Dissecting Tissue-Specific Transcriptomic Responses from Leaf and Roots under Salt Stress in Petunia hybrida Mitchell

One of the primary objectives of plant biotechnology is to increase resistance to abiotic stresses, such as salinity. Salinity is a major abiotic stress and increasing crop resistant to salt continues to the present day as a major challenge. Salt stress disturbs cellular environment leading to protein misfolding, affecting normal plant growth and causing agricultural losses worldwide. The advent of state-of-the-art technologies such as high throughput mRNA sequencing (RNA-seq) has revolutionized whole-transcriptome analysis by allowing, with high precision, to measure changes in gene expression. In this work, we used tissue-specific RNA-seq to gain insight into the Petunia hybrida transcriptional responses under NaCl stress using a controlled hydroponic system. Roots and leaves samples were taken from a continuum of 48 h of acute 150 mM NaCl. This analysis revealed a set of tissue and time point specific differentially expressed genes, such as genes related to transport, signal transduction, ion homeostasis as well as novel and undescribed genes, such as Peaxi162Scf00003g04130 and Peaxi162Scf00589g00323 expressed only in roots under salt stress. In this work, we identified early and late expressed genes in response to salt stress while providing a core of differentially express genes across all time points and tissues, including the trehalose-6-phosphate synthase 1 (TPS1), a glycosyltransferase reported in salt tolerance in other species. To test the function of the novel petunia TPS1 allele, we cloned and showed that TPS1 is a functional plant gene capable of complementing the trehalose biosynthesis pathway in a yeast tps1 mutant. The list of candidate genes to enhance salt tolerance provided in this work constitutes a major effort to better understand the detrimental effects of salinity in petunia with direct implications for other economically important Solanaceous species.


Introduction
Salinity rapidly reduces plant growth inducing a suite of metabolic changes in plant physiology [1]. Initially upon sudden increases in sodium chloride, hormonal signals generated by the roots can lead to rapid reductions in growth and ultimately loss of yield in agriculture crops [2]. Salt-affected soils have become a major concern worldwide due to its detrimental impact in agricultural crop productivity. High rhizosphere NaCl levels can cause plant osmotic stress, protein misfolding, ion toxicity, nutritional

Tissue Sample and RNA Isolation
From within the nine plants per treatment, three biological replicates were established by randomly grouping three sets of plants within each treatment condition. At each time point (as described below) samples from three plants per group were pooled together to create one biological replicate. Leaves were sampled at 0, 6, and 24 h and roots were sampled at 0, 6, 12, 24 and 48 h after treatment was applied (NaCl vs. Hoagland's solution with no added NaCl), yielding 48 samples total (8 time points and organs × 2 treatments × 3 biological replicates). At each time point roots were carefully dissected longitudinally (i.e., strands of full length roots from the base of the plant to the root tip). The most recently expanded leaf (fourth or fifth leaf from the lateral meristem) from a lateral branch was selected. It is important to note that plants did not have enough of the targeted mature leaves to take from more than three time points (while they had plenty of root samples to take more time points), which left us with more roots than leaves time points, in addition this is why plants were grouped into three sets within each treatment, with one plant within each set sampled at each time point. The 48 samples were frozen immediately in liquid nitrogen and stored at −80 • C before RNA isolation. Total RNA was isolated using Trizol Reagent (Invitrogen, Carlsbad, CA, USA) and purified through a Qiagen RNeasy Column (Qiagen, Hilden, Germany) according to the manufacturer's instructions. A 1% agarose gel was run to indicate the integrity of the RNA and ribosomal bands were used for total RNA quality control. Four root and leaf samples were further quantified in an Agilent 2100 Bioanalyzer (Agilent, Santa Clara, CA, USA) at the Cornell University Biotechnology Resource Center) to verify total RNA quality. RNA integrity number (RIN) for these samples analyzed were 8.5, 9.1, 8.9, 8.5 (roots), 8.7, 8.5, 8.7 and 6.7 (leaves).

Library Preparation and Deep Sequencing
Multiplexed libraries for the 48 samples were constructed using a High-Throughput Illumina Strand-Specific RNA Sequencing Library protocol [23]. The forty-eight double stranded complementary DNA (cDNA) libraries were pooled together (20 ng/library) and sent for sequencing to the Cornell University Biotechnology Resource Center (http://www.biotech.cornell. edu/biotechnology-resource-center-brc).
Paired-end sequencing was performed including 2 × 100 cycles + 7 cycle index-read in in 3 lanes of the HiSeq 2500 Illumina platform with 'TruSeq PE Cluster Kit v3' for the flow-cell and 'TruSeq SBS kit v3' for the sequencing reagents.

