Distinct branches of the N-end rule pathway modulate the plant immune response

(cid:1) The N-end rule pathway is a highly conserved constituent of the ubiquitin proteasome system, yet little is known about its biological roles. (cid:1) Here we explored the role of the N-end rule pathway in the plant immune response. We investigated the genetic inﬂuences of components of the pathway and known protein substrates on physiological, biochemical and metabolic responses to pathogen infection. (cid:1) We show that the glutamine (Gln) deamidation and cysteine (Cys) oxidation branches are both components of the plant immune system, through the E3 ligase PROTEOLYSIS (PRT)6. In Arabidopsis thaliana Gln-speciﬁc amino-terminal (Nt)-amidase (NTAQ1) controls the expression of speciﬁc defence-response genes, activates the synthesis pathway for the phytoalexin camalexin and inﬂuences basal resistance to the hemibiotroph pathogen Pseudomonas syringae pv tomato ( Pst ). The Nt-Cys ETHYLENE RESPONSE FACTOR VII transcription factor substrates enhance pathogen-induced stomatal closure. Transgenic barley with reduced HvPRT6 expression showed enhanced resistance to Ps. japonica and Blumeria graminis f. sp. hordei , indicating a conserved role of the pathway. (cid:1) We propose that that separate branches of the N-end rule pathway act as distinct components of the plant immune response in ﬂowering plants.


