Identification of a defense response gene involved in signaling pathways against PVA and PVY in potato

ABSTRACT Potato is the most important non-grain food crop in the world. Viruses, particularly potato virus Y (PVY) and potato virus A (PVA), are among the major agricultural pathogens causing severe reduction in potato yield and quality worldwide. Virus infection induces host factors to interfere with its infection cycle. Evaluation of these factors facilitates the development of intrinsic resistance to plant viruses. In this study, a small G-protein as one of the critical signaling factors was evaluated in plant response to PVY and PVA to enhance resistance. For this purpose, the gene expression dataset of G-proteins in potato plant under five biotic (viruses, bacteria, fungi, nematodes, and insects) and four abiotic (cold, heat, salinity, and drought) stress conditions were collected from gene expression databases. We reduced the number of the selected G-proteins to a single protein, StSAR1A, which is possibly involved in virus inhibition. StSAR1A overexpressed transgenic plants were created via the Agrobacterium-mediated method. Real-time PCR and Enzyme-linked immunosorbent assay tests of transgenic plants mechanically inoculated with PVY and PVA indicated that the overexpression of StSAR1A gene enhanced resistance to both viruses. The virus-infected transgenic plants exhibited a greater stem length, a larger leaf size, a higher fresh/dry weight, and a greater node number than those of the wild-type plants. The maximal photochemical efficiency of photosystem II, stomatal conductivity, and net photosynthetic rate in the virus-infected transgenic plants were also obviously higher than those of the control. The present study may help to understand aspects of resistance against viruses.


Introduction
Potato (Solanum tuberosum L.), as one of the most important agricultural crops in the world (FAOSTAT, http://faostat.fao.org), is central to global food security and ranks third after rice and wheat. 1 Crop growth and yield can be drastically reduced by virus diseases. The loss in potato yield caused by viruses can reach up to 90% depending on the type of the virus. 2 The main cause of seed degeneration was also found to be viruses, particularly potato virus Y (PVY) and potato virus A (PVA) of the genus Potyvirus. 3 In plants, virus infection induces dramatic morphological and physiological changes such as decreased plant vegetative performance, decreased photosynthesis, and increased respiration. These adverse effects of viral infection on host plants result in inferior performance such as decreased host biomass and crop yield loss. 4,5 Host resistance is one of the best strategies to control virus infection in plant species. [6][7][8] One of the considerable challenges of potato breeders is to interpret the resistance mechanisms against virus and develop virus-resistant potato cultivars. Hence, knowledge about how plants regulate their innate immune response and how it can be manipulated by pathogens is necessary in plant science to identify and define natural plant response to stresses. [9][10][11] Viruses must be capable of exploiting infected cell processes to replicate, translocate from one cell to another via plasmodesmata, and move systemically in the entire plant using the plant vascular system. 11,12 Owing to the very limited viral genome size, RNA viruses encode only a small number of essential proteins. Therefore, it is crucial for the virus to employ host factors to promote the infection cycle efficiently. 11,13 Defense signal transduction pathways involve various genes, which are critical for plant immune responses against diverse pathogens. Although several natural resistance genes against members of the genus Potyvirus have been reported in the recent decade in numerous plant species, few of them have been identified. 14 Resistance (R) genes, which generally encode nucleotide binding and leucine-rich repeat (NB-LRR) proteins, deliberate a strong resistance called a hypersensitive response (HR). This resistance involves specific recognition of the pathogen effector, i.e. race specificity (gene-forgene resistance). 15 Therefore, resistance is ineffective in protecting the plants when new pathogen strains appear after a few years of use. 16 Due to the high frequency of emerging resistance-breaking pathogens against R-genes, the focus now is on using signaling and defense-related genes for breeding.
To prepare an array of effective tools against viruses, plants have also evolved many other mechanisms, including production of biochemical compounds before or after the invasion of the pathogen into the plant tissues. Artificial induction of defense signaling and related genes in plants like potato has been shown to confer resistance to a broad-spectrum of pathogens using transgenes such as transcription factors (WRKY, ERF, TGA, MYB), kinases (MAPK kinases, CDPKs), negative regulators (RIN4, SNI1, SON1), and positive regulators (EDS1, PAD4, SGT1). [17][18][19] Unlike breeding by introducing of R genes, the resistance acquired by manipulating downstream signaling and defense-related genes has been documented that can be durable and effective. 17,[20][21][22] This is true also for the members of signal-transducing G-proteins.
Guanosine triphosphate (GTP)-binding proteins (G-proteins) are essential components participating in signal transduction pathways derived from various stresses, including plant defense signaling. 23,24 Upon binding with GTP, G-proteins are activated by the hydrolysis of the bound GTP to GDP. 25 As many G-proteins are stress-responsive, active G-proteins bind with downstream targets to affect and initiate signaling events upon stresses. [24][25][26] Three broad categories of signal-transducing G-proteins, including heterotrimeric G-proteins, small G-proteins (small GTPases), and other 'unconventional' G-proteins, have been identified in plants. 24,27 The small GTPase superfamily is divided into four main subfamilies in plant eukaryotic cells: ADP-ribosylation factor 1 (ARF1)/secretion-associated RAS super family 1 (SAR1), Ras-related proteins in brain (RAB), Rho-related protein from plants (ROP), and RAs-related nuclear protein (RAN). [28][29][30] Plants recognize the pathogen elicitors and produce a diverse array of primary and secondary signals in response to biotic and abiotic stresses to activate various plant protector and defense genes. Subsequently, the defense genes prevent pathogen invasion by producing glutathione S-transferases, proteinase inhibitors, peroxidases, PR proteins, and hydrolytic enzymes (downstream responses). 31 Previous studies, regarding the plant defense response to pathogens, have shown that modification or overexpression of a single signaling pathway gene regulating a large number of defenseresponsive genes confer resistance to a broad spectrum of pathogens. 18,19 There are many examples of successful engineered plants by transformation with different constructs to overexpress trans-and endogenous genes in transgenic plants. Overexpression of signaling and defense-related genes can lead to a constitutive expression resistance phenotype. 32 Small G-proteins play a vital role in plant disease resistance and subsequent cellular responses to pathogens such as bacteria, [33][34][35] fungi, 36 and viruses. 37,38 A number of G-proteins have already been introduced into different plants.
Overexpression of OsRAN2, a small G-protein in transgenic rice and Arabidopsis thaliana plants, exhibited higher survival rates under cold stress than the wild type did. 39 Transgenic tobacco plants overexpressing a small G-protein, RGPL, enhanced resistance to tobacco mosaic virus (TMV) infection. 40 It was also observed that expression of constitutively active (CA) OsRac1, a small GTPase Rho in transgenic rice plants, resulted in production of reactive oxygen species (ROS) and cell death. 41 In another research, heterologous expression of N resistance gene and OsRac1 in tobacco transgenic plants showed a failure to induce antioxidant genes against TMV infection resulting in impaired ROS production. 42 In potato, stable overexpression of AtRop1 (DN-AtRop1) increased resistance to fungus Phytophthora infestans infection. 43 Among small GTP-binding proteins, SAR1 is a unique family within the Ras superfamily. Endomembrane trafficking plays a key role in the maintenance of fundamental cellular functions (signal transduction, cellular homeostasis, etc.) and in response to environmental stresses. SAR1, as a molecular switch, controls the assembly of the coat protein complex II (COPII) directing vesicle budding from the endoplasmic reticulum (ER) 44 and integrates coat assembly with the cargo selection process. [45][46][47] The COPII machinery is involved in plant responses to biotic and abiotic stresses. 46 The SAR1 GTPase is an essential factor of COPII vesicle coats involved in plant responses to various stresses. 45,46 Although the SAR1 superfamily of small GTPases is a key regulatory GTPase, its functions in response to various stresses are not well understood.
Plant genetic engineering represents an applied approach to traditional breeding methods, since it can elevate resistance to diseases, while other desirable traits of the plant are maintained. Finding differentially expressed genes under virus infection is an initial step to search for the most important genes in plant antiviral defense response. In this regard, candidate genes are likely to be significantly activated in resistant cultivars. In this study, given the ability of G-proteins in the enhancement of stress resistance, G-proteins were collected in potato plants infected with PVY and PVA using meta-analysis of publicly available microarray datasets. In this screening, a gene, encoding a kind of small G-protein, which was originally named as StSAR1A-like protein (StSAR1A) was selected. The selected transgenic plants were tested to determine whether overexpressed StSAR1A enhanced resistance to viruses.