Processing of Illumina RNA-Seq Reads
Quality control and preprocessing of metagenomic data were performed using FastQC software [24]. Adapters and low-quality sequences were filtered out with Ea-Utils software [25]. Reads with phred-like quality score greater than 30 and read length greater than 50 bp were kept and aligned against the P. axillaris reference genome [26]. The P. axillaris genome was used as a reference as P. hybrida is more closely related to P. axillaris than its other parental species Petunia inflata.

Sequence Alignment to the Petunia Genome
The splice junction mapper TopHat2 [27] was used to align filtered RNA-seq reads to the Petunia axillaris genome. Default parameters for TopHat2 were used except for strand specificity (-library-type = fr-firststrand) to match to the first strand of cDNA synthesized (antisense to the mRNA). Uniquely mapped reads were extracted from the TopHat2 output binary (BAM) file using samtools [28] and selected for the "NH:i:1" two-character string tag. Only uniquely mapped reads were used for downstream analysis.

Table Counts
The HTSeq: Analyzing High-Throughput Sequencing Data with Pythons software [29] was used with default parameters except for the stranded = reverse and "-i ID" mode to generate tables-counts for downstream differential expression analysis for the R packages edgeR [30].

Gene Expression and Differential Gene Expression
We performed gene expression and differential gene expression analyses with the R packages edgeR [30] and the Linux-based Cufflinks program [31] (version 2.2.1) "-G" option. With edgeR, we discarded genes whose cpm (counts per million) was lower than a threshold of two reads per gene in at least three biological replicates, as suggested in the edgeR vignette [32]. To identify expressed gene with Cufflinks, we chose those genes that had a lower confidence interval bigger than 0 (conf_lo > 0) and whose status was "ok" (quant_status = "OK").

Sequencing Data
Illumina sequencing raw data (fastq) and table counts have been submitted to the Gene Expression Omnibus (GEO) database (accession no. GSE97428).

Truncating the Petunia Trehalose 6 Phosphate Synthase PhTPS1 (⊗PhTPS1)
One of the highly differentially expressed genes in our dataset was the TPS1, which has also been implicated in salt stress tolerance in other species [20,21]. To verify function of the TPS1 we cloned and tested its function in a yeast system. Analysis of the TPS1 N-terminus from A. thaliana and S. lepidophylla led to the identification of two conserved residues which decrease enzyme activity when present [20]. Therefore, we truncated the first~80 amino acids (aa) towards the 5 -end to increase the PhTPS1 catalytic activity. Two different primers were designed to truncate the 3 kb PhTPS1 while at the same time introducing sequence homologous to the promoter of the pDB20 vector, used in later cloning steps. A 56 bp forward primer (5 -GCTATACCAAGCATACAATCAACTTAAAGCGGCCGCATGCAACGACTCTTAGTTGT-3') was designed. The 5'-end of this primer has 23 bp of the promoter sequence of pDB20 vector (italics) followed by five bases of filler sequence (TTAAA), a NotI restriction site (underlined), an in-framed ATG start codon (bold case) and 17 bases complementary to the PhTPS1 sequence. A reverse primer (5 -TGGCGAAGAAGTCCAAAGCTTATTTGCGGCCGCTTAAGAAGAGGCTTCAGCTAGT-3') including a stop codon was used. This trimmed petunia TPS1 gene (⊗PhTPS1), which now included a yeast promoter sequence, was cloned into the pDB20 yeast vector and sequenced.