Introduction
The regulation of protein stability through the ubiquitin proteasome system (UPS) is a central component of cellular homeostasis, environment interactions and developmental programmes (Varshavsky, 2012), and an important component of the plant immune system (Zhou & Zeng, 2017). Plants have evolved to recognize the presence of a pathogen in two main ways. Basal (primary) defence is characterised by the recognition of pathogen elicitors called pathogen associated molecular patterns (PAMPs) by protein receptors known as pattern recognition receptors (PRR), activating PAMP-triggered immunity (PTI) (Boller & Felix, 2009). When this response is effective, pathogens can deliver effector molecules into the host cells to weaken PTI and facilitate infection triggering a second layer of defence (effector triggered immunity; ETI). ETI is typically a qualitative response based on interference with pathogen effector activity by plant resistance (R) gene products, localized inside the cell (Dangl & Jones, 2001). Both PTI and ETI induce similar immune responses but of different amplitude, with ETI often resulting in a hypersensitive response (HR). The specific set of mechanisms activated also depend to a large extent on the life strategy of the pathogen and how adapted they are to the host. Typically, the plant hormones jasmonic acid (JA) and ethylene (ET) mediate responses to nonadapted necrotrophs that cause host cell death to acquire nutrients from dead or senescent tissues (Grant & Jones, 2009;Pieterse et al., 2009) whilst salicylic acid (SA) plays a crucial role in activating defence against adapted biotrophs and hemibiotrophs. Recently, regulation of protein stability by the Arg/N-end rule pathway of ubiquitin-mediated proteolysis has been demonstrated to play a role in plant responses to biotic stress. The pathway is associated with increased development of clubroot caused by the obligate biotroph Plasmodiophora brassicae (Gravot et al., 2016). Induction of components of the hypoxia response, controlled by Group VII ETHYLENE RESPONSE FACTOR (ERFVII) transcription factor substrates (ERFVIIs), enhanced clubroot development, indicating that the protist hijacks the N-end rule ERFVII regulation system to enhance infection. In another study, inactivation of different components of the Arg/N-end rule pathway was shown to result in greater susceptibility of Arabidopsis to necrotrophic pathogens and altered timing and amplitude of response to the hemibiotroph Pseudomonas syringae pathovar tomato (Pst) AvrRpm1 (de Marchi et al., 2016). A correlation between Nt-Acetylation and the stability of a Nod-like receptor, Suppressor of NPR1, Constitutive 1 (SNC1) was also reported (Xu et al., 2015). Whilst these reports provide evidence that the N-end rule pathway is involved in the regulation of plant defence responses, the mechanisms, substrates or their function in resistance have not been investigated previously (Gibbs et al., 2014a). The N-end rule pathway of ubiquitin-mediated proteolysis is an ancient and conserved branch of the UPS (Gibbs et al., 2014a). This pathway relates the half-life of substrates to the amino-terminal (Nt-) residue, which forms part of an N-degron (Gibbs et al., 2014a). Destabilizing residues of the Arg/N-end rule are produced following endo-peptidase cleavage and may be primary, secondary or tertiary (Fig. 1a). Basic and hydrophobic primary destabilizing residues are recognized directly by N-recognin E3 ligases, in plants represented by two proteins, PROTEOLYSIS(PRT)6 and PRT1 (Gibbs et al., 2014a). Secondary destabilizing residues (Glu, Asp and oxidized Cys) can be N-terminally arginylated by arginyl-transferases (ATEs), and tertiary destabilizing residues (Gln, Asn and Cys) can undergo modifications to form secondary destabilizing residues (Gibbs et al., 2014a). Oxidation of Cys was shown in vitro to occur both nonenzymically (Hu et al., 2005) or enzymatically (Weits et al., 2014;White et al., 2017), whereas in higher eukaryotes deamidations of Gln and Asn are carried out by residue-specific N-terminal amidases (NTAQ1 (Wang et al., 2009) and NTAN1 (Grigoryev et al., 1996), respectively). This hierarchical structure is conserved in eukaryotes, and physiological substrates with N-terminal residues representing these destabilizing classes have been identified (Piatkov et al., 2014). The Usp1 deubiquitylase is targeted for degradation through the deamidation branch of the Arg/N-end rule via NTAQ1 as a consequence of auto-cleavage, that reveals N-terminal Gln (Piatkov et al., 2012). Proteins with similarities to mouse NTAN1 and NTAQ1 are encoded in higher plant genomes, in Arabidopsis by AT2G44420 (putative NTAN1) and AT2G41760 (putative NTAQ1). Expression of these in a deamidation deficient nta1 mutant of Saccharomyces cerevisiae could functionally restore degradation of the N-end rule reporters Asn-b-galactosidase (b-Gal) and Gln-b-Gal, respectively. ATE activity was required for this destabilization in yeast (Graciet et al., 2010). Although the Arg/N-end rule pathway is evolutionarily highly conserved in eukaryotes, few substrates or functions for different branches have been shown. In plants the Cys branch of the Arg/N-end rule pathway controls homeostatic response to hypoxia (low oxygen) and NO sensing through the Met-Cys initiating ERFVII transcription factor substrates (Gibbs et al., 2011(Gibbs et al., , 2014bLicausi et al., 2011).
In this paper, we investigated the role of distinct branches of the Arg/N-end rule pathway in the immune response in Arabidopsis and barley (Hordeum vulgare). We demonstrate that two branches of the pathway, Glu-deamidation and Cys-oxidation, regulate resistance to the hemibiotroph Pst and the biotroph Blumeria graminis f. sp. hordei (Bgh). We also show a significant role for Nt-Gln amidase NTAQ1 in the regulation of molecular components associated with basal responses to infection, and a role for both NTAQ1 and the known Nt-Cys ERFVII substrates in resistance related to stomatal function.

Construction of transgenic Arabidopsis lines ectopically expressing NTAQ1
To generate Arabidopsis NTAQ1 overexpressing lines, fulllength cDNA sequence (with and without the STOP codon) was amplified from 7-d-old seedling cDNA and recombined into pDONR221. The constructs were mobilized into pH7m34G and pH7m24GW2, with the GSrhino tag in C-terminal or Nterminal position of the NTAQ1, respectively (Karimi et al., 2007). Then the constructs were transformed into Agrobacterium tumefaciens (strain GV3101 pMP90) and Arabidopsis ntaq1-3 using standard protocols (Clough & Bent, 1998).

