https://doi.org/10.1093/plphys/kiac579 PLANT PHYSIOLOGY 2023: 191: 1934–1952 A mini-TGA protein modulates gene expression through heterogeneous association with transcription factors Research Article Downloaded from https://academic.oup.com/plphys/article/191/3/1934/6903602 by guest on 26 May 2023 Špela Tomaž ,1,2,* Marko Petek ,1 Tjaša Lukan ,1 Karmen Pogačar ,1 Katja Stare ,1 Erica Teixeira Prates ,3 Daniel A. Jacobson ,3 Jan Zrimec ,1 Gregor Bajc ,4 Matej Butala ,4 Maruša Pompe Novak ,1,5 Quentin Dudley ,6 Nicola Patron ,6 Ajda Taler-Verčič ,7,8 Aleksandra Usenik ,7,9 Dušan Turk ,7,9 Salomé Prat ,10 Anna Coll 1,† and Kristina Gruden 1,† 1 Department of Biotechnology and Systems Biology, National Institute of Biology, 1000 Ljubljana, Slovenia 2 Jožef Stefan International Postgraduate School, 1000 Ljubljana, Slovenia 3 Biosciences Division, Oak Ridge National Laboratory,, Oak Ridge, Tennessee 37831, USA 4 Department of Biology, Biotechnical Faculty, University of Ljubljana, 1000 Ljubljana, Slovenia 5 School for Viticulture and Enology, University of Nova Gorica, 5271 Vipava, Slovenia 6 Earlham Institute, Norwich Research Park, Norwich NR4 7UZ, UK 7 Department of Biochemistry and Molecular and Structural Biology, Jožef Stefan Institute, 1000 Ljubljana, Slovenia 8 Faculty of Medicine, Institute of Biochemistry and Molecular Genetics, University of Ljubljana, 1000 Ljubljana, Slovenia 9 Centre of Excellence for Integrated Approaches in Chemistry and Biology of Proteins, 1000 Ljubljana, Slovenia 10 Department of Plant Development and Signal Transduction, Centre for Research in Agricultural Genomics, 08193 Cerdanyola, Barcelona, Spain *Author for correspondence: spela.tomaz@nib.si † These authors contributed equally and share the last authorship. Š.T., A.C., and K.G. designed the research. Š.T., A.C., T.L., M.P., K.P., K.S., M.P.N., J.Z., Q.D., G.B., and E.T.P. performed the experiments and analyzed the data. Š.T. wrote the initial manuscript draft. All authors contributed to writing or revision of the manuscript. The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (https://academic.oup.com/plphys/pages/General-Instructions) is Anna Coll (anna.coll@nib.si). Abstract TGA (TGACG-binding) transcription factors, which bind their target DNA through a conserved basic region leucine zipper (bZIP) domain, are vital regulators of gene expression in salicylic acid (SA)-mediated plant immunity. Here, we investigated the role of StTGA2.1, a potato (Solanum tuberosum) TGA lacking the full bZIP, which we named a mini-TGA. Such truncated proteins have been widely assigned as loss-of-function mutants. We, however, confirmed that StTGA2.1 overexpression com­ pensates for SA-deficiency, indicating a distinct mechanism of action compared with model plant species. To understand the underlying mechanisms, we showed that StTGA2.1 can physically interact with StTGA2.2 and StTGA2.3, while its interaction with DNA was not detected. We investigated the changes in transcriptional regulation due to StTGA2.1 overexpression, iden­ tifying direct and indirect target genes. Using in planta transactivation assays, we confirmed that StTGA2.1 interacts with StTGA2.3 to activate StPRX07, a member of class III peroxidases (StPRX), which are known to play role in immune response. Finally, via structural modeling and molecular dynamics simulations, we hypothesized that the compact molecular architecture of StTGA2.1 distorts DNA conformation upon heterodimer binding to enable transcriptional activation. This study Received September 9, 2022. Accepted November 24, 2022. Advance access publication December 15, 2022 © The Author(s) 2022. Published by Oxford University Press on behalf of American Society of Plant Biologists. Open Access This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. The role of mini-TGAs in plant immunity PLANT PHYSIOLOGY 2023: 191; 1934–1952 | 1935 demonstrates how protein truncation can lead to distinct functions and that such events should be studied carefully in other protein families. Introduction then alleviated through NPR1 interaction with AtTGA2 N terminus, affecting the binding stoichiometry and leading Plants have developed efficient strategies to withstand the to the formation of a transcriptional activation complex invasion of surrounding microbes. Pathogen recognition is (Rochon et al., 2006; Boyle et al., 2009). Additionally, other re­ mediated by plant cell-surface and intracellular receptors, ports propose NPR1 interacts with TGAs not yet bound to triggering a cascade of intracellular reactions, orchestrated DNA and indicate it could facilitate TGA binding to target by phytohormones, ultimately leading to a finely modulated promoter (Johnson et al., 2008). Furthermore, regulatory pro­ transcriptional reprogramming (Zhou and Zhang, 2020). teins, such as WRKY50 (Hussain et al., 2018) and histone acyl­ Regulation of defense-related gene expression is among the transferase (HAC) transcription factors (Jin et al., 2018), have Downloaded from https://academic.oup.com/plphys/article/191/3/1934/6903602 by guest on 26 May 2023 most fundamental aspects of the immune response, involv­ also been shown to contribute to AtTGA2 transcriptional ing multiple transcription factors and cofactor proteins. function. Since their initial discovery in tobacco (Nicotiana tabacum) Although the results obtained in Arabidopsis provide a over 30 years ago (Katagiri et al., 1989), the importance of molecular framework for understanding the role of TGAs TGA (TGACG-binding) transcription factors in plant im­ in plant immunity, we know much less about their function munity, as well as modulation of other cellular processes, in crops. The involvement of TGAs in biotic stress response has been widely studied (Gatz, 2013). has been reported in several species, including rice (Oryza sa­ TGAs are a group of transcription factors belonging to the tiva) (Moon et al., 2018), soybean (Glycine max) (Lawaju et al., basic region leucine zipper (bZIP) protein family. Their mech­ 2018), strawberry (Fragaria × ananassa) (Feng et al., 2020), anism of action has been thoroughly studied in Arabidopsis tobacco (Thurow et al., 2005), and tomato (Solanum lycoper­ (Arabidopsis thaliana), where the 10 Arabidopsis TGAs sicum) (Ekengren et al., 2003). Potato (Solanum tuberosum) is (AtTGAs) group into five clades (Jakoby et al., 2002). Clade one of the most widely grown crops (FAO, 2020) and tuber II members, AtTGA2, AtTGA5, and AtTGA6, are essential production is severely threatened by pathogen infections. regulators of the salicylic acid (SA)-mediated defense re­ Several transcription factor families have been associated sponse, where they play a redundant, yet vital role in estab­ with the regulation of potato defense response lishing resistance following infection (Zhang et al., 2003; (Chacón-Cerdas et al., 2020), but the mechanisms underlying Zhou and Zhang, 2020). They coregulate the expression of potato TGA (StTGA) activity remain largely unexplored. key defense-related genes and genes involved in SA synthesis Here, we identify the mini-TGA StTGA2.1, a potato clade II through interaction with NON-EXPRESSOR OF PR (NPR) co­ TGA, which lacks most of the bZIP DNA-binding domain and factors (Zhang et al., 1999; Ding et al., 2018), while also par­ has a shorter N terminus. We hypothesize that StTGA2.1 can­ ticipating in jasmonic acid and ethylene-mediated signaling not bind DNA by itself because of the truncated bZIP and (Zander et al., 2010). Structurally, TGAs consist of an intrin­ therefore modulates gene expression through its interaction sically disordered N terminus of varying length, a conserved with additional DNA-binding StTGAs. By combining in vivo bZIP domain, which entails a basic region and a leucine zip­ and in vitro functional studies, we confirm the role of per, and a C-terminal region that contains a putative Delay of StTGA2.1 in potato immunity. Furthermore, using in silico Germination 1 (DOG1) domain (Tomaž et al., 2022). TGAs structural analysis and molecular dynamics (MD) simula­ bind their target DNA through the bZIP basic region, while tions, we provide insights into the molecular basis for a dif­ the leucine zipper is important for protein dimerization ferent mechanism of action of StTGA2.1 compared to (Thurow et al., 2005) and oligomerization (Boyle et al., other StTGAs. 2009). The TGACG core sequence is sufficient for TGA bind­ ing, although high-throughput DNA-binding studies revealed the TGACGTCA palindrome as the representative binding motif (Thibaud-Nissen et al., 2006; O’Malley et al., 2016). Results The molecular mechanisms of TGA-mediated regulation Potato encodes clade II TGAs with truncated bZIP involve complex interactions between TGAs and other pro­ domain teins (Gatz, 2013). For example, the SA-receptor NPR1 inter­ In previous work, we investigated gene expression in re­ acts with AtTGA2 to activate the expression of the sponse to viral infection in nontransgenic-resistant potato PATHOGENESIS-RELATED-1 (PR-1) gene expression (Zhang (NT) and its transgenic derivative (NahG), which is impaired et al., 1999; Backer et al., 2019), but the mechanistic aspect in SA accumulation and thus sensitive to infection (Baebler of this cooperation is not yet entirely clear. Several studies et al., 2014). To identify the TGA transcription factors in­ suggest that AtTGA2 acts as a constitutive repressor of volved in potato immunity, we examined the expression pat­ PR-1 in absence of biotic stress (Zhang et al., 2003; Rochon terns of the 14 StTGA genes, orthologs of AtTGAs et al., 2006; Kesarwani et al., 2007). Its repressive activity is (Supplemental Tables S1 and S2). Notably, Sotub10g022560 1936 | PLANT PHYSIOLOGY 2023: 191; 1934–1952 Tomaž et al. was up-regulated in infected NahG transgenic plants, but not Additionally, the size-exclusion chromatography (SEC) elu­ in the parental lines, suggesting that it may be an important tion volume of a recombinant His6-tagged StTGA2.1 corre­ component of SA signaling. sponded to the size of a dimer (Supplemental Figure S3, D To classify the StTGAs, we conducted a phylogenetic ana­ and E), while chemical cross-linking of a nontagged protein lysis of all candidate potato proteins, along with the 10 yielded monomers, dimers, and higher order complexes AtTGAs and 13 TGAs from tomato (SlTGAs) (Hou et al., (Supplemental Figure S3F). Overall, these results demon­ 2019; Lemaire-Chamley et al., 2022). Interestingly, the three strate that StTGA2.1 retains protein–protein interaction clade II AtTGAs are orthologous to five StTGAs and four ability. In addition, we examined whether StTGA2.1 can SlTGAs (Figure 1A). Three closely related members of this interact with two potato NPR cofactors, a putative ortholog clade, including Sotub10g022560, named StTGA2.1, of AtNPR1, StNPR1 (Sotub07g011600), and a putative ortho­ StTGA2.4 (Sotub10g022570), and SlTGA2.3 (Solyc10g080780) log of AtNPR3 and AtNPR4, StNPR3/4 (Sotub02g015550). Our (Hou et al., 2019; Lemaire-Chamley et al., 2022), have shorter results showed that StTGA2.1 as well as StTGA2.2 and protein sequences than other TGAs (Figure 1B). Domain pre­ StTGA2.3 interact with both StNPRs in yeast and that the Downloaded from https://academic.oup.com/plphys/article/191/3/1934/6903602 by guest on 26 May 2023 diction studies showed that they retain the putative addition of SA to the media promotes these interactions C-terminal DOG1 domain; however, the bZIP domain is al­ (Supplemental Figure S4). Thus, the ability to interact with most completely lost, retaining only a partial zipper region. NPR proteins is not perturbed in mini-TGA StTGA2.1. In addition, their N terminus is very short and dissimilar to Subcellular localization of green fluorescent protein the N termini