Genome-wide identification of sugar transporter gene family in Brassicaceae crops and an expression analysis in the radish

Sugar not only is an important biomacromolecule that plays important roles in plant growth, development, and biotic and abiotic stress tolerance but also provides a skeleton for other macromolecules, such as proteins and nucleic acids. Sugar transporter proteins (STPs) play essential roles in plant sugar transport and ultimately affect the abovementioned life processes. However, the evolutionary dynamics of this important gene family in Brassicaceae crops are still largely unknown, and the functional differentiation of radish STP genes remains unclear. In the present study, a comparative genomic study of STP genes in five representative Brassicaceae crops was conducted, and a total of 25, 25, 28, 36 and 49 STP genes were individually identified in Raphanus sativus (Rs), Brassica oleracea (Bo), B. rapa (Br), B. napus (Bn) and B. juncea (Bj), which were divided into four clades by phylogenetic analysis. The number of STP genes was no direct correlation with genome size and the total number of coding genes in Brassicaceae crops, and their physical and chemical properties showed no significant difference. Expression analysis showed that radish STP genes play vital roles not only in flower and seedpod development but also under heavy metal (cadmium, chromium and lead), NaCl and PEG-6000 stresses, Agrobacterium tumefaciens infection, and exogenous sugar treatment. RsSTP13.2 was significantly upregulated in the resistant radish cultivar by A. tumefaciens infection and induced by heavy metal, NaCl and PEG-6000 stress, indicating that it is involved in resistance to both biotic and abiotic stress in radish. The present study provides insights into the evolutionary patterns of the STP gene family in Brassicaceae genomes and provides a theoretical basis for future functional analysis of STP genes in Brassicaceae crops.


