Eleven StOSMs Genes in the Potato Genome Response to Water Deficiency

Osmotin plays an important role in plant response to low temperature and pathogen infection. To date little is known on existence and full function of OSM genes in genome-wide. In this study, based on an OSM expression sequence tag identified from a drought-treated subtractive library, StOSMs in the potato genome were identified and characterized in response to a water deficit gradient. BLAST and bioinformatic analysis revealed that there are at least eleven StOSMs in the potato genome. Among eleven StOSMs, eight StOSM mRNAs accumulated in leaf at the drought-lethal critical point (DLP) were at least 4-fold higher than in the control. The peak StOSM-8E mRNA accumulation was 49-fold higher than the control. Three StOSM mRNAs in leaf at DLP were 9.2-fold lower than in the control. This result of qRT-PCR quantification was identical with the FPKM values of all eleven StOSMs examined in the the Potato Genome Sequence database (PGSD). In conclusion, water deficiency does induce the regulation of the expression of the eleven StOSMs. Either up or downregulated by water deficiency, StOSMs play a positive role in enhancing potato tolerance to drought stress. Therefore, osmotins can be considered to be drought responsive molecules involved in withstanding water deficits. The result provides an evidence to reveal how potato adapts to drought stress.


Introduction
Osmotin (accession No. M29279), encoding the stress response molecule osmotin, was initially identified in cultured tobacco cells (Nicotiana tabacum L. cv Wisconsin 38) induced by low water potential (Singh NK et al 1989). Subsequently, it was showed that abscisic acid (ABA), low temperature and NaCl can induce the accumulation of three StOSM mRNAs (pA13, pA35, and pA81) in S. commersonii grown in vitro (Zhu et al 1995). Later on StOSM was shown to exhibit antifungal activity against a range of fungal pathogens (Rivero et al. 2012, Mani et al. 2012and Patade et al. 2013. Recently, StOSM overexpression in transgenic Arabidopsis and pepper results in H 2 O 2 accumulation and the response of hypersensitive cell death in the leaf and reduces susceptibility to pathogen infection (Choi et al. 2013).
Remarkably, it is showed that plant osmotin, by binding to the adiponectin receptor as an agonist, activates a pathway concerned with the resistance to animal disease. Osmotin reduces ethanol neurotoxicity via the upregulation of the antiapoptotic Bcl-2 protein, and reverses synaptic dysfunction and neuronal apoptosis. The Bcl-2 protein appears to block a distal step in a common pathway for apoptosis and programmed cell death (Naseer et al. 2014 andShah et al. 2014).
Osmotin is not only an intercellular component involved in stress tolerance but also an agonist that alternatively binds to the adiponectin receptor involved in animal apoptosis. However, although osmotin was identified in tobacco in 1989, little is known regarding the details of plant osmotin, particularly how many OSM members are present in a plant genome and how these members work together in response to stress.
In potato, nine StOSM genes were identified by screening a bacterial artificial chromosome (BAC) library. The nine StOSM genes are organised into two loci on chromosomes 08 and 11 (Castillo et al. 2005). Whether there are additional StOSM genes in the potato genome and the functional differences among the potato StOSM genes regarding the response to abiotic stress, such as drought, remain unknown.
With respect to the severity of the water shortage, each plant species is more or less tolerant to a certain drought level. A reliable and comparable level of plant tolerance to drought is known as the drought-lethal critical point (DLP). The availability of water in soil at a level lower than the DLP leads to plant death due to cell dehydration.
In this study, with a full-length StOSM cDNA isolated initially from a drought-treated subtractive library, eleven StOSM genes were identified using BLAST analysis of the Potato Genome Sequence database (PGSD). Leaf mRNA accumulation of StOSMs were assayed using RT-PCR and qRT-PCR and compared with FPKM (Fragments Per Kilobase of exon model per Million mapped reads) value of StOSM genes in the PGSD. The result indicated that expression of eight StOSMs in leafs at DLP was upregulated, three down-regulated. The result revealed that osmotins, as drought responsive molecules involved in withstanding water deficits in potato. The isolated cDNA in this study was still named after StOSM-3B because the cDNA sequence showed high identity with that of StOSM-3B. Analysis of the expression of StOSM-3B in E. coli indicated that as the StOSM-3B mRNA accumulates, the tolerance of recombinant E. coli, as described by colony livability under osmotic stress, was significantly improved, even as the osmotic stress generated by PEG6000 increased (Hu et al 2012).
Wounding, abscisic acid (ABA), low temperature, NaCl and salicylic acid (SA) increased the StOSM-3B (pA81) mRNA expression in S. commersonii cell cultures and in plants grown in vitro. The reason for why these conditions induce StOSM-3B mRNA accumulation is unknown yet (Zhu et al. 1995). In addition, the involvement of StOSM-3B in water stress has not yet been reported. All potato StOSM genes were collected from the NCBI GenBank using BLAST analysis and aligned with these eleven StOSM. No one beyond eleven StOSM was found according the sequence alignment. Therefore, seven genes, such as StOSM-3B and -3F, were designated according to names used previously and registered in the database (Castillo et al. 2005). Four new StOSM genes were identified and named based on the length of the putative encoded peptide. Overall, eleven StOSM genes were distributed on five of twelve potato chromosomes ( Figure 3). Seven of the eleven StOSM genes are near 54277k-54230k on chromosome 08 ( Figure 4). The other four are located on chromosomes 01, 03, 06 and 11 ( Figure 3 and Table 1). The alignment of the amino acid sequences of the eleven osmotins revealed four conserved regions: the NNCPYT, RIW and TGDCGG motifs, located in the N-terminal region, and the AYSY motif, located in the C-terminal region ( Figure 6A). A Jalview of the alignment of StOSM-5A, -3B, -3F, -1G and -8E based on their phylogenetic relationship showed that a tandem duplication event occurred in these five genes ( Figure 6B). Interestingly, the putative amino acid sequences of all osmotins range from 182 aa for StOSM-182 to 306 aa for StOSM-306. StOSM-182 and -306 show a greater divergence towards the C-terminal region ( Figure 6A). The sequences of four motifs were searched in the UniProt Knowledgebase (UniProtKB) by blastx. However, the result of zero hits from the search showed that the function of these conserved regions is unknown yet.
StOSM-182, with 182 amino acid residuals, is the shortest osmotin protein and does not contain the AYSY motif in its C-terminus ( Figure 6A). In addition to the common four motifs, StOSM-182 and -306 share two additional motifs, FCPMKGVKRPN and ILLPFPMLI. Neither an expression sequence tag nor gDNA/cDNA clone of these two genes was registered in the database. Meanwhile, there was no hit for annotated or predicted function of these motifs when these motif sequences BLAST was done in all available protein database. This means that function of these motifs is unknown yet.