Data filtering
In our previous study, we showed for the first time, that a three-step filtering method identified key genes with higher confidence in details. 48 In the present study, we reused the filtering method to identify the potential key genes by focusing on GTP binding genes. Briefly, the selected genes were filtered in a three-step process to detect the key genes that were effectively responsive to the PVY and PVA infection in potato plants. For this purpose, the differentially expressed genes (fold change ≥1.25 and P < .05) were obtained using Fisher's combined probability test 49 (Figure 1(a)). The genes exhibiting significant expression changes only in resistant plants were identified as follows. The pattern of gene expression after virus infection was separately compared in susceptible and resistant plants, as described previously. 48 Using the Venn diagram, the number of shared genes was found between susceptible and resistant plants. A list of genes that were uniquely expressed in resistant plants was obtained by subtracting those expressed in susceptible plants. Since we aimed to select genes involved in resistance to both viruses, we reduced the number of the selected genes commonly expressed in both PVY and PVA infection experiments (Figure 1(b)).

Cultivation and propagation of potato plants
Seed tubers of the potato (S. tuberosum L. cv. Desiree), which are susceptible to PVY and PVA, were planted in the greenhouse with natural light under a 16/8-h light/dark photoperiod and a temperature of 22°C. In this study, the plant S. tuberosum cv. Degima, which is resistant to PVY and PVA, was served as positive control plant. For propagation, plant nodes were cultivated on liquid MS 50 containing salts (Wao, Osaka, Japan), vitamins, and sucrose (

Preparation of viruses for inoculation tests
In this study, PVY strain, infecting transgenic and control plants, was the ordinary strain of PVY (PVY O ). The PVY isolate was from the laboratory collection at the University of Hokkaido, which was maintained as freeze-dried infected tobacco leaf material or as purified virus preparations at −20° C. The accession number of PVA used in this work was MAFF307028 from the NARO Genebank (https://www.gene.affrc.go.jp/databases-micro_ search_detail.php?maff=307028). 51 Infected leaves (100 mg) were ground well with 1 ml of 0. Seven days after inoculation (dpi), upper noninoculated leaves showing symptoms were used as an inoculum for mechanical inoculation into potato plants.