In vitro assay for NTAQ1 activity
The Arabidopsis NTAQ1 coding sequence was cloned from cDNA and flanked by an N-terminal tobacco etch virus (TEV) protease recognition sequence (ENLYFQ-X) using primers ss_ntaq1_tev and as_ntaq1_gw, followed by a second PCR with as_ntaq1_gw and adapter tev attaching a Gateway attB1 site for sub-cloning into pDONR201 (Invitrogen). An LR reaction into pVP16 (Thao et al., 2005) leads to an N-terminal 8xHis:MBP double affinity tag. An assay for NTAQ activity was performed as described previously (Wang et al., 2009)

Analysis of pathogen growth in plant material
The bacterial suspension was injected with a needleless syringe into the abaxial side of leaves or sprayed on the surface of the leaves of 3.5-wk-old plants. Pst DC3000 avrRpm1, Pst DC3000 and Pst DC3000 hrpA À were grown overnight at 28°C in Petri dishes on King's B medium. For analysis of bacterial growth, three leaves per plant of at least seven plants were injected with a bacterial suspension of 10 6 CFU ml À1 (OD 600 nm 0.1 = 10 8 CFU ml À1 ) or sprayed with a suspension of 10 8 CFU ml À1 . A disc of 0.28 cm 2 from each infected leaf was

Ó 2018 The Authors
New Phytologist Ó 2018 New Phytologist Trust New Phytologist (2018) www.newphytologist.com excised at 96 h, pooled in triplicate, homogenized, diluted and plated for counting. Inoculation of Botrytis cinerea was performed by pipetting a drop of 10 ll of a suspension of 5 9 10 5 spores ml À1 onto the surface of the leaves. The response was analyzed by measuring the diameter of the symptoms produced in three leaves of at least 20 independent plants.
Barley plants were infected with Fusarium spp. and Blumeria graminis f. sp. hordei as previously described (Ajigboye et al., 2016). Leaf material of 25-d-old barley plants grown under controlled conditions (20°C:15°C; 16-h photoperiod; 80% RH, 500 lmol m À2 s À1 metal halide lamps (HQI) and supplemented with tungsten bulbs) were syringe infiltrated with 0.1 OD Ps. pv japonica obtained from the National Collection of Plant Pathogenic Bacteria (NCPPB), UK. Leaf material was collected before treatment and 4 d after inoculation for conductivity assays and RNA extraction. Production of H 2 O 2 was visualized by staining with 3,3 0 -diaminobenzidine tetrachloride as described (Thordal-Christensen et al., 1997;Moreno et al., 2005).

Stomatal aperture analyses
For stomatal aperture in response to Pst assays, leaves from 3.5wk-old plants were used. In the morning after 2 h the lights were switched on and peels from the abaxial side of the leaves were placed in Petri dishes containing 10 mM MES/KOH pH 6.1, 50 mM KCl and 0.1 mM CaCl 2 for 2 h in continuous light. Then the buffer was replaced with a solution of Pst DC3000 (OD 0.2: 2 9 10 8 CFU ml À1 ). Stomatal aperture was measured after 0, 1 and 3 h of incubation with the bacteria. Stomatal aperture measurements for ABA sensitivity assays were carried out on detached leaf epidermis as described previously (McAinsh et al., 1991;Chater et al., 2011).

Protein extraction and Immunoblotting
Protein extractions and immunoblotting were carried out as described previously (Gibbs et al., 2011).

Gene expression analysis
RNA extraction, cDNA synthesis, semiquantitative and quantitative RT-PCR were performed as previously described for Arabidopsis (Gibbs et al., 2011(Gibbs et al., , 2014b and barley (Mendiondo et al., 2016). For primers used see Supporting Information  Table S1.

Analysis of nitrate reductase activity
Nitrate reductase was assayed as previously (Vicente et al., 2017) with modifications described elsewhere (Kaiser & Lewis, 1984).

Experimental statistical analyses
All experiments were performed at least in triplicate. Statistical comparisons were conducted using GraphPad PRISM 7.0 software. Horizontal lines represent standard error of the mean values in all graphs. For statistical comparisons we used Student's t-test, where statistically significant differences are reported as: ***, P < 0.001; **, P < 0.01; *, P < 0.05; and one-way analysis of variance (ANOVA) with Tukey's multiple comparisons test, where significant differences (a < 0.05) are denoted with different letters.