Introns and UTRs of StOSMs
The structure of all the StOSM genes was highly conserved. Only three StOSM genes, including StOSM-182 and -306, have two introns (Figure 7). Within the second intron of StOSM-182 is the first exon of a putative gene, which encodes ethylene-responsive proteinase inhibitor 1 (EPI). The second and third EPI exons are located in the 3' end of the third exon of StOSM-182. In addition, there is one more copy of EPI locating within the promoter region of StOSM-3B (Figure 4). The diagram is drawn to scale, according to the alignment of the cDNA sequences with the corresponding genomic sequences. All genomic sequences for StOSM were obtained from sequencing data from the Potato Genome Sequencing Consortium website (http://solanaceae.plantbiology.msu.edu) exon of these two genes have 5 nt and 7 nt in common, respectively. These findings imply that the two exonic splicing patterns are conserved between the two genes.
The lengths of the UTRs (untranslated region), which are important in the regulation of RNA mobility (Banerjee et al. 2006 and2009), were compared using sequences from either the clones obtained in this study or web-based RNA data. The StOSM-306 5'UTR is 120 nt, and the StOSM-251 3'UTR is 220 nt. Both were the longest UTRs among the eleven StOSMs (Table 1). StOSM-182 and StOSM-297, which contain two introns, have neither a 5'UTR nor a 3'UTR.  (Xu et al. 2011 andCory et al. 2014), the only available RNA-Seq data under water stress is from the RH leaf. Therefore, the RNA-Seq data for certain RH organs, including leaf with and without water stress, stem, roots, stolon and mature tuber, were selected and compared with respect to the expression of the eleven StOSM genes. The fragments per kb per million mapped reads (FPKM) in leaves were compared with the FPKM in the other four organs (Figure 8). StOSM-297 mRNA occurred at a low FPKM level in the control leaf and at high FPKM levels in the other four organs. The absence of a detectable FPKM level in leaves under water stress suggests that StOSM-297 is only one of eleven StOSMs and that its expression in leaves was supressed by the induction of water stress (Figure 8). Although StOSM-251 is present in both the treatment and control leaves, StOSM-182 was expressed at a very low level in the tuber and control leaf (Figure 8), and the eleven StOSM transcripts were ubiquitously observed in a variety of organs.