of other clade II TGAs. We named these three (GFP)-tagged StTGA2.1 in the N. benthamiana leaf epidermis proteins mini-TGAs. and mesophyll showed that it can localize to cell nuclei By targeted sequencing of a ∼36.5 kb region on chromo­ (Figure 3C). StTGA2.1 also showed a distinct localization some 10, where StTGA2.1, StTGA2.2 (Sotub10g022550), and pattern with intense fluorescence in the cytoplasm, which StTGA2.4 loci are colocated, we confirmed the reduced was enhanced around the chloroplasts (Supplemental length of StTGA2.1 and StTGA2.4 in a tetraploid cultivar Figure S5A). We also detected its fluorescence in the ER that was used for further analyses (Supplemental Figure S1). and in granular formations of about 0.5–1.0 μm in size (Supplemental Figure S5). In contrast, StTGA2.2 and StTGA2.1 improves immune response in SA-deficient StTGA2.3 showed predominantly nuclear localization and were organized into subnuclear formations of different sizes potato within the nuclei (Figure 3C, Supplemental Figure S6). SA signaling has proven vital for the establishment of an ef­ ficient defense response against potato virus Y (PVY) infec­ tion in resistant potato cultivars (Baebler et al., 2014; Lukan Identification of potential StTGA2.1 targets with et al., 2020). We thus investigated the role of StTGA2.1 in spatial transcriptomic profiling plant immunity using the potato-PVY pathosystem. We gen­ To gain insight into the mini-TGA mechanism of action in erated SA-deficient NahG transgenic potato plants inducibly plant immunity, we examined the expression profile of overexpressing StTGA2.1 (TGA2.1-NahG) using the NahG plants overexpressing StTGA2.1. By sampling tissue sec­ glucocorticoid-system (Aoyama and Chua, 1997), in which tions containing lesions and their immediate surrounding target gene expression is controlled by external application area after PVY infection (Supplemental Figure S7A), we of dexamethasone (DEX). Three transgenic TGA2.1-NahG were able to follow transcriptomic changes in lines, showing more than six-fold induction in StTGA2.1 ex­ PVY-responding cells (Lukan et al., 2020). RNA sequencing pression after DEX treatment (Supplemental Figure S2, A– results showed a regulation of 217 genes due to StTGA2.1 C), were selected for further analysis. We observed that viral overexpression in the NahG background (TGA2.1-NahG vs. replication was significantly reduced in TGA2.1-NahG com­ NahG plants comparison, Supplemental Table S3). pared to NahG at 10 days post infection (dpi) (Figure 2, However, over 1,800 genes were differentially expressed ex­ Supplemental Figure S2, D–F). As expected, little to no PVY clusively in TGA2.1-NahG, when plants were exposed to was detected in NT plants exhibiting a typical resistant pheno­ pathogen infection (Supplemental Figure S7B). Technical val­ type (Baebler et al., 2014). This shows that overexpression of idation of the RNA sequencing data by reverse transcription StTGA2.1 can compensate for the lack of SA in potato immune quantitative PCR (RT-qPCR) is shown in Supplemental response to PVY. Table S4. Gene set enrichment analysis (Supplemental Table S5) en­ StTGA2.1 retains its dimerization ability and shows a abled us to extract key differences in gene expression at the distinct localization pattern level of processes or functionally related gene groups (BINs), Protein interaction studies using the yeast two-hybrid assay as they are defined by the MapMan ontology (Ramšak et al., showed that StTGA2.1 can form both homodimers and het­ 2014). Important immunity-related or regulatory BINs, en­ erodimers with StTGA2.2 and StTGA2.3 (Sotub01g009430) riched in up- or down-regulated genes of PVY- vs. mock- (Figure 3A), further confirmed by in planta coimmunopreci­ inoculated plants for all three genotypes, are listed in pitation assay (Figure 3B, Supplemental Figure S3, A–C). Table 1. Up-regulated genes enriched uniquely in The role of mini-TGAs in plant immunity PLANT PHYSIOLOGY 2023: 191; 1934–1952 | 1937 Figure 1 Phylogenetic analysis and domain characterization of StTGAs. A, A rooted phylogenetic tree of potato, tomato, and Arabidopsis TGAs. The Downloaded from https://academic.oup.com/plphys/article/191/3/1934/6903602 by guest on 26 May 2023 mini-TGA branch is shaded and StTGA2.1 is marked (arrow). The branch length scale represents the number of amino acid substitutions per site. The rice OsTGA2.1 (Chern et al., 2001) serves as tree root. B, Protein sequence alignment of clade II TGAs, showing the position of the bZIP domain (box) and the shorter sequences of mini-TGA members, StTGA2.1, StTGA2.4, and SlTGA2.3. The alignment is colored with the Geneious Prime (https://www.geneious.com/) sequence similarity color scheme, based on the identity score matrix. Sequence numbering (aa) is shown above the alignment. TGA2.1-NahG included isoprenoid metabolism-related with the other two genotypes, as was the up-regulated ex­ genes, PHD finger and PHOR1 transcription factors, and per­ pression of WRKY transcription factors (Table 1). oxidases (Table 1). On the other hand, the C2C2(ZN) DOF transcription factors were enriched in down-regulated genes StTGA2.1 and StTGA2.3 activate the class III in TGA2.1-NahG plants. PVY-regulation of cytokinin and jas­ peroxidase StPRX07 monate metabolism was lost in TGA2.1-NahG compared As the expression of several peroxidases was up-regulated after StTGA2.1 overexpression in PVY-infected NahG plants 1938 | PLANT PHYSIOLOGY 2023: 191; 1934–1952 Tomaž et al. Downloaded from https://academic.oup.com/plphys/article/191/3/1934/6903602 by guest on 26 May 2023 Figure 2 StTGA2.1 attenuates PVY replication in SA-deficient plants. Relative expression levels of (A) PVY and (B) StTGA2.1 in PVY-infected leaves of DEX-treated TGA2.1-NahG line 12 (white), NahG (light grey), and NT (dark grey) plants at 3, 7, and 10 dpi. Average values ± SE from three biological replicates are shown. Significance (P < .05) was determined using a two-tailed t test comparing TGA2.1-NahG with NahG (p1) and TGA2.1-NahG with NT (p2). C, Phenotypic differences in PVY-infected leaves of DEX-treated TGA2.1-NahG line 12, NahG, and NT plants at 10 dpi. (Table 1, Supplemental Table S5), we recognized them as TGA-binding motifs in the StPRX07 promoter. Furthermore, ti­ potential direct targets of StTGA2.1. To test our tration of StTGA2.3 premixed with StTGA2.1 over PRX07p_1 hypothesis, we selected three class III peroxidases (StPRX, and PRX07p_2 fragments resulted in higher responses com­ Supplemental Table S6), StPRX07 (Sotub09g020950), StPRX15 pared with StTGA2.3 alone, whereas this was not the case for (Sotub02g035680), and StPRX46 (Sotub03g007840), which PRX15p_1 (Supplemental Figure S8B). These results support were up-regulated in TGA2.1-NahG compared with NahG the formation of a StTGA2.1-StTGA2.3 complex at the (Supplemental Tables S3 and S7), for further analysis. StPRX07 regulatory region. Analysis of their promoter regions revealed predicted Finally, we tested the ability of StTGA2.3 and StTGA2.1 to TGA-binding motifs between 450 and 750 bp upstream of activate the StPRX07 promoter in planta, using a transient the transcription start site (Figure 4A). transactivation assay (Lasierra and Prat, 2018). For this pur­ To investigate the ability of StTGAs to bind these motifs, we pose, the 2.95 kb long promoter region upstream of the first tested whether the StTGA2.3 and StTGA2.1 proteins could StPRX07 start codon, containing both predicted bind to four candidate DNA fragments from StPRX promoter TGA-binding motifs, was fused to a luciferase (LucF) coding regions, PRX07p_1, PRX07p_2, PRX15p_1, and PRX46p sequence. GFP-tagged StTGA2.1 and BFP-tagged StTGA2.3 (Figure 4A), using surface plasmon resonance. Titration of a re­ were then coexpressed with the reporter construct, con­ combinant StTGA2.3 over chip-immobilized PRX07p_1 and firmed by confocal microscopy. Coexpression of the reporter PRX07p_2 fragments, carrying the predicted TGA- construct with StTGA2.3 induced the expression of StPRX07:: binding motifs of the StPRX07 promoter, resulted in a dose- LucF by approximately 20% compared with basal promoter dependent increase in response, compared to reference activity, whereas coexpression with StTGA2.1 resulted in (Figure 4B). Interaction with PRX15p_1 and PRX46p fragments only minor induction (Figure 4C, Supplemental was negligible (Figure 4B). As predicted by the absence of the Figure S8C). In contrast, more than two-fold induction in basic region, we did not measure any interaction between promoter activity was observed when coexpressed with the His6-tagged StTGA2.1 and the tested DNA both StTGA2.1 and StTGA2.3, compared to control, meaning (Supplemental Figure S8A). These results suggest that the induction was about three to four-times stronger when StTGA2.3, but not StTGA2.1, binds specifically to the both StTGAs were overexpressed (Figure 4C, Supplemental The role of mini-TGAs in plant immunity PLANT PHYSIOLOGY 2023: 191; 1934–1952 | 1939 Downloaded from https://academic.oup.com/plphys/article/191/3/1934/6903602 by guest on 26 May 2023 Figure 3 StTGA2.1 can form homodimers, heterodimers, and localizes to diverse cellular compartments. A, StTGA2.1 interactions with itself, StTGA2.2, or StTGA2.3 in the yeast two-hybrid assay. Yeast were cotransformed with bait (BD) and prey (AD) construct combinations and selected on control media without Leu and Trp (-L-W). Positive interactions were determined by yeast growth on selection media without Leu, Trp, His and adenine, with added X-α-galactosidase and Aureobasidin A (-L-W-H-A + Xgal + Aur). B, StTGA2.1 interactions with itself, StTGA2.2, or StTGA2.3 in the coimmunoprecipitation assay. The combination of GFP and HA-tagged proteins expressed in N. benthamiana is indicated for each sample (+/−). Positive interactions were determined by detection of immunoprecipitated (GFP-IP) complexes with anti-HA antibodies. Arrows indicate expected bands. In case of HA-tagged StTGA2.3, two bands were detected, likely due to partial protein degradation. Controls are shown in Supplementary Figure S3. C, Subcellular localization of GFP-tagged StTGA2.1, StTGA2.2, and StTGA2.3 (yellow) with Histone 2B-monomeric red fluorescent protein 1 (H2B-RFP) nuclear marker (red) in N. benthamiana leaves. The p19 silencing suppressor was expressed as control. Protein fluorescence is repre­ sented as the z-stack maximum projection. Arrows indicate examples of StTGA2.1 nuclear localization. Scale bars, 100 μm. Figure S8C). These results indicate that strong activation of StTGA2.1 N terminus likely contributes to protein StPRX07 promoter is achieved only when both StTGA2.3 interactions and alters TGA binding to DNA and StTGA2.1 are present. Comparative structural analysis using AlphaFold (AF) (Jumper et al., 2021) revealed important singularities in the 1940 | PLANT PHYSIOLOGY 2023: 191; 1934–1952 Tomaž et al. Table 1 Selected functional groups (BINs) enriched in up- or down-regulated genes in TGA2.1-NahG, NahG, and NT plants after PVY infection. FDR corrected q value <.05. (+), enriched in up-regulated genes; (−), enriched in down-regulated genes BIN Functional group Size TGA2.1-NahG NahG NT Secondary metabolism 16.1.2 Mevalonate pathway 47 + 16.1.5 Terpenoids 155 + 16.10 Simple phenols 38 + Hormone metabolism 17.1.3 Abscisic acid-regulated 32 − − 17.2 Auxin 351 − − 17.2.3 Auxin-regulated 287 − − 17.4.1 Cytokinin 85 − − 17.7 Jasmonate 64 + + 17.8 Salicylic acid 22 + Downloaded from https://academic.oup.com/plphys/article/191/3/1934/6903602 by guest on 26 May 2023 Stress—biotic 20.1.7 PR proteins 208 + + + 20.1.7.1 PR-1 33 + 20.1.7.3 PR-3/PR-4/PR-8/PR-11 44 + + + Miscellaneous 26.12 Peroxidases 139 + 26.21 Protease inhibitor/seed storage/lipid transfer proteins 117 − Transcription