Background
Most plants on earth, except for parasitic plants, fix carbon through photosynthesis to produce soluble sugar as the main carbohydrate through various reactions in the cytoplasm [1]. These sugars have the functions of providing energy for cell life activities, providing a skeleton for macromolecules such as proteins and nucleic acids.
Sugars also participate in the regulation of various metabolic pathways and biotic and abiotic stress responses of plants as signal molecules [2,3].
Sugars synthesized from source cells were transported to the sink organs via the phloem mainly in the form of sucrose [4]. In the sink organs, sucrose can directly enter the cells by symplast pathways through plasmodesmata or via the apoplast pathway mediated by sugar transporters [1,4]. In addition, sucrose can also be hydrolysed to monosaccharides glucose and fructose in the apoplast by cell wall invertases, and thereafter, was transported into the cells across the plasma membrane mediated by Open Access *Correspondence: syauwjl@163.com monosaccharide transporters [4]. At present, three types of sugar transporters have been found in plants: sugars will eventually be exported transporters (SWEETs), monosaccharide transporters (MSTs) and sucrose transporters (SUTs) [5]. MSTs can be further divided into seven subfamilies, including sugar transporter protein (STP), vacuolar glucose transporter (VGT), tonoplast monosaccharide transporter (TMT), plastidic glucose transporter (GlcT)/suppressor of G protein beta 1 (SGB1), polyol transporter (PLT), inositol transporter (INT) and early response to dehydration-6-like (ERD6L) [6]. Among these subfamilies, the STPs are the best characterized subfamily and function in transporting hexose from the apoplastic space into the cell [6].
Previous studies have indicated that STPs are involved in regulating multiple growth and developmental processes and biotic and abiotic resistance by regulating the distribution and accumulation of soluble sugars in plants. Overexpression of AtSTP1 in A. thaliana significantly alters the extracellular sugar contents and inhibits its growth and branching [20]. AtSTP1 and AtSTP4 cooperate to import glucose to guard cells, providing carbon sources for light-induced stomatal opening and guarding cell starch accumulation [21]. AtSTP4/6/8/9/10/11 displays high expression levels in pollen tubes, and a sextuple knockout plant eliminates the inhibitory effect of glucose on pollen tube elongation in vitro [22]. The expression of AtSTP13 was significantly induced by Botrytis cinerea infection, and its overexpression enhanced the resistance to grey mould disease by improving glucose uptake [23], while its orthologous gene in wheat, TaSTP13, contributes to susceptibility to Puccinia striiformis f. sp. tritici (Pst), most likely by increasing the fungal sugar supply [24]. In addition, TaSTP6 was induced in leaves by Pst infection, and its expression promoted the susceptibility of wheat to stripe rust [25]. A more recent study indicated that overexpression of AtSTP8 in A. thaliana significantly induced the accumulation of hexose in the mature leaf and enhanced susceptibility to powdery mildew disease [17]. The expression of STPs is also significantly affected by abiotic stress. For example, both excess zinc and iron/zinc deficient stress significantly upregulate the expression of STP13 in shoot of Phaseolus vulgaris L., while excess zinc stress upregulated and iron/zinc deficient stress downregulated its expression in roots [26]. The expression of STP2, STP3, STP4, STP11, STP19 and STP25 in rice was significantly induced under salt, osmotic, and drought stress. STP11 was upregulated under ABA, IAA, 6-BA, SA and GA treatment, while STP1 and STP14 were upregulated under sucrose, glucose and fructose treatment [27]. These findings indicate that STPs play important roles in plant sugar transport, growth, development, and stress tolerance.
Brassicaceae includes many economically and nutritionally important crops, such as Raphanus sativus, Brassica oleracea, B. rapa, B. napus and B. juncea. The ancestor of diploid Brassica and Raphanus species, including B. oleracea, B. rapa, and R. sativus, has undergone Brassiceae lineage-specific whole-genome triplication (WGT) after its divergence from the A. thaliana lineage approximately 20 million years ago (MYA) [28]. However, the neotetraploids B. napus and B. juncea were allopolyploid between the ancestors of B. rapa (genome AA) and B. oleracea (genome CC), B. rapa (genome AA) and Brassica nigra (genome BB) [28]. The expansion and contraction of the STP gene family in these plants during their evolution are still largely unknown, even though high-quality genomes have been assembled. Thus, a genome-wide identification and analysis of the STP gene family in R. sativus, B. oleracea, B. rapa, B. napus and B. juncea was conducted in the present study. Additionally, the expression patterns in various organs and in response to biotic and abiotic stress were determined in the radish to characterize the functional differentiation of the STP gene family. The results of the present study provide insight into the expansion and contraction of the STP gene family in Brassicaceae and provide the basis for their functional study.

Genome-wide identification of STP genes from five Brassicaceae crops
Twenty-five, 25, 28, 36 and 49 STP-encoding genes were identified from the genomes of R. sativus (Rs), B. oleracea (Bo), B. rapa (Br), B. napus (Bn) and B. juncea (Bj), respectively (Table S1). There were also 14 A. thaliana STP genes identified in this study, which were identical to those described in a previous report [4], confirming the reliability of our results. No STP gene was found on the scaffolds of these six genomes, while the number of STP genes distributed on chromosomes (Chr) of each species varied greatly. In R. sativus, Chr1 and Chr5 had the highest number of STP genes. B. oleracea and B. rapa had the most STP genes on chromosomes C1 and A01, respectively. B. napus An-subgenomes had the most genes on A03, and Cn-subgenomes had the most genes on C05 and C07. The B. juncea An and Bn subgenomes had the most genes on A01 and B01, respectively (Fig. S1).
The physical and chemical properties of all STP proteins were analysed (Table S2). There were no significant differences in amino acid residue number, molecular weights, aromaticity, instability index, isoelectric point or gravy among the six species (Fig. 1). The predicted aromaticity ranged from 0.09-0.15, the instability index ranged from 28. 58-45.46, and the isoelectric point ranged from 5.44-9.65 (Table S2).