Expression of
The most striking feature of this StOSM mRNA expression profile, however, was that eight of the eleven StOSM genes in RH leaves under water stress exhibited 24-54-fold higher FPKM values than in the control leaves and the other four organs (Figure 8).
Overall, the average FPKM value of three StOSMs, StOSM-3F, -1G and -8E, in leaves under stress were 33-fold higher than in the control, exhibiting the highest FPKM values (Figure 8). In contrast, the FPKM values for StOSM-306 and -251 in leaves under water stress was approximately 3-fold lower than in the control.
The average FPKM value of the other 5 StOSMs that increased in response to water stress was 21-fold higher than in the control (Figure 8). StOSM-8E was the most abundant RNA of the eleven StOSMs, and the top three ranked StOSMs by expression are StOSM-3F, -1G and -8E (Figure 8).
Both the 21-33-fold increase and the 3-fold decrease of the FPKM values for the eleven StOSMs induced by drought suggest that the StOSMs not only are associated with low temperature, abscisic acid and NaCl, but also respond positively to water stress as drought-induced genes. RT-PCR analysis showed that eight of the eleven StOSMs, StOSM-8E, -1G, 3B, -3F, 5A, 3C, 2D and -182, were upregulated in response to the degressive gradient of water availability (Figure 9). The peak mRNA accumulation of all upregulated StOSMs was reached at the DLP. Among the eight upregulated StOSMs, StOSM-1G and -8E synchronously showed much stronger and higher increases than the others.
The expression attenuation of three StOSMs, StOSM-297, -251 and -306, was observed for the same case ( Figure 9). StOSM-1G and -8E synchronously showed much stronger and higher increases in mRNA accumulation than the other upregulated StOSMs over the degressive gradient. Figure 9 Dynamic quantification of leaf OSM mRNA accumulation over the decreasinggressive gradient. Under the degressive gradient of water stress from 70% ± 5% to 10% ±2% (DLP), OSM mRNA accumulation was assayed by reverse transcription PCR in the leaves of the potato species, S. tuberosum ssp. ZHB (WCM: water content of the media in which the plants grow) The sequences of certain StOSM members, such as StOSM-3B, -3F and -5A, share greater than 95% identity, which makes distinguishing each member precisely by RT-PCR difficult. A visible distinction of these StOSMs on the gel displaying the RT-PCR products confirmed the specificity and reliability of the eleven primer pairs to distinguish each gene by RT-PCR. Therefore, these primers were used to quantify the expression of the eleven StOSM mRNAs in leaves.

Real-time quantitative reverse transcription-PCR (qRT-PCR) for StOSM expression analysis
Real-time qRT-PCR was performed for all StOSM members to determine the leaf RNA accumulation patterns in response to decreasing water availability. The mRNA accumulation of eleven StOSMs in the leaf induced by 10% WCM (also DLP) as the treatment and by 70% WCM as the control was quantitatively assayed (Figure 10). Similar to the qualitative assay results, eight of eleven StOSMs under the 10% DLP condition displayed significant increases in RNA accumulation compared with the control condition.
The changes in the eleven StOSMs induced by drought varied from gene to gene (Figure 4). Leaf StOSM mRNA accumulation for the genes upregulated by DLP was 4-49-fold higher than in the control. The peak StOSM-8E mRNA accumulation was greater than 49-fold higher than in the control. StOSM-297 mRNA accumulation decreased 9-fold in response to drought compared with the control.
StOSM-251 showed exceptional leaf mRNA accumulation. No significant difference was observed in the relative levels of transcripts between the treatment and control leaves ( Figure 10). This finding was not in agreement with the FPKM value and qualitative assay results.
This comparison clearly revealed that the result of the qualitative assay is in agreement with the mRNA quantification and FPKM value results from the RNA-Seq analysis (Table 2). This observation indicates that drought does induce the regulation of the expression of the eleven StOSMs. With eight members of StOSM being upregulated and three being downregulated, StOSMs play a positive role in enhancing plant tolerance to stress.
Note:stands for down-regulated; ＋ for up regulated Therefore, osmotins can be considered to be drought responsive molecules involved in withstanding water deficits. Recent reports provide direct evidences to support osmotin as a drought responsive molecule. Expression of a osmotin-like SnOLP from Solanum nigrum, 93.0% identity on aminoacid sequence with OSM-3F (PGSC0003DMG400003044) in S. tuberosum confers drought tolerance in transgenic soybean (Weber et al., 2014). Aslo, overexpression of OSM-3F gene confers tolerance to salt and drought stresses in transgenic tomato (Solanum lycopersicum L.) (Goel et al., 2010).
Drought stress is an important environment factor limiting crop production in some areas. Up to date, a litter is known on how to enhance crop tolerance to drought stress. This study not only provides a method on how to screen isoforms of a gene from a genome database, but also disclose response of osmotin in drought tolerance. This is a practical option to dig a genome for member identification of a gene family.

