Role of the methionine cycle in the temperature‐sensitive responses of potato plants to potato virus Y

Abstract Plant–virus interactions are greatly influenced by environmental factors such as temperatures. In virus‐infected plants, enhanced temperature is frequently associated with more severe symptoms and higher virus content. However, the mechanisms involved in such regulatory effects remain largely uncharacterized. To provide more insight into the mechanisms whereby temperature regulates plant–virus interactions, we analysed changes in the proteome of potato cv. Chicago plants infected with potato virus Y (PVY) at normal (22 °C) and elevated temperature (28 °C), which is known to significantly increase plant susceptibility to the virus. One of the most intriguing findings is that the main enzymes of the methionine cycle (MTC) were down‐regulated at the higher but not at normal temperatures. With good agreement, we found that higher temperature conditions triggered consistent and concerted changes in the level of MTC metabolites, suggesting that the enhanced susceptibility of potato plants to PVY at 28 °C may at least be partially orchestrated by the down‐regulation of MTC enzymes and concomitant cycle perturbation. In line with this, foliar treatment of these plants with methionine restored accumulation of MTC metabolites and subverted the susceptibility to PVY at elevated temperature. These data are discussed in the context of the major function of the MTC in transmethylation processes.

that elevated temperatures may suppress a range of these antiviral responses, rendering plants more susceptible to virus attack. Such a phenomenon has been observed in incompatible plant-virus interactions such as R (resistance) gene-mediated defence responses (Zhu et al., 2010). For example, the resistance mediated in tobacco by the N gene (a resistance [R] gene encoding a nucleotidebinding site leucine-rich repeat [NBS-LRR] domain-containing protein) against tobacco mosaic virus (TMV) confers defence only at temperatures below 28 °C. Under these conditions, TMV triggers a hypersensitive response (HR), which is defined as a rapid necrosis at the site of virus entry that prevents further spread of the pathogen to cells surrounding the initial site of infection. At higher temperatures, resistance does not develop and TMV spreads systemically throughout the plant (Zhu et al., 2010). The observed deregulation of defence responses in a temperature-dependent manner occurs presumably due to temperature-sensitive conformational loss of function of the tobacco NBS-LRR protein, preventing its interaction with the TMV p50 effector protein (Zhu et al., 2010).
A variety of R genes conferring HR to PVY have been identified in potato species (Solanum spp.) (Solomon-Blackburn and Bradshaw, 2007). Many of these genes, including Ny in S. sparsipilum and S. sucrense, or Ny-1 in potato cv. Rywal, confer resistance only at low temperatures (16-20 °C). At higher temperatures (24-28 °C), resistance does not take place, and PVY infects plants systemically. In contrast, resistance genes Ry sto in S. stoloniferum and Ry chc in S. chacoense, which confer extreme resistance (inhibit virus replication without apparent HR) to the tobacco veinal necrosis strain of PVY (PVY N ), are functional at both low (16-20 °C) and elevated (above 24 °C) temperatures (Bradshaw and Ramsay, 2005;Solomon-Blackburn and Bradshaw, 2007).
Compatible virus infections, which are characterized by efficient systemic virus spread from the initially infected tissues, are also affected by higher temperatures (Anfoka et al., 2016;Obrępalska-Stęplowska et al., 2015;Prasch and Sonnewald, 2013). For instance, elevated temperatures significantly increase the susceptibility of Arabidopsis plants to turnip mosaic virus (Prasch and Sonnewald, 2013). Likewise, tomato plants subjected to higher temperatures were more susceptible to tomato yellow leaf curl virus (Anfoka et al., 2016).