Results
Nt-Gln amidase and Cys oxidation branches of the Arg/ N-end rule pathway increase basal resistance against Pst DC3000 The role for the Arg/N-end rule pathway in the plant immune response was assessed using the model bacterial pathogen P. syringae pv tomato DC3000 and T-DNA insertion null mutants of the putative Gln-specific amino-terminal amidase NTAQ1 (AT2G41760) (Fig. S1a-d) and N-recognin E3 ligase PRT6 (AT5G02310) genes, and a premature termination allele of the putative Asn-specific amino-terminal amidase NTAN1 (AT2G44420) (Q202*) (Fig. 1a). The entire effect of NTAQ1, NTAN1 and Cys branches of the Arg/N-end rule pathway on response to pathogen challenge can be assessed by analysis of the prt6 mutant, as this removes E3 ligase activity, thus stabilizing all substrates of NTAQ1, NTAN1 and substrates with Nt-Cys (Fig. 1a). Bacterial growth in leaves of prt6 was significantly lower by 4 d post-infiltration with virulent (Pst DC3000) or avirulent (Pst DC3000 avrRmp1) strains, indicating that substrates destabilized by PRT6 action contribute to the immune response (Figs 1b, S2a). In comparison, ntaq1 alleles also showed significantly lower bacterial growth (comparable with that of prt6 ) compared with both the ntan1-1 mutant or the wild type (WT) Col-0 for plants grown from seed in soil under neutral days (12 h : 12 h, light : dark). These results are opposite to those obtained by de Marchi et al. (2016), who found enhanced sensitivity to Pst DC3000 of N-end rule mutants prt6 and ate1 ate2 (which removes ATE Nt-arginylation activity, Fig. 1a). To investigate this difference, we assayed bacterial growth under conditions used by de Marchi et al. for plant growth and infection. In their case, germination and initial 7 d growth of seedlings was carried out on agar containing MS medium and 0.5% sucrose before transfer to soil and, following transfer, plants were grown under short-day conditions (9 h : 15 h, light : dark). We grew Col-0, prt6-1 and ate1 ate2 under these conditions and assayed bacterial growth at 2 d and 4 d post-infiltration. For plants grown under neutral days, we found that by 4 d post-infection, bacterial growth was significantly lower in N-end rule mutants than in the

Research
New Phytologist WT (Fig. S2b). All subsequent reported experiments were carried out using plants grown from seed under neutral-day conditions.
Tissue cellular leakage measured 4 d following infection was significantly lower in prt6 and ntaq1 mutants (Figs 1c, S1d). Expression in WT of NTAQ1 and PRT6 was not strongly affected by infection with either bacterial strain (Fig. S2c). Inoculation with the PTI inducer Pst DC3000 hrpA À (with a compromised type-three secretion system), resulted in reduced susceptibility of prt6 and ntaq1 mutants compared with WT or ntan1 (Fig. 1d). Ectopic expression of either Nt-or C-terminally tagged NTAQ1 removed enhanced resistance of ntaq1-3 (Fig. 1e), and the double mutant prt6-1 ntaq1-3 did not show significant difference compared with the single mutants prt6-1 or ntaq1-3 (Fig. 1f). It was previously suggested that formation of N-terminal pyroglutamate by glutaminyl cyclase (GC) might compete with NTAQ1 for Nt-Gln substrates (Wang et al., 2009), implying that a lack of GC activity could lead to enhanced susceptibility. We observed a similar response to Pst DC3000 of WT and a mutant of GLUTAMINYL CYCLASE1 (GC1) (Schilling et al., 2007) (Fig. S2d), indicating that competition for Nt-Gln substrates between NTAQ1 and GC1 is not relevant for the regulation of bacterial growth following infection. To define the biochemical action of NTAQ1, we analysed the Nt-deamidation capacity of recombinant Arabidopsis NTAQ1 that showed high specificity for Nt-Gln in comparison with Nt-Asn, -Gly and-Lys (Fig. 1g). Col-0 prt6-1 erfVII prt6 erfVII