Discussion
From a metabolic point of view, the plant response to water stress involves the generation of intercellular components concerned with the regulation of water potential. Osmotin is a component that accumulates substantially when induced by water stress. Previous studies revealed that the enhancement of osmotin expression can increase plant tolerance to fungi and abiotic stress. However, little is known regarding the mechanism of osmotin in the response to stress. More importantly, plant genomes annotated have revealed that a gene has diverse functions. Therefore, to reveal a specific role of a gene or its product and its interaction with other genes/proteins in a stress response pathway, a prerequisite is to clarify and specify the function(s) of the gene and any differences from other genes with the same or similar functions.
Based on the consideration above and the genome sequence database, starting with an EST from a StOSM gene isolated from the stress subtractive library for water stress, the drought responses of eleven StOSM genes were revealed. Eight StOSMs were upregulated and three were downregulated in response to water deficits. Using the RegSite Plant Database and the online software TSSP, the binding sites for cis-elements in the StOSM promoter were analysed, revealing that the pattern of binding sites in the StOSM promoters varies among the eleven StOSM genes. The resulting expression of the different StOSM genes varies during the response to water stress.
The drought tolerance capacity of a plant, similar to its tolerance to other stresses, has been formed during the long process of evolution. No water conditions except extreme drought induce the expression of potential drought responsive genes and the accumulation of drought responsive molecules. Based on the potato drought-lethal critical point (DLP) defined in this study according to the phenotype under water deficit, StOSM mRNA accumulation reaches its peak at DLP along a degressive gradient toward the DLP. This observation has not been previously reported.
The qualitative and quantitative expression of the StOSM genes in this study is in agreement with the data analysis of the potato genome sequence (Xu et al. 2011). Osmotins, as response molecules, are clearly involved in drought tolerance in the potato. However, the extent to which the osmotins are associated with drought tolerance in the potato, such as the specific stages in the adult plant and during the restoration of water after DLP, has yet to be determined. The answers to these questions are indispensable to establishing the dynamic network involved in drought response.
In addition, osmotins, as drought responsive players, surely interact or crosstalk with other molecules in the drought response network. In this study, the qualitative and quantitative results of StOSM in response to drought provide a background in which the interaction or crosstalk of StOSM with other response molecules can be revealed to establish the network.
Although the first StOSM (accession No. M29279) was isolated almost 25 years ago, eleven-StOSM in S. tuberosum identified in this study is the first one to disclose OSM members in a plant genome. This is necessary not only for understanding StOSM function in the drought response network, but also for engineering improvements in plant tolerance to drought.

Plant materials
S. tuberosum genotype ZHB is a tetriploid cultivar. ZHB seedling can survival at the level of relative soil water content as low as 10%. RH (RH89-039-16) genotype is a diploid clone derived from a cross between a S. tuberosum 'dihaploid' (SUH2293) and a diploid clone (BC1034). RH was selected as a diploid model for potato genome sequencing. In this study ZHB minitubers were sown in a pot containing peat as the medium (Pindstruo Mosebrug, DK) and grown under the following conditions: water content of 70% ± 5% in the medium, light intensity of 100 µmol m -2 s -1 , temperature of 20-25°C and relative humidity of 60-80%, as it was done by Anne J. K (Anne et al., 1996, Liu et al. 2012).
On the 35th day after minituber germination (when the stored starch and other nutrition in the minituber is completely exhausted as observed in our case), the seedlings were kept under the same conditions, with the exception that no water was provided. The water content in the medium (WCM) was assayed every 12 hours until seedling death. Leaf sampling with three replicates was performed along with a water deficit gradient above the drought-lethal critical point (DLP) ranging 70, 60, 50, 40, 30, 20 and 10 %. Ten-gram leaf samples were collected, frozen in liquid nitrogen and stored at -80°C for subsequent experiments.