With regard to PVY, we have recently shown that susceptibility of potato cv. Chicago plants to systemic PVY infection (virus accumulation and symptom production in systemically infected leaves) is significantly enhanced by elevated temperature compared with normal conditions (Makarova et al., 2018). Interestingly, such an enhanced susceptibility was correlated with reduced expression of genes encoding pathogenesis-related (PR) proteins, which are hallmarks of salicylic acid (SA)-mediated plant defence responses. We have also demonstrated that SA pretreatment subverts enhanced susceptibility to PVY in cv. Chicago at higher temperature (Makarova et al., 2018), implicating SA as a key regulator of mechanisms determining susceptibility/resistance in potato (Baebler et al., 2011;Carr et al., 2018;Kogovšek and Ravnikar, 2013;Love et al., 2005;Vlot et al., 2009).
RNA interference (RNAi), which is also called RNA silencing, is a versatile, evolutionarily conserved and sequence-specific mechanism for controlling endogenous gene expression and degrading foreign nucleic acids. RNAi-based defence responses involve processing of virus-derived double-stranded RNAs (dsRNAs) into small interfering RNAs (siRNAs), which in a complex with some plant proteins trigger the sequence-specific inactivation/degradation of viral RNAs (Baulcombe, 2005;Ding, 2010;Guo et al., 2019;Mlotshwa et al., 2008;Yang and Li, 2018). Remarkably, these RNAi-mediated mechanisms may also be controlled by temperature.
Interestingly, antiviral defence mechanisms may also be regulated via the interplay of viruses with the plant methylation cycle (MTC; Mäkinen and De, 2019). In the MTC, S-adenosyl methionine synthase (SAMS) converts methionine (MET) and adenosine triphosphate to S-adenosyl methionine (SAM), which is a universal methyl donor for numerous methylation reactions. As a result of transferring its methyl groups to target molecules, SAM becomes S-adenosyl homocysteine (SAH), which is then converted to homocysteine (HCY) by SAH hydrolase (SAHH). MET synthase (MS) completes the cycle by converting HCY back to MET with 5-methyltetrahydrofolate (5-MTHF) as methyl group donor (reviewed by Mäkinen and De, 2019). 5-MTHF is a product of the MTC-coupled folate cycle synthesized from 5,10-MTHF by methylenetetrahydrofolate reductase (MTHFR). In turn, 5,10-MTHF is converted from tetrahydrofolate by serine hydroxymethyltransferase (SHM).
The MTC plays a central role in various biological processes such as metabolism, signal transduction, and gene expression. With regard to virus-plant interactions, the MTC is closely linked to RNAi pathways, in which siRNAs are stabilized by the MTC-based methylation process (Li et al., 2005). Plant host DNA methylation, as an epigenetic mechanism driven by the MTC, has also been suggested as playing an important role in modulating host responses to viruses by modifying functions of host genes and affecting gene expression (Corrêa et al., 2020;Wang et al., 2019). Another factor released as a product of the MTC-related pathway is ethylene, a plant hormone that, like other hormones, plays important roles in plant responses to biotic and abiotic stress responses (reviewed in Müller and Munné-Bosch, 2015).
To provide more insight into the mechanisms of how temperature regulates plant antiviral defence responses, this study used an isobaric tag for relative and absolute quantitation (iTRAQ) labelling to comprehensively analyse changes in the proteome of potato plants (cv. Chicago) infected with PVY at normal (22 °C) and elevated temperature (28 °C) conditions. We showed that at 14 days postinoculation (dpi) infection at 28 °C (which significantly facilitates susceptibility to the virus) induced greater changes in the abundance of proteins (152 differentially expressed proteins, DEPs) than at 22 °C (23 DEPs). Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses were applied to characterize the protein functions and the significant pathways associated with the DEPs, and protein-protein interaction networks were constructed using STRING. It was not surprising that biotic and abiotic stress responsive proteins were among the groups up-regulated during PVY infection at both normal and higher temperature.
Most strikingly, the main MTC enzymes such as MS, SAMS, and SAHH, as well as enzymes of the MTC-coupled folate cycle, namely SHM and MTHFR, were found to be down-regulated in potato plants infected with PVY (at 8 dpi), but only at the higher and not at normal temperatures. In good agreement, we found that these conditions (PVY + higher temperature) triggered consistent and