The pDB20 Yeast Vector
The backbone of the pDB20 yeast expression vector comes from the pUC18 vector. pDB20 carries the ADCI (ADHJ) promoter which drives high levels of gene expression. In the polylinker region this vector has six cloning sites; HindIII/NotI/BstXI/BstXI/NotI/HindIII. The vector has URA3 as a yeast selectable marker and amp R as an E. coli selectable marker. The vector also has HpaI restriction site outside the polylinker region.
2.13. Cloning ⊗PhTPS1 into the pDB20 Yeast Expression Vector 0.25 µg of pDB20 digested with NotI (Invitrogen was treated with phosphatase following the protocol provided by the New England BioLabs kit (Antarctic Phosphatase, New England BioLabs, Ipswich, MA, USA). The trimmed ⊗PhTPS1 (0.23 µg), was digested with the same NotI enzyme. Ligation was carried out with T4 DNA ligase enzyme (Lucigen, Middleton, WI, USA) and the ligation mix was used to transform competent E. coli cells (Lucigen, E. cloni 10G CLASSIC electrocompetent cells). Plasmid was fully sequenced to check for correct open reading frame (ORF) and rule out PCR artifacts.
Yeast strains were grown in complete liquid medium containing glucose (YPD: 10 g peptone, 20 g yeast extract, 20 g dextrose and 1L dH 2 O). Cells were transformed (Frozen-EZ Yeast Transformation II Kit, Zymo research, Irvine, CA, USA) with 0.25 µg of the NotI-digested pDB20 vector plus 0.23 µg of ⊗PhTPS1 DNA. Another aliquot of the competent cells was transformed separately with NotI-digested pDB20 vector. The transformed yeast was plated on complete supplement mixture-Ura (CSM-Ura; Yeast nitrogen base without amino acids, glucose, agar and Complete Supplement Mixture minus uracil; Sunrise Science Products, San Diego, CA, USA) media to select for Ura + cells. Cells from W303-1A (WT), YSH290 (tps1⊗), YSH450 (tps2⊗) and YSH652 (tps1⊗tps2⊗) with pDB20 + ⊗PhTPS1 and empty pDB20 vector alone were streaked onto Min + Leu + His + Ade-Ura + Galactose and onto Min + Leu + His + Ade-Ura + Glucose to assess if the ⊗PhTPS1 would restore the function of mutant yeast while growing on glucose.

Results and Discussion
To investigate the transcriptomic profile of petunia under salt stress, we performed high-throughput RNA-seq from leaves and roots across 150 mM NaCl. We expected that the identification of differentially expressed genes (DEGs) between the tissues and time points will provide insight into the set of transcripts expressed at early and late stages of salt stress, as well as those DEGs expressed in all samples.
Three lanes of the HiSeq2500 Illumina sequencing platform yielded~700 million paired-end raw reads, with an average of~14 million reads per library. Nearly 20 million reads were filtered out after removing barcode adapters and low-quality sequences. The remaining 672 million reads were aligned against the newly sequenced P. axillaris reference genome [26] with more than 75% of them successfully mapping to the genome sequence. Among the~600 million reads mapped to the genome, 532 million were mapped uniquely to only one location and used for subsequent downstream analyses. A detailed breakdown is shown in Table S1.
We used two independent and open-source programs for RNA-seq analysis to determine expressed and DEGs in our data set: Cufflinks [31] and edgeR [30]. Our choice of these tools reflects significant differences in the underlying algorithms; while Cufflinks uses isoform expression estimates, edgeR compute gene level-based expression estimates. Furthermore, as they use different normalization approaches and dispersion estimates, we expected each program to return a somewhat different set of DEGs. Indeed, edgeR was generally more conservative in identifying DEGs at the same cutoff than Cufflinks ( Figure S1).

Transcriptomic Relationship between the Samples
To assess the transcriptomic relationship between the samples, we used hierarchy clustering with gene-expression values. Hierarchical clustering by Pearson correlation distance method shows that samples cluster, first, by tissue (leaf vs. roots) and then based on responses to salt stress rather than time point ( Figure S2A). We also employed a Spearman rank correlation (as it is less sensitive than the Pearson correlation to strong outliers) and does not inflate type I error rates [34,35]. This approach is in accordance with Pearson; samples cluster based on tissue and responses to salt stress ( Figure S2B). To facilitate future use of these datasets, all the expressed genes identified by both program in all time points and tissue samples are included (Tables S2-S9).

Differentially Expressed Genes across Time Point and Tissues
For a gene to be selected as DEGs, we required the transcript to be identified by both edgeR and Cufflinks at a false discovery rate (FDR) < 0.05. The number of DEGs is shown in the overlap sections in the Venn diagram, Figure S1. Note that hereafter we focused on those eight comparisons where genes from a given tissue at a specific time point in the control treatment (CTR) is compared with the same tissue at the same time point under salt stress (STR). Thus, differential gene expression is based on treatment (control vs. salt) and not other variables (see Materials and Methods section). Throughout the manuscript we refer to these comparisons as: LF_00h, LF_06h, LF_24h, RT_00h, RT_06h, RT_12h, RT_24h and RT_48h. The number of up-and downregulated genes in each comparison is shown in Figure 1 and all DEGs can be found in Tables S10-S17.
genes from a given tissue at a specific time point in the control treatment (CTR) is compared with the same tissue at the same time point under salt stress (STR). Thus, differential gene expression is based on treatment (control vs. salt) and not other variables (see Materials and Methods section). Throughout the manuscript we refer to these comparisons as: LF_00h, LF_06h, LF_24h, RT_00h, RT_06h, RT_12h, RT_24h and RT_48h. The number of up-and downregulated genes in each comparison is shown in Figure 1 and all DEGs can be found in Tables S10-S17.