Phylogenetic and gene structure analysis
The sequence of the cloned StOSM cDNA was compared against the NCBI files and the PGSC_DM_v3.4_gene.fasta files from the Potato Genome Sequencing Consortium (PGSC) website (http://potatogenomics.plantbiology.msu.edu) (Xu et al. 2011 andCory et al. 2014). Using the downloaded sequences for the StOSM and promoter from the PGSC database, two phylogenetic trees for the StOSM peptides and promoters were constructed using the neighbour-joining method in the MEGA 5.02 software package developed by (Saitou andNei1987 andTamura et al. 2007

Reverse transcription PCR for StOSM expression analysis
A total of 300-350 ng of RNA was extracted from leaves at WCM concentrations of 70%, 60%, 50%, 40%, 30%, 20% and 10%. Following the instructions for Premix Ex TaqW Version 2.0 (TaKaRa), reverse transcription PCR was performed using primers (Table 3) designed specifically for each of the eleven StOSMs. The PCR product amplified from the treatment templates using beta-actin-specific primers was used as an internal control. Reverse transcription reactions with template-free were used as a negative control to exclude DNA contamination. Using a standardised procedure, all PCRs were optimised with respect to the denaturing temperature and extension time to yield a product of the expected size.

Real-time quantitative reverse transcription-PCR (qRT-PCR) for StOSM expression analysis
Total RNA was treated with RNase-free DNase (Qiagen). The quantity and quality of the RNA samples were estimated using a Nano spectrophotometer (ND-1000; Thermo Scientific). RNA samples with a 260:280 ratio ranging from 1.9 to 2.1 and a 260:230 ratio ranging from 2.0 to 2.5 were used for the qRT-PCR analyses.
Real-time quantitative reverse transcription-PCR analysis was performed using the PrimeScript™ RT Master Mix and SYBR® Premix Ex Taq™ II (RR820A and RR036A, TaKaRa Biotechnology, Dalian, China) following the manufacturer's protocol. Briefly, 50-ng aliquots of the total RNA template were subjected to each qRT-PCR in a final volume of 20 µL containing 4 µL of the PrimeScript RT Master Mix and 0.4 µL of qScript Reverse Transcriptase along with target-specific primers (200 nm).
All reactions were performed in triplicate using 1 μl of template cDNA in a final volume of 20 μl in a Bio-Rad Real Time PCR System (Bio-Rad, CA, USA). The Illumina Eco qPCR machine with fast quantitative PCR cycling parameters was used for complementary DNA synthesis with the following conditions: 37°C for 15 min followed by 85°C for 5 s; initial denaturation − 95°C for 30 s for 1 cycle; 2 step PCR − 95°C for 5 s and 60°C for 30 s for 40 cycles.
Beta-actin (accession number DQ252512) was used as an internal control. The relative gene quantification method was used to calculate the expression levels of different target genes (Livak and Schmittgen 2001). Primer specificity was determined by melting curve analysis and qualification analysis by RT-PCR. A standard curve was generated based on serial dilutions of cDNA to calculate the gene-specific PCR.
The relative expression levels of each StOSM gene were analysed using the IQ5 software and are presented as the fold-difference from the baseline expression levels. Using Origin Pro 8.0 software (Origin Lab Corp., Northampton, MA, USA)，paired t-tests were performed to assess the significance of expression level differences between the treatments and the controls.
Author's contributions YXL designed the experiment and primers used, performed data analysis, wrote and revised the manuscript and supervised performance of the project. Performed the experiments: Qiao LB, Dai HX, Fan HH, Zhang XP and Wu F. Analysed the data: Yao XL, Qiao LB, and Fan HH. Wrote the paper: YAO XL and Qiao LB.