Pst DC3000 injection
Pst DC3000 spray Pst DC3000 spray Using mutants in which ERFVII activity was removed (Abbas et al., 2015) (rap2.12 rap2.2 rap2.3 hre1 hre2 pentuple mutant, hereafter erfVII, and the prt6 erfVII sextuple mutant), analysis of infections of Pst DC3000 following infiltration showed no significant influence of ERFVIIs in affecting apoplastic growth of either virulent or avirulent Pst strains (Figs 2a, S3a). Bacterial growth 4 d following foliar spray application of Pst DC3000 revealed greater resistance of both prt6-1 and ntaq1-3 mutants compared with WT or ntan1-1 (Figs 2b, S3b), which for both foliar spray and injection required SA, analysed in double mutant combinations of prt6-1 or ntaq1-3 with sid2-1. SID2 is an isochorismate synthase required for SA synthesis (Nawrath & Metraux, 1999) (Fig. S3c). Stomatal closure is a key component of early defence response following pathogen attack (Arnaud & Hwang, 2015). We found that, in response to Pst, WT initially closed and then, induced by the pathogen, reopened its stomata, as did prt6-1 and ntaq1-3. The erfVII and prt6 erfVII mutants failed to close stomata at any point (Fig. 2c). ERFVIIs have previously been shown to regulate stomatal ABA sensitivity via the Nend rule pathway (Vicente et al., 2017), and we also found ntaq1-3 stomata were hypersensitive to ABA (Fig. S3d). In response to Pst DC3000 infection following foliar spray application, resistance was significantly lower in the absence of ERFVII transcription factors (either erfVII or prt6 erfVII) compared with WT or prt6 (Fig. 2d), respectively. Response to the foliar spray application of Pst DC3000 was associated with a large decrease in activity and expression of NITRATE REDUCTASE (NR) (Fig. 2e,f). This reduction has been previously linked with increased basal resistance against Pst (Park et al., 2011), whereas expression of ADH1, a marker for hypoxia, was only increased immediately following pathogen challenge (Fig. S3e). Infection with Pst DC3000 was associated by 24 h with increased stabilization of an artificial Cys-Arg/N-end rule substrate derived from the construct 35S:MC-HA GUS, that following constitutive MetAP activity is expressed as C-HA GUS (Gibbs et al., 2014b;Vicente et al., 2017) (Fig. 2g). To clarify whether plant-derived factors were solely responsible for the control of the stability of C-HA GUS, we injected the PAMP peptide flg22, and showed that injection of flg22 was able to stabilize C-HA GUS (Fig. 2h).

The Arg/N-end end rule pathway has a conserved function in the immune response
To determine the conservation of Arg/N-end rule pathway role in the immune response, we tested responses to pathogens in barley, a monocot species distantly related to Arabidopsis, in which the expression of the PRT6 orthologue gene HvPRT6 was reduced by RNAi (Mendiondo et al., 2016). Following inoculation with a strain of P. syringae pv japonica with known pathogenicity to barley (Dey et al., 2014), significantly lower bacterial load was observed in HvPRT6 RNAi leaves compared with the WT (Fig. 3a). Similarly, HvPRT6 RNAi plants exhibited reduced development and severity of mildew caused by Bgh (Fig. 3b,c). By contrast, susceptibility of HvPRT6 RNAi to Fusarium graminearum or F. culmorum, tested on detached leaves was increased compared with the WT (Fig. 3d). To assess the response of prt6-1 in Arabidopsis to a necrotroph we inoculated the mutant and WT with the fungal pathogen B. cinerea but we failed to observe any significant differences in disease severity, measured as diameter of necrotic lesions (Fig. S3f). Infection of barley with Ps pv japonica or Bgh also resulted in accumulation of the artificial Nt-Cys substrate CGGAIL-GUS (from pUBI: MCGGAIL-GUS, containing the first highly conserved seven residues of ERFVIIs; Gibbs et al., 2014b;Mendiondo et al., 2016;Vicente et al., 2017), therefore Nt-Cys stabilization in response to infection is conserved in flowering plants (Fig. 3e).

