∼600 aa in length, with 11–13 introns and 12–14 exons, whereas there were several genes with only 4–5 introns/exons. These findings are consistent with the findings of the present study. Detailed information on the SUT genes, including their name, encoded protein, CDS length, molecular weight and PI value, is shown in Table 1.

Phylogenetic relationships of SUT proteins in major plant species

In the present study, the evolution of SUT gene families among the representative plant species was systematically investigated. A phylogenetic tree comprising 24 plant species was constructed, including green algae, mosses, lycophytes, gymnosperms, monocots and dicots. The SUT domain sequence and neighbor-joining method were used to construct the phylogenetic tree, with 1,000 bootstrap replicates. In this study, the SUT genes of several eukaryotic chlorophytes clustered on a unique branch, which was defined as an outgroup. All SUTs were classified into three groups and five subgroups—A, B1, B2.1, B2.2, and C (Fig. 1). Group A contained at least one member from mosses, lycophytes and angiosperms, including both monocots and dicots. Group B was the largest group and was divided into three subgroups; subgroup B1 comprised SUTs exclusively from dicot species, corresponding to the SUT1 clade (Lalonde & Frommer, 2012). Subgroup B2.2 contained SUTs from both monocot and dicot species that were also present in the SUT2 group (Lalonde & Frommer, 2012). Subgroup B2.1 was a monocot-specific expansion clade containing SUT3 and SUT5, as reported by Kühn & Grof (2010). Group C contained SUTs from mosses, lycophytes and angiosperms, including both monocots and dicots, corresponding to the SUT4 clade (Lalonde & Frommer, 2012). Type I SUTs have typically been proposed to be specific to eudicots. Notably, no orchid SUTs were found in clade B1; however, they were found in both clades B2.1 and B2.2.

Sucrose transporters have been identified in lower terrestrial plants, including both lycophytes and mosses, with six SUTs in Selaginella lepidophylla and 7 SUTs in Physcomitrella patens. There were 6-10 SUT genes in monocot species such as rice (six genes), maize (10 genes) and sorghum (eight genes). In contrast, in another monocot species, Ananas comosus, only three SUTs were identified. For most dicot species, 4-9 SUTs were identified. These results revealed that the number of sucrose transporters remained largely stable during the evolution from lower plants to terrestrial plants. However, the SUTs expanded in several species, such as Triticum aestivum (18 genes) and Glycine max (14 genes), which may be the result of whole-genome polyploidization. The SUT s of some monocot species expanded in subgroup B2.1; for example, there were five ZmaSUT s in subgroup B2.1, whereas 3 ZmaSUT s were identified in subgroup A, and only one was identified in subgroups B2.2 and C. Likewise, the SUTs from dicot species, such as GmaSUT s, AtSUT s and DcaSUT s, expanded in subgroup B1. The characean alga Chlorokybus atmosphyticus contains one SUT homolog that is basal to all streptophyte SUTs (Reinders, Sivitz & Ward, 2012). We also identified one SUT (VcaSUT01) in the chlorophyte Volvox carteri. Therefore, the origin of sucrose transporters predates the divergence between green algae and the ancestors of terrestrial plants.

Conserved motif analyses of SUT genes

The diversity of motif compositions among sucrose transporters of Orchidaceae species was assessed using the MEME program; a total of 10 conserved motifs were identified. The distribution of these 10 motifs in the SUT proteins is shown in Fig. 2. Motif 2 was the most conserved SUT domain and was identified in all of the SUT proteins except PeqSUT08 and DoSUT06.

In addition, motif 10 was observed in 17 SUT proteins but was absent in PeqSUT08, AsSUT05, DoSUT06, DoSUT08, and PeqSUT02. All three members in group A contained the same four motifs: motif 10, motif 2, motif 5 and motif 9. Moreover, except for DoSUT07, all group B members shared the same motif, motif 5; likewise, motif 4 was also common to all group B SUTs except for AsSUT04 (Fig. 2). Among the 12 SUTs in subgroup B2.1, three motifs were commonly present: motif 2, motif 3, and motif 5. There were eight sucrose transporters that had all 10 motifs, including five in P. equestris (PeqSUT03, PeqSUT04, PeqSUT05, PeqSUT06 and PeqSUT07) and two in A. shenzhenica (AsSUT02 and AsSUT06), whereas D. officinale had only one motif (DoSUT05). The sucrose transporters in each subgroup shared several unique motifs, indicating that the SUT proteins within the same subgroups may have certain functional similarities. In addition, the motif distribution of the SUTs suggested that these genes were largely conserved during evolution.

Water-soluble sugar content in D. officinale

To understand sucrose partitioning and the functions of sucrose transporters in Orchidaceae species, we measured the water-soluble sugar content in different tissues of D. officinale, including leaves, stems, flowers and roots, using the GC-MS/MS method. The results showed that the content of water-soluble polysaccharides varied significantly among the different tissues (Table 2, Fig. 3). The amount of total water-soluble polysaccharides was highest in the stems of D. officinale, at approximately 116.17 mg/g, followed by the leaves, at approximately 113.23 mg/g. The flowers had approximately 88.08 mg/g, whereas the roots had a significantly lower level of water-soluble polysaccharides, at ∼26.66 mg/g (Fig. 3A). These results indicated that the water-soluble polysaccharides were mainly deposited in the stems of D. officinale. The sucrose content also varied greatly among the different tissues. Nonetheless, the sucrose content was highest in the flowers, at approximately 28.1 mg/g, followed by leaves (∼18.13 mg/g), which are the major source tissues for photosynthetically assimilated sucrose. The amount of sucrose in the stems was ∼13.77 mg/g, and that in roots was the lowest, at only ∼7.82 mg/g (Fig. 3B). Taken together, these results showed that although the total sugar content was highest in the stems, sucrose was mainly transported to the floral organs of D. officinale.

Expression patterns of SUT genes in different tissues of D. officinale

To further understand the roles of SUT genes in orchids, the expression profiles of DoSUT genes in D. officinale were investigated. RNA sequencing (RNA-seq) was performed on different tissues, including the leaves, stems, flowers and roots, of D. officinale. The FPKM expression levels of the DoSUT genes in the four different tissues are provided in Table S2. Moreover, in Fig. 4, the expression levels of different DoSUT genes in the four D. officinale tissues are represented as different colors.

In the present study, RNA-seq showed that most of the DoSUTs were expressed in the flowers, among which three genes, DoSUT01, DoSUT08, and DoSUT06, had significantly high expression levels. Phylogenetically, DoSUT01, DoSUT08, and DoSUT06 were classified as members of subgroup A and of the monocot-specific expansion subgroup B2.1. Only one gene, DoSUT02, was significantly expressed in the leaves and was also expressed in the flowers and roots. We deduce that DoSUT02 may play a role in phloem loading in D. officinale. Nonetheless, other sugar transporters, such as SWEETs and MSTs, are also likely involved in sucrose transport.

Three genes, DoSUT03, DoSUT05 and DoSUT07, were expressed in the Dendrobium stems. Both DoSUT05 and DoSUT07 were moderately expressed in the stems and flowers, whereas DoSUT03 was slightly expressed in the stems and significantly expressed in the roots. In addition, DoSUT01 and DoSUT08 were expressed at low levels in the roots. The expression of DoSUTs was also measured in the flowers, stems and leaves of D. officinale using qRT-PCR (Table S4, Fig. 5). The results were largely consistent with those from RNA-seq analysis. Specifically, DoSUT01, DoSUT06, DoSUT07 and DoSUT08 were significantly expressed in flowers. Nonetheless, DoSUT02 was also expressed at significantly high levels in the stems. Three genes, DoSUT03, DoSUT05 and DoSUT07, were expressed in Dendrobium stems. DoSUT05 was expressed at significantly higher levels in the leaves than in the flowers, whereas DoSUT03 was expressed at lower levels in the stems and at significantly higher levels in the roots.