Expanding from Previous Transcriptomic Studies
We previously performed a similar experiment where petunia plants were exposed to NaCl stress and leaf samples were sequenced [15]. Although an RNA-seq de novo assembly approach was used to map reads instead of the petunia reference genome, some of the DEGs reported here were also identified in our previous study. In both approaches, genes involved in sugar synthesis were highly expressed in leaves. For example, the bidirectional sugar transporter (Peaxi162Scf01337g00018), and α-glucan water dikinase (Peaxi162Scf00192g00217) were identified in leaves with a fold induction of >4 (24 h) and 8-fold induction in leaf at 6 h of salt stress. Moreover, sugar synthesis-related genes were identified, such as galactinol synthase (Peaxi162Scf00366g00813) as well as chaperone genes from the 60 and 70 kDa family were differentially expressed in leaf tissue and detected with both approaches.
To gain insight into salt stress in a tissue specific manner, in addition to leaves, we added root samples and used the petunia genome to map reads. Interestingly, a highly expressed root specific gene was identified in this work. The phosphatidylinositol-4-phosphate 5-kinase (PIP5K) (Peaxi162Scf00258g00045) was significantly induced in all root time points; 12-fold at 6 and 12 h, 60-fold at 24 h, and 18-fold at 48 h. Notably, chemically related to PIP5K, the inositol polyphosphate 5-phosphatase I (Peaxi162Scf00422g00510) was highly expressed (>10-fold, FDR > 0.01) only in roots at 48 h, suggesting that the phosphatidylinositols family of lipids are an essential class of lipids with important roles in early and late salt stress [36]. Although not well characterized in plant cells, phosphoinositide signaling pathways have been linked to abiotic stress such as salinity and drought [37].
These results for PIPK5 are in accordance with those of DeWald et al. [36], who demonstrated that plants respond to salt and osmotic stress by synthesizing phosphoinositides. In their work, 2-week-old A. thaliana plants were treated (immersed) in osmotic-adjusting solutions with 250 mM NaCl for 1 h. High performance liquid chromatography (HPLC) analysis revealed that glycerophosphoinositol phosphate compounds increased by approximately 20-fold in immersed plants vs. non-stressed plants. Apse et al. showed that overexpressing a vacuolar antiport in Arabidopsis increased the ability to withstand high levels of salt (up to 200 mM NaCl) [42]. In our dataset, the cation/H + antiporter 15 (Peaxi162Scf00284g00220), was highly induced in roots at 24 h (>12-fold) and at RT_48h (>7-fold). We also identified the Peaxi162Scf00192g00922 (K + efflux antiporter 5) that significantly increased its expression in RT_48h (>3-fold).
Master regulators such as MYB transcription factors also play a key role in salinity tolerance, as suggested in this work. Peaxi162Scf00147g00136 (myb domain protein 12) was significantly induced (FDR < 0.01) (Tables S2-S17). Different MYB members mediate signal transduction and regulate some stress-responsive genes involved in NaCl stress coping mechanism [43,44].
As some of these DEGs are exclusively expressed at early (i.e., 6 h) and late (i.e., 48 h) time points in the 48 h continuum of NaCl stress in roots, we sought to classify which DEGs were up-and downregulated as early/late response. Thus, we identified 597 DEGs solely expressed at 6 h and 788 DEGs as late (48 h) response ( Figure 2A). We provide the top 10 most highly up-/downregulated DEGs from roots at 6 h and 48 h in Table 1. All these DEGs can be found in Tables S18-S19.