regulation 27.3.32 WRKY transcription factors 97 + + 27.3.63 PHD finger transcription factors 49 + 27.3.64 PHOR1 transcription factors 21 + 27.3.8 C2C2(Zn) DOF transcription factors 41 − DNA synthesis—chromatin structure 28.1.3 Histone 83 + + + 28.1.3.2 Histone core 76 + + + 28.1.3.2.1 Histone core H2A proteins 26 + + 28.1.3.2.3 Histone core H3 proteins 20 + + + Protein degradation 29.5.11.20 Ubiquitin-proteasome 73 + + Signaling—receptor kinases 30.2.8.1 Leucine-rich repeat VIII (type 1) 20 − 30.2.16 Catharanthus roseus-like RLK1 79 + 30.2.17 DUF26 94 + 30.2.19 Legume-lectin 38 + + 30.2.99 Miscellaneous 206 + + molecular architecture of StTGA2.1, mostly contained in its MD simulations, identified the key DNA-binding residues N terminus. In the AF models of StTGA2.2 and StTGA2.3, (Figure 5, C and D, Supplemental Tables S8 and S9). the intrinsically disordered N terminus is followed by an Persistent interactions in the StTGA2.2 homodimer mostly α-helical bZIP domain, which includes several basic residues correspond to salt bridges, formed between the bZIP basic re­ and three heptads that comprise the leucine zipper sidues and the DNA phosphate groups, and do not explain the (Figure 5, A and B, Supplemental Figure S9). In contrast, StTGA2.2 motif-binding specificity. In contrast, sequence spe­ StTGA2.1 has a short N terminus with low helical propensity cificity is provided by hydrophobic contacts between the due to the Pro25 α-helix breaker, harboring only the basic re­ StTGA2.2 Ala172 residue and the DNA T2 methyl group or be­ sidues Arg20, Arg30, and Arg24, and lacking the first and tween Ala171 and T-4 or T-5. Indeed, most of the predicted most of the second leucine zipper heptad (Figure 5, A and TGA-binding motifs (Figure 4A) have a thymine or adenine B). The conserved hydrophobic residues Leu34, Val31, and in these positions. While the contacts involving StTGA2.2 in Phe38 may contribute to protein dimerization by forming the StTGA2.1-StTGA2.2 heterodimer are highly preserved, a partial zipper that could be stabilized by the hydrophobic the StTGA2.1–DNA interactions are dramatically reduced, Val28 and Val29. Persistent contacts identified in MD simu­ with only Arg11, Arg20, and Arg24 forming persistent con­ lations of StTGA2.2 and StTGA2.3 homo- and heterodimers tacts with the DNA phosphate groups (Figure 5D, are preserved in StTGA2.1 and involve the said hydrophobic Supplemental Table S9). Moreover, the StTGA2.1 partial zip­ residues (Figure 5B, Supplemental Figure S10). per positions the last basic residue (Arg24) more distant to the We then inquired if the StTGA2.1 N terminus binds to DNA. DNA compared to StTGA2.2 (Lys180), breaking the protein– Based on prior knowledge of AtTGA2 and its cognate TGACG DNA complex symmetry. motif (Boyle et al., 2009), we modeled the DNA-bound Another important singularity of StTGA2.1 is that its par­ StTGA2.2-StTGA2.2 and StTGA2.2-StTGA2.1 dimers and, via tial zipper connects directly to the putative DOG1 domain, The role of mini-TGAs in plant immunity PLANT PHYSIOLOGY 2023: 191; 1934–1952 | 1941 Downloaded from https://academic.oup.com/plphys/article/191/3/1934/6903602 by guest on 26 May 2023 Figure 4 StTGA2.1, together with its interactor StTGA2.3, activates expression of StPRX07. A, TGA-binding motifs in selected StPRX promoters. The predicted motifs are boxed and the nucleotides, differing from the core TGACG(T) sequence, its reverse complement or the TGACGTCA palin­ drome, are underlined. Numbers indicate position upstream of transcription start site. B, Surface plasmon resonance results, showing the interaction between StTGA2.3 and chip-immobilized PRX15p_1, PRX46p, PRX07p_1, or PRX07p_2 DNA fragments, bound to the chip at ∼38, 41, 65, or 53 response units (RU), respectively. Representative sensorgrams are shown. C, Transactivation assay results, showing in planta StPRX07 promoter ac­ tivation by GFP-tagged StTGA2.1 (light orange), BFP-tagged StTGA2.3 (dark orange) or a combination of both (black). BFP or GFP-tagged controls and their combination (control) were used to detect the basal promoter activity (grey). Average values ± SE of 18 biological replicates in the first 10 h of measurement are shown. Significance (P < .05) was determined using a two-tailed t test and is shown below the response curve for GFP-tagged StTGA2.1 (light orange), BFP-tagged StTGA2.3 (dark orange), or a combination of both (black) compared with control. The experiment was repeated twice with similar results (Supplemental Figure S8C). RLU, relative light units. while the two domains are connected by a 13 aa peptide link­ Discussion er in StTGA2.2 and StTGA2.3. This may greatly influence its TGAs are involved in modulation of various cellular processes, interdomain conformational flexibility. Pro51 in StTGA2.1 acting as positive or negative regulators of gene expression (Ala227 and Ala110, in StTGA2.2 and StTGA2.3, respectively), (Gatz, 2013). Their structural features provide the basis for disrupts the DOG1 α-helix 1 (α1) and forms a shorter, disor­ their functional variability, defining their subcellular localiza­ dered linker that could somehow compensate for this ab­ tion, target recognition, DNA-binding, as well as their ability sence. These results indicate that the compact molecular to form dimers, oligomers, or interact with other proteins architecture of StTGA2.1, which causes an asymmetric distri­ (Tomaž et al., 2022). In Arabidopsis, all 10 AtTGAs share a bution of basic residues in the StTGA2.1–StTGA2.2 heterodi­ highly conserved bZIP domain, essential for establishing spe­ mer, substantially distorts the DNA conformation near the cific interactions with DNA. Here, we report on the structural- binding site, supporting strong promoter activation upon functional relationship of a mini-TGA from potato, StTGA2.1, binding of the heterodimer compared with binding of the which lacks a full bZIP but still acts at target gene activation. homodimer. Mini-TGAs were identified in potato (this study), tomato (Hou et al., 2019; Lemaire-Chamley et al., 2022), and straw­ berry (Feng et al., 2020), but not in Arabidopsis. 1942 | PLANT PHYSIOLOGY 2023: 191; 1934–1952 Tomaž et al. Downloaded from https://academic.oup.com/plphys/article/191/3/1934/6903602 by guest on 26 May 2023 Figure 5 Comparative structural analysis and simulations of StTGA2.1 N terminus interactions with StTGA2.2 and StTGA2.3 bZIP domains. A, Protein sequence alignment of StTGA2.1 N terminus with AtTGA2, StTGA2.2, and StTGA2.3 bZIP domains. The basic region (orange box) and the leucine zipper heptads (grey dashed boxes) are indicated. Conserved amino acids, in respect to the consensus sequence, are marked with dots. StTGA2.1 contains hydrophobic residues (Val31 and Phe38) in two out of three Leu positions in the heptads (grey) and has a completely conserved third heptad. B, Molecular architectures of StTGA2.1 (orange) and StTGA2.3 (green). The StTGA2.3 bZIP domain (aa 40–95), and StTGA2.1 N terminus (aa 1–45) are shown. The C-terminal region is highly conserved between StTGA2.1 (aa 47–240), StTGA2.2 (aa 222–446), and StTGA2.3 (aa 105–327). Helices in the C-terminal part are labelled (α1–α6). Amino acid residues that are discussed in this study are represented as liquorice and labelled for StTGA2.1 (orange) and StTGA2.3 (green). Those forming persistent contacts in the leucine zipper, according to MD simulations, are shown in bold (green). Basic amino acid residues that may contribute to DNA-binding, in the regions aa 40–63 of StTGA2.3 and aa 20–28 of StTGA2.1, are depicted and labelled. Fully nonconservative substitution sites in the putative DOG1 domains are also represented as liquorice. C, Representative snapshot of the MD simulations of the DNA-bound StTGA2.2 homodimer and D, StTGA2.2-StTGA2.1 heterodimer. The DNA double helix is represented in violet. The DNA sequence is shown at the bottom and the binding site core is underlined. A dot is used as a reference at the center of the sequence for numbering the nucleotide residues. StTGA2.1 (orange) and StTGA2.2 (cyan) are represented as cartoon. Salt bridges and hydrogen bonds between protein and DNA are indicated with black dashed lines. Hydrophobic contacts are indicated as dotted lines. Amino acid residues forming persistent interactions are labelled in bold. The presence of a hexacoordinated Mg2+ was assumed (yellow sphere), based on its importance for CREB-bZIP (Moll et al., 2002). Homo- and heterodimerization of StTGA2.1 (Figure 3, NPR1-AtTGA3 complex revealed the formation of stable A and B, Supplemental Figure S3), which contains only a homodimer contacts between two AtTGA3 C-terminal re­ part of an already short and presumably unstable TGA zipper gions (Kumar et al., 2022). StTGA2.1 nuclear localization al­ region (Deppmann et al., 2004), corroborates the findings of lows its role in gene regulation; however, its unusually Boyle et al. (2009), who established that the leucine zipper is broad localization pattern (Figure 3C, Supplemental not crucial for dimerization of AtTGA2 (Boyle et al., 2009). Figure S5) suggests that it might perform different tasks, as Instead, StTGA2.1 dimerization is likely mediated by interac­ has been shown for other plant transcription factors with in­ tions involving its C-terminal region, previously reported to trinsically disordered regions (Powers et al., 2019). TGAs also contain a dimer stabilization region (Katagiri et al., 1992). interact with different proteins present in both nuclei and In line with this, the recent cryo-EM structure of an cytoplasm, including NPR cofactors (Boyle et al., 2009; Ding The role of mini-TGAs in plant immunity PLANT PHYSIOLOGY 2023: 191; 1934–1952 | 1943 et al., 2018), which was already confirmed for StTGA2.1 (Ramšak et al., 2014) contain predicted secretory signal peptides (Supplemental Figure S4), glutaredoxins (Li et al., 2009), (Supplemental Table S6), indicating StTGA2.1 could affect apo­ and calmodulins (Popescu et al., 2007). Thus, StTGA2.1 could plastic ROS production in plant defense. modulate the function of these partners by sequestering Transcription factor cooperativity is essential in eukaryotic them into inactive complexes, as was proposed for transcription regulation and can arise through various mechan­ AtTGA2 and NPR1 (Fan and Dong, 2002). isms, involving protein–protein and/or protein–DNA interac­ Multiple functions of StTGA2.1 are supported also by the tions (Morgunova and Taipale, 2017). For example, the diversity of detected transcriptional changes following expos­ interaction between two NPR1 proteins bridges two AtTGA3 ure to pathogen infection (Table 1, Supplemental Tables S3 homodimers bound to separate DNA-binding motifs, creating and S5). StTGA2.1 may directly affect the activity of several a complex with an AtTGA32-NPR12-AtTGA32 stoichiometry transcriptional regulators and its effects on gene expression (Kumar et al., 2022). Previous studies have shown that TGA mu­ are likely amplified by further secondary regulation. Thus, tants, impaired in DNA-binding through diverse modifications we hypothesize StTGA2.1 is an important player in shaping of the bZIP domain, prevent DNA-binding of wild type homo­ Downloaded from https://academic.oup.com/plphys/article/191/3/1934/6903602 by guest on 26 May 2023 the transcriptional landscape during infection. Interestingly, logs (Rieping, 1994; Niggeweg et al., 2000; Pontier et al., 2001), our results show StTGA2.1 improves potato immunity even which somewhat opposes the cooperative activation via an in the absence of SA, even though clade II TGAs are known StTGA2.1-StTGA2.3 complex. Compared to homodimers of its for their regulatory role in the SA pathway (Zhang et al., putative paralogs, our MD simulations suggest that the asym­ 2003; Zhou