| Impact of elevated temperature on the susceptibility of potato plants to PVY
In previous work, we studied the effect of moderately elevated temperature (28 °C) on the dynamics of PVY infection in potato plants (Makarova et al., 2018). This temperature was chosen to mimic the conditions that may arise during mild heat stress under global climate changes. In inoculated leaves of cv. Chicago, PVY was detected by 3 dpi in inoculated leaves at relatively low levels, which did not increase significantly over time and were not affected by temperature ( Figure 2b in Makarova et al., 2018). Starting at approximately 5-8 dpi, PVY spread systemically, invading upper leaves at both normal (22 °C) and elevated (28 °C) temperatures. With time, a significant increase in virus levels was observed in the systemically infected leaves of plants grown at 22 °C (up to seven-fold); however, virus levels were found to be significantly higher in corresponding tissues of plants grown at 28 °C (Makarova et al., 2018), suggesting that elevated temperature greatly enhances susceptibility of cv.

| Protein profiles of PVY-infected potato plants at normal and elevated temperature
To explore the underlying mechanisms that lead to increased susceptibility of potato to PVY at elevated temperature compared with normal conditions, iTRAQ-based quantitative comparative proteomic analysis was conducted at 8 and 14 dpi (Figure 2a).
A total of 22,615 peptides were identified with a 1% false discovery rate (FDR) (Table S2), which were assigned to 5,756 proteins from a custom database (Methods S1). The numbers of peptides and corresponding protein groups identified by liquid chromatography tandem mass spectrometry (LC-MS/MS) analysis for both control mock-inoculated and plants infected with PVY at F I G U R E 1 Accumulation of PVY RNA (measured using quantitative reverse transcription PCR) in systemically infected leaves of potato plants over 3-21 days postinoculation (dpi) time periods at 22 or 28 °C as shown. Analysis of variance and Tukey's HSD post hoc tests were performed for quantitative reverse transcription PCR data. ***p < .001 22 or 28 °C are indicated in Table 1. Differential protein screening (determined by the ratio in the PVY-infected samples and corresponding untreated control (mock-inoculated plants at 22 °C) identified 642 DEPs.
In PVY-infected plants grown at normal temperature, we identified only 16 DEP groups including five up-regulated and 11 down-regulated proteins at 8 dpi and 23 DEP protgroups (18 up-regulated and five down-regulated) at 14 dpi (Table S3). This suggests F I G U R E 2 (a) The strategy for comparative quantitative analysis of protein expression in potato leaves by 8-plexisobaric tags for relative and absolute quantitation (iTRAQ). For detailed iTRAQ data see Tables S2 and S3  Abbreviation: dpi, days postinoculation.

TA B L E 1
Numbers of peptides and corresponding protein groups identified by liquid chromatography tandem mass spectrometry (LC-MS/MS) that at this temperature the response of the plants to PVY was relatively moderate. At 8 dpi, among the population of up-regulated DEPs, we found biotic and abiotic stress-responsive proteins such as superoxide dismutase (SOD), which plays a vital role in protecting plants from reactive oxygen species (ROS)-mediated injury (Younus, 2018), and fibrillin 8, a plastid-associated lipid-binding protein induced by abscisic acid (ABA) (Yang et al., 2006), whereas the down-regulated DEPs included ribosomal proteins, glutamine cyclotransferase, and monodehydroascorbate reductase (Table S3). Both glutamine cyclotransferase and monodehydroascorbate reductase have been shown to mitigate damage in plants under abiotic stress (Johnston et al., 2015;Paulose et al., 2013).

At 14 dpi in virus-infected plants, we identified 18 up-regulated
DEPs, which included a group of ribosomal proteins and also α-tubulin, the latter of which constitutes part of the cytoskeleton (Table S3).
This is consistent with previous findings that have implicated cytoskeletal components in intra-and intercellular trafficking of many viruses (Pitzalis and Heinlein, 2017). In addition, we also observed that SOD was down-regulated in infected plants, which presumably means reduced protection against ROS damage (Younus, 2018) at this stage of infection.
At elevated temperature, the proteome changes were more pronounced. In virus-infected plants at 8 dpi, we identified 64 DEP groups, 15 of which were up-regulated and 49 down-regulated.
We then used the STRING database to create functional protein association networks of DEPs down-regulated at elevated temperature.
Based on this analysis, two distinct clusters were identified: ribosomal proteins and proteins involved in the MTC (Figure 2b). In the MTC cluster, the main enzymes of the MTC were identified: MS, SAMS, and SAHH. Moreover, this cluster included some MTC-associated proteins such as those involved in the folate cycle: SHM and MTHFR. were enhanced (Table S4). We also identified 73 down-regulated DEPs.
Among the top 10 down-regulated DEPs, chloroplast proteins and histone H2A were observed. According to GO term analysis, down-reg- The chaperone DnaJ (PGSC0003DMP400040462) was down-regulated at 8 dpi but up-regulated at 14 dpi.
One of the most striking findings of the proteomic analysis is therefore that all key enzymes of the MTC and MTC-related folate cycle, including MS, SAMS, SAHH, SHM, and MTHFR, were down-regulated at the protein level by PVY infection at the higher temperature ( Figure 2d).

| RNA expression levels of key MTC-related genes
To examine whether the MTC-related proteomic changes identified above were due to transcriptional regulation, we examined how In contrast, at normal temperature (22 °C) PVY did not affect transcription of SAMS, SAHH, SHM, and MTHFR, but moderately enhanced MS transcript accumulation (at 7 dpi). Thus, under these conditions, there is a disparity between MS gene transcription and translation levels (compare Table S3 and Figure 3). In fact, inconsistency between proteomic and transcriptomic data have

| Methionine subverts high susceptibility to PVY at elevated temperature in potato plants
To Collectively, these data confirm the role of the MTC in modulating PVY susceptibility in potato plants at the higher temperature.

| D ISCUSS I ON
The warming global climate is leading to rising temperatures, which may modulate plant-virus interactions, potentially further incurring significant decreases in crop yield and quality (Pandey et al., 2017). It is therefore crucial to study the effect of elevated temperatures on plant responses to virus infections.
We have recently shown that susceptibility to PVY dramatically increases in systemically infected leaves of potato at higher temperatures (Makarova et al., 2018). Similar effects have also been observed for many other viruses (e.g., Anfoka et al., 2016;Prasch and Sonnewald, 2013).
Several reports have demonstrated successful approaches using proteomics to elucidate virus resistance mechanisms in plants reductase, which are associated with glutathione metabolism (Hasanuzzaman et al., 2019). Glutathione is known to be an antioxidant and may play an important role in plant defence signalling against biotic and abiotic stresses (Hasanuzzaman et al., 2019). At 14 dpi, the overall degree of proteomic modulation in response to PVY infection was at similar magnitude to that observed at 8 dpi, but there were some differences in the levels of some proteins.
For example, at 14 dpi, SOD was down-regulated and ribosomal proteins were up-regulated in comparison to the levels observed at 8 dpi, but the biological relevance for their up-regulation remains unclear.
At elevated temperature, the proteome changes were much more pronounced. Proteins in the up-regulated group were mainly related to stress responses, such as calreticulin, salt tolerance protein 4, acidic endochitinase, patatin 3, glutathione S-transferase, glyoxysomal fatty acid β-oxidation multifunctional protein, and small heat-shock proteins (Table S3) Therefore, our future efforts will be focused on the elucidation of the regulatory role of specific small RNA and DNA methylation activities in the mechanisms of plant-virus resistance/susceptibility.
Protein/histone methylation may also be involved in the response of potato plants to PVY at higher temperatures ( Figure 7b).
An additional factor that may contribute to the increased susceptibility of potato plants to PVY at higher temperatures is a potential change in ethylene production, which is also associated with the MTC. Ethylene, like other phytohormones, plays an important role in plant responses to various pathogens, including viruses. Moreover, it is widely accepted that ethylene may play a significant role in triggering different types of acquired resistance (Alazem and Lin, 2015;Van Loon et al., 2006). Thus, greater susceptibility of potato plants to PVY at the higher temperature may be determined, for example, by the lack of the MTC metabolites needed for ethylene production.
Our data show that neither PVY infection at normal temperature (on its own) nor elevated temperature alone induce alterations in the expression of MTC-related genes. Only combined stress caused by PVY and the higher temperature show a significantly different expression pattern for these genes (down-regulation) and consequent relevant changes in the levels of MTC metabolites. These data are in good agreement with many other reports showing that a quite distinct gene expression programme may be established in response to combined stress as a result of integration of individual stress-responsive pathways (Makarova et al., 2018;Prasch and Sonnewald, 2013;Rizhsky et al., 2004). Thus, we suggest that responses to combined heat stress and PVY infection in potato cv. Chicago are integrated and reprogrammed in a way to specifically affect the MTC in a manner which increases plant sensitivity to PVY.
We have also previously shown that SA, another essential component of regulatory defence signalling pathways, affects susceptibility to PVY in cv. Chicago plants maintained at the higher temperature (Makarova et al., 2018). The molecular mechanisms underlying the functional links between MTC and SA-dependent responses remain unknown. A possibility is that both MTC and SA are interconnected with regulation of RNAi (Alamillo et al., 2006;Canizares et al., 2013;Ding et al., 2001;González et al., 2012;Ivanov et al., 2016;Jovel et al., 2011;Lee et al., 2016;Li et al., 2005;Mäkinen and De, 2019). Some other findings point to intimate interplay between SA-mediated defence and signalling pathways directed by ethylene

| E XPERIMENTAL PROCEDURE S
Additional details can be found in the Supporting Information Methods S1.

| Protein extraction and trypsin digestion
Proteins were extracted using the phenol extraction method (Faurobert et al., 2007), quantified by Bradford protein assays (Bio-Rad), and re-

| Bioinformatic analysis
Protein-protein interaction networks were constructed using STRING v. 10 (www.strin g-db.org; Szklarczyk et al., 2015). The strength of protein interactions was set at 0.4, the default option.
Cytoscape software (Sannon et al., 2003) was applied to visualize the protein interaction networks. The functions and pathway enrichment of candidate DEGs were analysed using g:Profiler (Raudvere et al., 2019).
However, for simplicity, in the main body of the manuscript the "St" nomenclature is not used for genes, proteins, or mRNAs. Total RNA was isolated as described previously (Makarova et al., 2018).
Aliquots of DNase-treated RNA were reverse transcribed into cDNA using the SuperScript First-Strand Synthesis System for RT-PCR (Invitrogen), in conjunction with either an oligo-dT primer (for host plant-specific mRNAs) or a PVY-specific primer (see Table S1).
The primer pairs for SYBR Green-based real-time PCR analysis of PVY RNA and host mRNAs were designed using Plant Genomics Resource Phytozome 12 (https://phyto zome.jgi.doe.gov/pz/portal. html) and PRIMER EXPRESS software, and are listed in Table S1. The C t values for PVY RNA and each mRNA of interest were normalized using two internal reference genes encoding cytochrome c oxidase subunit 1 (StCOX; Baebler et al., 2011) and StEF-1α (Nicot et al., 2005); primers are listed in Table S1. The average of the C t values of the two reference genes was used to analyse PVY and host mRNA levels. More detailed information on the RT-qPCR procedure is provided in Methods S1.

| Analysis of MTC-related metabolites
To

| Statistics
Statistical analysis was performed across four biological repeats.
Statistical analyses and bar plots were made in Python v. 3.7.5 (G. van Rossum, Python tutorial, Technical Report CS-R9526, Centrum voor Wiskunde en Informatica (CWI, Amsterdam, May 1995). For two-or more-way analysis of variance (ANOVA), Tukey's honestly significant difference (HSD) tests based on multiple comparisons of means were applied to determine which pairwise comparisons were statistically significant. Differences were considered to be significant at p ˂ .05.

| Foliar (exogenous) application of MET
A solution of 1.5 mM l-methionine (Merck) or water (as control) was sprayed every second day onto the leaves of mock-and PVYinfected plants maintained at 22 or 28 °C. PVY RNA accumulation and content of MTC-related metabolites in these plants were quantified as described above.

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The mass spectrometry proteomics data have been deposited to the

S U PP O RTI N G I N FO R M ATI O N
Additional Supporting Information may be found online in the Supporting Information section.