Candidate Genes to Enhance Salt Stress
To this end, we sought to identify specific candidate genes that can potentially increase salt stress tolerance in plants. We reasoned that good candidate genes were those whose steady-state transcript levels are significant at the early onset of salt stress (i.e., 6 h) and whose expression remained active through days of exposure to NaCl stress (i.e., 48 h) in both leaves and roots.
Thus, a stringent criterion was used to select a subset of leaf and roots DEGs for downstream analysis; we required the transcript to be identified as differentially expressed in all six comparisons (LF_06h, LF_24h, RT_06h, RT_12h, RT_24h, RT_48h) by both edgeR and Cufflinks. Using this approach, as suggested by others [45], led to the identification of 828 root DEGs and 2170 leaf DEGs (Figure 2). We then overlapped the 828 and 2170 DEGs to generate a final common list of 178 genes differentially expressed in both roots and leaves over a 48 h period ( Figure S3B).
As no salt stress is induced at 0 h, we sought to confirm that this common list of 178 genes were not differentially expressed as a result of stresses other than NaCl, such as mechanical damage and/or changes in growing conditions (adapting to hydroponic system). As a conservative estimate of the control gene set, we identified transcripts in the union of all the DEGs at 0 h in both roots and leaves. Thus, we would consider a "control" gene even if it was differentially expressed in only one tissue (Figure 4a). This lead to 1291 "control" DEGs expressed at 0 h (Table S20).
We expected a small overlap between the set of control DEGs identified either in roots and/or leaves at 0 h versus those DEGs identified from tissues exposed to NaCl stress. Indeed, only 18 transcripts were found in common when overlapping these datasets ( Figure S3C.). We then removed these 18 transcripts from our analysis to eliminate any gene that might be differentially expressed as a result of other stresses than NaCl, yielding a final "cleaned" list of 160 DEGs (Table S21).
Although many of these 160 DEGs were up/down regulated based on time points and tissue, we found 17 DEGs in the dataset of 160 genes that are induced at all time points (6, 12, 24, and 48 h in roots; and 6 and 24 h in leaves) and 20 DEGs whose transcript levels were downregulated at all time points in all tissues. The expression profile of these up (17) and down (20) regulated DEGs is represented in a heatmap ( Figure 3) and the top 10 most up-and downregulated from the 160 DEGs are shown in Table 2.

Candidate Genes to Enhance Salt Stress
To this end, we sought to identify specific candidate genes that can potentially increase salt stress tolerance in plants. We reasoned that good candidate genes were those whose steady-state transcript levels are significant at the early onset of salt stress (i.e., 6 h) and whose expression remained active through days of exposure to NaCl stress (i.e., 48 h) in both leaves and roots.
Thus, a stringent criterion was used to select a subset of leaf and roots DEGs for downstream analysis; we required the transcript to be identified as differentially expressed in all six comparisons (LF_06h, LF_24h, RT_06h, RT_12h, RT_24h, RT_48h) by both edgeR and Cufflinks. Using this approach, as suggested by others [45], led to the identification of 828 root DEGs and 2170 leaf DEGs (Figure 2). We then overlapped the 828 and 2170 DEGs to generate a final common list of 178 genes differentially expressed in both roots and leaves over a 48 h period ( Figure S3B).
As no salt stress is induced at 0 h, we sought to confirm that this common list of 178 genes were not differentially expressed as a result of stresses other than NaCl, such as mechanical damage and/or changes in growing conditions (adapting to hydroponic system). As a conservative estimate of the control gene set, we identified transcripts in the union of all the DEGs at 0 h in both roots and leaves. Thus, we would consider a "control" gene even if it was differentially expressed in only one tissue (Figure 4a). This lead to 1291 "control" DEGs expressed at 0 h (Table S20).
We expected a small overlap between the set of control DEGs identified either in roots and/or leaves at 0 h versus those DEGs identified from tissues exposed to NaCl stress. Indeed, only 18 transcripts were found in common when overlapping these datasets ( Figure S3C.). We then removed these 18 transcripts from our analysis to eliminate any gene that might be differentially expressed as a result of other stresses than NaCl, yielding a final "cleaned" list of 160 DEGs (Table S21).
Although many of these 160 DEGs were up/down regulated based on time points and tissue, we found 17 DEGs in the dataset of 160 genes that are induced at all time points (6, 12, 24, and 48 h in roots; and 6 and 24 h in leaves) and 20 DEGs whose transcript levels were downregulated at all time points in all tissues. The expression profile of these up (17) and down (20) regulated DEGs is represented in a heatmap ( Figure 3) and the top 10 most up-and downregulated from the 160 DEGs are shown in Table 2.

Petunia Trehalose-6-Phosphate Synthase 1 Gene, a Candidate Gene to Enhance Salt Tolerance in Solanaceae Plants
Using this approach, we were able to select a suite of potential salt gene regulators. From the list of the 17 upregulated DEGs (Figure 3 and Table 2), the petunia trehalose-6-phosphate synthase (PhTPS1) gene increased its expression across all the time points, with maximum expression values in roots at 24 h of salt stress (Figure 4).