and Zhang, 2020). Our results thus indicate that metrical distribution of basic residues in the bZIP-like domain the TGA mechanism of action in potato may differ substantial­ in the StTGA2.1-StTGA2.2 heterodimer substantially distorts ly from that in Arabidopsis. It is important to note that the DNA conformation near its binding site (Figure 5). We hy­ StTGA2.1 overexpression in SA-deficient plants may not reflect pothesize that StTGA2.1 dramatically affects the overall con­ its function in nontransgenic potato completely, as several formation of the regulatory complex due to its compact parts of immune signaling are under the control of SA and molecular architecture. are thus not fully functional in our transgenic plants. We In conclusion, we show that, although mini-TGAs are not have, however, focused on overexpression in a SA-deficient able to bind DNA on their own, their unusual structure sup­ background to check whether resistance towards PVY can ports diverse functionalities, such as allowing the induction be improved. Overexpression of StTGA2.1 in a nontransgenic of class III peroxidases in immune signaling. We thus provide background would not result in observable phenotypic differ­ evidence that truncation in evolution of genes does not ne­ ences, because the plants are already fully resistant to PVY. cessary lead to a loss-of-function phenotype. Instead, add­ Furthermore, as we did not find StTGA2.1 transcriptionally itional functions can be attained. Through this, we shed regulated during PVY infection in nontransgenic plants additional perspective on immune signaling in nonmodel (Supplemental Tables S2, S4, and S7), the preparation of knock­ species, as Arabidopsis does not encode such proteins. out nontransgenic plants is unlikely to affect plant immunity. Several studies have evaluated the influence of clade II TGA dominant-negative bZIP mutants on plant immunity during bacterial infection, leading to contradictory results (Niggeweg Materials and methods et al., 2000; Pontier et al., 2001; Fan and Dong, 2002). Reactive In silico sequence and structural analysis oxygen species (ROS) act as signaling molecules in biotic stress TGA transcription factor orthologs from potato (Solanum tu­ (Bleau and Spoel, 2021). Early ROS production is central to plant berosum) were identified based on ortholog information in­ defense and TGAs have previously been associated with cellular cluded in the GoMapMan database (Ramšak et al., 2014). The redox control, physically interacting with or regulating the ex­ initial list was further pruned based on protein sequence pression of CC-type glutaredoxins (Ndamukong et al., 2007; alignments created with Geneious Alignment in Geneious Li et al., 2009; Hou et al., 2019). Furthermore, clade II TGAs Prime 2020.1.1 (https://www.geneious.com/) and BLAST re­ modulate the expression of glutathione-S-transferases in sults to exclude technical errors of ortholog detection and se­ ROS-processing responses to UV-B stress (Herrera-Vásquez quencing. Identified StTGAs are listed in Supplemental et al., 2021), while clade IV AtTGAs are regulated by Table S1. Basic protein information was calculated using flg22-induced ROS production (Noshi et al., 2016). Here we the ExPASy ProtParam tool (Gasteiger et al., 2005). Protein show that the synergistic activity of TGAs regulates the expres­ domain prediction was performed with ExPASy Prosite (de sion of yet another group of enzymes involved in Castro et al., 2006). Protein sequences of SlTGAs (Hou ROS-metabolism, the class III peroxidases (Figure 4, B and C, et al., 2019; Lemaire-Chamley et al., 2022) were retrieved Supplemental Figure S8). Class III peroxidases are heme- from the Sol Genomics Network (Fernandez-Pozo et al., containing glycoproteins, secreted to the apoplast or localized 2015) and sequences of AtTGAs from The Arabidopsis in vacuoles (Almagro et al., 2009). Among them, AtPRX33 Information Resource (Berardini et al., 2015). and AtPRX34 proved vital for apoplastic ROS production in re­ For the phylogenetic analysis, the sequences were aligned sponse to flg22 and elf26 (Daudi et al., 2012). Most of the StPRX with MAFFT (Katoh and Standley, 2013), using the L-INS-I itera­ protein sequences from the peroxidase functional group tive refinement method, and the alignment used for a 1944 | PLANT PHYSIOLOGY 2023: 191; 1934–1952 Tomaž et al. maximum-likelihood phylogenetic tree construction in respectively. CHARMM-formatted topology and parameter MEGA-X (Kumar et al., 2018), using the Jones-Taylor-Thorton files were converted to GROMACS input files using the VMD matrix-based model (Jones et al., 1992) and 1,000 bootstrap re­ plugin TopoGromacs (Vermaas et al., 2016). petitions. The rice (Oryza sativa) OsTGA2.1 (Chern et al., 2001) The MD simulations were performed on the Summit protein sequence (Q7X993) was retrieved from UniProtKB supercomputer at the Oak Ridge Leadership Computing (https://www.uniprot.org/) and used as tree root. For sequence Facility. Energy minimization was performed for all systems similarity visualization, the protein sequences were aligned with with steepest descent. Periodic electrostatic interactions Geneious Alignment in Geneious Prime 2020.1.1 (https://www. were treated with the particle mesh Ewald method geneious.com/). Potato peroxidases were identified with pro­ (Darden et al., 1993). LINCS (Hess et al., 1997) was used to tein sequence BLAST against the RedoxiBase database (Savelli constrain bonds involving hydrogen atoms. et al., 2019) and the secretory signal peptides were predicted Similar protocols of simulation were applied for the free and with SignalP 5.0 (Almagro Armenteros et al., 2019) DNA-bound TGA dimers. Preceding the classical simulations, (Supplemental Table S6). Predictions of transcription factor we performed long equilibration runs of 315 ns as part of our Downloaded from https://academic.oup.com/plphys/article/191/3/1934/6903602 by guest on 26 May 2023 binding motifs in promoter sequences were performed with protocol adapted from the MD simulation-based method of TRANSFAC (Matys et al., 2006) and predictions of transcription structural refinement described by Heo et al. (2021). In this start sites with TSSFinder (de Medeiros Oliveira et al., 2021). protocol, potential sampling is accelerated with hydrogen Structural models of StTGA2.1, StTGA2.2, and StTGA2.3 mass repartitioning and by applying fairly high temperatures. were generated with AlphaFold (Jumper et al., 2021). The Weak position-restraint potentials were applied for minimum top-ranked models were selected. The VMD (Visual bias and to compensate for the high thermal energy. Velocity Molecular Dynamics, version 1.9.4a48) molecular visualiza­ Langevin dynamics was performed using a friction constant of tion program was used for visual analysis and structural align­ 1 ps−1. During the equilibration phase, position restraints ap­ ment of protein models. plied to Cα atoms in the leucine heptads were gradually re­ leased and the temperature gradually increased, reaching MD simulations the maximum of 360 K (Supplemental Table S10). After The initial homo- and heterodimeric configurations of long sampling at 340 and 320 K, a final phase of equilibration StTGA2.1-StTGA2.2, StTGA2.2-StTGA2.2, and StTGA2.2-StT is conducted at 298.15 K, the temperature of the following GA2.3 N-terminal fragments were defined using the crystal production runs. During the final equilibration phase, flat- structure of CREB-bZIP-CRE (PDB id: 1DH3) as template bottom harmonic-restraint potentials were applied, using a (Schumacher et al., 2000). Corresponding amino acid changes force constant of 0.25 kcal/mol/Å2 and a flat-bottom width to the template were done using the VMD psfgen plugin, pre­ of 4 Å. To adjust box size, part of the equilibration phase serving the coordinates of the backbone and Cβ atoms. StT was conducted in the NpT ensemble, using the Berendsen GA2.1, StTGA2.2, and StTGA2.3 are truncated, keeping the barostat (Berendsen et al., 1984) applying a compressibility amino acids 1–43, 159–206, and 42–89, respectively. The N of 4.5 × 10−5 bar−1 and a time constant of 1.0 ps. In the final termini of StTGA2.2 and StTGA2.3 are capped with phase of equilibration, the atomic velocities were assigned N-methylamide and the C termini of the three proteins from a Maxwell–Boltzmann distribution using random num­ with acetyl. For simulations of DNA-bound StTGA2.1-StT bers of seed. The production runs of free and DNA-bound di­ GA2.2 and StTGA2.2-StTGA2.2, the DNA fragment from the mers consisted of five unbiased independent simulations of template crystal structure was kept and the nucleotides 128 and 200 ns, respectively. The position-restraint potential were modified using psfgen. The final DNA sequence corre­ applied to Mg2+ and its coordinated water molecules was kept sponds to the TGACGT motif, complementary to the linker during these simulations. scan 5 element and its adjacent regions of Arabidopsis For simulation analysis, the VMD plugin Hbonds was used (Arabidopsis thaliana) PR-1 promoter as-1-like sequence to count hydrogen bonds formed during the production (Lebel et al., 1998) (Figure 5, C and D). The crystal Mg2+ cation runs. The geometric criteria adopted are a cut-off of 3.0 Å and the six coordinated water molecules were kept. for donor–acceptor distance and 20° for acceptor–donor-H GROMACS-2020 (Abraham et al., 2015) was used to prepare angle. In Figure 5, C and D, salt bridges and hydrogen bonds inputs and run MD simulations. The simulation boxes were gen­ between protein–DNA occurring during more than 10% of erated as an octahedron, defining a solvation layer of 10 Å min­ the simulation time are shown. Persistent contacts were imum thickness around the molecular complex. NaCl of 0.15 M identified using the VMD plugin Timeline. In Figure 5, was used to establish electroneutrality. Protonation states were C and D, amino acid residues involved interactions or hydro­ defined for pH 7.0. Amber ff99SB (Lindorff-Larsen et al., 2010) phobic contacts persisting for more than 30% of the simula­ and ff14SB9 (Maier et al., 2015) were used to describe the pro­ tion time are shown. Grace was used for plots (https:// tein in the free and DNA-bound TGA dimers, respectively, and plasma-gate.weizmann.ac.il/Grace/). PARMBSC1 (Ivani et al., 2015) was used to describe the DNA. TIP4P-D (Piana et al., 2015) or simple point charge (SPC) Plant material (Berendsen et al., 1981) was used to describe water molecules Potato nontransgenic cultivar Rywal (NT) and Rywal-NahG in the simulations of the free and DNA-bound TGA dimers, (NahG), a transgenic line impaired in SA accumulation due The role of mini-TGAs in plant immunity PLANT PHYSIOLOGY 2023: 191; 1934–1952 | 1945 to salicylate hydroxylase expression (Baebler et al., 2014), recombined through LR reaction into the pGWB435 were used in this study. Plants were propagated from stem Gateway vector (Nakagawa et al., 2007), as described above, node tissue cultures and transferred to soil 2 weeks after inserting the promoter upstream of a luciferase reporter node segmentation, where they were kept in growth cham­ (LucF). bers under controlled environmental conditions at 22/20°C For recombinant protein production, the StTGA2.1 cds was with a long-day (16 h) photoperiod of light (light intensity inserted into the pMCSG7 bacterial expression vector 4,000 lm/m2) and 60%–70% relative humidity. Nicotiana (Eschenfeldt et al., 2009) by ligation-independent cloning benthamiana plants were grown from seeds and kept in (Aslanidis and Jong, 1990) to produce a protein with an growth chambers under the same conditions. N-terminal hexahistidine (His6) tag. The cds of StTGA2.3 was amplified using primers, enabling the digestion–ligation