Cloning of StSAR1A in a binary vector
The PCR was carried out using the gene-specific primers designed from 5ʹ untranslated region (UTR) and 3ʹ UTR ( Table 2) to obtain the fulllength coding sequence (CDS) of the StSAR1A gene (GenBank accession number KY196464) and the Max Taq DNA polymerase (mixture of Taq and Pfu in a 16:1 ratio; Vivantis Technologies, Malaysia) according to the manufacturer's instructions. The purified PCR products were cloned into a pTZ57R/T cloning vector. The pTZ57R/ T-StSAR1A vector was digested with the restriction enzyme (XhoI) to confirm the presence and length of the inserted fragment. After confirmation by sequencing, the correct pTZ57R/T-StSAR1A vector was cloned into the downstream of the cauliflower mosaic virus (CaMV) 35S promoter of pRI201-AN (10432 bp) as a plant binary vector (Takara Bio., Tokyo, Japan). Agrobacterium tumefaciens strain LBA4404 was then transformed with the constructed pRI201-AN-StSAR1A by the freeze-thaw method. 52

Genetic transformation of potato plants
Potato plants cv. Desiree was transformed with A. tumefaciens 53 containing the pRI201-AN-StSAR1A binary vector. The internode pieces (4-6 mm) of potato tissues were used as explants for transformation. The explants were maintained on a co-culture medium (   In this study, the regeneration medium without antibiotic was used as a control medium. Kanamycin-resistant fully rooted plants were further tested for the presence of transgene using PCR.

Molecular characterization of the transgenic plants
The kanamycin-resistant transformants were confirmed by PCR using the forward primer for the CaMV 35S promoter (35S-F) and the StSAR1A specific reverse primer (StSAR1A-R1) ( Table 2) with the product 788 bp. Oligonucleotide primers ( Table 2) were designed using the Primer Analysis software, Oligo Ver. 7.54 (Molecular Biology Insights; Wojciech and Piotr Rychlik, USA). The PCR-positive plants were further verified by Southern blot analysis. Total genomic DNA was separately isolated from the potato leaves of transgenic and non-transgenic plants, as the control, following the plant DNAzol reagent (Invitrogen, USA) according to the manufacturer's instructions. The genomic DNA isolated from untransformed potato plants and pRI201-AN plasmids containing the transgene (StSAR1A) were used as negative and positive controls, respectively. The probe sequences were designed using the Oligo 7 Primer Analysis Software. To prepare hybridization probes, the pRI201-AN vector (pRI201-AN-StSAR1A) was used as a template in the PCR reaction. The T7 promoter-sequence was fused to the 5′-end of the StSAR1A gene-specific primer (StSAR1A-T7-R**, Table 2). The PCR was carried out using the genespecific reverse primer with T7 (StSAR1A-T7-R**), the gene-specific forward (StSAR1A-F) primer (Table 2), and using the KOD FX Neo DNA polymerase (Toyobo, Osaka, Japan). The amplicons of the expected size (808 bp) were gel purified and quantified using a Nanodrop spectrophotometer (ND-1000, USA). Approximately 200 ng of PCR product was used for DIG-labeled RNA. Transcription was carried out using the T7 RNA polymerase, the DIG RNA Labeling Mix, and the purified PCR fragment (template DNA). The reaction was incubated at 37°C for 2 h. The genomic DNA from the transgenic and non-transgenic plants (20 μg) was digested with restriction endonuclease EcoRI, which was not cut within the T-DNA region. The digested DNA was fractionated in a 0.6% (w/v) agarose gel and blotted onto a nylon membrane (Roche, Germany), and then it was UV cross-linked. DNA blotting and hybridization were conducted as previously described. 54 The hybridization signals were detected using the Detection Starter Kit I, according to the manufacturer's instruction (Roche, Basel, Switzerland). Table 2. Primer sequences and amplified length for G-protein (StSAR1A), PVY (potato virus Y), PVA (potato virus A), 18S (18S ribosomal RNA), 35S (CaMV 35S promoter) genes, HSP (heat shock protein terminator), and Vir B6 (vir genes of Agrobacterium).
Primers-sequence used for real-time PCR and b primer-sequence used for Southern blot analysis; whereby T7 promoter-sequence is underlined.

Gene expression analysis and detection of viral infection
Virus-inoculated leaf tissues (1 g) of N. benthamiana were carefully ground in 1 ml phosphate buffer (0.5 M, pH 8.0) and used to inoculate transgenic and control plants with similar viral doses. One week later, the expression patterns of each targeted gene in the course of PVY and PVA inoculation in transgenic and control plants were monitored by real-time PCR. Conserved regions of viral coat protein (CP) were used to monitor each virus (PVY-F2* and R2* or PVA-F2* and R2*). The expression level of the StSAR1A gene was evaluated using gene-specific primers (StSAR1A-F2* and R2*) ( Table 2). Each biological replicate was obtained via the average of three technical replicates. Two leaves from three potato plants were sampled and pooled to produce three biological replicates per treatment. Total RNA was extracted using a Plant Total RNA Purification Kit (cat#TR02-150, Molecular Biology Tools). The reverse transcriptase enzyme (MmuLV, Fermentas) was used to synthesize cDNA by following the manufacturer's instructions. Additionally, 1 μl of diluted cDNA (1:20 ratio for targeted genes and viruses, 1:100 ratios for internal control genes) was tested with iQ SYBR Green Supermix (Bio-Rad) in a final volume of 20 μl reaction under the following conditions: 95°C for 3 min (hot start), followed by 40 cycles of 95°C for 5 s, 60°C for 10 s, 72°C for 20 s (amplification) and 95°C for 30 s, 65°C for 30 s, 95°C for 30 s (melt). The 18S ribosomal RNA (X67238) was used as an internal control to estimate the relative expression levels. To detect primer-dimer artifacts and ensure amplification specificity, primers specificity was assessed using a melting curve analysis. Primer efficiency was confirmed using serial dilutions of cDNA at a final concentration of 10 −1 , 10 −2 , 10 −3 , 10 −4 , 10 −5 . The 2 −ΔΔCt method was used to calculate the relative level of gene expression. 55,56 Three independent experiments were performed to confirm the results. In this study, Wt (non-inoculated potato plants), Wt/Y and Wt/A (wild-type plants inoculated with PVY and PVA, respectively), R (Resistant plants), R/Y and R/A (resistant plants infected with PVY and PVA, respectively), and Wt/M, R/M, and T/M (wild-type, resistant, and transgenic plants mockinoculated for both viruses) were grown under the same conditions and served as controls. The internode segments without Agrobacterium treatment were grown in unselective medium with no antibiotic as control plants.