Petunia Trehalose-6-Phosphate Synthase 1 Gene, a Candidate Gene to Enhance Salt Tolerance in Solanaceae Plants
Using this approach, we were able to select a suite of potential salt gene regulators. From the list of the 17 upregulated DEGs (Figure 3 and Table 2), the petunia trehalose-6-phosphate synthase (PhTPS1) gene increased its expression across all the time points, with maximum expression values in roots at 24 h of salt stress (Figure 4). Trehalose sugar, widely present in bacteria, yeast and some plants, prevents physical and chemical instability in proteins that occurs upon desiccation when exposed to high concentration of salt [46]. In the first step of the trehalose biosynthesis pathway, the trehalose-6-phosphate synthase gene (TPS) form α,α-trehalose 6-phosphate (T6P) intermediate sugar which is then converted to trehalose by a trehalose-phosphatase gene (TPP) [47]. It has also been hypothesized the T6P as a sugar signaling molecule whose synthesis and degradation is regulated for coordinated pathways in response to prevailing carbon supply [48].
Inactivation of the S. cerevisiae TPS gene (ScTPS1) causes a growth defect when grown in the presence of glucose in the medium associated with deregulation of the initial part of glycolysis. Because glycolysis is essential in all eukaryote organisms [49], we next verified if the PhTPS1 was a functional gene capable of complementing the trehalose biosynthesis in yeast.

The Petunia Trehalose-6-Phosphate Synthase 1 Gene Rescue Mutant Yeast Phenotype
We cloned the novel PhTPS1 allele and performed a phylogenic analysis based on amino acid sequences. The predicted amino acid sequence from the petunia cDNA confirms that the PhTPS1 protein is most closely related to the Solanaceae family, as shown in the phylogenetic tree ( Figure S4).
We next performed a functionality assay with the PhTPS1 to assess if the plant allele could directly rescue yeast mutant phenotypes. We truncated the first 80 aa towards the 5′-end of the PhTPS1 to increase its catalytic activity, as it has been previously shown for bacterial genes (see Materials and Methods section). We used four different yeast strains; two single knockout mutants ("tps1⊗" and "tpp⊗"), a double knockout "tps1⊗tpp⊗" and a wild type yeast strain (see genotypes in Material and Methods section). We reasoned that the TPS1 gene would rescue the lethal phenotype from the tps1⊗. Trehalose sugar, widely present in bacteria, yeast and some plants, prevents physical and chemical instability in proteins that occurs upon desiccation when exposed to high concentration of salt [46]. In the first step of the trehalose biosynthesis pathway, the trehalose-6-phosphate synthase gene (TPS) form α,α-trehalose 6-phosphate (T6P) intermediate sugar which is then converted to trehalose by a trehalose-phosphatase gene (TPP) [47]. It has also been hypothesized the T6P as a sugar signaling molecule whose synthesis and degradation is regulated for coordinated pathways in response to prevailing carbon supply [48].
Inactivation of the S. cerevisiae TPS gene (ScTPS1) causes a growth defect when grown in the presence of glucose in the medium associated with deregulation of the initial part of glycolysis. Because glycolysis is essential in all eukaryote organisms [49], we next verified if the PhTPS1 was a functional gene capable of complementing the trehalose biosynthesis in yeast.