DNA constructs reaction with the BsaI restriction enzyme. Three silent muta­ Full-length coding sequences (cds) of StTGA2.1, StTGA2.2, tions were introduced into its sequence, to remove two na­ StTGA2.3, StNPR1, and StNPR3/4 were amplified from potato tive BsaI restriction sites. The amplified fragment was Downloaded from https://academic.oup.com/plphys/article/191/3/1934/6903602 by guest on 26 May 2023 cultivar Rywal cDNA and inserted into the pJET1.2/blunt subsequently ligated into the pEPQD0KN0025 acceptor cloning vector using the CloneJET PCR Cloning Kit (Thermo backbone (Addgene plasmid #162283) (Dudley et al., 2021), Scientific, USA), following the manufacturer’s instructions. together with pEPQD0CM0030 (Addgene plasmid The selected genes were subsequently cloned into pENTR #162312) (Dudley et al., 2021), which adds an additional D-TOPO vector using pENTR Directional TOPO Cloning Kit GS peptide to the protein C terminus. (Invitrogen, USA) and recombined through LR reaction using All primer pairs used in the cloning procedure are listed in the Gateway LR Clonase II Enzyme Mix (Invitrogen, USA) into Supplemental Table S11. Sequence verification was per­ several Gateway destination vectors (VIB, Belgium). For formed with Sanger sequencing (Eurofins Genomics, coimmunoprecipitation experiments, localization studies, Germany). and transactivation assays, the StTGA2.1, StTGA2.2, and StTGA2.3 cds were inserted into pH7FWG2 and pJCV52 ex­ Transient expression assays pression vectors (Karimi et al., 2002) to produce proteins Homemade electrocompetent Agrobacterium tumefaciens with C-terminal enhanced GFP and hemagglutinin A (HA) GV3101 cells were transformed with prepared constructs fusions, respectively. For transactivation assays, StTGA2.3 by electroporation. Transformants were used for agroinfiltra­ cds was fused with the mTagBFP2 cds (from Addgene plas­ tion of the bottom three fully developed leaves of 3–4- mid # 102638) (Stark et al., 2018), to produce a protein weeks-old N. benthamiana plants, as described previously with a C-terminal blue fluorescent protein (BFP) tag, prior (Lazar et al., 2014). In cases of cotransformation with agro­ to cloning into pENTR D-TOPO vector (Invitrogen, USA) bacteria carrying different constructs, the 1:1 ratio was ap­ and subsequently recombined into the pK7WG2 vector plied. An equal volume of agrobacteria carrying p19 (Karimi et al., 2002). A short linker encoding six Gly residues silencing suppressor (kindly provided by prof. Jacek Hennig, was introduced between the StTGA2.3 and BFP sequence. BFP PAS, Poland) was added to the mixture. Agrobacteria carry­ fused with a short sequence encoding an N7 nuclear localiza­ ing p19 only were used as controls. tion signal (Ghareeb et al., 2016) (N7-BFP) was recombined into pK7WG2 (Karimi et al., 2002) as control. Confocal microscopy For overexpression experiments, the StTGA2.1 cds was amp­ Protein fluorescence was visualized 3–5 d after transient N. lified with primers harboring XhoI and SpeI restriction enzyme benthamiana transformation. For protein localization, the cleavage sites and inserted into the pTA7002 vector (Aoyama Leica TCS SP5 laser scanning confocal microscope mounted and Chua, 1997), enabling glucocorticoid-inducible gene ex­ on a Leica DMI 6000 CS inverted microscope with an HC PL pression in planta, through restriction–ligation cloning. FLUOTAR 10× objective or HCX PL APO lambda blue 63.0 For the yeast two-hybrid assays, the cds of StTGA2.1, × 1.40 oil-immersion objective (Leica Microsystems, StTGA2.2, StTGA2.3, StNPR1, and StNPR3/4 were amplified Germany) was used, using the settings described previously and inserted into the pGBKT7 (bait) yeast expression vector (Lukan et al., 2018b). The Histone 2B-monomeric red fluores­ through in vivo cloning with Matchmaker Gold Yeast cent protein 1 (H2B-RFP) nuclear marker (Federici et al., 2012) Two-Hybrid System (Clontech, USA), to produce proteins was used to visualize cell nuclei. For coimmunoprecipitation with an N-terminal Gal4 DNA-binding domain. StTGA2.1, and transactivation assays, the protein fluorescence was con­ StNPR1, and StNPR3/4 were inserted also into the pGADT7 firmed with the Leica TCS LSI macroscope with Plan APO 5× (prey) vector (Clontech, USA), to produce proteins with an and 20× objectives (Leica Microsystems, Germany), using the N-terminal Gal4 activation domain, using the same cloning settings described previously (Lukan et al., 2018a). The green, system. blue or red fluorescent protein fluorescence was excited using Promoter sequences of StPRX07, StPRX15, and StPRX46 488 nm, 405 and 543 nm laser lines, respectively. The emission were amplified from potato cultivar Rywal genomic DNA was measured in the window of 505–520 nm for GFP, 450– and inserted into the pENTR D-TOPO vector (Invitrogen, 465 nm for BFP, 570–630 nm for H2B-RFP and 690–750 nm USA). The StPRX07 promoter sequence was subsequently for autofluorescence. In Figure 3C, gain 550–665 was used 1946 | PLANT PHYSIOLOGY 2023: 191; 1934–1952 Tomaž et al. for measuring RFP emission, gain 500–800 for measuring GFP as described previously (Lukan et al., 2022). Plantlets grown emission and gain 550 for measuring autofluorescence. In on regeneration media plates with hygromycin selection Supplemental Figures S5 and S6, gain 675–920 was used for were sub-cultured in order to generate independent trans­ measuring RFP emission, gain 500–800 for measuring GFP genic lines. Transgenic lines were confirmed with PCR emission and gain 640–740 for measuring autofluorescence. (Supplemental Table S11). Lines 7, 12, and 13 were selected Laser intensity was set to 60% in all figures. The Leica LAS for further analysis. AF Lite software (Leica Microsystems, Germany) was used for image processing. Virus inoculation and plant treatments Three- to four-weeks-old potato plants were inoculated with Yeast two-hybrid assay GFP-tagged infectious PVY clones PVYN605-GFP (Rupar et al., Bait (containing StTGA2.1, StTGA2.2, StTGA2.3, StNPR1 or 2015) or PVYN605(123)-GFP (Lukan et al., 2022) or mock in­ StNPR3/4 cds), and prey (containing StTGA2.1, StNPR1 or oculum, as described previously (Baebler et al., 2009). To in­ StNPR3/4 cds) construct combinations were transformed Downloaded from https://academic.oup.com/plphys/article/191/3/1934/6903602 by guest on 26 May 2023 duce gene overexpression, plants were treated with DEX into the Y2H Gold strain using the Matchmaker Gold foliar spray solution containing 30 μM DEX and 0.01% (v/v) Yeast Two-Hybrid System (Clontech, USA) and the transfor­ Tween-20 or control spray solution without DEX (control), mants selected on control SD media without Leu and Trp 3 h prior to virus inoculation, 3 h after virus inoculation, (-L-W). Interactions were analyzed on selection SD media and every day post inoculation until sampling. without Leu, Trp, His and adenine, with added X-α-Gal and Aureobasidin A (-L-W-H-A + Xgal + Aur). The proteins were tested for autoactivation through cotransformation of bait Gene expression analysis with RT-qPCR constructs with an empty prey vector. To evaluate the For gene expression analysis, total RNA isolation and RT-qPCR strength of interaction, saturated yeast culture dilutions were performed as described previously (Baebler et al., 2014). (10−1, 10−2 and 10−3) were spotted onto selection media. DEX-induced StTGA2.1 overexpression in fully developed To evaluate the effect of SA on the strength of interaction, leaves of TGA2.1-NahG transgenic lines was confirmed 3 h the dilutions were spotted onto selection media containing after DEX treatment using a RT-qPCR assay targeting 0.1 mM or 1.0 mM SA. StTGA2.1 cds. The leaves of three DEX-treated plants and two or three nontreated plants were sampled, one leaf per Coimmunoprecipitation assay plant. For PVY abundance analysis, PVY-infected leaves of HA or GFP-tagged StTGA2.1, StTGA2.2 and StTGA2.3 were tran­ DEX-treated TGA2.1-NahG, NahG and NT genotypes were siently expressed in N. benthamiana leaves in different combina­ sampled at 3, 5, and 7 dpi or 3, 7, and 10 dpi. For each genotype tions. The empty pB7WGF2 vector (Karimi et al., 2002), and treatment, three plants were analyzed, sampling one leaf expressing the GFP protein, was used as control. The fluores­ per plant per dpi. PVY abundance and StTGA2.1 expression cence of GFP and GFP-tagged proteins was confirmed with con­ were quantified using two sample dilutions and a relative focal microscopy after 4 d. Total proteins were extracted from standard curve method by normalization to the endogenous ∼500 mg leaf material with immunoprecipitation (IP) buffer, control StCOX1 with quantGenius (http://quantgenius.nib.si) containing 10 mM Tris-HCl, pH 7.5, 150 mM NaCl, 2 mM (Baebler et al., 2017). A two-tailed t test was used to compare MgCl2, 1 mM dithiothreitol and 1x EDTA-free Protease treatments, when applicable. The RT-qPCR analysis was per­ Inhibitor Cocktail (Roche, Switzerland), followed by 1 h incuba­ formed for TGA2.1-NahG transgenic lines 7 and 12. tion with GFP-Trap Magnetic Agarose beads (ChromoTek, RNA sequencing results were validated technically and bio­ Germany) at 4°C. The beads were washed three times with IP logically with RT-qPCR, as described above. For technical valid­ buffer and eluted into SDS-PAGE loading buffer, containing ation, the expression of StACX3, StCS, StPti5, StPRX28, and 100 mM Tris-HCl, pH 6.8, 4% (w/v) SDS, 0.2% (w/v) bromophe­ StTGA2.1 was followed. Biological validation was performed in nol blue, 20% (v/v) glycerol and 200 mM dithiothreitol. The im­ an independent experiment repetition with TGA2.1-NahG munoprecipitated proteins and protein extracts were analyzed transgenic lines 7 and 12, following gene expression of by SDS-PAGE and Western blot, using anti-GFP (diluted StPRX07, StPRX15, StPRX46, StTGA2.1, and PVY. For biological 1:3,000 or 1: 5,000, Invitrogen, USA) and anti-HA (diluted validation, 2–15 early visible lesions and their immediate sur­ 1:1,000, ChromoTek, Germany) antibodies. roundings were sampled from PVY-inoculated leaves of DEX-treated TGA2.1-NahG, NahG, and NT plants at 4 dpi, as de­ Generation of StTGA2.1 overexpression plants scribed previously (Lukan et al., 2020). About 20–30 sections of Transgenic TGA2.1-NahG plants were obtained by stable comparable size were sampled from mock-inoculated leaves as transformation of the Rywal-NahG potato genotype controls. Three plants per genotype per treatment were ana­ (Baebler et al., 2014). Electrocompetent A. tumefaciens strain lyzed, pooling together all lesions or mock sections from one LBA4404 was electroporated with the pTA7002 vector leaf per plant. Total RNA was isolated as described previously (Aoyama and Chua, 1997) carrying the StTGA2.1 cds, as de­ (Lukan et al., 2020). StCOX1 and StEF-1 were used for normaliza­ scribed above. Agrobacteria were used for stable transform­ tion in both cases, as described above. A two-tailed t test was ation of sterile plantlet stem internodes from tissue culture, used to compare treatments, when applicable. The role of mini-TGAs in plant immunity PLANT PHYSIOLOGY 2023: 191; 1934–1952 | 1947 All primers and probes used for RT-qPCR analysis together basecalling was performed using Guppy 4.2.2. The reads with the target gene IDs are listed in Supplemental Table S12. were error corrected with NECAT (Chen et al., 2021) set­ RT-qPCR assays, targeting StPRX07, StPRX15, StPRX46, ting GENOME_SIZE = 100,000,000 and PREP_OUTPUT_ StTGA2.1, and StPti5 were designed with Primer Express COVERAGE = 20,000. Chimeric reads were split using v2.0 (Applied Biosystems, USA), using the sequences from Pacasus (Warris et al., 2018) and all reads designated the potato reference genome (Xu et al., 2011), cultivar as “passed” were mapped to the group Phureja DM1-3 Rywal cds, and cultivar Rywal and cultivar Désirée reference potato genome v6.1 (Pham et al., 2020) using Minimap2 transcriptomes (Petek et al., 2020). (Li, 2018). The obtained BAM file was indexed and sorted using SAMtools (Danecek et al., 2021). Raw Nanopore RNA sequencing analysis reads were deposited at SRA under accession number For RNA sequencing, 2–25 early visible lesions and their im­ PRJNA803339. mediate surroundings were sampled from PVY-inoculated leaves of DEX-treated TGA2.1-NahG line 7, NahG and NT Downloaded from https://academic.oup.com/plphys/article/191/3/1934/6903602 by guest on 26 May 2023 Transactivation assay plants and control-treated TGA2.1-NahG line 7 plants at GFP-tagged StTGA2.1, BFP-tagged StTGA2.3 and their com­ 4 dpi, as described previously (Lukan et al., 2020). About bination were transiently expressed in N. benthamiana leaves, 20–30 sections of comparable size were sampled from mock- with N7-BFP and either a GFP-tagged SNF-related serine/ inoculated leaves as controls. Three plants per genotype per threonine-protein kinase (StSAPK8) or an empty pH7FWG2 treatment were analyzed, pooling together all lesions or vector (Karimi et al., 2002) as controls. Protein fluorescence mock sections from one leaf per plant. Total RNA was iso­ was confirmed with confocal microscopy after 3–5 d. The lated as described previously (Lukan et al., 2020). transactivation assays were performed as described previous­ Strand-specific library preparation and sequencing were per­ ly (Lasierra and Prat, 2018). In brief, 0.5-cm-diameter leaf discs formed by Novogene (HongKong), using the NovaSeq plat­ were sampled at 4 dpi and preincubated in MS liquid media form (Illumina) to generate 150-bp paired-end reads. Read with 35 μM D-luciferin substrate for 4 h before analysis. quality control was performed using FastQC (Andrews, Luminescence was measured in 10 min intervals with 2010). The presence of contaminant organism reads was de­ Centro LB963 Luminometer (Berthold Technologies, termined using Centrifuge (Kim et al., 2016). Reads were Germany). Leaf discs of 17–18 per construct combination mapped to the reference group Phureja DM1-3 potato gen­ were analyzed. A two-tailed t test was used to compare sam­ ome v4.04 (Xu et al., 2011) using the merged PGSC and ITAG ples. The experiment was repeated twice. genome annotation (Petek et al., 2020) and counted using STAR (Dobin et al., 2013) with default parameters. Differential expression analysis was performed in R using Protein production, purification, characterization, the limma package (Smyth et al., 2018). Raw and normalized and antibody preparation read counts as well as a processed data table were depos­ For recombinant production of His6-tagged StTGA2.1, ited at GEO under accession number GSE196078. Genes Escherichia coli BL21(DE3) cells were transformed with the with Benjamini–Hochberg FDR adjusted P values <.05 pMCSG7 vector (Eschenfeldt et al., 2009) carrying the and |log2FC| ≤ −1 were considered statistically significant­ StTGA2.1 cds, grown overnight and subsequently transferred ly differentially expressed. The Venn diagram was drawn, to the liquid autoinduction media (Studier, 2005), where according to results obtained with the Gene List Venn they were incubated for 4 h at 37°C and further 20 h at Diagram tool (http://genevenn.sourceforge.net/). 20°C to produce the protein. Cells were harvested by centri­ Gene Set Enrichment Analysis (Subramanian et al., 2005) fugation, lysed and the protein purified by nickel affinity was performed using nonfiltered normalized counts to search chromatography using the His-Trap HP column coupled for regulated processes and functionally related gene groups, with SEC using the HiPrep 26/60 Sephacryl S-200 column altered significantly by virus inoculation in different geno­ (GE Healthcare Life Sciences, UK). The protein was eluted types (FDR corrected q value <.05) using MapMan ontology into a buffer containing 30 mM Tris, pH 7.5, and 400 mM (Ramšak et al., 2014) as the source of gene groups. NaCl, and used for rabbit polyclonal anti-StTGA2.1 antibody preparation, provided by GenScript (USA). Targeted genomic sequencing The protein oligomeric state was determined based on Genomic DNA was isolated from potato cultivar Rywal leaves SEC elution volume and Gel Filtration LWM Calibration Kit using the DNeasy Plant Mini Kit (Qiagen, Germany). Two sets (standard sizes: conalbumin 75 kDa, ovalbumin 44 kDa, carbon­ of primers were designed to target the region of interest ic anhydrase 29 kDa, ribonuclease A 13.7 kDa and aprotinin (Supplemental Figure S1A, Supplemental Table S13). 6.5 kDa, GE Healthcare, USA). Additionally, the His6-tag was re­ Droplet-based PCR-free target region enrichment, library moved by His6-tagged TEV protease cleavage and a secondary preparation using the SQK-LSK109 kit (Oxford Nanopore nickel affinity chromatography followed by an additional SEC, Technologies, United Kingdom) and long-read sequencing as well as an anion-exchange chromatography purification on the MinION platform using the R9.4.1-type flow cell step. Chemical cross-linking was performed after His6-tag was performed by Samplix (Denmark). Nanopore read removal, for which the protein buffer was exchanged to 1948 | PLANT PHYSIOLOGY 2023: 191; 1934–1952 Tomaž et al. 30 mM Hepes, pH 7.5, 400 mM NaCl using ultrafiltration with injected over DNA at a flow rate of 30 µL/min with an associ­ Amicon Ultra centrifugal filter units (Merck, Germany). The re­ ation time of 120 s and followed by dissociation for 300 s. action was performed using the BS3 crosslinker according to the Regeneration of the sensor surface was performed with manufacturer’s instructions (Thermo Scientific, USA) and the 50 mM NaOH solution for 10 and 300 mM NaCl for 10 s at protein oligomeric state evaluated by SDS-PAGE. a flow rate of 30 µL/min. The sensorgrams for the The E. coli cell-free protein synthesis (CFPS) was used for StTGA2.3 and/or the StTGA2.1 proteins were double sub­ the production of StTGA2.3. All CFPS reactions (total volume tracted for the response of the reference flow cell 1 and for 30 or 75 μL) were performed as described previously (Dudley the response of the CFPS components reference or of the et al., 2021), with 20–24 h incubation at 16, 20, or 25°C. Either running buffer, respectively. the empty pEPQD0KN0025 vector (Addgene plasmid #162283) (Dudley et al., 2021) or water was added to the re­ agent mixture to prepare a CFPS components reference. Accession numbers Proteins were detected by SDS-PAGE and Western blot, using Downloaded from https://academic.oup.com/plphys/article/191/3/1934/6903602 by guest on 26 May 2023 Sequence data from this article can be found in the GenBank anti-StTGA2.1 antibodies (diluted 1:4,000, GenScript, USA). data library under accession numbers: StTGA2.1 cds Additionally, the protein identity was confirmed with mass (OM569617), StTGA2.2 cds (OM569618), StTGA2.3 cds spectrometry, performed at the Department of (OM569619), StNPR1 cds (OM569620), StNPR3/4 cds Biochemistry and Molecular and Structural Biology at the (OM569621), StPRX07 promoter (OM569622), StPRX15 pro­ Jožef Stefan Institute (Slovenia). moter (OM569623), and StPRX46 promoter (OM569624). Surface plasmon resonance Surface plasmon resonance measurements were performed Supplemental data on Biacore T200 (GE Healthcare, USA) at 25°C at the The following materials are available in the online version of Infrastructural Centre for Analysis of Molecular this article. Interactions, University of Ljubljana (Slovenia). To prepare Supplemental Figure S1. Targeted long-read sequencing the DNA, the PRX07p_1, PRX07p_2, PRX15p_1, and confirms the presence of mini-TGAs in potato genome. PRX46p complementary primers (Integrated DNA Supplemental Figure S2. PVY replication in the second sali­ Technologies, Belgium, Supplemental Table S14) were mixed cylic acid-deficient transgenic line overexpressing StTGA2.1. in a 2:3 molar ratio (long:short primers) and annealed by Supplemental Figure S3. Protein interaction analysis cooling the mixtures from 95°C to 4°C. The resulting DNA shows the mini-TGA StTGA2.1 can form homodimers in fragments carried the selected 20 bp promoter regions planta and in vitro. with a 15-nucleotide overhang that allowed hybridization Supplemental Figure S4. Protein interactions between with the complementary biotinylated S1 primer (Caveney StTGAs and StNPR cofactors in yeast. et al., 2019), immobilized on the streptavidin sensor chip Supplemental Figure S5. Diverse localization patterns of (GE Healthcare, USA). StTGA2.3, either alone or premixed StTGA2.1. with a His6-tagged StTGA2.1, protein–DNA-binding experi­ Supplemental Figure S6. StTGA2.2 and StTGA2.3 sub­ ments were performed in a running buffer containing nuclear formations. 25 mM Tris, pH 7.4, 140 mM NaCl, 1 mM MgCl2, and Supplemental Figure S7. RNA sequencing sampling pro­ 0.005% (v/v) P20. For His6-tagged StTGA2.1 DNA-binding as­ cedure and gene expression analysis Venn diagram. says, the running buffer contained 180 mM instead of Supplemental Figure S8. Interaction between StTGA2.1 140 mM NaCl. Flow cell 1 was used as a reference and the or StTGA2.1 premixed with StTGA2.3 and selected DNA fragments were injected across the flow cell 2 at a TGA-binding sites and transactivation assay repetition. flow rate of 5 µL/min to immobilize 42–105 response units. Supplemental Figure S9. Comparative structural analysis A kinetic titration approach was used to study the interac­ and persistent contacts in dimers of StTGA2.1 and StTGA2.2. tions between the CFPS-produced StTGA2.3 protein, the Supplemental Figure S10. Probability density of residues CFPS components reference that lacked StTGA2.3 or the of StTGA2.2 forming contacts with a dimer partner. His6-tagged StTGA2.1 (18.75, 37.5, 75, 150, or 300 nM) and Supplemental Table S1. A list of identified StTGA ortho­ the DNA fragments. The highest concentration of total pro­ logs including basic protein information. tein (264 µg/mL) and four sequential 1.5-fold dilutions were Supplemental Table S2. Differential expression of StTGAs used for the CFPS-produced StTGA2.3 and the CFPS compo­ in NT and NahG genotypes after viral infection. nents reference. The proteins were injected across DNA at five Supplemental Table S3. Differential gene expression in concentrations, with no dissociation time between protein salicylic acid-deficient transgenic plants overexpressing injections, at a flow rate of 30 µL/min. We used the multi cycle StTGA2.1 after viral infection (available as an Excel file). kinetics approach to study the interaction between StTGA2.1 Supplemental Table S4. Technical validation of RNA se­ (300 nM) premixed with CFPS-produced StTGA2.3 (total quencing results with RT-qPCR. protein concentration of 130 µg/mL) with DNA fragments Supplemental Table S5. Enrichment of differentially regu­ PRX07p_1, PRX07p_2, or PRX15p_1. The proteins were lated genes in salicylic acid-deficient transgenic plants The role of mini-TGAs in plant immunity PLANT PHYSIOLOGY 2023: 191; 1934–1952 | 1949 overexpressing StTGA2.1 after viral infection (available as an Department of Energy will provide public access to these re­ Excel file). sults of federally sponsored research in accordance with the Supplemental Table S6. Classification of selected potato DOE Public Access Plan (http://energy.gov/downloads/doe- peroxidases. public-access-plan, last accessed September 16, 2020). Supplemental Table S7. Biological validation of RNA se­ Conflict of interest statement. None declared. quencing results with RT-qPCR. Supplemental Table S8. Interactions between the StTGA2.2 homodimer and DNA in molecular dynamics simulations. References Supplemental Table S9. Interactions between the Abraham MJ, Murtola T, Schulz R, Páll S, Smith JC, Hess B, Lindah E StTGA2.1 in heterodimer and DNA in molecular dynamics (2015) GROMACS: high performance molecular simulations through simulations. multi-level parallelism from laptops to supercomputers. SoftwareX 1–2: 19–25 Supplemental Table S10. Molecular dynamics Downloaded from https://academic.oup.com/plphys/article/191/3/1934/6903602 by guest on 26 May 2023 Almagro L, Gómez Ros LV, Belchi-Navarro S, Bru R, Ros Barceló A, simulations-based protocol for structure refinement of free Pedreño MA (2009) Class III peroxidases in plant defence reactions. J and DNA-bound TGA dimers. Exp Bot 60(2): 377–390 Supplemental Table S11. Primers used for cloning and se­ Almagro Armenteros JJ, Tsirigos KD, Sønderby CK, Petersen TN, quencing (available as an Excel file). Winther O, Brunak S, von Heijne G, Nielsen H (2019) SignalP 5.0 improves signal peptide predictions using deep neural networks. Supplemental Table S12. Primers and probes used for Nat Biotechnol 37(4): 420–423 RT-qPCR analysis. Andrews S (2010) FastQC: a quality control tool for high throughput Supplemental Table S13. Primers used for targeted gen­ sequence data. https://www.bioinformatics.babraham.ac.uk/ ome sequencing. projects/fastqc Supplemental Table S14. Complementary primers used Aoyama T, Chua NH (1997) A glucocorticoid-mediated transcriptional induction system in transgenic plants. Plant J 11(3): 605–612 for preparation of promoter DNA fragments. Aslanidis C, de Jong PJ (1990) Ligation-independent cloning of PCR products (LIC-PCR). Nucleic Acids Res 18(20): 6069–6074 Backer R, Naidoo S, van den Berg N (2019) The NONEXPRESSOR OF Acknowledgments PATHOGENESIS-RELATED GENES 1 (NPR1) and related family: The authors thank Barbara Jaklič, Valentina Levak, Tjaša mechanistic insights in plant disease resistance. Front Plant Sci 10: 102 Mahkovec Povalej, Lidija Matičič, Nastja Marondini, and Baebler Š, Krečič-Stres H, Rotter A, Kogovšek P, Cankar K, Kok EJ, Rebecca Vollmeier for technical support and laboratory as­ Gruden K, Kovač M, Žel J, Pompe-Novak M, et al. (2009) sistance. They also thank prof. Dr. Jim Haseloff and Dr. PVYNTN elicits a diverse gene expression response in different potato Fernán Federici (University of Cambridge, UK) for providing genotypes in the first 12 h after inoculation. Mol Plant Pathol 10(2): the plasmid containing H2B-RFP. Template to amplify N7 263–275 was a kind gift from Prof. Volker Lipka Baebler Š, Svalina M, Petek M, Stare K, Rotter A, Pompe-Novak M, Gruden K (2017) quantGenius: implementation of a decision sup­ (Georg-August-Universität, Göttingen). port system for qPCR-based gene quantification. BMC Bioinformatics 18(1): 276 Baebler Š, Witek K, Petek M, Stare K, Tušek-Žnidarič M, Funding Pompe-Novak M, Renaut J, Szajko K, Strzelczyk-Żyta D, This work was supported by the Slovenian Research Agency Marczewski W, et al. (2014) Salicylic acid is an indispensable compo­ nent of the Ny-1 resistance-gene-mediated response against Potato through the research core funding no. P4-0165, the projects virus Y infection in potato. J Exp Bot 65(4): 1095–1109 no. J4-1777 and J1-2467, the contract no. 1000-22-0105, in ac­ Berardini TZ, Reiser L, Li D, Mezheritsky Y, Muller R, Strait E, Huala cordance with the agreement on (co) financing research ac­ E (2015) The Arabidopsis information resource: making and mining tivity in 2022, as well as by the European Union Horizon 2020 the “gold standard” annotated reference plant genome. Genesis Framework Programme under grant agreement no. 862858, 53(8): 474–485 Berendsen HJC, Postma JPM, Van Gunsteren WF, Dinola A, Haak JR project ADAPT. Funding was also provided by the (1984) Molecular dynamics with coupling to an external bath. J Plant-Microbe Interfaces (PMI) Scientific Focus Area in the Chem Phys 81(8): 3684–3690 Genomic Sciences Program of from the US Department of Berendsen HJC, Postma JPM, Van Gunsteren WF, Hermans J (1981) Energy, Office of Science, Office of Biological and Interaction models for water in relation to protein hydration. In B Environmental Research. This manuscript has been coau­ Pullman, ed, Intermolecular Forces. Reidel, Dordrecht, pp 331–342 Bleau JR, Spoel SH (2021) Selective redox signaling shapes plant- thored by UT-Battelle, LLC under contract no. pathogen interactions. Plant Physiol 186(1): 53–65 DE-AC05-00OR22725 with the US Department of Energy. Boyle P, Le Su E, Rochon A, Shearer HL, Murmu J, Chu JY, Fobert PR, The United States Government retains and the publisher, Despré s C (2009) The BTB/POZ domain of the Arabidopsis disease by accepting the article for publication, acknowledges that resistance protein NPR1 interacts with the repression domain of the United States Government retains a nonexclusive, paid- TGA2 to negate its function. Plant Cell 21(11): 3700–3713 Caveney NA, Pavlin A, Caballero G, Bahun M, Hodnik V, de Castro L, up, irrevocable, world-wide license to publish or reproduce Fornelos N, Butala M, Strynadka NCJ (2019) Structural insights into the published form of this manuscript, or allow others to bacteriophage GIL01 gp7 inhibition of host LexA repressor. Structure do so, for United States Government purposes. The 27(7): 1094–1102.e4 1950 | PLANT PHYSIOLOGY 2023: 191; 1934–1952 Tomaž et al. Chacón-Cerdas R, Barboza-Barquero L, Albertazzi FJ, Gasteiger E, Hoogland C, Gattiker A, Duvaud S, Wilkins MR, Appel Rivera-Méndez W (2020) Transcription factors controlling biotic RD, Bairoch A (2005) Protein identification and analysis tools on the stress response in potato plants. Physiol Mol Plant Pathol 112: ExPASy server. In JM Walker, ed, The Proteomics Protocols 101527 Handbook, Ed 1. Humana Press, Totowa, NJ, pp 571–607 Chen Y, Nie F, Xie SQ, Zheng YF, Dai Q, Bray T, Wang YX, Xing JF, Gatz C (2013) From pioneers to team players: TGA transcription fac­ Huang ZJ, Wang DP, et al. (2021) Efficient assembly of nanopore tors provide a molecular link between different stress pathways. reads via highly accurate and intact error correction. Nat Commun Mol Plant–Microbe Interact 26(2): 151–159 12(1): 60 Ghareeb H, Laukamm S, Lipka V (2016) COLORFUL-circuit: a plat­ Chern MS, Fitzgerald HA, Yadav RC, Canlas PE, Dong X, Ronald PC form for rapid multigene assembly, delivery, and expression in plants. (2001) Evidence for a disease-resistance pathway in rice similar to the Front Plant Sci 7: 246 NPR1-mediated signaling pathway in Arabidopsis. Plant J 27(2): Heo L, Arbour CF, Janson G, Feig M (2021) Improved sampling strat­ 101–113 egies for protein model refinement based on molecular dynamics Danecek P, Bonfield JK, Liddle J, Marshall J, Ohan V, Pollard MO, simulation. J Chem Theory Comput 17(3): 1931–1943 Whitwham A, Keane T, McCarthy SA, Davies RM, et al. (2021) Herrera-Vásquez A, Fonseca A, Ugalde JM, Lamig L, Seguel A, Twelve years of SAMtools and BCFtools. Gigascience 10(2): giab008 Moyano TC, Gutiérrez RA, Salinas P, Vidal EA, Holuigue L Downloaded from https://academic.oup.com/plphys/article/191/3/1934/6903602 by guest on 26 May 2023 Darden T, York D, Pedersen L (1993) Particle mesh Ewald: an N⋅log(N ) (2021) TGA class II transcription factors are essential to restrict oxi­ method for Ewald sums in large systems. J Chem Phys 98(12): dative stress in response to UV-B stress in Arabidopsis. J Exp Bot 10089–10092 72(5): 1891–1905 Daudi A, Cheng Z, O’Brien JA, Mammarella N, Khan S, Ausubel FM, Hess B, Bekker H, Berendsen HJC, Fraaije JGEM (1997) LINCS: a linear Bolwell GP (2012) The apoplastic oxidative burst peroxidase in constraint solver for molecular simulations. J Comput Chem 18(12): Arabidopsis is a major component of pattern-triggered immunity. 1463–1472 Plant Cell 24(1): 275–287 Hou J, Sun Q, Li J, Ahammed GJ, Yu J, Fang H, Xia X (2019) de Castro E, Sigrist CJA, Gattiker A, Bulliard V, Glutaredoxin S25 and its interacting TGACG motif-binding factor Langendijk-Genevaux PS, Gasteiger E, Bairoch A, Hulo N (2006) TGA2 mediate brassinosteroid-induced chlorothalonil metabolism ScanProsite: detection of PROSITE signature matches and in tomato plants. Environ Pollut 255(Pt2): 113256 ProRule-associated functional and structural residues in proteins. Hussain RMF, Sheikh AH, Haider I, Quareshy M, Linthorst HJM Nucleic Acids Res 34(suppl_2): W362–W365 (2018) Arabidopsis WRKY50 and TGA transcription factors synergis­ de Medeiros Oliveira M, Bonadio I, Lie de Melo A, Mendes Souza G, tically activate expression of PR1. Front Plant Sci 9: 930 Durham AM (2021) TSSFinder—fast and accurate ab initio predic­ Ivani I, Dans PD, Noy A, Pérez A, Faustino I, Hospital A, Walther J, tion of the core promoter in eukaryotic genomes. Brief Bioinform Andrio P, Goñi R, Balaceanu A, et al. (2015) Parmbsc1: a refined 22(6): bbab198 force field for DNA simulations. Nat Methods 13(1): 55–58 Deppmann CD, Acharya A, Rishi V, Wobbes B, Smeekens S, Jakoby M, Weisshaar B, Dröge-Laser W, Vicente-Carbajosa J, Taparowsky EJ, Vinson C (2004) Dimerization specificity of all 67 Tiedemann J, Kroj T, Parcy F (2002) bZIP transcription factors in B-ZIP motifs in Arabidopsis thaliana: a comparison to Homo sapiens Arabidopsis. Trends Plant Sci 7(3): 106–111 B-ZIP motifs. Nucleic Acids Res 32(11): 3435–3445 Jin H, Choi SM, Kang MJ, Yun SH, Kwon DJ, Noh YS, Noh B (2018) Ding Y, Sun T, Ao K, Peng Y, Zhang Y, Li X, Zhang Y (2018) Opposite Salicylic acid-induced transcriptional reprogramming by the roles of salicylic acid receptors NPR1 and NPR3/NPR4 in transcrip­ HAC-NPR1-TGA histone acetyltransferase complex in Arabidopsis. tional regulation of plant immunity. Cell 173(6): 1454–1467.e15 Nucleic Acids Res 46(22): 11712–11725 Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, Batut Johnson C, Mhatre A, Arias J (2008) NPR1 Preferentially binds to the P, Chaisson M, Gingeras TR (2013) STAR: ultrafast universal DNA-inactive form of Arabidopsis TGA2. Biochim Biophys Acta RNA-seq aligner. Bioinformatics 29(1): 15–21 1779(10): 583–589 Dudley QM, Cai YM, Kallam K, Debreyne H, Carrasco Lopez JA, Jones DT, Taylor WR, Thornton JM (1992) The rapid generation of Patron NJ (2021) Biofoundry-assisted expression and characteriza­ mutation data matrices from protein sequences. Bioinformatics tion of plant proteins. Synth Biol 6(1): ysab029 8(3): 275–282 Ekengren SK, Liu Y, Schiff M, Dinesh-Kumar SP, Martin GB (2003) Jumper J, Evans R, Pritzel A, Green T, Figurnov M, Ronneberger O, Two MAPK cascades, NPR1, and TGA transcription factors play a Tunyasuvunakool K, Bates R, Žídek A, Potapenko A, et al. (2021) role in Pto-mediated disease resistance in tomato. Plant J 36(6): Highly accurate protein structure prediction with AlphaFold. Nature 905–917 596(7873): 583–589 Eschenfeldt WH, Stols L, Sanville Millard C, Joachimiak A, Donnelly Karimi M, Inzé D, Depicker A (2002) GATEWAYTM vectors for MI (2009) A family of LIC vectors for high-throughput cloning and Agrobacterium-mediated plant transformation. Trends Plant Sci purification of proteins. In SA Doyle, ed, High Throughput Protein 7(5): 193–195 Expression and Purification. Humana Press, Totowa, NJ, pp 105-115 Katagiri F, Lam E, Chua NH (1989) Two tobacco DNA-binding pro­ Fan W, Dong X (2002) In vivo interaction between NPR1 and transcrip­ teins with homology to the nuclear factor CREB. Nature 340(6236): tion factor TGA2 leads to salicylic acid-mediated gene activation in 727–730 Arabidopsis. Plant Cell 14(6): 1377–1389 Katagiri F, Seipel K, Chua N-H (1992) Identification of a novel dimer FAO (2020) World Food and Agriculture—Statistical Pocketbook 2020. stabilization region in a plant bZIP transcription activator. Mol Cell Food and Agriculture Organization of the United Nations, Rome Biol 12(11): 4809–4816 Federici F, Dupuy L, Laplaze L, Heisler M, Haseloff J (2012) Integrated Katoh K, Standley DM (2013) MAFFT Multiple sequence alignment genetic and computation methods for in planta cytometry. Nat software version 7: improvements in performance and usability. Methods 9(5): 483–485 Mol Biol Evol 30(4): 772–780 Feng J, Cheng Y, Zheng C (2020) Expression patterns of octoploid Kesarwani M, Yoo J, Dong X (2007) Genetic interactions of TGA tran­ strawberry TGA genes reveal a potential role in response to scription factors in the regulation of pathogenesis-related genes and Podosphaera aphanis infection. Plant Biotechnol Rep 14(1): 55–67 disease resistance in Arabidopsis. Plant Physiol 144(1): 336–346 Fernandez-Pozo N, Menda N, Edwards JD, Saha S, Tecle IY, Strickler Kim D, Song L, Breitwieser FP, Salzberg SL (2016) Centrifuge: rapid SR, Bombarely A, Fisher-York T, Pujar A, Foerster H, et al. (2015) and sensitive classification of metagenomic sequences. Genome The Sol Genomics Network (SGN)-from genotype to phenotype to Res 26(12): 1721–1729 breeding. Nucleic Acids Res 43(D1): D1036–D1041 The role of mini-TGAs in plant immunity PLANT PHYSIOLOGY 2023: 191; 1934–1952 | 1951 Kumar S, Stecher G, Li M, Knyaz C, Tamura K (2018) MEGA X: mo­ Morgunova E, Taipale J (2017) Structural perspective of cooperative lecular evolutionary genetics analysis across computing platforms. transcription factor binding. Curr Opin Struct Biol 47: 1–8 Mol Biol Evol 35(6): 1547–1549 Nakagawa T, Suzuki T, Murata S, Nakamura S, Hino T, Maeo K, Kumar S, Zavaliev R, Wu Q, Zhou Y, Cheng J, Dillard L, Powers J, Tabata R, Kawai T, Tanaka K, Niwa Y, et al. (2007) Improved Withers J, Zhao J, Guan Z, et al. (2022) Structural basis of NPR1 in Gateway binary vectors: high-performance vectors for creation of fu­ activating plant immunity. Nature 605(7910): 561–566 sion constructs in transgenic analysis of plants. Biosci Biotechnol Lasierra P, Prat S (2018) Transient transactivation studies in Nicotiana Biochem 71(8): 2095–2100 benthamiana leaves. In L Oñate-Sánchez, ed, Two-Hybrid Systems. Ndamukong I, Al AA, Thurow C, Fode B, Zander M, Weigel R, Gatz C Humana Press, New York, NY, pp 311-322 (2007) SA-inducible Arabidopsis glutaredoxin interacts with TGA Lawaju BR, Lawrence KS, Lawrence GW, Klink VP (2018) factors and suppresses JA-responsive PDF1.2 transcription. Plant J Harpin-inducible defense signaling components impair infection by 50(1): 128–139 the ascomycete Macrophomina phaseolina. Plant Physiol Biochem Niggeweg R, Thurow C, Kegler C, Gatz C (2000) Tobacco transcription 129: 331–348 factor TGA2.2 is the main component of as-1-binding factor ASF-1 Lazar A, Coll A, Dobnik D, Baebler Š, Bedina-Zavec A, Žel J, Gruden and is involved in salicylic acid- and auxin-inducible expression K (2014) Involvement of potato (Solanum tuberosum L.) MKK6 in re­ of as-1-containing target promoters. J Biol Chem 275(26): Downloaded from https://academic.oup.com/plphys/article/191/3/1934/6903602 by guest on 26 May 2023 sponse to potato virus Y. PLoS One 9(8): e104553 19897–19905 Lebel E, Heifetz P, Thorne L, Uknes S, Ryals J, Ward E (1998) Noshi M, Mori D, Tanabe N, Maruta T, Shigeoka S (2016) Arabidopsis Functional analysis of regulatory sequences controlling PR-1 gene ex­ clade IV TGA transcription factors, TGA10 and TGA9, are involved in pression in Arabidopsis. Plant J 16(2): 223–233 ROS-mediated responses to bacterial PAMP flg22. Plant Sci 252: Lemaire-Chamley M, Koutouan C, Jorly J, Assali J, Yoshida T, 12–21 Nogueira M, Tohge T, Ferrand C, Peres LEP, Asamizu E, et al. O’Malley RC, Huang SSC, Song L, Lewsey MG, Bartlett A, Nery JR, (2022) A chimeric TGA repressor slows down fruit maturation and Galli M, Gallavotti A, Ecker JR (2016) Cistrome and epicistrome fea­ ripening in tomato. Plant Cell Physiol 63(1): 120–134 tures shape the regulatory DNA landscape. Cell 165(5): 1280–1292 Li H (2018) Minimap2: pairwise alignment for nucleotide sequences. Petek M, Zagorščak M, Ramšak Ž, Sanders S, Tomaž Š, Tseng E, Bioinformatics 34(18): 3094–3100 Zouine M, Coll A, Gruden K (2020) Cultivar-specific transcriptome Li S, Lauri A, Ziemann M, Busch A, Bhave M, Zachgo S (2009) Nuclear and pan-transcriptome reconstruction of tetraploid potato. Sci Data activity of ROXY1, a glutaredoxin interacting with TGA factors, is re­ 7(1): 249 quired for petal development in Arabidopsis thaliana. Plant Cell Pham GM, Hamilton JP, Wood JC, Burke JT, Zhao H, Vaillancourt B, 21(2): 429–441 Ou S, Jiang J, Buell CR (2020) Construction of a chromosome-scale Lindorff-Larsen K, Piana S, Palmo K, Maragakis P, Klepeis JL, Dror long-read reference genome assembly for potato. Gigascience 9(9): RO, Shaw DE (2010) Improved side-chain torsion potentials for giaa100 the Amber ff99SB protein force field. Proteins 78(8): 1950–1958 Piana S, Donchev AG, Robustelli P, Shaw DE (2015) Water dispersion Lukan T, Baebler Š, Pompe-Novak M, Guček K, Zagorščak M, Coll A, interactions strongly influence simulated structural properties of dis­ Gruden K (2018a) Cell death is not sufficient for the restriction of ordered protein states. J Phys Chem B 119(16): 5113–5123 potato virus Y spread in hypersensitive response-conferred resistance Pontier D, Miao ZH, Lam E (2001) Trans-dominant suppression of in potato. Front Plant Sci 9: 168 plant TGA factors reveals their negative and positive roles in plant Lukan T, Machens F, Coll A, Baebler Š, Messerschmidt K, Gruden K defense responses. Plant J 27(6): 529–538 (2018b) Plant X-tender: an extension of the AssemblX system for the Popescu SC, Popescu GV, Bachan S, Zhang Z, Seay M, Gerstein M, assembly and expression of multigene constructs in plants. PLoS One Snyder M, Dinesh-Kumar SP (2007) Differential binding of 13(1): e0190526 calmodulin-related proteins to their targets revealed through high- Lukan T, Pompe-Novak M, Baebler Š, Tušek-Žnidarič M, Kladnik A, density Arabidopsis protein microarrays. Proc Natl Acad Sci U S A Križnik M, Blejec A, Zagorščak M, Stare K, Dušak B, et al. (2020) 104(11): 4730–4735 Precision transcriptomics of viral foci reveals the spatial regulation Powers SK, Holehouse AS, Korasick DA, Schreiber KH, Clark N, Jing of immune-signaling genes and identifies RBOHD as an important H, Emenecker R, Han S, Tycksen E, Sozzani R, et al. (2019) player in the incompatible interaction between potato virus Y and Nucleo-cytoplasmic partitioning of ARF proteins controls auxin re­ potato. Plant J 104(3): 645–661 sponses in Arabidopsis thaliana. Mol Cell 76(1): 177–190.e5 Lukan T, Županič A, Mahkovec Povalej T, Brunkard JO, Kmetič M, Ramšak Ž, Baebler Š, Rotter A, Korbar M, Mozetič I, Usadel B, Juteršek M, Baebler Š, Gruden K (2023) Chloroplast redox state Gruden K (2014) GoMapMan: integration, consolidation and visual­ changes mark cell-to-cell signalling in the hypersensitive response. ization of plant gene annotations within the MapMan ontology. New Phytol 237(2): 548–562 Nucleic Acids Res 42(D1): D1167–D1175 Maier JA, Martinez C, Kasavajhala K, Wickstrom L, Hauser KE, Rieping M, Fritz M, Prat S, Gatz C (1994) A dominant negative mutant Simmerling C (2015) ff14SB: improving the accuracy of protein of PG13 suppresses transcription from a cauliflower mosaic virus 35S side chain and backbone parameters from ff99SB. J Chem Theory truncated promoter in transgenic tobacco plants. Plant Cell 6(8): Comput 11(8): 3696–3713 1087–1098 Matys V, Kel-Margoulis O V, Fricke E, Liebich I, Land S, Barre-Dirrie Rochon A, Boyle P, Wignes T, Fobert PR, Després C (2006) The coac­ A, Reuter I, Chekmenev D, Krull M, Hornischer K, et al. (2006) tivator function of Arabidopsis NPR1 requires the core of its TRANSFAC® and its module TRANSCompel®: transcriptional BTB/POZ domain and the oxidation of C-terminal cysteines. Plant gene regulation in eukaryotes. Nucleic Acids Res 34(suppl_1): Cell 18(12): 3670–3685 D108–D110 Rupar M, Faurez F, Tribodet M, Gutiérrez Aguirre I, Delaunay A, Moll JR, Acharya A, Gal J, Mir AA, Vinson C (2002) Magnesium is re­ Glais L, Kriznik M, Dobnik D, Gruden K, Jacquot E, et al. (2015) quired for specific DNA binding of the CREB B-ZIP domain. Nucleic Fluorescently tagged potato virus Y: a versatile tool for functional Acids Res 30(5): 1240–1246 analysis of plant–virus interactions. Mol Plant–Microbe Interact Moon SJ, Park HJ, Kim TH, Kang JW, Lee JY, Cho JH, Lee JH, Park DS, 28(7): 739–750 Byun MO, Kim BG, et al. (2018) OsTGA2 confers disease resistance Savelli B, Li Q, Webber M, Jemmat AM, Robitaille A, Zamocky M, to rice against leaf blight by regulating expression levels of disease re­ Mathé C, Dunand C (2019) Redoxibase: a database for ROS homeo­ lated genes via interaction with NH1. PLoS One 13(11): e0206910 stasis regulated proteins. Redox Biol 26: 101247 1952 | PLANT PHYSIOLOGY 2023: 191; 1934–1952 Tomaž et al. Schumacher MA, Goodman RH, Brennan RG (2000) The structure of Tomaž Š, Gruden K, Coll A (2022) TGA Transcription factors— a CREB bZIP·somatostatin CRE complex reveals the basis for selective Structural characteristics as basis for functional variability. Front dimerization and divalent cation-enhanced DNA binding. J Biol Plant Sci 13: 935819 Chem 275(45): 35242–35247 Vermaas JV, Hardy DJ, Stone JE, Tajkhorshid E, Kohlmeyer A Smyth GK, Ritchie ME, Law CW, Alhamdoosh M, Su S, Dong X, Tian (2016) TopoGromacs: automated topology conversion from L (2018) RNA-seq analysis is easy as 1-2-3 with limma, Glimma and CHARMM to GROMACS within VMD. J Chem Inf Model 56(6): edgeR. F1000Res 5: 1408 1112–1116 Stark JC, Huang A, Nguyen PQ, Dubner RS, Hsu KJ, Ferrante TC, Warris S, Schijlen E, van de Geest H, Vegesna R, Hesselink T, te Lintel Anderson M, Kanapskyte A, Mucha Q, Packett JS, et al. (2018) Hekkert B, Sanchez Perez G, Medvedev P, Makova KD, de Ridder BioBitsTM Bright: a fluorescent synthetic biology education kit. Sci D (2018) Correcting palindromes in long reads after whole-genome Adv 4(8): eaat5107 amplification. BMC Genomics 19(1): 798 Studier FW (2005) Protein production by auto-induction in high dens­ Xu X, Pan S, Cheng S, Zhang B, Mu D, Ni P, Zhang G, Yang S, Li R, ity shaking cultures. Protein Expr Purif 41(1): 207–234 Wang J, et al. (2011) Genome sequence and analysis of the tuber Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, crop potato. Nature 475(7355): 189–195 Gillette MA, Paulovich A, Pomeroy SL, Golub TR, Lander ES, Zander M, La Camera S, Lamotte O, Métraux JP, Gatz C (2010) Downloaded from https://academic.oup.com/plphys/article/191/3/1934/6903602 by guest on 26 May 2023 et al. (2005) Gene set enrichment analysis: a knowledge-based ap­ Arabidopsis thaliana class-II TGA transcription factors are essential proach for interpreting genome-wide expression profiles. Proc Natl activators of jasmonic acid/ethylene-induced defense responses. Acad Sci U S A 102(43): 15545–15550 Plant J 61(2): 200–210 Thibaud-Nissen F, Wu H, Richmond T, Redman JC, Johnson C, Zhang Y, Fan W, Kinkema M, Li X, Dong X (1999) Interaction of NPR1 Green R, Arias J, Town CD (2006) Development of Arabidopsis with basic leucine zipper protein transcription factors that bind se­ whole-genome microarrays and their application to the discovery quences required for salicylic acid induction of the PR-1 gene. Proc of binding sites for the TGA2 transcription factor in salicylic acid- Natl Acad Sci U S A 96(11): 6523–6528 treated plants. Plant J 47(1): 152–162 Zhang Y, Tessaro MJ, Lassner M, Li X (2003) Knockout analysis of Thurow C, Schiermeyer A, Krawczyk S, Butterbrodt T, Nickolov K, Arabidopsis transcription factors TGA2, TGA5, and TGA6 reveals Gatz C (2005) Tobacco bZIP transcription factor TGA2.2 and related their redundant and essential roles in systemic acquired resistance. factor TGA2.1 have distinct roles in plant defense responses and Plant Cell 15(11): 2647–2653 plant development. Plant J 44(1): 100–113 Zhou JM, Zhang Y (2020) Plant immunity: danger perception and sig­ naling. Cell 181(5): 978–989