Enzyme-linked immunosorbent assay (ELISA)
The extent of viral infection in transgenic and nontransgenic plants was determined by doubleantibody sandwich ELISA (DAS-ELISA) 57 using polyclonal antibodies raised against PVY and PVA. Three weeks after inoculation with PVY and PVA, non-inoculated upper leaf samples from the same positions of potato plants were collected and examined for virus accumulation. Two leaves from three potato plants were sampled and pooled together to produce three biological replicates per treatment. Absorbance at 405 nm optical density was determined using an ELISA plate reader (ARVO MX 1420 MULTILABEL COUNTER). The triplicate absorbance readings for each sample were averaged and corrected by subtracting these values from the buffer blank triplicate readings. Virus-inoculated samples were those with average absorbance values greater than R (R = [mean ± 3 × standard deviation] of the negative controls). 58 Blank, mock-inoculated, and resistant potato plants were used as negative controls, and the virusinoculated leaves of potato wild-type plants were considered positive controls.

Gas exchange and chlorophyll fluorescence measurements
The virus-infected transgenic potato and control plants were used to determine the possible effects of PVY and PVA on photosynthetic parameters. Analyses of the various photosynthetic parameters were simultaneously performed at the second and third leaf counting from the top of the plants. Two leaves of each plant and three to nine plants per treatment were used to perform measurements. The gas exchange parameters were taken with an LI-6400 portable photosynthesis system (LI-COR Inc., Lincoln, N.E., USA), and analyses of the major fluorescence parameters (F 0 , F m , and F v ) of Chl were simultaneously conducted using a portable modulated fluorimeter (Plant Stress Meter, PSM Mark II, Biomonitor S.C.I AB, Umea, Sweden). Leaves were dark-adapted for 20 min before measuring of the fluorescence transient over 2 s and with an actinic stimulation at PPFD of 400 µmol photon m −2 s −1 . Maximal fluorescence (F m ') and minimal fluorescence (F 0ʹ ) were measured for light-adapted leaves (for 15 min). Variable fluorescence (F v ) was calculated from the difference between F o and F m in dark-adapted conditions and variable fluorescence (F v ') from the difference between Fo' and Fm' in light-adapted conditions. The non-photochemical quenching of variable ChlF (qN) was determined in accordance with the equation

Data analysis
Data filtering was accomplished with J-Express 2012 and Excel software programs to reduce the data. GoMapMan online tool (http://www.gomap man.org/) was used to match the gene functional annotations. Furthermore, the MapMan version 3.5.1R2 software (http://mapman.gabipd.org) was run to visualize the results and regulated pathways during virus-host interactions of potato gene annotations. Moreover, the Venn diagrams were created with Bioinformatics and Evolutionary Genomics online (http://bioinformatics.psb.ugent.be/webt ools/Venn/). All real-time PCR-MCA assays were performed using AriaMx Real-Time PCR System software. Prediction and analysis of transcription factor binding sites were conducted using the CiiiDER tool to predict potential transcription factor binding sites within endogenous StSAR1A and CaMV 35S promoter sequences. CiiiDER identifies those transcription factors that are significantly enriched in both promoters. Statistical analyses were conducted on the obtained dataset using the SAS software (version 8.2; SAS Institute, Cary, NC). Duncan's multiple range test 60 was used to identify significant differences (P < .05) in gene expression analysis among tested plants into homogenous subsamples of means that were not different from each other. One-way ANOVA was also used for statistical analyses of multiple groups and physiological parameters. In addition, the cluster heatmap (distance measure: Euclidean; clustering algorithm: Ward) was obtained through online web-based tools (https://www.metaboanalyst.ca/).

Identification of candidate genes involved in resistance to PVY and PVA
The expression analysis of 113 G-proteins in potato plants under virus stress conditions indicated that the total number of differentially expressed genes under PVY and PVA treatments was estimated to be 83 genes. These genes were mapped using the MapMan software to organize and display differentially expressed genes in biotic stress pathways (Figure 1(a)). The number of genes regulated in potato resistant cultivars was 32 and 18 for PVY and PVA, respectively. To recognize the G-proteins involved in viral resistance response (PVY and PVA), the shared genes were also found between them in potato resistant cultivars. Based on the results, the number of common differentially expressed genes in resistant cultivars was estimated to be two G-proteins, namely stmcq88 and stmhx51 (Figure 1(b)). Unlike stmhx51, the stmcq88 gene produced no antagonistic response to virus stresses and other biotic and abiotic stresses in this study (Figure 1(c)). Therefore, stmcq88, a G-protein (referred to as StSAR1A gene in this study), was selected to evaluate the effect of StSAR1A overexpression on PVY and PVA resistance.