The Petunia Trehalose-6-Phosphate Synthase 1 Gene Rescue Mutant Yeast Phenotype
We cloned the novel PhTPS1 allele and performed a phylogenic analysis based on amino acid sequences. The predicted amino acid sequence from the petunia cDNA confirms that the PhTPS1 protein is most closely related to the Solanaceae family, as shown in the phylogenetic tree ( Figure S4).
We next performed a functionality assay with the PhTPS1 to assess if the plant allele could directly rescue yeast mutant phenotypes. We truncated the first 80 aa towards the 5 -end of the PhTPS1 to increase its catalytic activity, as it has been previously shown for bacterial genes (see Materials and Methods section). We used four different yeast strains; two single knockout mutants ("tps1⊗" and "tpp⊗"), a double knockout "tps1⊗tpp⊗" and a wild type yeast strain (see genotypes in Material and Methods section). We reasoned that the TPS1 gene would rescue the lethal phenotype from the tps1⊗.
Yeast transformations were performed with the truncated petunia TPS1 allele in the pDB20 yeast expression vector and the vector alone (as control) via yeast homologous recombination. Transformants were selected by plating on Min + their respective amino acids minus Ura + media to select for those that became prototrophic ( + ) for the ability to synthesize Ura + marker (Material and Methods section). All four strains transformed with empty vector and the PhTPS1 gene grew on galactose sugar when provided as carbon source, as expected ( Figure 5A). However, when the carbon source was changed to glucose, neither the tps1⊗ single mutant nor the tps1⊗tps2⊗ that were transformed with the empty vector were able to grow. Conversely, the tps1⊗ and tps1⊗tps2⊗ complemented with the petunia TPS1 gene was able to grow in glucose, showing that PhTPS1 is a functional gene capable of restoring the influx of glucose into glycolysis. Empty vector and vector plus gene did not have a growth effect in the wild type W303-1A and tps2⊗ yeast background, as shown in Figure 5B. Yeast transformations were performed with the truncated petunia TPS1 allele in the pDB20 yeast expression vector and the vector alone (as control) via yeast homologous recombination. Transformants were selected by plating on Min + their respective amino acids minus Ura + media to select for those that became prototrophic ( + ) for the ability to synthesize Ura + marker (Material and Methods section). All four strains transformed with empty vector and the PhTPS1 gene grew on galactose sugar when provided as carbon source, as expected ( Figure 5A). However, when the carbon source was changed to glucose, neither the tps1⊗ single mutant nor the tps1⊗tps2⊗ that were transformed with the empty vector were able to grow. Conversely, the tps1⊗ and tps1⊗tps2⊗ complemented with the petunia TPS1 gene was able to grow in glucose, showing that PhTPS1 is a functional gene capable of restoring the influx of glucose into glycolysis. Empty vector and vector plus gene did not have a growth effect in the wild type W303-1A and tps2⊗ yeast background, as shown in Figure 5B. Interestingly, the rate-limiting step in the trehalose biosynthesis appears to be the TPS1 gene, as the double knockout vector plus TPS gene was able to grow in glucose. This in turn, suggests that the T6P intermediate can be dephosphorylated by nonspecific phosphatases. We believe that future research should aim to preform analysis in transgenic plants overexpressing the PhTPS1, as plants will most likely enhance their tolerance salt stress. In summary, this analysis reveals a suite of thousands of genes that are differentially expressed in P. hybrida in roots and leaves upon perceiving and responding to salt stress. For example, calcium-dependent protein kinases expression increased significantly upon acute salt stress, indicating that calcium plays an important role in early steps of the transduction pathway of salt stress signaling. Expression of genes such as the root specific PIP5K appear to provide a quick way to relay stress signals leading to downstream gene expression to mitigate salt damage. Master regulators such as MYB transcription factors also play a key role in Only YSH290 (tps1⊗) V + G and YSH652 double knockout (tps1⊗tps2⊗) V + G were able to grow in glucose when looking at the TPS1 allele showing that PhTPS1 it is capable of restoring TPS1 function in a mutant yeast. Empty vector alone failed to rescue function and YSH290 (tps1⊗) EV and YSH652 double knockout (tps1⊗tps2⊗) EV cells were not able to grow in glucose.
Interestingly, the rate-limiting step in the trehalose biosynthesis appears to be the TPS1 gene, as the double knockout vector plus TPS gene was able to grow in glucose. This in turn, suggests that the T6P intermediate can be dephosphorylated by nonspecific phosphatases. We believe that future research should aim to preform analysis in transgenic plants overexpressing the PhTPS1, as plants will most likely enhance their tolerance salt stress. In summary, this analysis reveals a suite of thousands of genes that are differentially expressed in P. hybrida in roots and leaves upon perceiving and responding to salt stress. For example, calcium-dependent protein kinases expression increased significantly upon acute salt stress, indicating that calcium plays an important role in early steps of the transduction pathway of salt stress signaling. Expression of genes such as the root specific PIP5K appear to provide a quick way to relay stress signals leading to downstream gene expression to mitigate salt damage. Master regulators such as MYB transcription factors also play a key role in salinity tolerance, as suggested in this work. Different MYB members mediate signal transduction and regulate some stress-responsive genes involved in NaCl stress coping mechanism. Importantly, the TPS1 gene, widely described in the literature for its involvement in abiotic stress tolerance in other species, was differentially expressed at all time points in all tissue and, upon functional assay we showed is a functional gene capable of rescuing mutant yeast phenotype. Although we focused on a subset of genes, it is important to note that other DEGs identified in this work should not be discounted as potential salt stress regulators. Other approaches in the near future may lead to the discovery of other putative enhancer of salt stress.  