Phylogenetic relationships of STP family members from six Brassicaceae species
To analyse the possible evolutionary characteristics of the STP gene family in Brassicaceae, we conducted a phylogenetic tree based on 177 STP amino acid sequences from six species. All STP proteins were clustered into four groups (Fig. 2). In comparison, group II contained the most STP gene family members, followed by group III and group IV, and Group I had the fewest members ( Fig. 3). For group III and IV, R. sativus, B. oleracea and B. rapa had the same STP gene number, which was between

Conserved motif distribution and structural analysis of the STP family
To gain insight into potential functions and diversification among STPs, the encoded conserved motifs and exon-intron organizations were compared. As expected, most phylogenetically closely related STPs shared similar motifs and structures (Fig. 4). Fifteen predicted motifs were identified throughout the STP protein sequences. Motifs 1, 2, 5, 7, 8, 9, and 14 were present in all analysed STPs. The length of motifs ranged from 15 to 92, and part of the putative sugar_tr domain was predicted in motifs 1-7 and motif 11 ( Table 1). The exon/intron structures exhibited a highly conserved organization in STP genes. Most STPs (67%)

Tandem duplications and synteny of STP genes
Segmental and tandem duplications provide critical sources of primitive genetic material for genome complexity and evolutionary novelty. We investigated the syntenic and tandem relationships of STP genes. In A. thaliana, AtSTP4 and AtSTP10, distributed on chromosome 1, were demonstrated to be tandem duplications. For the other species, STP4 and STP10 on chromosome 1 were located in tandem duplicated regions. STP10 had 3-5 copies in the tandem duplication cluster. However, this tandem duplication cluster was not found in B. napus.
Of the 14 AtSTPs, most have a syntenic relationship with STP genes in other species, except for AtSTP3, which exhibited no syntenic genes in B. juncea, AtSTP5, which lacked a syntenic gene in R. sativus, AtSTP10, which had no syntenic relationship with B. napus, AtSTP12, which lacked a syntenic gene in R. sativus and B. oleracea ( Fig. 5 and Table S1). Additionally, most AtSTP genes were associated with more than one gene pair with other species. For instance, both AtSTP4 and AtSTP6 have more than four syntenic genes in other species. In addition, most STP genes in R. sativus, B. oleracea, B. rapa, B. napus and B. juncea have found syntenic STP genes in A. thaliana (Fig. S3).

Cis-acting element analysis of RsSTPs
To investigate the mechanism of transcriptional control of RsSTP genes, the cis-acting elements in the 1.5 kb potential promoter region of these genes were identified using the PlantCARE program. A total of 49 types of cisacting elements were identified in the promoter regions of RsSTP genes, except for common cis-acting elements in promoters (e.g., CAAT boxes and TATA boxes) (Fig.  S4). The cis-acting elements involved in light responsiveness were most abundant in the promoter regions of RsSTPs, indicating that they might participate in light regulatory pathways. Additionally, cis-acting elements responded to phytohormones such as abscisic acid (ABRE), auxin (TGA-element, AuxRE, TGA-box and AuxRR-core), methyl jasmonate (CGTCA-motif and TGAGG-motif ), salicylic acid (TCA-element) and gibberellin (P-box, GARE-box and TATC-box). Furthermore, stress-related cis-acting elements were also found in the promoter regions, including anaerobic (ARE), drought (MBS), low-temperature (LTR), defence and stress (TC-rich repeats) and anoxic (GC motif ) regions.

Expression profiles of RsSTPs in different tissues
Gene expression patterns are always associated with functional divergence in a gene family [7,29]. Therefore, public RNA-seq data were used to analyse the expression patterns of RsSTPs in various tissues (taproot, leaf, bolting, flower, seedpod and callus) [30]. In the present study,   (Fig. 8). Therefore, these three significantly induced genes are most likely candidates confer radish resistance to heavy metal stress.

Expression profiles of RsSTPs in response to NaCl and PEG-6000 stress and glucose, sucrose and fructose treatment
A quantitative real-time PCR (RT-qPCR) assay was performed to examine RsSTP expression levels in root tissue under salinity and simulated drought (PEG-6000) stress and their response to exogenous sugar treatment, including glucose, fructose and sucrose (Fig. 9). The RT-qPCR results indicated that RsSTP9.2 and RsSTP14

Discussion
STP proteins in plants play vital roles in monosaccharide transport and are involved in regulating multiple growth and developmental processes and biotic and abiotic resistance. In recent years, with the availability of various plant genomes, genome-wide and expression analysis of the STP gene family has been reported in many plants, such as Arabidopsis thaliana [6], Oryza sativa [27,31], Solanum lycopersicum [32], Pyrus bretschneideri Rehd) [33], Brassica  oleracea var. capitata L. [34], Fragaria vesca [35], Capsicum annuum L. [29], Manihot esculenta [36], Triticum aestivum L. [37] and son on. However, the evolutionary dynamics and functional analysis of the STP gene family in Brassicaceae crops are still largely unknown.
In the present study, a total of 25, 25, 28, 36 and 49 STP-encoding genes were identified from the genomes Fig. 9 Quantitative real-time PCR analysis of RsSTP expression levels in the roots in response to 1.5% NaCl and 20% PEG stress and 2% glucose, 2% fructose and 2% sucrose treatment. The presented gene expression levels are relative to the expression of the reference gene RsGAPDH. Data are presented as the mean ± standard error of three independent experiments. CK: control treatment with distilled water of R. sativus, B. oleracea, B. rapa, B. napus and B. juncea, respectively. The same STP gene numbers were identified in R. sativus and B. oleracea even though R. sativus (460 Mb and 44,109 coding genes, respectively) [38] and B. oleracea (648 Mb and 54,475 coding genes, respectively) [39] have different genome sizes and gene numbers. B. rapa (442.9 Mb and 45,985 coding genes, respectively) [40] and R. sativus have similar genome sizes and gene numbers, but variant in STP gene numbers. Thus, the number of STP genes is no direct correlation with genome size and the total number of coding genes in Brassicaceae crops. The same result was also reported by a previous report in Gramineae crops [7]. Tetraploid B. juncea (920 Mb and 101,959 coding genes, respectively) [41], which have undergone an allopolyploidization event, as well as B. napus (1008 Mb and 100,919 coding genes, respectively) [42], have an almost twice larger genome size and number of coding genes than diploid R. sativus, B. oleracea and B. rapa, and have significantly more STP gene numbers were identified. This result makes us speculated that a large number of STP genes retained after the allopolyploidization event, even though accompanied by gene losses occur in this process.
Segmental and tandem duplication events play a critical role in the expansion and increased functional diversity of the STP gene family. Previous studies indicated that a WGT occurred in the common ancestor of Brassicaceae crops following its divergence from A. thaliana [28]. In this study, we revealed that most AtSTP genes have syntenic pairs in other Brassicaceae species with more than one copy, which is consistent with the polyploidization of these species. Here, segmental duplication was the major force driving the expansion of STP genes in Brassicaceae. Additionally, several genes were lost or retained one copy, suggesting that there may have been some variability in the gene loss events during evolution. Previous studies concluded that functionally redundant genes prefer to be lost in the diploidization process occurring after paleopolyploidy events [43]. The tandem duplication STP10 genes were present in R. sativus, B. rapa, B. oleracea and two subgenomes of B. juncea, indicating that the associated tandem duplication event occurred in the common ancestor.
We revealed physical and chemical properties, diverse gene structures, conserved motifs and phylogenetic analysis of STP genes in Brassicaceae crops. The physical and chemical properties of STPs showed that there were no significant differences among the six species, which indicated that the STP proteins were conserved among different species. Furthermore, most of the STP genes contained four exons and three introns, which is similar to other plant species, such as rice, tomato, and pears [27,[31][32][33]. A phylogenetic analysis revealed that the STPs in Brassicaceae crops were classified into four groups, consistent with the classification in cassava and Gramineae [7,36], suggesting that STP proteins are highly conserved across lineages. These results indicated that STP functions are mainly conserved in different species.
Previous studies have produced evidence that STPs are involved in plant growth and developmental processes and biotic and abiotic stress responses. The cis-acting elements in the promoter regions are always associated with their transcriptional control and function. In the present study, cis-acting elements involved in light responsiveness, responding to phytohormones and stresses were also found in the predicted promoter regions of RsSTPs (Fig. S4), suggest that these genes might be involved in radish growth and developmental processes and biotic and abiotic stress responses.
In the process of evolution, most of the homologous genes of different species retain the same or similar biological functions, and the expression patterns of genes are often closely related to their functions. In the present study, expression analysis of RsSTPs in different tissues showed that RsSTP4.1 and RsSTP4.3 were expressed in all tissues and their expression levels were significantly induced under Pb and Cd stress, but not by Cr stress and A. tumefaciens infection. AtSTP4, which is orthologous in A. thaliana, could transport a broad spectrum of monosaccharides [11], and induced by fungal biotroph Erysiphe cichoracearum infection [44]. Therefore, RsSTP4.1 and RsSTP4.3 might play vital roles in soluble sugar transporters in different tissues of radish and also involved in Pb and Cd stress resistance. However, their biological function needs further experimental verification. AtSTP4/6/8/9/10/11 were found to be highly expressed in pollen tubes by a previous study [22]. Our present results show that eight genes (RsSTP2.1/2.2/4.2/6.1/7.2 /9.1/9.2/14) were specifically expressed in flowers, and we conjectured that these genes might be responsible for glucose uptake into pollen tubes. RsSTP3 and RsSTP6.2 were highly expressed in seedpods, suggesting that they might be involved in soluble sugar accumulation in this tissue. STP genes were reported participate in response to biological stress in other plants. The expression of BoSTP4b and BoSTP12 were up-regulated in cabbage with Plasmodiophora brassicae infection [34]. RsSTP6.2 and RsSTP13.2 most likely confer radish resistance to A. tumefaciens infection by RNA-seq data analysis in the present study. Earlier studies showed that STP genes play important roles in various sugar transportation processes [4,6,[8][9][10][11][12][13][14][15][16][17][18][19]. We detected that the expression of most genes was not significantly affected by 2% glucose, fructose and sucrose treatment, and five genes were significantly downregulated by at least one treatment (Fig. 9).
Only RsSTP8 and RsSTP13.1 were upregulated by fructose treatment. Deng et al. also indicated that only OsST8 was upregulated by fructose treatment in the roots of rice [27]. The transporter of glucose and sucrose in radish roots might be responsible for other sugar transporters.
RsSTP13.2 was significantly upregulated in resistant plants by A. tumefaciens infection but undetectable in susceptible plants and induced by Cd, Cr, Pb, NaCl and PEG-6000 stress, indicating that RsSTP13.2 is involved in resistance to both biotic and abiotic stress in radish. A previous study also indicated that STP13 is involved in biotic and abiotic responses and resistance in other plants. In A. thaliana, STP13 maintains low expression under normal conditions, but STP13 is induced by MYB96 and reabsorbs the monosaccharides that are released by damaged cells under saline conditions [45,46]. The upregulation of AtSTP13 deprivation of extracellular sugar levels, which is used as an energy source for pathogens, enhances antibacterial defence [47]. In addition, AtSTP13 is also involved in resistance to Botrytis cinerea by affecting glucose transport [23]. In Phaseolus vulgaris L., both excess zinc and iron/zinc deficient stress significantly upregulate the expression of STP13 in the shoot, while excess zinc stress induced and iron/zinc deficient stress decreased its expression in roots [26].

Conclusions
The present study provides insights into the evolutionary patterns of the STP gene family in Brassicaceae genomes and provides a theoretical basis for future functional analysis of STP genes in Brassicaceae crops. RsSTP13.2 may serve as a candidate gene to improve the biotic and abiotic resistance of plants through transgenic technology.

The phylogenetic tree construction
For phylogenetic tree construction, first, all the STP full-length protein sequences of six Brassicaceae species were aligned by using the MUSCLE program [49]. Then, MEGAX [50] was used to construct a neighbour-joining tree using the Jones-Taylor-Tornton (JTT) model with 1000 bootstrap replicates. Additionally, uniform rates and homogeneous lineages were adopted, and partial deletion with a site coverage threshold of 70% was given for gaps/missing data.

Sequence properties, conserved motifs and gene structure analyses
The Biopython module Bio.SeqUtils. ProtParam of Python language was used to calculate the molecular weight, aromaticity and other physical and chemical properties of STP proteins. The R script was used to compare the differences among different species using ggpubr with 'anova' methods. The MEME suite [51] was used to identify the conserved motifs of STP proteins with the following parameters: the maximum number to be found was set to 15, and the motif length was set to 8-100 bp. The Pfam domains of motifs were identified in the Pfam database (http:// pfam. xfam. org). The gene structures information containing extron and intron position were obtained from GFF file using python script. The TBtools program [52] was used to visualize gene structures and conserved motifs.

Tandem duplications and syntenic analysis of STP genes
Tandem genes in A. thaliana and other species were defined as those genes that were separated by ten or fewer genes. The Multiple Collinearity Scan toolkit (MCScanX) [53] was used to identify syntenic duplication events between A. thaliana and other species, with the default parameters. The synteny relationship of STP genes was visualized using TBtools software [52]. The subgenomes of B. napus and B. juncea were calculated separately.

Cis-acting element analyses of RsSTPs and its transcriptional profiles in RNA-seq data
The cis-acting elements in the potential promoter region (the upstream 1.5 kb sequence starting from the start codon) of the radish STP genes were identified using the PlantCARE program (http:// bioin forma tics. psb. ugent. be/ webto ols/ plant care/ html/). The RsSTP expression profiles of different tissues (root, leaf, bolting, flower, silique and callus), their response to heavy metals (cadmium, chromium and lead) and Agrobacterium tumefaciens infection were analysed based on public transcriptome data [30,[54][55][56][57]. The fragments per kilobase of transcript per million mapped reads (FPKM) data were log 2 transformed, and a heatmap was created using TBtools [52].

Plant materials and stress treatments
Seeds of radish cultivar 'Xin-li-mei' purchased from Jingyan Yinong (Beijing) Seed Sci-Tech Co., Ltd. were surfacesterilized in 1% NaClO and incubated at 22 °C for 2 d in darkness. The germinated seeds were sown into plastic pots and incubated in a growth chamber at a 16 h day (22 °C)/8 h night (20 °C) cycle. Seedlings were watered as needed with half-strength Hoagland's nutrient solution. Plants at the three true leaf stages were used for subsequent abiotic stress and sugar treatment. For the simulated salinity and drought stress treatments, the seedlings were subjected to 1.5% NaCl and 20% PEG-6000 for 3 h, respectively. For sugar treatments, the seedlings were subjected to 2.0% glucose, sucrose and fructose solution for 3 h. The seedlings treated with sterile water were used as a control. Eight plants were used as biological replicates, and the roots were collected. The samples were frozen in liquid nitrogen immediately after collection and stored at -80 °C for RNA extraction. The collection of plant material, is in compliance with relevant institutional, national, and international guidelines and legislation.

RNA extraction, reverse transcription quantitative polymerase chain reaction (RT-qPCR) analysis of RsSTPs
Total RNA was extracted from different radish samples with a Quick RNA Isolation Kit (Huayueyang, Beijing, China) according to the manufacturer's instructions. A total of 800 ng of high-quality total RNA was used to synthesize first-strand cDNA with PrimeScript ® Reverse Transcriptase (Takara Biotechnology, Dalian, China). The SYBR Green qPCR kit (Takara Biotechnology) was used for RT-qPCR in a Stratagene Mx3000P thermocycler (Agilent, Santa Clara, CA, USA). RsGAPDH was utilized as an internal control, and the relative expression levels of RsSTPs were calculated with the 2 −ΔΔCt method [58,59]. The primer used for RT-qPCR is shown in Table S1.