Genetic transformation and molecular characterization of transgenic plants
To investigate the effect of StSAR1A overexpression on resistance to PVY and PVA, we generated the overexpressed transgenic potato plants (referred to as StSAR1A-OE in this study). Node cultivation of potato plants was propagated on the liquid MS medium. All untransformed internode pieces were regenerated in the control medium without antibiotics, while no regeneration occurred in the presence of kanamycin but without cefotaxime. The transformation efficiency was approximately 33%. Finally, 15 independent transformants were obtained for the StSAR1A gene after regenerating shoots and rooting transformed plantlets. Among them, ten regenerated plants were proven to have the StSAR1A gene by PCR. These transgenic potato plants were also found to be negative for the presence of the Agrobacterium confirmed using Vir B6 primers.
Southern blot using the Digoxigenin-labeled full-length cDNA sequence of StSAR1A as a probe confirmed that the StSAR1A gene was integrated into the potato genome (Fig. S1). Independent integration pattern of transgenes into the different transgenic potato genome caused a various size of the transgene on the blot. The number of integrated points into the transgenic potato genome ranged from one to five as compared to the negative control (nontransgenic potato plant). As Fig. S1 shows, two copies and one copy of the transgene StSAR1A were presented in lanes T 1 and T 2 (red arrows), respectively in the genome of potato. Many factors influence transgene expression, including the transgene copy number. Integration of multiple copies of the transgene into the plant genome has mostly been considered a disadvantage and strongly related to inactivation/silencing of the transgene. 61,62 Previous studies have also suggested for screening of single-copy transgene insertion events in potato plants. 63,64 In the present study, approximately 50% of the insertions appeared to be single copy. Therefore, five transgenic events with one transgene copy number were chosen for subsequent analysis. The pRI20-AN vector with target genes (StSAR1A) was examined as a positive control.

Assay of overexpression of StSAR1A transgene in transgenic plants
The expression of the StSAR1A transgene in nontransgenic and screened transgenic plants was analyzed using the real-time PCR. Analysis of melting curves for primer specificity showed good specificity with a single sharp peak. The results also demonstrated good primer efficiency for the internal control gene (95.1%), the StSAR1A gene (106.5%), and the coat protein gene of PVY (94.0%) and PVA (91.1%) (a slope -4.4, -2.9, -4.5, and -4.6, respectively).
No statistical differences were detected between the non-inoculated (Wt, R, and T) and mockinoculated (Wt/M, R/M, and T/M, respectively) plants in the expression level of the potato endogenous StSAR1A (Figure 2(a)). A low basal level of the endogenous StSAR1A was shown in potato Wt plants under non-stress conditions (Figure 2(a), green box), which can be used by plants in the routine activity. The ability of the wild-type plants to induce the endogenous StSAR1A upon viral infection was determined by comparing the gene expression levels of the endogenous StSAR1A in Wt/Y and Wt/A to those of Wt plants. Similar to Wt plants, there were extremely low expression levels of the endogenous StSAR1A gene in Wt/Y and Wt/A (2.2-and 1.9-fold, respectively) (Figure 2(a)). The results indicated that the endogenous StSAR1A in Wt/Y and Wt/A was slightly induced (2.2-and 1.9fold, respectively) upon viral infection (Figure 2(a), red boxes). The results in the present study showed that the expression levels of StSAR1A in T plants were not significantly different from those of R plants. Significant differences were detected between R/Y and R plants or R/A and R plants in the expression level of the endogenous StSAR1A gene (Figure 2(a), red boxes). The results also revealed that the expression level of StSAR1A was significantly higher in T/Y and T/A plants than in Wt/Y and Wt/A plants (4.2-and 3.9-fold, respectively). To assess the increased expression levels of the transgene StSAR1A under the control of the CaMV 35S promoter, the gene expression levels in T and Wt were compared to each other. The results showed that the fold induction of the transgene StSAR1A under the CaMV 35S promoter was significantly higher in T than in Wt (2.53-fold) ( Figure  2(a), yellow boxes). With the aim to evaluate the induction of the endogenous StSAR1A in transgenic potato plants after virus infection, the gene expression levels of the endogenous StSAR1A in T/Y and T/A were compared to those in T. Interestingly, the results indicated that the fold induction of the endogenous StSAR1A was significantly higher in T/Y and T/A than in T (1.62-and 1.07-fold, respectively (Figure 2(a), red boxes). Therefore, the induction of the endogenous StSAR1A after virus infection was higher in T/Y and T/A than in Wt/Y and Wt/ A, respectively (3.15-and 2.88-fold, respectively) ( Figure 2(a), red boxes). Analysis of transcription factor binding sites was also conducted to predict the potential transcription factor binding sites within endogenous StSAR1A and CaMV 35S promoter sequences. The results revealed that these promoter sequences shared ten different transcription factors (Figure 2(c)).