Figure S4: Phylogenic tree (including non-plant organisms) showing the Petunia Hybrida cv. 'Mitchell Diploid' TPS1 relative to TPS1 protein sequences from other species: Nicotiana tabacum, Solanum lycopersicum, Vitis vinifera, Populus trichocarpa, Arabidopsis thaliana, Ricinus communis, Sorghum bicolor, Zea mays, Oryza sativa, Triticum aestivum, Selaginella lepidophylla, Candida albicans, Saccharomyces cerevisiae, Clavispora lusitaniae, Apis mellifera, Drosophila melanogaster, Escherichia coli and Ginkgo biloba; Table S1: Summary of RNA-seq data. LF = Leaf, RT = Roots. CTR = No NaCl, STR = 150 mM NaCl. 00h = 0 h of treatment or control, 06h = 6 h of treatment or control, 12h = 12 h of treatment or control, 24h = 24 h of treatment or control and 48h = 48 h of treatment or control. Three biological replicates are indicated as B1, B2 and B3; Table S2: All expressed protein-coding and non protein-coding genes identified at 0 h (00h) of 150 mM salt stress (STR) and control (CTR) treatments in Petunia leaf. Gene ID, annotation and Fragments Per Kilobase of transcript per Million mapped reads (FPKM) are included; Table S3: All expressed protein-coding and non-protein-coding genes identified at 6 h (06h) of 150 mM salt stress (STR) and control (CTR) treatments in petunia leaf, Gene ID, annotation and Fragments Per Kilobase of transcript per Million mapped reads (FPKM) are included; Table S4: All expressed protein-coding and non-protein-coding genes identified at 24 h (24 h) of 150 mM salt stress (STR) and control (CTR) treatments in petunia leaf, Table S5: All expressed protein-coding and non-protein-coding genes identified at 0 h (00h) of 150 mM salt stress (STR) and control (CTR) treatments in Petunia roots. Gene ID, annotation and Fragments Per Kilobase of transcript per Million mapped reads (FPKM) are included; Table S6: All expressed protein-coding and non-protein-coding genes identified at 6 h (06h) of 150 mM salt stress (STR) and control (CTR) treatments in petunia roots. Gene ID, annotation and Fragments Per Kilobase of transcript per Million mapped reads (FPKM) are included; Table S7: All expressed protein-coding and non-protein-coding genes identified at 12 h (12 h) of 150 mM salt stress (STR) and control (CTR) treatments in petunia roots, Gene ID, annotation and Fragments Per Kilobase of transcript per Million mapped reads (FPKM) are included; Table S8: All expressed protein-coding and non-protein-coding genes identified at 24 h (24 h) of 150 mM salt stress (STR) and control (CTR) treatments in petunia roots. Gene ID, annotation and Fragments Per Kilobase of transcript per Million mapped reads (FPKM) are included; Table S9. All expressed protein-coding and non-protein-coding genes identified at 48 h (48 h) of 150 mM salt stress (STR) and control (CTR) treatments in petunia roots, Gene ID, annotation and Fragments Per Kilobase of transcript per Million mapped reads (FPKM) are included; Table S10: Differentially expressed genes (DEGs) commonly identified by Cufflinks and edgeR between Control (CTR) and Salt (STR) in leaf at 0 h (00h) of treatment. A cutoff of 0.05 was used. LF = leaf, RT = roots; Table S11: Differentially expressed genes (DEGs) commonly identified by Cufflinks and edgeR between control (CTR) and salt (STR) in leaf at 6 h (06h) of treatment. A cutoff of 0.05 was used. LF = leaf, RT = roots; Table S12: Differentially expressed genes (DEGs) commonly identified by Cufflinks and edgeR between Control (CTR) and Salt (STR) in leaf at 24h (24 h) of treatment. A cutoff of 0.05 was used. LF = leaf, RT = roots; Table S13: Differentially expressed genes (DEGs) commonly identified by Cufflinks and edgeR between control (CTR) and salt (STR) in root at 0 h (00h) of treatment. A cutoff of 0.05 was used. LF = leaf, RT = roots; Table S14: Differentially expressed genes (DEGs) commonly identified by Cufflinks and edgeR between control (CTR) and salt (STR) in root at 6 h (06h) of treatment. A cutoff of 0.05 was used. LF = leaf, RT = roots; Table S15: Differentially expressed genes (DEGs) commonly identified by Cufflinks and edgeR between control (CTR) and salt (STR) in root at 12 h (12 h) of treatment. A cutoff of 0.05 was used. LF = leaf, RT = roots; Table S16: Differentially expressed genes (DEGs) commonly identified by Cufflinks and edgeR between control (CTR) and salt (STR) in root at 24 h (24 h) of treatment. A cutoff of 0.05 was used. LF = leaf, RT = roots; Table S17: Differentially expressed genes (DEGs) commonly identified by Cufflinks and edgeR between control (CTR) and salt (STR) in root at 48h (48 h) of treatment. A cutoff of 0.05 was used. LF = leaf, RT = roots; Table S18: All 597 identified DEGs that are expressed as an early (06h) response to salt stress (150 mM) in roots, Table S19: All 788 identified DEGs that are expressed as an early (48 h) response to salt stress (150 mM) in roots, Table S20: Differentially expressed genes (control DEGs) from roots (RT) and leaf (LF) at 0 h (00h) of salt stress, Table S21: All the 160 "cleaned" differentially expressed genes DEGs.