NTAQ1 regulates expression of the camalexin biosynthesis pathway
A shotgun proteomic analysis of total proteins from untreated ntaq1-3 and WT adult leaves revealed 13 proteins which were significantly differentially regulated, 12 exhibited increased and one decreased abundance in ntaq1-3 (Table S2). The functions of most ntaq1 upregulated proteins are related to oxidative, biotic and abiotic stresses, including a 2-OXOGLUTARATE OXYGENASE (AT3G19010) potentially involved in quercetin biosynthesis and targeted by bacterial effectors (Truman et al., 2006) and DJ-1 protein homolog E (DJ1E) involved in response to PAMPs (Lehmeyer et al., 2016). Not all ntaq1 upregulated proteins were also upregulated at the level of RNA (Fig. S4). Several ntaq1 over-accumulated proteins are involved in the regulation of reactive oxygen species (ROS). However, analysis of gene expression of a ROS accumulation marker, the antioxidant enzyme CATALASE1 (CAT1), and histochemical analysis of the accumulation of the ROS hydrogen peroxide (H 2 O 2 ) during infections with Pst failed to reveal significant differences between the mutants ntaq1 and prt6 and WT (Fig. S5). Increased tolerance of the mutants which was associated with less cellular damage required SID2, an isochorismate synthase required for SA synthesis (Nawrath & Metraux, 1999), as double mutant combinations of prt6-1 or ntaq1-3 with sid2-1 showed susceptibility similar to the sid2 single mutant (Fig. S3c). Analysis of phytohormone levels indicated that there were no differences between ntaq1-3, prt6-1 or WT in untreated or infected leaves for SA, JA or IAA (Figs 4, S6). These results together suggest a functional redundancy of ntaq1 upregulated proteins with other antioxidant mechanisms, already documented in the case of the GLUTATHIONE S-TRANSFERASEs (GSTs) (Sappl et al., 2009), or alternative roles for ntaq1 upregulated proteins in plant defence.
One of the identified proteins upregulated in ntaq1, the phi class GSTF6, functions in secondary metabolism related to the synthesis of the major Arabidopsis phytoalexin, camalexin (Su et al., 2011), as do the upregulated proteins PUTATIVE ANTHRANILATE PHOSPHORIBOSYLTRANSFERASE (involved in the synthesis of the camalexin precursor tryptophan; Zhao & Last, 1996) and IAA-AMINO ACID HYDROLASE (ILL4), that generates indole-3-acetic acid (IAA) from its conjugated form (Davies et al., 1999). Another upregulated protein, GSTF7 was hypothesized to play a role in camalexin synthesis based on its induction in the constitutively active MKK9 mutant  (Su et al., 2011). Our analysis of previously published transcriptome data (de Marchi et al., 2016) comparing gene expression in ate1 ate2 with WT, and comparing gene expression during Pst infection in Col-0 and ate1 ate2 also showed increased expression of RNAs encoding camalexin synthesis genes (Tables S3, S4). Analysis of transcript expression indicated greater accumulation for most genes of camalexin synthesis in mature uninfected leaves of ntaq1 and prt6 compared to WT (Figs 4, S7), including PAD3 (CYP71B15), that catalyzes the final two steps of camalexin synthesis. Interestingly, during a time course following infiltration with Pst DC3000, levels of camalexin-associated transcripts, including GSTF6 and PAD3, as well as GSTF7 increased in WT but to a lesser extent in mutant leaves (Figs 4, S7). Whilst basal levels of camalexin in uninfected leaves were similar in mutants and WT they increased to a greater degree in mutants than WT in response to infection (Fig. 4). Mutant plants showed greater basal levels of indole-3-carboxylic acid (I3CA), a compound synthesized during the defence response and a potential precursor of camalexin through the action of GH3.5 (Forcat et al., 2010;Wang et al., 2012) that was also upregulated at the RNA level in untreated leaves of ntaq1-3 (Fig. 4). Camalexin synthesis is highly interconnected with other pathways of secondary metabolism, for example it has been reported that vte2 and cyp83a1, mutants of key steps of tocopherol and aliphatic glucosinolate synthesis pathways respectively, show increased levels of camalexin (Sattler et al., 2006;Liu et al., 2016). VTE2 and CYP83A1 showed decreased expression in ntaq1-3 and prt6-1 in both basal and infected conditions (Figs 4, S8). Combination of a null pad3 allele with prt6-1 resulted in a loss of the prt6 enhanced resistance to injected Pst DC3000 (Fig. 5).
The Arg/N-end rule pathway regulates an age-dependent primed state in uninfected plants Previous work showed that hypoxia-associated genes are ectopically upregulated in prt6 and ate1 ate2 mutant seedlings (Gibbs et al., 2011;Licausi, 2013). However, it was recently shown that this is age-dependent, that in mature mutant plants these genes  (Giuntoli et al., 2017). We also observe a large reduction in expression of hypoxia genes in older prt6 plants and saw a similar trend in WT for some genes (Fig. S9a). No age-related differences were found in NTAQ1 expression in either WT or prt6 backgrounds (Fig. S9b), however GSTF6/7 and PAD3 showed increased expression with age in prt6-1 and ntaq1-3 plants compared with WT (Fig. 6a). In N-end rule mutants, compared to WT we found age-related increases for the SAresponsive PATHOGENESIS RELATED (PR) protein genes PR1 and PR5, whilst JA and ET responsive PR3 and PR4 showed no differences (Fig. 6b). In barley, constitutive increase in expression of the SA-responsive genes HvPR1 and Hvß1-3 glucanase (Horvath et al., 2003;Rostoks et al., 2003) was found in leaves of HvPRT6 RNAi plants, and infection with Bgh did not result in an increase in expression in HvPRT6 RNAi plants, that was observed in WT plants (Fig. 6c).