Assay of virus resistance in transgenic plants
To evaluate the effect of StSAR1A overexpression on resistance to PVY and PVA, we compared responses of transgenic plants against each virus and compared them to those of wild-type plants using the real-time PCR analysis. Overexpression of StSAR1A led to a reduction in the expression level of PVY and PVA CP genes up to 0.22-and 0.34-fold in T/Y and T/A plants relative to those in Wt/Y and Wt/ A, respectively (Figure 2(b)). These results suggested an inverse relationship between the expression level of the StSAR1A gene and the viral CP gene. Hence, the difference in the expression level of the viral CP gene in viral-infected transgenic plants was not statistically significant compared to resistant plants. The results of this comparison showed that resistance of T/Y and T/A had the least differences compared to those of R/Y and R/A plants, respectively.
Virus CP accumulation in the leaves of studied plants was also investigated by ELISA at 21 d.p.i (Figure 3). The ELISA results were consistent with

Assessment of morphological traits
We monitored plant growth under controlled conditions to assess phenotypic differences between transgenic and control potato plants. Although Wt/ Y and Wt/A plants displayed chlorosis and less growth, PVY showed the most severe symptoms compared to PVA in the wild-type control ( Figure  4). No apparent morphological differences existed between the non-inoculated (Wt, R, and T) and mock-inoculated (Wt/M, R/M, and T/M, respectively) plants in morphological traits (Table S1). Morphological responses of potato plants to virus stresses showed a significant decline in plant growthrelated traits of Wt/Y and Wt/A compared to healthy control plants (Wt) ( Table S1). The T/Y and T/A plants also showed phenotypic differences in plant growth traits compared to Wt/Y and Wt/A plants, which exhibited a quicker growth rate, a greater stem length and a larger diameter, a larger leaf size, a higher fresh/dry weight, and a greater node number. Similar data were obtained in R/Y and R/A plants than in Wt/Y and Wt/A plants, respectively (Table S1 and Figure 4). Moreover, T/Y and T/A plants showed a shorter internode length than Wt/Y and Wt/A plants did, respectively (Table S1 and Figure 4). No deleterious consequences of genetic transformation (e.g. pleiotropic or insertional effects) were observed on the plant morphology or growth in T/Y and T/A plants (Figure 4).

Assessment of photosynthesis and gas exchange variables
Mock-inoculated plants (Wt/M, R/M, and T/M, respectively) did not show any changes compared to non-inoculated plants (Wt, R, and T) in terms of gas-exchange and photosynthesis parameters (Table S1). PVY and PVA infection of wild-type plants (Wt/Y and Wt/A) had severe impacts on photosynthetic responses compared to healthy control plants (Wt) ( Table S1). Photosynthetic responses of StSAR1A-OE plants showed that various traits and parameters related to gas-exchange and photosynthesis responses were enhanced in both T/Y and T/A plants compared to Wt/Y and Wt/A plants (Table S1). The results revealed that transgenic inoculated plants exhibited reduced susceptibility to both PVY and PVA compared to wild-type plants. Like R/Y and R/A plants, T/ Y and T/A plants, compared to Wt/Y and Wt/A plants, showed higher rates of Fv/Fm, stomatal conductivity and transpiration, net photosynthetic rate, intercellular CO 2 concentration, and leaf temperature. On the contrary, for Wt/Y and Wt/A plants the opposite effect (i.e., higher F'v/ F'm and qN) was observed compared to other plants.

Physiological and morphological evaluation of the non-transgenic and transgenic potato responses to viruses by heatmap
The effect of virus infection on plant physiological and morphological traits and the overexpression of StSAR1A on the overall growth and development of transgenic plants were evaluated via heatmap hierarchical clustering ( Figure 5). Clustering analysis (top) showed three major groups: Group I includes Wt/Y and Wt/A, Group II includes Wt, Wt/M, T/Y, and T/A, while Group III includes R, R/M, R/Y, R/A, T, and T/M. The clustering analysis (left) of different parameters showed two main groups where the Group a represented four studied physiological and morphological parameters (Tleaf, F´v/F´m, internode length, and qN), while Group b represented the other physiological and morphological parameters ( Figure 5).

Discussion
Plants have the ability to integrate divergent signaling pathways to allow appropriate defense responses against various stresses. 65,66 Therefore, overexpression of a single gene can have a negative (i.e., susceptibility) or positive effect (i.e., tolerance) on the plant response to other stresses. In this study, the response evaluation of two selected genes; stmhx51 and stmcq88 to different biotic and abiotic stresses showed that unlike the virus stresses, the fungi, cold, and heat stresses caused a significant decrease in the expression level of the stmhx51 gene (synergistic action with virus stresses) in potato resistant cultivars against every stress. There is an exception in potato resistant cultivar against nematode attack, which induced a significant increase of transcript levels of stmhx51 (antagonistic action with virus stresses).
Regarding stmcq88, the results of the heatmap support the idea that there is a synergistic action between virus and different stresses. Since the virus response gene, stmcq88, is also regulated by fungi, nematodes and cold stresses, previous studies have suggested that many components of the regulatory networks were antagonistically involved in responses to biotic and abiotic stresses' function. These components can induce more or less susceptibility to a certain type of stress. 66,67 In this study, analysis of gene expression in response to different biotic and abiotic stresses revealed that the stmcq88 gene might not act antagonistically against the virus infection stresses. Moreover, the overexpression of the stmcq88 gene may also increase resistance to fungi, nematodes, and cold stresses at the same time.
The R/Y and R/A have shown high ability to induce the expression of endogenous StSAR1A to Wt/Y and Wt/A, respectively. This comparison confirmed the inability of Wt plants to adequately induce the internal gene after virus infection. The results in the present study demonstrated that similar to R/Y and R/A plants, the expression levels of StSAR1A in T/Y and T/A plants were significantly different from those of Wt/Y and Wt/A, respectively. Therefore, overexpression of the StSAR1A transgene not only increases the capability of transgenic potato plants in adequate induction of endogenous StSAR1A, but also confers the virus resistance to PVY and PVA in T/Y and T/A plants, respectively. A positive interaction between a strong and weak promoter has been documented wherein the weakened promoter and powerful and contiguous promoter (such as CaMV 35S) synergistically affect each other. It has also been shown that the activity of CaMV 35S and weak promoters positively would influence each other when the weak and strong contagious promoters share the same transcription factors. 68 The promoter analysis of CaMV 35S and StSAR1A gene showed that they shared some different transcription factors (Figure 2(c)). These observations suggest that the endogenous StSAR1A promoter might have been positively upregulated by the CaMV 35S promoter. Therefore, dramatic induction of overexpression StSAR1A in the transgenic potato after infection with virus (T/Y and T/A) relative to the infected wild type (Wt/Y and Wt/A, respectively) can be outcome of the positive interaction between a strong transgene promoter (CaMV 35S) and a weak endogenous promoter sharing the same transcription factors.
According to our results, the overexpression of the StSAR1A gene, as G-protein, significantly enhanced resistance to both PVY and PVA, suggesting the involvement of StSAR1A in antiviral defense potato plants. Considerable evidence indicates that plant endomembrane trafficking system is closely associated with stress-related signaling pathways to meet the requirements of rapid changes in cellular processes and to ensure the correct localization of stress-related cargo molecules. 69,70 However, the detailed molecular mechanisms are unknown. The function of SAR1A is critical for the assembly and organization of COPII, an important complex for protein transport from ER to Golgi. [45][46][47] The COPII machinery is associated with plant responses to biotic and abiotic stresses. 46 Our data suggest that SAR1A expression likely enhances virus stress resistance with the involvement of the COPII network, but this requires further study. Small G-proteins have also been proven to function in cell signaling events, 26 especially in the regulation of plant immunity, as a key molecular switch for signaling pathways and their related functions in plants. 24,71 Previous studies demonstrated the role of G-protein in defense against pathogens. Overexpression of the OsGAP1 gene, encoding a small G-protein in rice genome, in transgenic A. thaliana and rice led to enhanced resistance to the bacterial pathogens in both monocots and dicots. 33 Additionally, overexpression of StRab, a small GTP-binding protein gene from potato, significantly reduced the lesion area of the inoculated transgenic potato plants with mixture races of P. infestans. 36 Small G-proteins were also found to be involved in the interaction between plant resistance genes and pathogen avirulence genes, activation of the other G-proteins to initiate defense responses, accumulation of ROS, and expression of PR proteins. 72,73 Furthermore, the physical association of GTPase-activating proteins (GAPs) with NB-LRR resistance protein Rx has been shown to regulate against plant viruses. 37,38 However, the way a small G-protein StSAR1A is involved in antiviral defense is an interesting question for further research. The overexpression result of StSAR1A in transgenic plants was the reduced rate of viral accumulation. Previous findings demonstrated that the small G-protein was involved in activating immune responses by activating the ROS signaling. 42,74 Therefore, the small G-protein StSAR1A may enhance antiviral resistance possibly by regulating the ROS production in plants that could be explored in future studies. One of the earliest plant's defense responses during an incompatible host-pathogen interaction is the generation of hydrogen peroxide (H 2 O 2 ) and other ROS. 75 H 2 O 2 and ROS are important to prevent pathogen accumulation through antimicrobial activities. 42 Modification or overexpression of a single signaling pathway gene may cause growth repression and/or yield loss due to constitutive overexpression of a large number of genes at a time. 19 Virus infection causes a significant decline in plant growthrelated traits, leading to crop yield loss. In this study, it can be concluded that overexpression of StSAR1A gene promotes the plant vegetative growth than wild-type plants when infected with viruses. In the present study, the potato transgenic plants were slightly affected by virus infection and showed a greater plant height, a larger node number, and a shorter internode length than wild-typeinfected plants. Similar trends were observed in these characters between transgenic plants and resistant cultivars under virus infection. This is in agreement with previous reports in which virus infection drastically reduced the plant growth of susceptible cultivars, and ultimately resulted in crop losses compared to resistant cultivars. 76,77 However, economic injury to plants due to virus infection depends on some important factors such as virus strain, resistance of the infected cultivar, the plant age at the time of infection, and environmental conditions. 78 As demonstrated on the hierarchical clustering and heat map in Figure 5, Wt/Y and Wt/A were affected by the virus more than others were, and they fell in the separated clade. In addition, transgenic plants under stress conditions (T/Y or T/A) tended to cluster with wild-type under non-stress conditions (Wt). Therefore, transgenic potato plants can be considered PVY and PVA resistance owing to their better performance under stress conditions for all the studied traits. Moreover, under non-stress conditions, transgenic plants (T) tended to cluster with resistant plants, suggesting that transgenic and resistant plants have more or less similar response to virus according to the measured traits in Table S1. This result also indicated that overexpression of StSAR1A had no discernible effect on the growth and development of transgenic plants. Therefore, these plants can be considered for use in future breeding programs.
The heatmap of variables suggests that the leaf temperature, qN, Internode length, and F´v/F´m have been more affected in Wt upon virus infection (Wt/Y and Wt/A). On the contrary, these traits have been less affected by virus infection in resistant and transgene plants ( Figure 5). These traits have been demonstrated as important markers of the plant. The internode length, for example, was positively correlated with the plant performance, 79 and longer internodes help the plant to increase light harvest. 80 Potato performance is reduced by increasing the temperatures above 25°C. 81 Higher leave temperatures are also known to reduce potato canopy photosynthesis. 82 Furthermore, it has been shown that leave high temperature negatively affects the leaf area development in potato plants. 83 The effective photochemical efficiency in light-adapted state (FV'/FM') can be used to estimate the PSII operational efficiency. 84 Moreover, FV'/FM' were positively correlated with lower plant photosynthesis and total yield efficiency in potato crop. 85 The results demonstrated in Figure 5 indicated that over-performance of T/Y and T/A plants could improve the leaf temperature, qN, internode length, and F´v/F´m factors relative to Wt/Y and Wt/A.
Virus-infected sensitive plants often exhibit severe morphological and physiological changes, with modifications in the structure and function of the chloroplast. 4,5,86 Reduction in the photosynthetic rate in sensitive plants is commonly reduced by virus infection. 4,87,88 It has also been found that reduced plant growth rates in virus-infected plants can be associated with impaired photosynthetic apparatus. 4,86,87 This study revealed improvement in the photosynthetic rate, which may lead to increased growth and fitness in T/Y and T/A plant than in Wt/Y and Wt/A plants, respectively. In line with this result, it has been reported that overexpression of small G-proteins can regulate stress resistance, promote plant growth, and increase yield production. 89,90 Our results suggest that the photosynthetic activity of infected plants was more affected in PVY than in PVA. Consistent observations were reported previously. 5 The findings of this research are also consistent with previous studies revealing that overexpression of AtSAR1A did not affect the plant growth negatively. 91,92 Identification and understanding of the host cell factors would increase the knowledge of molecular mechanisms underlying the manipulation by viruses and subsequently may facilitate the development of a defense strategy against viruses. Identification of defense-related genes induced against virus infection would be useful to breed crop plants for antivirus resistance. Overall, our results revealed that StSAR1A was involved in defense responses against PVY and PVA in potato plants. Furthermore, it was found that we could confer enhanced antiviral resistance on potato plants by overexpression of the StSAR1A gene. This study demonstrates that our systems biology approach can be successfully applied for selection genes to confer beneficial traits and enhance resistance against various abiotic and biotic stresses like antiviral resistance on potato plants.

Conclusion
In this study, we presented a strategy to select effective candidate genes and to examine their efficiency in practice. Screening of a list of G-proteins using microarray meta-analysis based on the systems biology approach suggested one candidate gene (StSAR1A). Virus-inoculation tests with transgenic plants revealed that constitutive expression of StSAR1A enhanced the resistance of transgenic potato plants against PVY and PVA. The experimental data were in agreement with meta-analysis regarding the enrollment of StSAR1A in the virus resistance process. In addition, morphological investigations revealed that StSAR1A-OE plants promoted plant growth and photosynthetic responses and did not show any negative impacts on the growth of transgenic plants. Since the StSAR1A gene was significantly up-regulated in response to PVY and PVA infection in all microarray experiments, StSAR1A may be one of the key regulatory genes during various plant-virus interactions, including PVY and PVA infection. Furthermore, the virus-infected transgenic plants exhibited a promoted growth rate, indicating the functionality of StSAR1A in potato growth. Morphological and physiological assessments and expression analysis of StSAR1A genes together demonstrated that StSAR1A worth to be implemented in the breeding programs not only because of it simple genetic regulation but also because it reacts to the virus-like resistant lines. Moreover, not only similarity of morphological and physiological responses between the resistant plants and transgenic plants, but also results of the expression analysis of StSAR1A gene ( Figure 2) suggest the same regulatory and subsequently physiological programing during confrontation with the viruses. This issue leads to higher plant performance relative to Wt plants, demonstrating that selection of StSAR1A based on data analysis is an effective strategy to increase plant performance against viruses. Hence, StSAR1A would be a proper candidate for breeding potato plants highly resistant to viruses, perhaps with additional beneficial traits. Additionally, it is speculated that StSAR1A is important to modulate plant resistance to fungi, nematodes, and cold stresses as demonstrated by the meta-analysis method. Considering the involvement of small G-proteins in diverse biotic and abiotic stress responses, it is worth examining whether StSAR1A-OE potato plants have enhanced resistance or tolerance to multiple biotic and abiotic stresses. Further investigation of the physical interaction of the StSAR1A gene with downstream stressresponsive and growth and developmental-related genes is also necessary at the protein level.

Acknowledgments
We would like to thank Dr Tatsuji Hataya (Hokkaido University, Japan) to kindly provide us the monoclonal antibodies of PVA and Dr Tetsuya Yamada (Hokkaido University, Japan) for useful comments to construct binary vectors. We also thank the genetic resources center in Japan, NARO for providing potato virus A. This study was financially supported by Tarbiat Modares Univ., Tehran, IRN (No. 9230042002 to M.S. Sabet) and Japan Society for the Promotion of Science KAKENHI (No. 25450055 and 16H04879 to K.S. Nakahara).

Key Message
To explore additional genetic resources to confer antiviral resistance on potato, we developed a screening and evaluation system using microarray meta-analysis and transgenic plants and exemplified it with small G-proteins.

Disclosure of Potential Conflicts of Interest
The authors declare that they have no conflict of interest.

Author's Contribution
Zhila Osmani, performed all the experiments and data analysis and wrote the article; Mohammad Sadegh Sabet and Kenji S. Nakahara, supervised the project and provided editorial input into the writing; Ali Mokhtassi-Bidgoli, supported and verified the analysis of morphological traits and photosynthesis-related parameters; Ahmad Moieni, advised the tissue culture of potato plants part, Kenji S. Nakahara and Masoud Shams-Bakhsh, advised the plant-virus interaction part; Khabat Vahabi, conceived the microarray meta-analysis research plan. All authors discussed the results and contributed to the final manuscript.