Discussion
We show here that a role for Arg/N-end rule pathway-mediated immunity is conserved in flowering plants. In Arabidopsis we demonstrate physiological, biochemical and molecular roles for Nend rule component NTAQ1 in influencing basal defence by enhancing expression of defence proteins and synthesis of camalexin, and a role for the ERFVII known substrates in influencing stomatal response, against the hemibiotroph Pst. We show a role in barley of the Arg/N-end rule in response to the biotroph Bgh and hemibiotroph Ps japonica. We suggest that benefits of increased immunity may not be realized against necrotrophic pathogens (as shown in the interaction between Fusarium spp. and barley). It has been documented that camalexin is part of the defence response against the necrotroph fungus B. cinerea,inhibiting its growth in a dose-dependent manner (Ferrari et al., 2 0 0 3 ) . In our experiments, there were no differences in responses of WT and prt6 to B. cinerea suggesting that independently of other mechanisms activated, an increase in camalexin in prt6 may not reach a level necessary for reduction in fungal growth. A recent report showed N-end rule mutants, including alleles of prt6, ate1 ate2 and ntaq1 to be in general equal or more sensitive than WT Arabidopsis to a wide range of bacterial and fungal pathogens with diverse infection strategies and lifestyles (de Marchi et al.,2 0 1 6 ) . Our results, in which plants were grown under either neutral days or under the short-day condition used by de Marchi et al. showed the opposite results (of increased resistance). Our results provide a consistent pattern across different levels of expression (including enhanced defence gene transcripts and increased levels of camalexin synthesis proteins in untreated plants, and consistent phenotypes between Arabidopsis and barley) that indicate a role for NTAQ1 substrates and ERFVIIs as components of the immune response that enhance resistance. Therefore, differences in observed phenotypes of N-end rule mutants in response to infection between our studies remain to be resolved.
A specific effect for ERFVIIs was observed in the stomatal response to Pst. ABA is an important component of stomatal response to pathogens (McLachlan et al., 2014) and stabilized ERFVIIs enhance ABA sensitivity of stomata (Vicente et al., 2017). We observed a large increase in stability of artificial Nt-Cys reporters in both Arabidopsis and barley. Stabilisation could be caused by shielding of the Nt, or a reduction of either NO or oxygen. We did not observe an increase in hypoxia-related gene expression (of ADH1) at the same time as GUS stabilization, however we did observe a decline in NR activity. Seemingly contradictory to this assertion is the well known burst of NO in response to Pst infection (Delledonne et al., 1998). However, this burst occurs early following infection, well before the reduction in NR activity and stabilization of artificial Nt-Cys reporters in both Arabidopsis and barley. It has previously been shown that in the NR null mutant nia1 nia2, which produces very low NO levels, the NO burst in response to infection is highly reduced (Modolo et al., 2006;Chen et al., 2014). Further experiments would be required to determine a causative role of reduced NR activity leading to enhanced stabilization. Regardless of the mechanism of stabilization, the observation of increased stability of Nt-Cys substrates following infection in both Arabidopsis and barley indicates a conserved role for modulation of the Cys-Arg/N-end rule pathway, and function for Nt-Cys substrates, in response to pathogen infection that deserves further investigation. Enhanced ABA sensitivity and stomatal response to Pst of the ntaq1 mutant also suggests that Nt-Gln substrate(s) contribute to the stomatal ABA response to pathogens, and explains why erfVII is more sensitive to Pst than prt6 erfVII (where NTAQ1 substrates are still stabilized). An opposite effect of ERFVIIs was shown for interactions of Arabidopsis with the biotroph P. brassicae, as ERFVIIs enhanced infection indirectly by influencing fermentation (Gravot et al., 2016).  These observations and others  indicate an important role for ERFVIIs in the plant immune response.
Analysis of the response to Pst DC3000 hrpA À , together with increased expression of SA-associated defence genes and increased camalexin synthesis, suggests a role for NTAQ1 in the onset of general and inducible PTI defence. An age-related increase in SArelated defence gene expression in N-end rule mutants was not matched by increased SA levels. This suggests a possible role for immune-related MAPK cascade activating MPK3/6 that are sufficient for SA-independent induction of most SA-responsive ***

Research
New Phytologist genes, including PR1 (Asai et al., 2002). Concomitantly, it has been demonstrated that both MPK3 and MPK6 activation trigger GSTF6, 7 (and DJ1E) protein accumulation, which produces an increase in camalexin (Xu et al., 2008;Su et al., 2011). The observed increased accumulation of camalexin in ntaq1 and prt6 provides one explanation for the increased resistance of these mutants. Although expression of camalexin synthesis genes was ectopically upregulated in uninfected mature leaves of mutants, enhanced camalexin accumulation was only observed in response to infection. This may be the result of shunting of intermediate(s) to other secondary metabolism pathways. In line with this, unchallenged ntaq1 and prt6 plants show greater levels of I3CA. The observation that mutation of pad3 reverts the enhanced resistance of prt6 highlights the role of N-end rule regulated camalexin synthesis in enhancing the immune response.
How might NTAQ1 function during development and in response to pathogen attack? NTAQ1 and PRT6 expression do not change in response to pathogen attack. NTAQ1 function influences defence gene expression and camalexin synthesis. We demonstrate that downstream responses to NTAQ1, measured as responsive gene expression, are modified during development (although the expression of NTAQ1 (and PRT6 ) transcripts were not affected by ageing), suggesting that NTAQ1 substrate(s) may show an age-dependent increase in abundance. Following protease cleavage their activity would be revealed in the ntaq1 mutant, where they would remain ectopically stabilized. Following protease cleavage to reveal Nt-Gln, NTAQ1 substrates should be degraded in WT plants. In this case, in mature WT leaves down-regulation of NTAQ1-linked protease activity (or NTAQ1 activity) in response to pathogen attack could result in substrate stabilization. Stabilized NTAQ1 substrate(s) (or uncleaved protease targets that provide substrates) may then function to enhance gene expression associated with defence genes and camalexin synthesis, both resulting in an enhanced basal immune response.
Our data support a conserved role of the Arg/N-end rule pathway in influencing plant immune responses. Barley contains one NTAQ1 gene (MLOC_70886) (Mayer et al., 2012). Manipulation of expression or activity of this gene will be required to understand whether NTAQ1 activity is also required for defence in barley. An important goal of future work will be to identify Nt-Gln substrates that influence the immune response. Although NTAQ1-related genes are present in all major groups of eukaryotes, only a single example exists of a biochemical role for this enzyme and its associated substrate (Usp1) (Piatkov et al., 2012). There is already evidence for Nt-Gln-bearing peptide fragments derived from proteins of diverse functions present in the plant METACASPASE-9 degradome (Tsiatsiani et al., 2013), suggesting that substrates for NTAQ1 exist. Our results establish new components of the plant immune response, and offer new targets to enhance resistance against plant pathogens.

Supporting Information
Additional Supporting Information may be found online in the Supporting Information section at the end of the article: