THE ROLE OF PHOTORESPIRATION IN PLANT RESPONSE TO HIGH LIGHT AND PATHOGEN INFECTION By Xiaotong Jiang A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Plant Biology – Doctor of Philosophy Molecular Plant Sciences – Dual Major 2024 ABSTRACT Plants interact with the dynamic environment constantly. Photorespiration is a physiological process that occurs simultaneously with photosynthesis and is often considered wasteful. However, photorespiration involves the coordination of multiple organelles and connects with various primary metabolic pathways, thus its response to the environment greatly influences the fate of plants. The roles of photorespiration in plant interaction with the dynamic environment are still not very clear. To investigate the underlying mechanisms, I applied various strategies to dissect the roles of photorespiration in plant response to both abiotic and biotic environmental factors. Previous studies from my lab found that some Arabidopsis photorespiratory mutants only exhibit obvious photosynthetic phenotypes under high and dynamic light conditions, suggesting that photorespiration is regulated by high light. To identify potential modulators of photorespiration under such conditions, I performed a genetic screen for suppressors of the Arabidopsis photorespiratory mutant, hpr1, which is defective in the peroxisomal hydroxypyruvate reductase 1. A suppressor that partially rescued the small rosette of the hpr1 mutant was mapped to GLYR1, which encodes the cytosolic glyoxylate reductase 1 that converts glyoxylate to glycolate. Independent loss-of-function alleles of GLYR1 also recapitulated the partial rescue of hpr1 in plant appearance and photosynthetic and photorespiratory activities. Interestingly, glyr1 also suppressed the phenotypes of the photorespiratory mutant catalase 2, but not a null allele of the PLGG1 (Plastidic Glycolate Glycerate Transporter 1) gene. Further investigations using metabolic and genetic tools provided evidence of a possible cytosolic glyoxylate shunt, which is triggered under high light conditions and in the absence of a properly functional main photorespiratory pathway. This shunt reduces the accumulated cytosolic glyoxylate to hydroxypyruvate, thus helping with carbon recycling through the cytosolic HPR2 enzyme. These findings support the metabolic flexibility of the photorespiration network under stress conditions. My transcriptomic analysis of hpr1 and its suppressor glyr1 hpr1 also supports the existence of this non-canonical cytosolic pathway. Drastic transcriptional reprogramming that involves broad cellular functions was found in the hpr1 mutant, which can be largely reverted by defective GLYR1. The rescuing effect of glyr1 is only prominent when HPR1 is absent, supporting the view that the accumulation of photorespiratory intermediates in hpr1 causes a stressful cellular environment that disrupts biological processes globally, and that glyr1 partially prevents this metabolic accumulation. To investigate the role of photorespiration in plant response to biotic stress, I analyzed the performance of photorespiratory mutants in plant immune response. My data showed that deficiencies of the peroxisomal photorespiratory enzyme HPR1 and the chloroplastic transporter PLGG1 compromised response in both layers of immunity, pattern-triggered immunity and effector-triggered immunity, and these defects can be rescued when the plants were grown under high CO2 conditions. These findings suggest that HPR1 and PLGG1 contribute to plant immune response via the photorespiratory pathway. My research broadens our understanding of the role of photorespiration in plants under stress conditions, which may help with agricultural efforts to improve crop performance in response to the changing environment. ACKNOWLEDGEMENTS First, I really appreciate my mentor, Dr. Jianping Hu, for her generous guidance and support throughout my whole Ph.D. She is always patient, thoughtful, and professional. She gives me a lot of freedom to explore my own interests in research, treats my thoughts and questions seriously, and is always around when I need help. I learned numerous research skills including experiments, writing and presentations from her, which are valuable for me. I always feel lucky to have her as my mentor. Next, I thank my committee members, Drs. Berkley Walker, Sheng Yang He and Gregg Howe, for their technical support, helpful instructions and insightful feedback. Dr. Walker’s expertise in photorespiration is essential for my research, which I heavily rely on in the last 6 years. I’ve learned a lot from Dr. He, who is very experienced in plant immunity and continues supporting me after he left MSU. Dr. Howe also generously shares his thoughts and lab resources all the time. I also thank previous and current members in the Hu Lab. I am happy to have overlaps and communications with many postdocs in the lab, including Drs. Stefanie Tietz, Ronghui Pan, Anne Rea, Linqu Han, Amanda Koenig, and Hiruni Weerasooriya. They all kindly share their knowledge and experiences of research and life, and provide me with vital help in designing experiments and troubleshooting. I have been obtaining a lot of assistance from undergraduate students and rotating students, including Linnea Hartz, Joy Li, Michael Nikolovski, Krishen Patel, Lillian Bieszke, Asia Hawkins, Carter Pasternak and Emily Schley. I also had many nice interactions with visiting student Katarzyna Krawczyk. Special thanks to lab members who directly contributed to my work: Drs. Stefanie Tietz and Jiying Li for performing EMS mutagenesis and the initial screening; Dr. Amanda Koenig and Joy Li for performing the GLYR1 localization experiment. People from other labs and research facilities also help me a lot. I thank David Hall for performing DEPI experiments; Cassandra Johnny (MSU Mass Spectrometry and Metabolomics Core) and Drs. Xinyu Fu and Kelem Alamrie for technical support in metabolic measurements; Drs. John Froehlich and Sarah Stainbrook for technical support in immunoblot; Nicholas Panchy (MSU Bioinformatics Core) for analyzing the RNA-seq data; Drs. Ludmila Roze and Luke Gregory for technical support in experiments related to enzyme activity; Drs. Li Zhang and Pai Li for technical support in experiments related to immunity; Melinda Frame (MSU Center for iv Advanced Microscopy) for technical support in confocal microscope; Dr. Alex Costa (Università degli Studi Milano) for sharing seeds and constructs of HyPer-sensor; Dr. Brad Day for sharing rps2-101c seeds and Pst DC3000 (avrRpt2) stain; and Jim Klug, Cody Keilen and Nick Deason for growth chamber support. Drs. Qiang Guo, Deserah Strand, Kaila Smith, Hussien Alameldin, Andre Velasquez and many other people in the MSU community all contributed to my research at a certain point, which I am grateful for. I also appreciate the support from my family and friends over the last 6 years. No matter where they are and if they understand what I am doing, they have been using their own ways to constantly back me up. I always feel power, warmth and love from them, which I am deeply appreciated. I thank my cat Bellatrix for her intimate company during most of my Ph.D., making me feel that I have a home here. Finally, I want to thank myself for persistently overcoming all the difficulties and eventually reaching the destination of the Ph.D. journey. v TABLE OF CONTENTS LIST OF ABBREVIATIONS ....................................................................................................... vii CHAPTER 1. Background .............................................................................................................. 1 1.1 Photorespiration .................................................................................................................... 1 1.2 The role of photorespiration under high light stress ............................................................. 2 1.3 The role of photorespiration in plant immunity .................................................................... 8 1.4 Summary ............................................................................................................................. 16 Figures ...................................................................................................................................... 17 Tables ........................................................................................................................................ 18 CHAPTER 2. A photorespiratory glyoxylate shunt in the cytosol supports photosynthesis and plant growth under high light conditions in Arabidopsis .............................................................. 19 2.1 Introduction ......................................................................................................................... 19 2.2 Results ................................................................................................................................. 22 2.3 Discussions ......................................................................................................................... 29 2.4 Methods .............................................................................................................................. 32 Figures ...................................................................................................................................... 38 CHAPTER 3. Defective GLYR1 largely reverts the broad transcriptional reprogramming of the hpr1 mutant under high light conditions ....................................................................................... 45 3.1 Introduction ......................................................................................................................... 45 3.2 Results ................................................................................................................................. 45 3.3 Discussions ......................................................................................................................... 49 3.4 Methods .............................................................................................................................. 50 Figures ...................................................................................................................................... 52 Tables ........................................................................................................................................ 62 CHAPTER 4. The role of photorespiration in plant immunity ..................................................... 66 4.1 Introduction ......................................................................................................................... 66 4.2 Results ................................................................................................................................. 67 4.3 Discussion and future directions ......................................................................................... 71 4.4 Methods .............................................................................................................................. 72 Figures ...................................................................................................................................... 76 CHAPTER 5. Summary and future perspectives .......................................................................... 86 REFERENCES ............................................................................................................................. 88 APPENDIX ................................................................................................................................. 103 vi LIST OF ABBREVIATIONS 2-OG 2-PG 3-PGA AAN 2-Oxoglutarate 2-Phosphoglycolate 3-Phosphoglycerate Aminoacetonitrile BASS6 Bile acid sodium symporter 6 CAT CEF CFU DAMP DEG dpi/hpi ETI FOC G6P Catalase Cyclic electron flow Colony-forming units Damage-associated molecular pattern Differentially expressed gene Day(s)/hour(s) post infiltration Effector-triggered immunity Fusarium oxysporum f. sp. cucumerinum Glucose-6-phosphate GABA γ-Aminobutyric acid GDC GGAT GLYK GLYR1 GOX HPR HR INH JA MAMP NLR NPQ PGLP Glycine decarboxylase complex Glutamate:glyoxylate aminotransferase Glycerate kinase Glyoxylate reductase 1 Glycolate oxidase Hydroxypyruvate reductase Hypersensitive response Isonicotinic acid hydrazide Jasmonate Microbe-associated molecular pattern Nucleotide-binding/leucine-rich-repeat receptor Non-photochemical quenching 2-PG phosphatase PLGG1 Plastidial glycolate/glycerate transporter 1 PMDH Peroxisomal malate dehydrogenase vii PR1 PRR PS Pst PTI Pathogenesis-related 1 Pattern recognition receptor Photosystem Pseudomonas syringae pv. tomato Pattern-triggered immunity PTM Post-translational modification qE ROS Energy-dependent quenching Reactive oxygen species Rubisco RuBP carboxylase/oxygenase RuBP Ribulose-1,5-bisphosphate SA SCN SGAT SHMT THF Salicylic acid Soybean cyst nematode Serine:glyoxylate aminotransferase Serine hydroxymethyltransferase Tetrahydrofolate viii 1.1 Photorespiration CHAPTER 1. Background Photorespiration, a biological pathway that is closely linked to photosynthesis, is initiated by the oxygenation of ribulose 1,5-bisphosphate (RuBP) catalyzed by ribulose 1,5-bisphosphate carboxylase-oxygenase (rubisco), producing 2-phosphoglycolate (2-PG) which can inhibit cellular functions when accumulated. In this pathway, 2-PG is first dephosphorylated by 2-PG phosphatase (PGLP) to produce glycolate, which is then transported out of the chloroplast by plastidial glycolate/glycerate transporter 1 (PLGG1) and bile acid sodium symporter 6 (BASS6). Upon entering the peroxisome, glycolate is converted to glyoxylate by glycolate oxidase (GOX), producing H2O2 that is then scavenged by catalases (CATs). Both glutamate:glyoxylate aminotransferase (GGAT) and serine:glyoxylate aminotransferase (SGAT) catalyze the conversion of glyoxylate to glycine. After transporting to the mitochondrion, glycine is converted to serine by the glycine decarboxylase complex (GDC) and serine hydroxymethyltransferase (SHMT), accompanying the tetrahydrofolate (THF) cycle and releasing CO2 and NH3. Serine is then transported back to the peroxisome, converted to hydroxypyruvate by SGAT, and subsequently reduced to glycerate by hydroxypyruvate reductase 1 (HPR1). HPR2 is an HPR isoform that reduces hydroxypyruvate to glycerate in the cytosol. Finally, glycerate is imported into the chloroplast through PLGG1 and phosphorylated to 3-phosphoglycerate (3-PGA) by glycerate kinase (GLYK) to recycle back to the Calvin-Benson cycle. Photorespiration consumes ATP in the chloroplast and NAD(P)H in the peroxisome and the cytosol, and releases NADH in the mitochondrion (Eisenhut et al., 2019) (Fig. 1.1). Because photorespiration itself, the regeneration of RuBP from 3-PGA, and the re- fixation of released CO2 and NH3 consume massive energy, this pathway significantly reduces photosynthetic efficiency (Walker et al., 2016b). Therefore, the photorespiratory pathway has become a major target for genetic engineering with the goal to increase crop yield (Betti et al., 2016; South et al., 2018). However, photorespiration is essential to C3 plants and even vital for C4 plants such as maize (Zelitch et al., 2009) and Flaveria bidentis (Levey et al., 2019), highlighting its importance to plant survival. Photorespiration has tight connections with nitrogen, sulfate and one-carbon (C1) metabolisms (Shi & Bloom, 2021), and has also been  This chapter is partially adapted from Jiang, X., Walker, B. J., He, S. Y., & Hu, J. (2023). The role of photorespiration in plant immunity. Frontiers in Plant Science, 14(February), 1–8. 1 shown to play roles in plant response to both abiotic (Voss et al., 2013) and biotic stresses (Sørhagen et al., 2013). 1.2 The role of photorespiration under high light stress 1.2.1 Plant response to high light conditions Light is an essential energy source and a critical environmental cue for plants, but its intensity often fluctuates beyond the ranges optimal for plant growth. For example, plants frequently experience high light conditions that exceed 2000 μmol m-2 s-1 on sunny days, posing significant stress to plants in the field (Ort, 2001; Mishra et al., 2012). To cope with this stress, plants have developed responses at various levels, from molecular, cellular, to organismal (Szymańska et al., 2017; Shi et al., 2022a). When light intensities exceed photosynthetic capacity, the generation of reactive oxygen species (ROS) from the photosynthetic apparatus is enhanced (Shi et al., 2022a). In chloroplasts, superoxide (O2 •-) and single oxygen (1O2) produced by Photosystems (PSs) I and II can cause photodamage to protein complexes and subsequently inactivate the electron transport chain (Roach & Krieger-Liszkay, 2014; Krieger-Liszkay & Shimakawa, 2022; Sharma et al., 2023). To mitigate photodamage, the antioxidant system in chloroplasts is activated to scavenge ROS. This system includes nonenzymatic components such as carotenoids, ascorbate, and glutathione, as well as enzymes like catalase and superoxide dismutase (Szymańska et al., 2017; Foyer, 2018; Bassi & Dall’osto, 2021). Plants also regulate photosynthetic electron transport to suppress ROS generation. Energy-dependent quenching (qE), the main and rapidly reversing component of non-photochemical quenching (NPQ), can safely dissipate the excessive energy absorbed by light-harvesting complexes as heat (Ruban, 2016; Bassi & Dall’osto, 2021). Additionally, cyclic electron flow (CEF) around PSI is activated under high light as a photoprotective strategy, increasing proton release into the thylakoid lumen and subsequently prompting ATP synthesis and energy dissipation as qE (Shikanai, 2007; Chaux et al., 2015). Despite their damaging nature, chloroplastic ROS also function as retrograde signaling molecules, transmitting environmental cues to the nucleus to coordinate whole-cell response (Li & Kim, 2022; Foyer & Hanke, 2022). Gene expression analyses have shown dynamic and temporal transcriptional reprogramming under high light and indicated the involvement of hormones, light signaling, metabolites, and developmental processes, in addition to ROS and photosynthesis (Rossel et al., 2002; Kleine et al., 2007; Suzuki et al., 2015; Zhao et al., 2016; 2 Crisp et al., 2017; Huang et al., 2019b). Another protective mechanism employed by the plant at cellular and organismal levels is the reduction of light absorption through processes such as chloroplast and leaf movement (Takahashi & Badger, 2011; Wada, 2013), and anthocyanin accumulation (Zheng et al., 2021; Ma et al., 2021). Overall, plants employ various strategies at different levels to survive high light conditions through complicated responses. A deeper understanding of the underlying mechanisms of these responses will ultimately enable us to develop crops that are more resilient to high light. 1.2.2 The role of photorespiration under high light conditions As a metabolic process closely related to photosynthesis, photorespiration has been shown to support the performance of photosynthesis, especially under stress conditions. Under regular growth conditions, many photorespiratory mutants exhibit compromised photosynthesis and growth in ambient air (Timm & Bauwe, 2013; Timm et al., 2016). Under high CO2 environments where rubisco oxygenation is inhibited, these photorespiratory mutants are largely recovered (Timm & Bauwe, 2013; Timm et al., 2016). Although the exact reasons for these air-grown phenotypes are not clear, the accumulated photorespiratory intermediates in these mutants, such as 2-PG, glyoxylate, and glycerate, can inhibit the activities of photosynthesis-related enzymes (Timm et al., 2016). Additionally, several Arabidopsis photorespiratory mutants, including hpr1, plgg1, cat2 and gox1, display much more pronounced photosynthetic phenotypes under high and dynamic light conditions than low and constant light conditions, highlighting the increased importance of photorespiration under high light (Li et al., 2019a). During high light and other abiotic stresses, increased photorespiratory flux rates and enzyme activities, and accumulations of photorespiratory metabolites are often observed, suggesting that photorespiration plays an important role in maintaining photosynthetic performance and helping with plant survival under these conditions (Muraoka et al., 2000; Voss et al., 2013; Ma et al., 2014; Huang et al., 2014, 2015; Sunil et al., 2019; Osei-Bonsu et al., 2021; Shi et al., 2022b). Under stresses, photorespiration is believed to function as an alternative electron sink that consumes excessive energy produced by the photosynthetic light reactions, preventing over-reduction of the electron transport chain and ROS-induced photodamage (Wingler et al., 2000; Ort & Baker, 2002; Voss et al., 2013; Huang et al., 2019a). However, an 3 alternative perspective suggests that photorespiration facilitates the synthesis of new PSII proteins during the repair of photodamage rather than preventing photodamage itself (Takahashi et al., 2007; Wang et al., 2022), a function not necessarily due to energy consumption in photorespiration. A recent work also challenged the idea of photorespiration as an alternative sink, as it observed decreased PSII efficiency (ΦII) and increased photoinhibition under high O2 conditions that cause increased energy demand of photorespiration, contradicting the protective role of photorespiration (Smith et al., 2023). Further investigations indicated a new role of photorespiration in avoiding substrate limitation in ATP synthesis (Smith et al., 2023). In addition, photorespiration can upregulate CEF and the alternative oxidase pathway in mitochondria to protect photosynthesis indirectly (Sunil et al., 2019). Finally, since photorespiration generates hydrogen peroxide (H2O2), an important signaling molecule, this pathway may directly contribute to signal transduction in high light response (Voss et al., 2013). In summary, photorespiration has a notable role in supporting photosynthesis under high light and other abiotic stresses, but the underlying mechanisms require further investigation. 1.2.3 The role of the hydroxypyruvate reductase (HPR) family in photorespiration and other processes As a major enzyme in photorespiration, HPR catalyzes the reduction of hydroxypyruvate to glycerate (Fig. 1.1). In Arabidopsis, three members of the HPR family have been shown to function in photorespiration (Timm et al., 2008, 2011). The peroxisomal HPR1 plays a major role in reducing hydroxypyruvate with NADH as the co-factor, whereas HPR2 and HPR3 in the cytosol have low activities and prefer NADPH as the co-factor (Timm et al., 2008, 2011; Xu et al., 2018a; Wang et al., 2022). All three enzymes can also reduce glyoxylate using NADH and NADPH, and HPR2 and HPR3 are additionally involved in the tyrosine conversion pathway, catalyzing the reduction of 4-hydroxyphenylpyruvic acid to 4-hydroxyphenyllactic acid (Timm et al., 2008, 2011; Xu et al., 2018a). Based on phylogenetic analysis, HPR1 is present across green plants, HPR2 is conserved in land plants, and HPR3 is angiosperm-specific (Xu et al., 2018a). HPR enzymes play an important role in photosynthesis. The knockout mutant of HPR1 shows compromised photosynthetic performance and growth phenotypes in the air, with additive effects observed in the hpr1 hpr2 double mutant and hpr1 hpr2 hpr3 triple mutant (Timm et al., 2008, 2011). Under high light conditions, hpr1 exhibits stronger phenotypes, including a 4 decrease in PSII efficiency, an increase in NPQ, CEF activation, H2O2 accumulation, and diminished levels of chlorophyll and anthocyanin (Li et al., 2019a; Wang et al., 2022). Further investigation of hpr1 in high light found that 2-PG accumulation inhibited the activity of triose phosphate isomerase, an enzyme of the Calvin-Benson cycle that converts glyceraldehyde 3- phosphate to dihydroxyacetone phosphate, resulting in a cytosolic bypass and glucose-6- phosphate (G6P) shunt in the Calvin-Benson cycle (Li et al., 2019b). The G6P shunt consumes ATP, which triggers high rates of CEF to balance energy demand (Li et al., 2019b). Consistent with the activation of the G6P shunt, increased CO2 release was also observed in hpr1 (Cousins et al., 2011; Timm et al., 2021). However, there are other possible explanations for this extra CO2 release, as the non-enzymatic decarboxylation of hydroxypyruvate and serine consumption through serine decarboxylase also seem to occur in hpr1 (Cousins et al., 2011; Timm et al., 2021). Additionally, HPR1 was found to play a role in maintaining the repair of PSII under HL (Wang et al., 2022). HPR enzymes have a broad influence on the level of metabolites in the plant. Deficiencies in one or more HPRs increase the levels of most photorespiratory intermediates, including glycolate, glycine, serine, hydroxypyruvate and glycerate, and these metabolic phenotypes are affected by photoperiods (Timm et al., 2008, 2011, 2021). Consistent with the impaired photosynthesis, carbohydrate levels are largely decreased in the hpr1 mutant (Timm et al., 2021). Levels for other metabolites related to photorespiration, such as intermediates in the tricarboxylic acid cycle and many amino acids, are elevated or reduced in hpr1 (Timm et al., 2008, 2011, 2021). Interestingly, different from its daytime-dependent accumulation pattern in the wild-type plants, serine was found to be constitutively elevated in the hpr1 mutant, inhibiting the expression of photorespiratory genes and reducing the level of the corresponding enzymes (Timm et al., 2013). Additionally, the accumulated glycolate can replace the bicarbonate ligand in PSII in hpr1, shifting the midpoint potential of the quinone acceptor and reducing 1O2 generation (Messant et al., 2018). HPR1 is closely connected in function with peroxisomal malate dehydrogenase (PMDH). PMDH is a component of the malate valve, a powerful system that transfers reducing equivalents (Selinski & Scheibe, 2019). In photosynthetic tissue, PMDH was believed to provide the NADH required by HPR1 to reduce hydroxypyruvate in the peroxisome (Reumann & Weber, 2006). However, an Arabidopsis mutant lacking both PMDH genes showed only a weak decrease in 5 photosynthesis despite more CO2 release from photorespiration, indicating that hydroxypyruvate reduction is not totally dependent on PMDH (Cousins et al., 2008). Additionally, it was suggested that in germinating seeds, PMDH oxidizes the NADH produced by β-oxidation, which is opposite to its function in photosynthetic tissue, and HPR1 can partially compensate for this role when PMDH is absent (Pracharoenwattana et al., 2010). Other than high light, HPR1 also seems to play a positive role in other abiotic stresses. The level of NADH-dependent HPR is elevated in barley during drought stress (Wingler et al., 1999), and the Arabidopsis hpr1 mutant is more susceptible to drought compared to the wild- type (Li & Hu, 2015). Additionally, defective HPR1 in Arabidopsis leads to decreases in ozone tolerance (Saji et al., 2017). Taken together, HPRs not only are important enzymes in photorespiration, but also show functions in processes related to photosynthesis, primary metabolism, energetics, and stress response. Mutants of the HPRs are valuable tools to examine the role of photorespiration in various pathways. 1.2.4 Regulation of photorespiration As photorespiration is tightly linked to primary metabolic pathways and crucial to plant survival under certain environmental conditions, this pathway is expected to be regulated. Although research in this area is still scarce, some regulatory mechanisms of photorespiration are emerging (Timm & Hagemann, 2020; Timm, 2020; Aroca et al., 2023). Photorespiratory flux is determined by the ratio of CO2 to O2. A low CO2/O2 ratio around Rubisco favors photorespiration, while high CO2 or low O2 concentrations inhibit it (Timm & Bauwe, 2013; Fu et al., 2023). A shift from high to low CO2 levels alters the level of photorespiratory metabolites, mostly inducing their accumulation (Timm et al., 2012; Eisenhut et al., 2017). Surprisingly, this reduction in CO2 does not cause obvious changes in the expression of photorespiratory genes (Pérez-Delgado et al., 2013; Eisenhut et al., 2017), suggesting that photorespiratory enzymes may have high capacities to handle variable fluxes (Timm, 2020). Light seems to regulate photorespiration. Most of the photorespiratory genes are up- regulated by light and during photomorphogenesis, and show diurnal changes in transcript and protein levels (Lutziger & Oliver, 2001; Foyer et al., 2009; Kaur et al., 2013; Timm et al., 2013; Wang et al., 2022). Light-responsive elements were also found in the upstream regions of photorespiratory genes (Aroca et al., 2023). Additionally, light triggers the alternative splicing of 6 HPR in pumpkin, preferentially producing the cytosolic over the peroxisomal localized isoform (Mano et al., 1999), possibly to reduce the level of the hydroxypyruvate that escapes from peroxisomes. Photorespiration can also regulate itself. As mentioned above, high levels of serine disrupt the diel fluctuation of photorespiratory gene expression and their protein levels in Arabidopsis (Timm et al., 2013). In addition, applying glycine to Arabidopsis enhances the accumulation of GDC and SHMT1 transcripts during the day (Timm et al., 2013). Moreover, GDC was suggested to be a key enzyme that controls photorespiratory flux, as overexpressing the H- or L-subunit of GDC decreased the level of photorespiratory intermediates (Timm & Hagemann, 2020). Besides the above-mentioned light-responsive elements in the promoters and the alternative splicing of HPR, more evidence shows that photorespiration genes are regulated by upstream regions and introns. Multiple regulatory elements were found in a nucleosome-depleted region within the Arabidopsis CAT2 promoter, regulating its transcript abundance (Laxa, 2017). The intron within the 5’UTR of Arabidopsis GGAT1 was shown to boost gene expression by recruiting RNA polymerase II (Laxa et al., 2016). Interestingly, the expression of the P-protein of GDC in the C4 species Flaveria trinervia is tissue-specific and regulated by both the two tandem sub-promoters and alternative splicing (Wiludda et al., 2012). Additionally, bioinformatic analysis predicted that upstream regulatory elements and 5’UTR introns are prevalent in Arabidopsis photorespiratory genes (Laxa & Fromm, 2018), suggesting that photorespiration is under active transcriptional and post-transcriptional regulation. Post-translational modifications (PTMs) also occur in photorespiratory proteins. Phosphorylation plays an important role in maintaining enzyme activity or cofactor binding in GOXs (Jossier et al., 2020), SHMT1 (Liu et al., 2019) and HPR1 (Liu et al., 2020). Thioredoxins in mitochondria can modify the redox status of the GDC L-protein, thereby regulating glycine decarboxylation (Reinholdt et al., 2019; da Fonseca-Pereira et al., 2020). Furthermore, many potential sites for phosphorylation, ubiquitination, acetylation, and redox modifications have been identified in photorespiratory enzymes (Hodges, 2022; Aroca et al., 2023), suggesting the possible involvement of additional PTMs in regulating photorespiration. In conclusion, photorespiration is clearly regulated through various mechanisms. Current evidence indicates that photorespiration responds to environmental factors such as CO2, O2 and 7 light, receives feedback from its own metabolites and enzymes, and is regulated at transcriptional, post-transcriptional and post-translational levels. These findings highlight the flexibility and importance of photorespiration, underscoring the need for further research in this important area. 1.3 The role of photorespiration in plant immunity 1.3.1 Plant immune system In nature, plants are constantly exposed to a dynamic external biotic environment, which drives the development of the plant immune system. As the first layer of immunity, elicitors from pathogenic and nonpathogenic microbes, known as microbe-associated molecular patterns (MAMPs), are recognized by plasma membrane-localized receptors known as pattern recognition receptors (PRRs) to activate pattern-triggered immunity (PTI) (Yu et al., 2017). Flg22, a peptide from the conserved domain of the bacterial flagellin, is one of the MAMPs. PTI also comprises plant responses to plant-derived endogenous elicitors generated in response to wounding or infection, such as small peptides and nucleotides, which are called damage- associated molecular patterns (DAMPs) (Yu et al., 2017). During PTI, intracellular signaling, transcriptional reprogramming, and other physiological responses culminate to limit pathogen growth. These events include increases in cytosolic Ca2+ concentration, reactive oxygen species (ROS) burst, and biosynthesis of phytohormones such as salicylic acid (SA) and jasmonate (JA) (Yu et al., 2017). To infect successfully, most pathogens can secrete virulent effectors into plant cells to suppress plant defense (Deslandes & Rivas, 2012). As the second layer of immunity, plants use intracellular nucleotide-binding/leucine-rich-repeat (NLR) receptors to recognize effectors, either directly or indirectly, leading to the activation of effector-triggered immunity (ETI) (Cui et al., 2015). ETI responses are similar to, but stronger than, those of the PTI, and often cause local programmed cell death called the hypersensitive response (HR) (Cui et al., 2015) (Fig. 1.1). Recent studies reveal that PTI and ETI are not simply two independent and distinct pathways but work together to regulate immune responses (Ngou et al., 2021; Pruitt et al., 2021; Yuan et al., 2021).  The majority of this section was published: Jiang, X., Walker, B. J., He, S. Y., & Hu, J. (2023). The role of photorespiration in plant immunity. Frontiers in Plant Science, 14(February), 1–8. Only minor modifications were made from the original publication. 8 The plant immune system appears to take advantage of photorespiration. For example, tightly connected with plant primary metabolism (Shi & Bloom, 2021), photorespiration can provide signals, substrates, or energy for immunity in face of pathogen invasion. In addition, the coupled response of photorespiration to environmental signals like dynamic light intensities (Fu & Walker, 2022) may represent a way for immunity to integrate environmental cues for optimal response. There have been no unequivocal conclusions so far on how the level of photorespiratory enzymes is regulated in response to pathogen infections. Some studies show that photorespiratory genes are generally suppressed by pathogen infection (Zabala et al., 2015; Giraldo – González et al., 2021; Kalapos et al., 2021; Yue et al., 2021), whereas in other studies certain photorespiratory genes show increased expression (Mitsuya et al., 2009; Ahammed et al., 2018; Silva et al., 2023). At the protein level, both up- and down-regulation of the photorespiratory enzymes in presence of pathogens have been observed (Segarra et al., 2007; Zhao et al., 2013; Ma et al., 2020; He et al., 2021). These discrepancies are likely due to the different plant-pathogen systems used and may indicate the complex nature of the response of various photorespiratory genes to stress at the gene expression and protein levels. 1.3.2 Photorespiratory ROS: important players in immune response ROS such as H2O2 are crucial signaling molecules during plant-pathogen interactions (Camejo et al., 2016). Photorespiration is a major source of H2O2 in photosynthetic cells (Foyer et al., 2009), and photorespiratory organelles such as peroxisomes also contain H2O2-scavenging systems such as catalases (see below). Not surprisingly, studies of the roles of photorespiration in plant immunity have been mainly focused on H2O2 (Table 1.1). GOX (Fig. 1.1) contributes to disease resistance through its H2O2-producing capability. GOX-silenced tobacco plants show compromised non-host resistance to bacterial pathogens Pseudomonas syringae pv. tomato (Pst) strain T1, P. syringae pv. glycinea and Xanthomonas campestris pv. vesicatoria, as well as reduced ETI responses to the effector AvrPto (Rojas et al., 2012). Consistently, GOX-deficient Arabidopsis mutants show compromised non-host resistance to P. syringae pv. syringae strain B728A and P. syringae pv. tabaci, and reduced ETI responses to the effectors AvrB and AvrRps4 (Rojas et al., 2012). Null mutants of HAOX (hydroxy-acid oxidase), the enzyme that belongs to the same L-2-HAOX family as GOX (Esser et al., 2014), exhibit gox-like phenotypes in response to pathogens (Rojas et al., 2012). The Arabidopsis gox 9 and hoax mutants also have decreased H2O2 levels after P. syringae pv. tabaci infection, which is independent of the H2O2-producing enzyme, NADPH oxidase (Rojas & Mysore, 2012; Rojas et al., 2012). In addition, reducing GOX2 expression in tomato lowers H2O2 levels in the leaf and increases plant susceptibility to the compatible pathogen Pst DC3000, a phenotype that can be rescued by H2O2 pre-treatment (Ahammed et al., 2018). Similarly, decreases in the level of H2O2 and increases in Pst DC3000 susceptibility were seen after application of isonicotinic acid hydrazide (INH), an inhibitor that blocks the conversion of glycine to serine in photorespiration and suppresses GOX activity (Ahammed et al., 2018). These results suggest that the H2O2 produced by GOX family members is important to immunity. However, silencing GOX1 in rice results in enhanced resistance to the compatible pathogen X. oryzae pv. oryzae (Chern et al., 2013). Additionally, three members from the tobacco GOX family contribute differently to H2O2 levels and defense (Xu et al., 2018b), yet all five members of the Arabidopsis GOX family work additively to increase resistance (Rojas et al., 2012). These inconsistent results regarding the function of different GOX members may be due to distinct plant-pathogen systems utilized and the functional divergence of family members in different plant lineages. The function of the H2O2-scavenging enzyme CAT (Fig. 1.1) in immune response has been investigated extensively. Without pathogen infection, CAT-deficient mutants show SA accumulation, induced expression of the SA-pathway marker gene PR1 (pathogenesis-related 1), cell death, along with H2O2 accumulation in tobacco (Takahashi et al., 1997; Chamnongpol et al., 1998; Mittler et al., 1999) and Arabidopsis (Chaouch et al., 2010; Chaouch & Noctor, 2010). In addition, SA was found to bind to CAT and inhibit CAT activity to increase the level of H2O2 in a variety of plant species (Chen et al., 1993; Sánchez-Casas & Klessig, 1994). The inhibition of CAT activity by SA analogs correlates with the induction of the PR1 gene and plant resistance to tobacco mosaic virus (Conrath et al., 1995). Suppression of CAT2 by SA in Arabidopsis also leads to decreases in auxin and JA biosynthesis (Yuan et al., 2017). This is consistent with the increased biotroph resistance that is dependent on SA and repressed by auxin, and decreased JA- dependent necrotroph resistance in the cat2 mutant (Yuan et al., 2017). This data supports the role of CAT2 as a mediator between SA and auxin/JA signaling pathways in response to different pathogens. CAT2 also seems to connect Ca2+ signaling to the JA pathway, as the calmodulin- binding protein IQM1 (IQ-Motif Containing Protein 1) positively regulates JA biosynthesis by enhancing CAT2 function at both the transcription and enzymatic activity levels (Lv et al., 10 2019). The transcription factor GBF1 (G-box binding factor 1) downregulates CAT2 expression during pathogen response, leading to high H2O2 levels (Giri et al., 2017), reinforcing the view that photorespiratory H2O2, whose level is modulated by CATs, may act as a hub in coordinating defense responses. Moreover, pathogens often target CAT to help with infection, which also suggests the importance of photorespiratory H2O2 in immunity. Effectors from the bacterial pathogen Ralstonia solanacearum (Sun et al., 2017) and the root-knot nematode Meloidogyne incognita (Zhao et al., 2021) inhibit CAT activity via physical interaction with the enzyme, and the 2b protein from the Cucumber mosaic virus induces CAT3 degradation in Arabidopsis (Murota et al., 2017). However, some pathogens seem to regulate the level of CAT positively. For example, the Pepino mosaic virus utilizes Triple Gene Block Protein 1 (TGBp1) to promote the activity of CAT1 and reduce H2O2 levels in tomato (Mathioudakis et al., 2013). Interestingly, the oomycete pathogen Phytophthora sojae has two effectors that interact with CATs and regulate H2O2 homeostasis in opposite directions (Zhang et al., 2015). Evidence suggesting that CAT and GOX act together to regulate H2O2 homeostasis in defense has been reported. Under sub-ambient CO2 conditions, enhanced resistance to the biotrophic oomycete Hyaloperonospora arabidopsidis and high intracellular ROS content were observed in Arabidopsis (Williams et al., 2018). This resistant phenotype is abolished in the gox1 or haox1 mutants under the same low CO2 conditions after pathogen inoculation, and the CAT2 gene is down-regulated by infection (Williams et al., 2018), suggesting that both boosted GOX and suppressed CAT contribute to ROS accumulation. More direct evidence comes from rice, where SA treatment disrupts the physical interaction between GOX and CAT and induces H2O2 accumulation (Zhang et al., 2016). These results suggest that H2O2 homeostasis during plant- pathogen interaction is possibly regulated by the association and disassociation of GOX and CAT. Besides peroxisomal H2O2, mitochondrial ROS can be influenced by photorespiration and are involved in defense as well. The P-protein and H-protein of GDC, the mitochondrial multienzyme complex that catalyzes glycine decarboxylation (Fig. 1.1), are repressed in activity by the victorin toxin produced by the fungus Cochliobolus victoriae (Navarre and Wolpert, 1995). Victorin treatment triggers mitochondrial ROS burst and subsequent apoptotic response in oat, a similar result to that caused by the GDC inhibitor aminoacetonitrile (AAN) (Yao et al., 11 2002). In addition, silencing GDC-T or GDC-P in tobacco suppresses victorin-triggered cell death and ETI response to the effector AvrPto (Gilbert & Wolpert, 2013). Furthermore, the bacterial elicitor harpin also inhibits GDC activity in Arabidopsis, resembling the inhibition by AAN treatment (Cristina Palmieri et al., 2010). Therefore, it is likely that GDC plays a role in reducing the level of ROS during plant-pathogen interaction to avoid damages caused by excess ROS. Moreover, the degradation of SHMT1, the enzyme that synthesizes serine in mitochondrion (Fig. 1.1), can induce mitochondrial ROS accumulation and other defense responses in tobacco and rice plants, conferring broad-spectrum resistance to the rice stripe virus, the rice blast fungus Magnaporthe oryzae, and the bacterial leaf blight pathogen X. oryzae pv. oryzae (Fu et al., 2022). The peroxisomal aminotransferase GGAT, which converts glyoxylate to glycine (Fig. 1.1), is also connected with H2O2. Compared to wild-type plants, the Arabidopsis ggat1 mutant is more resistant to the necrotrophic fungal pathogen Botrytis cinerea and contains lower H2O2 concentrations upon infection, whereas a higher H2O2 level is observed when the plants are uninfected (González‐lópez et al., 2021). How GGAT regulates H2O2 and whether this change in H2O2 levels imposes significant impacts on immune responses remains unknown. The impact of photorespiration on ROS levels may differ among the three photorespiratory organelles during plant-pathogen interactions. In the chloroplast, photorespiration may actually prevent ROS production during plant immune response. As the major source of chloroplastic ROS, the photosynthetic electron transport chain produces excessive reducing equivalents and ATP under stress conditions (Voss et al., 2013). Therefore, photorespiration may function as an alternative sink for these reducing equivalents and ATP to decrease ROS accumulation in the chloroplast and protect photosystems from photodamage (Voss et al., 2013). Meanwhile, it is likely that the high photorespiratory rate under stress conditions enhances H2O2 production in the peroxisome, and increases NADH production by GDC in mitochondria to increase the level of mitochondrial ROS. Nonetheless, these hypotheses remain to be tested under pathogen defense conditions. In conclusion, extensive evidence has demonstrated the key roles of ROS in plant immune response. The level of H2O2 is impacted by photorespiratory enzymes such as GOXs and CATs in peroxisomes and GDC in mitochondria, and potentially other photorespiratory proteins as well. 12 1.3.3 Involvement of photorespiratory metabolites in immunity Photorespiration involves a variety of metabolites connected to several primary metabolic pathways, including photosynthesis, C1 metabolism, amino acid metabolism, and nitrogen and sulfate assimilation (Shi & Bloom, 2021). Metabolite analysis of Arabidopsis suspension cultured cells in which immunity was activated by Pst DC3000, mutant Pst DC3000 (D28E), or flg22, revealed large-scale metabolic changes, including the glyoxylate and dicarboxylate metabolism and the amino acid metabolism that partially overlap with the photorespiratory pathway (Misra et al., 2016). In cucumber, nitrate-induced resistance to the fungus Fusarium oxysporum f. sp. cucumerinum (FOC), along with the accumulation of most of the photorespiratory intermediates except serine, was observed (Sun et al., 2021). As discussed below, specific photorespiratory metabolites have also been shown to be involved in plant- pathogen interactions (Table 1.1). Catalyzing the bidirectional conversion of serine and THF to glycine and 5,10- methylene-THF, the photorespiratory enzyme SHMT (Fig. 1.1) is also a crucial enzyme in C1 metabolism (Hanson & Roje, 2001). GmSHMT08c, which encodes a cytosolic SHMT in soybean, was identified to be a resistant gene to the soybean cyst nematode (Heterodera glycines, SCN) (Liu et al., 2012; Kandoth et al., 2017). The resistance is resulted from two amino acid substitutions in the GmSHMT08c protein that impede THF binding and reduce catalytic activity of the enzyme (Liu et al., 2012; Korasick et al., 2020). GmSHMT08c confers SCN-resistance in soybean roots (Liu et al., 2012), so it is less likely that photorespiration is involved in this resistance. Other members of the GmSHMT family do not seem to function in SCN resistance individually (Lakhssassi et al., 2019). However, considering the probable functional redundancy of the five mitochondrial GmSHMT members, folate metabolism is a possible point at which photorespiration affects plant immunity. Moreover, the Arabidopsis shmt1 mutant exhibits compromised defense responses to both biotrophic and necrotrophic pathogens (Moreno et al., 2005). Silencing tomato SHMT1 dampens resistance to P. syringae independent of H2O2, whereas overexpressing the gene enhances the resistance (Ahammed et al., 2018). Further, Arabidopsis SHMT4 binds to SA (Manohar et al., 2015), and rice SHMT1 interacts with the disease-resistance protein RPM1 (Wang et al., 2021), although their roles in immunity in these contexts have not been shown. Taken together, SHMT plays a role in defense response in several plant species. Except for the potential connection to folate metabolism in 13 soybean, and the mitochondrial ROS triggered by SHMT1 degradation in tobacco and rice, the underlying mechanisms of the SHMTs in immunity are still unknown in most species. The peroxisomal HPR enzyme that converts hydroxypyruvate to glycerate (Fig. 1.1) engages in immunity through photorespiratory metabolites. A soybean HPR interacts with P34, the receptor of the P. syringae elicitor syringolide, and applying glycerate and 3-PGA, products of the HPR-catalyzed reaction and the downstream step, respectively, restrains syringolide- triggered HR (Okinaka et al., 2002). Additionally, the cytosolic Arabidopsis HPR2 protein binds to SA, but evidence for its role in immunity is lacking (Manohar et al., 2015). The role of photorespiration-associated amino acids in plant immunity has been illustrated in several studies. In rice, 18 different amino acids, among which glutamate, glycine and serine are photorespiratory intermediates (Fig. 1.1), can induce systemic resistance against rice blast when individually applied to roots (Kadotani et al., 2016). Soaking tomato fruits in glutamate solution reduces colonization of the fungal pathogen Alternaria alternata and activates several primary metabolic pathways such as nitrogen metabolism, the γ-aminobutyric acid (GABA) shunt, and SA signaling (Yang et al., 2017). Consistently, glutamate can serve as a DAMP to induce Ca2+ signaling and thereafter defense responses in plants (Toyota et al., 2018). Taken together, current data provide evidence for the influence of photorespiratory metabolites on plant defense response. Further and in-depth studies are needed to elucidate the underlying mechanisms. 1.3.4 Influence of photorespiration on the biosynthesis of defense hormones SA and JA are the two major phytohormones in plant defense (Pieterse et al., 2012). SA is synthesized in plastids and in the cytosol (Lefevere et al., 2020), and the biosynthesis and activation of JA involve plastids, peroxisomes and the cytosol (Wasternack & Song, 2017). Recently, CAT2-promoted JA biosynthesis in Arabidopsis was shown to be achieved by the direct interaction between the N-terminus of CAT2 and the JA biosynthetic enzymes acyl-CoA oxidase 2 (ACX2) and ACX3, without the requirement of H2O2 (Zhang et al., 2021). Another study demonstrated that the JA-activated defense to the necrotrophic pathogen Erwinia amylovora is partially dependent on GOX2 and does not involve obvious changes in the level of H2O2 (Launay et al., 2022), indicating that other mechanisms independent of H2O2 may exist in this immune response. Given the overlap of the locations for photorespiration and defense hormone biosynthesis in several subcellular compartments, it is possible that one or multiple 14 photorespiratory enzymes or metabolites serve as mediators or signals in the biosynthesis of SA and JA. Although evidence for the connection between photorespiration and defense hormone biosynthesis is still scarce, it is a promising research direction that merits further investigations. 1.3.5 Other photorespiratory components involved in defense A few other photorespiratory enzymes are also involved in immunity, yet the mechanisms behind are inconclusive (Table 1.1). In a Pseudoperonospora cubensis-resistant melon cultivar, genes encoding two aminotransferases - homologs of the Arabidopsis peroxisomal aminotransferase SGAT, which converts glyoxylate to glycine and serine to hydroxypyruvate (Fig. 1.1), were found among the resistance genes (Taler et al., 2004). Overexpressing either gene confers resistance to the pathogen in the susceptible cultivar (Benjamin et al., 2009). That the resistant melon cultivar also exhibits high GOX activities indicates that this SGAT-regulated resistance may be attributed to high H2O2 levels (Taler et al., 2004). However, the positive role of SGAT in plant resistance to P. syringae in tomato was shown to be independent of H2O2 (Ahammed et al., 2018). Additionally, Arabidopsis SGAT was identified as an SA-binding protein, with unknown consequences in defense (Manohar et al., 2015). Further studies are needed to dissect the precise mechanism of the role of SGAT in immunity. The chloroplast photorespiratory kinase GLYK, which phosphorylates glycerate to make 3-PGA (Fig. 1.1), appears to play a positive role in immunity at multiple levels. Full-length GLYK in potato is a target for the Irish potato famine pathogen Phytophthora infestans effector protein AVRvnt1 through protein binding, resulting in the impediment of GLYK trafficking into chloroplasts and enhancement of GLYK degradation, as well as the activation of the ETI response mediated by Rpi-vnt1.1, the NLR that recognizes AVRvnt1 (Gao et al., 2020). GLYK silencing results in increased plant susceptibility to P. infestans lacking AVRvnt1 via an unknown mechanism (Gao et al., 2020). Interestingly, the full-length GLYK protein is mainly produced under light (Gao et al., 2020), when photorespiration operates, indicating that the function of GLYK in immunity likely depends on photorespiration. 1.3.6 Measurement of photorespiration rate in defense response Measuring physiological parameters of photorespiration in plants after pathogen infection provides new perspectives in dissecting the relationship between photorespiration and defense. Photorespiration rate, which can be estimated by the difference of net CO2 assimilation rate 15 between 2% and 21% O2, is increased upon Pst DC3000 infection, whereas INH, the inhibitor that blocks the conversion of glycine to serine in photorespiration and suppresses GOX activity, suppresses this increase (Ahammed et al., 2018). Other indicators of photorespiration rate used in the measurements include the photorespiratory CO2 compensation point (Γ*) and the ratio of glycine to serine (Gly/Ser). FOC-inoculated banana seedlings contain higher Γ* than untreated plants (Dong et al., 2016). In nitrate-induced FOC resistance cucumber plants, both Γ* and Gly/Ser are increased (Sun et al., 2021). Further studies are needed to determine whether the increased photorespiration rate reported in these studies contributes to defense responses. This quantitative approach may also be extended to additional studies aimed at dissecting the interplay between photorespiration and immunity. 1.4 Summary Plants employ multiple layers of mechanisms, including photorespiration, to respond to high light conditions. Photorespiration has been shown to support photosynthetic performance, especially under stress conditions like high light. Research on the photorespiratory enzyme HPR further supports the impacts of photorespiration on photosynthesis, primary metabolism, energetics, and stress response. Emerging evidence for the regulation of photorespiration signifies the flexibility of this pathway and its role in plant response to the dynamic environment. Plants also use the immune system to protect themselves against pathogens. Studies demonstrate the key role of photorespiration in plant immunity through changes in ROS homeostasis, while other mechanisms such as the participation of photorespiratory metabolites, the direct impact of photorespiration on defense hormone biosynthesis, and so on, are also emerging. A more precise understanding of the contribution of photorespiration to plant physiology and plant interaction with the environment is vital for developing crops with both high yield and stress resilience. 16 Figures Figure 1.1. The known photorespiratory pathway and a working model for the connections between photorespiration and plant immunity. ROS, photorespiratory metabolites, defense hormones, and possibly other mechanisms connect the photorespiratory pathway to key components of the immune network. See main text for detailed information of the photorespiratory pathway and plant immune response pathways, as well as mechanisms/potential mechanisms for their connections. Overlaps between some subcomponents of the immune response network and the photorespiratory organelles indicate the involvement of the particular organelles. Abbreviations: 2-OG, 2-oxoglutarate; 2-PG, 2-phosphoglycolate; 3-PGA, 3- phosphoglycerate; BASS6, bile acid sodium symporter 6; CAT, catalase; GGAT, glutamate:glyoxylate aminotransferase; GDC, glycine decarboxylase complex; GLYK, glycerate kinase; GOX, glycolate oxidase; HPR, hydroxypyruvate reductase; PGLP, 2-PG phosphatase; PLGG1, plastidial glycolate/glycerate transporter 1; Rubisco, RuBP carboxylase/oxygenase; RuBP, ribulose-1,5-bisphosphate; SGAT, serine:glyoxylate aminotransferase; SHMT, serine hydroxymethyltransferase; THF, tetrahydrofolate; MAMP, microbe-associated molecular pattern; DAMP, damage-associated molecular pattern; PRR, pattern recognition receptor; NLR, nucleotide-binding/leucine-rich-repeat receptor; PTI, pattern-triggered immunity; ETI, effector- triggered immunity; ROS, reactive oxygen species; SA, salicylic acid; JA, jasmonate. 17 Table 1.1. Photorespiratory enzymes that participate in defense response. Enzyme Full name Function in immunity References Tables GOX Glycolate oxidase Impacts ROS homeostasis and JA biosynthesis CAT Catalase Impacts ROS homeostasis; suppressed by SA; promotes JA biosynthesis and mediates crosstalk between SA and JA/auxin, and between Ca2+ and JA (AtCAT2) (Rojas & Mysore, 2012; Rojas et al., 2012; Chern et al., 2013; Zhang et al., 2016; Ahammed et al., 2018; Williams et al., 2018; Xu et al., 2018b; Launay et al., 2022) (Chen et al., 1993; Sánchez- Casas & Klessig, 1994; Conrath et al., 1995; Takahashi et al., 1997; Chamnongpol et al., 1998; Mittler et al., 1999; Chaouch et al., 2010; Chaouch & Noctor, 2010; Mathioudakis et al., 2013; Zhang et al., 2015, 2016, 2021; Giri et al., 2017; Murota et al., 2017; Sun et al., 2017; Yuan et al., 2017; Williams et al., 2018; Lv et al., 2019; Zhao et al., 2021) GGAT Glutamate: glyoxylate aminotransferase SGAT Serine:glyoxylate aminotransferase GDC Glycine decarboxylase complex SHMT Serine hydroxymethyl- transferase HPR Hydroxypyruvate reductase Connects with H2O2 (González‐lópez et al., 2021) Positive role in resistance in melon and tomato; bound by SA (AtSGAT) Impacts ROS homeostasis Impacts ROS homeostasis; contributes to resistance possibly through folate metabolism; bound by SA (AtSHMT4); interacts with the disease-resistance protein RPM1 (OsSHMT1) Interacts with syringolide receptor (GmHPR); bound by SA (AtHPR2) (Taler et al., 2004; Benjamin et al., 2009; Manohar et al., 2015; Ahammed et al., 2018) (Navarre & Wolpert, 1995; Yao et al., 2002; Cristina Palmieri et al., 2010; Gilbert & Wolpert, 2013) (Moreno et al., 2005; Liu et al., 2012; Manohar et al., 2015; Kandoth et al., 2017; Ahammed et al., 2018; Lakhssassi et al., 2019; Korasick et al., 2020; Wang et al., 2021; Fu et al., 2022) (Okinaka et al., 2002; Manohar et al., 2015) (Gao et al., 2020) GLYK Glycerate kinase Positive role in resistance in potato 18 CHAPTER 2. A photorespiratory glyoxylate shunt in the cytosol supports photosynthesis and plant growth under high light conditions in Arabidopsis 2.1 Introduction Light is an essential energy source and a critical environmental cue for plants, but its intensity often fluctuates beyond the ranges optimal for plant growth. For example, plants frequently experience high light conditions that exceed 2000 μmol m-2 s-1 on sunny days, posing significant stress to plants in the field (Ort, 2001; Mishra et al., 2012). When light intensities surpass the photosynthetic capacity, generation of reactive oxygen species (ROS) from the photosynthetic apparatus is enhanced, which can cause photodamage to protein complexes and subsequently inactivate the electron transport chain (Krieger-Liszkay & Shimakawa, 2022; Sharma et al., 2023). To cope with this stress, plants have developed strategies to respond at various levels, such as the antioxidant system to scavenge ROS, dissipation of excessive energy as non-photochemical quenching (NPQ), cyclic electron flow (CEF) to balance ATP/NADPH production, transcriptional reprogramming, chloroplast and leaf movement, and anthocyanin accumulation (Szymańska et al., 2017; Shi et al., 2022a). Photorespiration, a metabolic process closely related to photosynthesis, is initiated after the oxygenation of ribulose 1,5-bisphosphate (RuBP) by the photosynthetic enzyme ribulose 1,5- bisphosphate carboxylase-oxygenase (rubisco) (Eisenhut et al., 2019). Through a series of reactions residing sequentially in the chloroplast, peroxisome, mitochondrion, and the cytosol, the oxygenation product 2-phosphoglycolate (2-PG), which inhibits cell functions, is eventually converted to 3-phosphoglycerate (3-PGA) to be recycled back to the Calvin-Benson cycle in the chloroplast (Eisenhut et al., 2019) (Fig. 2.1). Although photorespiration is often considered a sub-optimal process because it consumes energy and releases pre-fixed carbon as CO2, a properly functional photorespiratory pathway supports photosynthetic performance, especially under stress conditions. In ambient air and under regular growth conditions, many photorespiratory mutants exhibit compromised photosynthesis and growth, phenotypes that are largely recovered (Timm & Bauwe, 2013; Timm et al., 2016) under high CO2 environments where rubisco oxygenation is inhibited (Timm & Bauwe, 2013; Timm et al., 2016). Although the exact reasons for these air-grown phenotypes are  This chapter has been submitted for publication. All the experiments were conducted by myself, except for the GLYR1 localization study, which was performed by Dr. Amanda Koenig and Joy Li. 19 unclear, the accumulated photorespiratory intermediates in these mutants, such as 2-PG, glyoxylate, and glycerate, can inhibit the activities of photosynthesis-related enzymes (Timm et al., 2016). Additionally, several Arabidopsis photorespiratory mutants, including hydroxypyruvate reductase 1 (hpr1), plastidial glycolate/glycerate transporter 1 (plgg1), catalase 2 (cat2) and glycolate oxidase 1 (gox1), have much more severe photosynthetic phenotypes under high and dynamic light compared with low and constant light conditions, supporting the increased importance of a properly functional photorespiratory pathway under high light (Li et al., 2019a). Under stress, photorespiration is believed to function as an alternative electron sink that consumes excessive energy produced by the photosynthetic light reactions (Wingler et al., 2000; Ort & Baker, 2002; Voss et al., 2013; Huang et al., 2019a), although a recent work challenged the idea that photorespiration’s role as an alternative electron sink is photoprotective (Smith et al., 2023). As a core enzyme in photorespiration, HPR catalyzes the reduction of hydroxypyruvate to produce glycerate (Fig. 2.1). In Arabidopsis, three HPR family members have been shown to function in photorespiration (Timm et al., 2008, 2011). Peroxisomal HPR1 plays a major role in reducing hydroxypyruvate with NADH as the co-factor, whereas HPR2 and HPR3 in the cytosol have low activities and higher affinities for NADPH (Timm et al., 2008, 2011; Xu et al., 2018a; Wang et al., 2022). HPR enzymes play an important role in photosynthesis, because the knockout mutant of HPR1 shows compromised photosynthetic performance and growth in the air, with additive effects observed in hpr1 hpr2 double mutant and hpr1 hpr2 hpr3 triple mutant (Timm et al., 2008, 2011). Under high light, hpr1 exhibits stronger phenotypes, including decreases in the efficiency of photosystem II (PSII), increases in NPQ, activation of CEF, accumulation of H2O2, and a marked reduction in chlorophyll and anthocyanin (Li et al., 2019a; Wang et al., 2022). Further investigation of hpr1 under high light found that 2-PG accumulation inhibited the activity of triose phosphate isomerase, an enzyme of the Calvin-Benson cycle that converts glyceraldehyde 3-phosphate to dihydroxyacetone phosphate. This inhibition results in a cytosolic bypass and glucose-6-phosphate (G6P) shunt in the Calvin-Benson cycle where the G6P shunt consumes ATP, triggering high rates of CEF to balance energy demand (Li et al., 2019b). Consistent with the activation of the G6P shunt, increased CO2 release was also observed in hpr1 (Cousins et al., 2011; Timm et al., 2021). However, there are other possible explanations for this extra CO2 release, as the non-enzymatic decarboxylation of 20 hydroxypyruvate and serine consumption through serine decarboxylase also seem to occur in hpr1 (Cousins et al., 2011; Timm et al., 2021). Finally, HPR1 was found to maintain the repair of PSII under high light (Wang et al., 2022). HPR enzymes have a broad influence on the level of metabolites in the plant. Deficiency in one or more HPRs increases the level of most photorespiratory intermediates, including glycolate, glycine, serine, hydroxypyruvate and glycerate, and these metabolic phenotypes are affected by photoperiods (Timm et al., 2008, 2011, 2021). Consistent with the impaired photosynthesis, carbohydrate levels are largely decreased in the hpr1 mutant (Timm et al., 2021). Levels for other metabolites related to photorespiration, such as intermediates in the tricarboxylic acid cycle and many amino acids, are elevated or reduced in hpr1 (Timm et al., 2008, 2011, 2021). Interestingly, different from its daytime-dependent accumulation pattern in the wild-type plants, serine was found to be constitutively elevated in the hpr1 mutant, inhibiting the expression of photorespiratory genes and reducing the level of the corresponding enzymes (Timm et al., 2013). Additionally, the accumulated glycolate can replace the bicarbonate ligand in PSII in hpr1, shifting the midpoint potential of the quinone acceptor and reducing the generation of single oxygen (Messant et al., 2018). While the main flux of photorespiration is well known, there are examples of alternative fluxes, such as the aforementioned cytosolic HPR2 and HPR3 enzymes that are partially redundant in function with HPR1. Understanding other routes of carbon flux associated with photorespiration is important to fully deciphering this metabolic network. In addition, photorespiration is tightly linked to primary metabolic pathways and crucial to plant survival under certain environmental conditions, thus this pathway is expected to be regulated. Although research in this area is still scarce, current evidence indicates that photorespiration responds to environmental factors such as CO2, O2 and light, receives feedback from its own metabolites and enzymes, and is regulated at transcriptional, post-transcriptional and post-translational levels (Timm & Hagemann, 2020; Timm, 2020; Aroca et al., 2023). Taken together, mechanistic research into the flexibility and regulation of photorespiration will be highly valuable to completely elucidating the role of this pathway in plant physiology and plant interaction with the environment. In this work, we investigated the regulation and flexibility of photorespiration by characterizing a genetic suppressor of the Arabidopsis hpr1 mutant under high light, mapping the 21 underlying gene, and conducting follow-up genetic and metabolic flux analyses. We found that defective glyoxylate reductase 1 (GLYR1), a cytosolic enzyme that catalyzes the conversion of glyoxylate to glycolate, can rescue the mutant phenotypes of hpr1 in plant growth, photosynthesis and levels of photorespiratory metabolites under high light conditions. Further examination showed that loss of function of GLYR1 can also partially suppress the phenotype of cat2, but not plgg1. Combining transitional metabolic profiling, glyoxylate feeding, and genetic analyses, we provided evidence for a cytosolic photorespiratory shunt that converts accumulated glyoxylate to hydroxypyruvate, which can act, at least partially, through the cytosolic HPR2 enzyme to enhance carbon recycling. This cytosolic shunt seems to be especially critical under high light intensities when a high rate of photorespiratory flux is required to deal with increased rates of total rubisco oxygenation reaction and in the absence of a functional major photorespiratory pathway. Our findings suggest that the metabolic flexibility of photorespiration can help plants adjust to stress conditions, thus may provide help with future efforts in improving plant performance under high light conditions. 2.2 Results 2.2.1 Loss of function of glyoxylate reductase 1 (GLYR1) partially rescues the growth and metabolic phenotypes of the hpr1 mutant To identify proteins with a regulatory or modulatory role in photorespiration, we performed EMS mutagenesis on Arabidopsis hpr1-1 mutant seeds and screened for genetic suppressors based on their growth and photosynthetic phenotypes under high light conditions (~700 μmol m-2 s-1). One suppressor, shpr7 (hpr1 suppressor number 7), was found to partially suppress the small rosette size of hpr1-1 under high light relative to normal light conditions (100 μmol m-2 s-1) (Fig. 2.2a and b). To characterize and map the causal mutation in shpr7, we backcrossed shpr7 to hpr1-1. The segregation ratio of hpr1-like vs. suppressor-like plants in the BC1F2 generation was 3:1, suggesting that the suppression is caused by a recessive mutation. To map the gene responsible for the suppression, genomic DNA extracted from 80 suppressor-like individuals in the BC3F2 generation was pooled for whole genome sequencing. We identified a point mutation in the exon of the Glyoxylate Reductase 1 (GLYR1) gene, which encodes an NADPH-dependent glyoxylate/succinic semialdehyde reductase (Hoover et al., 2007; Zarei et al., 2017), causing a glutamate (E)-to-lysine (K) substitution at amino acid 117 (Fig. 2.2c). RT-PCR analysis detected 22 similar levels of the GLYR1 transcripts in shpr7 and the wild-type (Fig. A1a), suggesting that this mutation does not lead to obvious changes in gene expression. To determine the effect of the E117K mutation at the protein level, peptide antibodies against amino acid 229-243 were generated, which detected an apparently decreased level of the GLYR1 protein (30.7kDa) in high light-grown shpr7 (Fig. 2.2d), indicating that this point mutation may cause the protein to be less stable. The antigen peptide is from a highly identical region between AtGLYR1 and its homolog, AtGLYR2, where the two proteins differ only by one amino acid (Simpson et al., 2008). We therefore reasoned that the weak signal in glyr1-1 in the immunoblot might be derived from the mature form of GLYR2, after its N-terminal transit peptide for chloroplast and mitochondrial dual targeting is removed upon import into the organelles (Fig. 2.2d). To confirm that loss-of-function mutations in GLYR1 can lead to the suppression of hpr1- 1 like that shown in shpr7, we obtained three independent T-DNA insertion mutant lines of GLYR1 (Fig. 2.2c). All three lines lacked detectable full-length GLYR1 transcript (Fig. A1b) and showed comparable growth and morphologies to the wild-type (Col-0) plants (Fig. A1c). Double mutants generated by crossing these lines individually with hpr1-1 all exhibited very similar phenotypes as shpr7 (Fig. 2.2a and b, Fig. A1c), confirming that loss of function of GLYR1 can partially rescue the hpr1-1 mutant phenotypes. Overexpressing the GLYR1 gene (35S::GLYR1) in shpr7 reverted its rosette size back to hpr1-like under high light (Fig. 2.2e, Fig. A1d), further confirming GLYR1 as the causal gene for the suppression of hpr1. Since the three T-DNA mutants were indistinguishable from each other in their capability to suppress hpr1-1, we selected the glyr1-1 allele for follow-up experiments. GLYR1 was shown to catalyze the conversion of glyoxylate to glycolate, and succinic semialdehyde to γ-hydroxybutyrate (Zarei et al., 2017). Because glyoxylate and glycolate are photorespiratory intermediates, we tested if the deficiency in GLYR1 also causes metabolic changes in the photorespiratory pathway (Fig. 2.3). Mutant hpr1-1 grown under high light conditions had elevated levels of all the 6 photorespiratory metabolites tested, consistent with previous reports that knockout mutant of HPR1 accumulates photorespiratory intermediates under normal growth conditions (Timm et al., 2008, 2021). The suppressor shpr7 and the double mutant glyr1-1 hpr1-1 showed partial rescue of the accumulation of at least 5 of these metabolites, including glycolate, glyoxylate, serine, hydroxypyruvate, and glycerate, and 23 consistent with its wild-type-like plant appearance, the glyr1-1 single mutant contained similar levels of these metabolites as Col-0 (Fig. 2.3). Taken together, our results proved that reduced or loss of function of the GLYR1 protein in Arabidopsis can partially suppress the mutant phenotypes of hpr1 in both plant growth and the content of photorespiratory metabolites. 2.2.2 Defects in GLYR1 also partially rescue the phenotypes of the photorespiratory mutant cat2, but not those of plgg1 To investigate if the suppression of hpr1 by glyr1 is specifically linked to HPR1 or broadly connected to components of the photorespiratory pathway, we crossed glyr1-1 into two other photorespiratory mutants, cat2-1 and plgg1-1 (Fig. 2.1). Similar to hpr1-1, cat2-1 and plgg1-1 are also compromised in growth and photosynthesis under high light (Li et al., 2019a). Interestingly, under high light, the double mutant glyr1-1 cat2-1 exhibited a bigger rosette size than cat2-1, whereas the lack of a functional GLYR1 was unable to improve the growth of plgg1- 1 (Fig. 2.4a). To determine if glyr1 can also rescue the reduced photosynthetic efficiency in cat2-1 and plgg1-1, as well as hpr1-1, we measured quantum efficiency of photosystem II (ΦII), a critical parameter of photosynthesis, in mutant and Col-0 seedlings. A 3-day light regime, with normal light on the first day and light gradients on days 2 and 3, was applied (Fig. 2.4b). Consistent with their growth phenotypes, hpr1-1, cat2-1 and plgg1-1 had lower ΦII values, especially under higher light intensities, whereas glyr1-1 resembled Col-0 (Fig. 2.4b). As expected, GLYR1 deficiency helped to improve the photosynthetic performance of hpr1-1 and cat2-1, but not plgg1-1 (Fig. 2.4b). 2.2.3 GLYR1 localizes to the cytosol Since the subcellular location of GLYR1 underlies the potential mechanism by which defective GLYR1 suppresses hpr1 and cat2 phenotypes, we sought to definitively determine its subcellular localization. While several studies purport GLYR1 localization in the cytosol, these findings did not sufficiently account for the putative peroxisomal targeting signal type 1 (PTS1) tripeptide (SRE>) at GLYR1’s extreme C-terminus, which implies its potential localization in peroxisome matrix. Apple (Malus domestica) GLYR1 localizes to the cytosol, but MdGLYR1 lacks the C-terminal putative PTS1 and therefore cannot be used to precisely infer the localization of AtGLYR1 (Brikis et al., 2017). An AtGLYR1-GFP fusion protein transiently 24 expressed in tobacco localized to the cytosol (Simpson et al., 2008), but again peroxisomal targeting cannot be discounted as the C-terminal PTS1 was blocked in this construct. Indirect evidence of cytosolic localization was shown in pea where the majority of glyoxylate-reducing activity was found in the cytosolic fraction, yet a small amount was detected in the chloroplast (Givan et al., 1988). Moreover, a study using an N-terminal fluorescent tag showed the cytosolic localization of AtGLYR1 in tobacco BY-2 and Arabidopsis cells (Ching et al., 2012), dispelling its peroxisomal localization. However, the fluorescence signals also appeared in chloroplast-like structures in Arabidopsis mesophyll cells. Because these data did not include chlorophyll fluorescence as a marker, we cannot definitively rule out chloroplast localization of GLYR1. Therefore, it was important to obtain evidence in which AtGLYR1’s cytosolic localization is unequivocally shown. To this end, we generated a 35S::eYFP-GLYR1 construct and co-expressed it with the peroxisome marker mScarlet-SRL (Koenig et al., 2023) in tobacco leaves. GLYR1 appeared diffused throughout the cytosol, and no GLYR1 signal was observed overlapping peroxisomal or chloroplast (visualized by chlorophyll autofluorescence) signals (Fig. 2.4c), confirming that GLYR1 is solely localized to the cytosol. 2.2.4 Transitional metabolic profiling uncovers a close link between hydroxypyruvate and GLYR1 Based on GLYR1’s mutant phenotypes, its protein localization, and the previously reported enzyme activity, we hypothesized a photorespiratory glyoxylate shunt in the cytosol. This shunt drives the conversion of glyoxylate to hydroxypyruvate and triggers carbon flux back to the Calvin-Bensen cycle, when the primary photorespiratory pathway is deficient and when plants are exposed to high light conditions. Serine was found to accumulate in the hpr1-1 and cat2-1 mutants (Timm et al., 2008; Bao et al., 2021), likely increasing the level of free serine in the cytosol. Likewise, we posited that impaired photorespiration in hpr1-1 and cat2-1 also causes increased glyoxylate leakage to the cytosol. The lack of a functional GLYR1 protein leads to an increase in the cytosolic level of glyoxylate, but in plants containing a functional primary photorespiratory pathway, this increase may not reach a level high enough to drive an aminotransferase activity with serine. Only in photorespiratory mutants such as hpr1 and cat2, which already accumulate cytosolic glyoxylate, can GLYR1 deficiency further increase the level of glyoxylate to a threshold to react with serine, producing hydroxypyruvate and glycine. The 25 hydroxypyruvate generated from this reaction can then be directly catalyzed by a cytosolic HPR (such as HPR2) to produce glycerate, which returns to the Calvin-Benson cycle in the chloroplast following phosphorylation by glycerate kinase (Fig. 2.1). Under this hypothesis, the lack of GLYR1 function can activate a cytosolic pathway in hpr1-1 and cat2-1 to recycle photorespiratory carbon more efficiently back to photosynthesis, decreasing the inhibitory effects of photorespiratory intermediates and subsequently improving the performance of these two mutants to compensate for the lack of HPR1 or CAT2. By contrast, loss of PLGG1 blocks the transport of glycerate into chloroplasts, preventing carbon recycling by HPR2 and therefore failing to rescue plgg1-1. To test this hypothesis, we first determined how quickly photorespiratory metabolites respond to high light. The partial rescue of photorespiratory metabolites observed earlier in this study (Fig. 2.3) was shown in plants after two weeks of growth under high light. Here we instead employed short-term treatment of photorespiratory conditions to avoid secondary effects from early metabolic events. Plants were grown under high CO2 (2,000 ppm) and normal light conditions for 3 weeks before transfer to ambient CO2 and high light conditions. Plant tissue was sampled for metabolite measurements after ~10 h illumination, as photorespiratory intermediates were reported to accumulate to high levels at the end of the day (Pick et al., 2013; Timm et al., 2013). When growing under high CO2, hpr1-1 only had increased serine and slightly decreased glycerate concentrations (Fig. A2), consistent with previously reported measurements of the photorespiratory intermediates under 1% (10,000 ppm) CO2 (Timm et al., 2008). The cat2-1 mutant performed similarly to Col-0 and importantly, glyr1-1 did not alter the levels of the photorespiratory metabolites in hpr1-1 or cat2-1 (Fig. A2), indicating that the influence of glyr1- 1 is well suppressed under high CO2. After 10 h of ambient CO2 and high light conditions, the levels of 5 out of the 6 photorespiratory intermediates, including glycolate, glyoxylate, serine, hydroxypyruvate and glycerate, were significantly higher in hpr1-1 compared to Col-0 (Fig. 2.5). In cat2-1, glycolate, serine and glycerate were increased, and hydroxypyruvate also had a small but obvious accumulation (Fig. 2.5, Fig. A3). Intriguingly, the metabolites best rescued by glyr1- 1 in both hpr1-1 and cat2-1 were glycolate and hydroxypyruvate (Fig. 2.5). Considering the enzymatic activity of GLYR1 in converting glyoxylate to glycolate, it is expected that glyr1-1 hpr1-1 and glyr1-1 cat2-1 have lower levels of glycolate. However, the dramatic rescue of 26 hydroxypyruvate in hpr1-1 and cat2-1 by glyr1-1 after only a 10-h treatment suggested that hydroxypyruvate is closely related to the function of GLYR1, which supports the cytosolic shunt we proposed. As the growth and photosynthetic phenotypes of cat2-1 were weaker than hpr1-1, hydroxypyruvate accumulation in cat2-1 was relatively lower and rescued better than hpr1-1, with glyr1 cat2 hydroxypyruvate content returning to wildtype levels at 22% that observed in cat2 (Fig. 2.5, Fig. A3). Glycerate abundance in the mutants showed a similar trend as that of hydroxypyruvate, although the difference between glyr1-1 hpr1-1 and hpr1-1 was smaller, which is probably because the response of glycerate requires hydroxypyruvate as the primary responder and is therefore indirect. In summary, transitional metabolic profile suggests a close connection between hydroxypyruvate and GLYR1 under high light conditions. It supports the role of the proposed cytosolic photorespiratory pathway, in which the accumulated glyoxylate caused by defective GLYR1 in the hpr1 and cat2 background is converted to hydroxypyruvate. 2.2.5 Feeding glyoxylate to the plant strongly inhibits growth of wild-type but can benefit hpr1 Based on our hypothesis, the availability of glyoxylate in the cytosol plays an important role in activating the non-canonical pathway in hpr1-1. To obtain further support for this role, we increased the level of free glyoxylate in the cytosol by directly supplying glyoxylate to the growth medium for hpr1-1. Although the mechanism is still unclear, glyoxylate was reported to inhibit RuBP regeneration and rubisco activation (Mulligan et al., 1983; Cook et al., 1985; Chastain & Ogren, 1989; Campbell & Ogren, 1990; Lu et al., 2014) and therefore is toxic to plants. In agreement with this, Col-0 plants grown in 0.4 mM glyoxylate exhibited strong growth inhibition and decreased fresh weight under normal or high light conditions (Fig. 2.6). However, hpr1-1 maintained similar fresh weight after glyoxylate feeding (Fig. 2.6b), despite some suppression in root elongation (Fig. 2.6a), indicating that glyoxylate can be metabolized more quickly in hpr1- 1, possibly through the cytosolic pathway we proposed. The glyr1-1 hpr1-1 double mutant showed a small growth inhibition upon glyoxylate treatment (Fig. 2.6), possibly because the total glyoxylate amount from both internal and external sources exceeds the capacity of this cytosolic pathway. By contrast, 0.1 mM serine had no influence on plant growth in all the lines under normal or high light conditions (Fig. A4), indicating that the cytosolic serine in hpr1-1 is 27 adequate for the cytosolic shunt we proposed. In summary, results from the glyoxylate feeding experiment showed that increasing glyoxylate availability in the cytosol can benefit hpr1-1, thus supporting our hypothesis. 2.2.6 The rescuing effect of glyr1 in hpr1 largely depends on HPR2 Our proposed photorespiratory glyoxylate shunt requires the activity of cytosolic HPR, so we investigated the role of HPR2 because it has a stronger role in photorespiration than HPR3 (Timm et al., 2011). Null mutant hpr2-3 (Fig. A5) was used to generate a triple knockout line glyr1-1 hpr1-1 hpr2-3. Because mutants defective in both HPR1 and HPR2 genes are already stunted, grow poorly in ambient air, and are intolerant to high light treatment, we first grew all the lines under high CO2 and normal light conditions, during which all mutants showed comparable morphologies as Col-0 (Fig. 2.7a, Day 0). After three weeks and at the end of the dark period, plants were moved to ambient CO2 and high light conditions, where they were kept for 9 days. This treatment led to much smaller rosettes in hpr1-1 compared to Col-0, which was partially rescued in glyr1-1 hpr1-1 and the original suppressor shpr7 (Fig. 2.7a). At Day 9, although hpr1-1 hpr2-3 and glyr1-1 hpr1-1 hpr2-3 performed poorly compared with other lines, they looked similar to each other, supporting our conclusion that, at least at this time point, glyr1-1’s role in helping carbon recycle back to the chloroplast largely depends on a functional HPR2. At Day 18, glyr1-1 hpr1-1 hpr2-3 became slightly bigger than hpr1-1 hpr2-3 (Fig. 2.7a), indicating that glyr1-1 may also act through other proteins (such as HPR3) for its role under longer photorespiratory conditions. Transitional metabolic profiling was also performed on mutants grown under high CO2 and normal light for 3 weeks followed by 10-h treatment with ambient CO2 and high light. Similar to the rescued glycolate level in glyr1-1 hpr1-1, glyr1-1 hpr1-1 hpr2-3 also showed a lower glycolate level than hpr1-1 hpr2-3 (Fig. 2.7b), confirming that the glycolate production via GLYR1 is greatly lost. By contrast, the level of hydroxypyruvate was only slightly rescued in glyr1-1 hpr1-1 hpr2-3 (Fig. 2.7b), supporting the importance of HPR2 in this cytosolic shunt. The cytosolic shunt we proposed indicates an increased flux through HPR2 in glyr1-1 hpr1-1 and shpr7 compared to hpr1-1, which may require a higher activity of the HPR2 enzyme. To test this possibility, the maximal enzymatic activities of HPR in plants were measured, which showed that glyr1-1 hpr1-1 and shpr7 had comparable enzymatic activities of HPR to hpr1-1 in the transitional (Fig. A6a) or stable stage (Fig. A6b) of high light treatment, using NADH or 28 NADPH as the co-factor. However, this assay measures the total maximal enzymatic activities of HPR in protein extract from plants, which is more correlated to the level of the HPR enzymes. Since the enzymes in the plant cell usually do not operate at their maximal capacities, the possible regulation of enzymatic levels to increase fluxes cannot be captured by this assay. 2.3 Discussions In this study, we provided evidence for a cytosolic glyoxylate shunt of photorespiration that can improve carbon recycling in the absence of enzymes in the major photorespiratory pathway under high light, supporting the metabolic flexibility of the photorespiration network. Under high light, deficiency in GLYR1, a cytosolic enzyme that converts glyoxylate to glycolate, can partially rescue the phenotypes of the photorespiration mutants hpr1 and cat2 but not those of the glycolate/glycerate transporter mutant plgg1. Further investigations showed that hydroxypyruvate is closely connected to GLYR1’s function, and the availability of glyoxylate and HPR2 in the cytosol are important for the suppression of hpr1. These results led us to propose a novel cytosolic photorespiratory shunt, which seems to be critical when the major photorespiratory pathway is compromised and when plants are exposed to photorespiratory conditions such as high light. Specifically, a reaction between glyoxylate and serine is catalyzed by an aminotransferase to produce hydroxypyruvate and glycine (Fig. 2.1). Via HPR2, cytosolic hydroxypyruvate is subsequently converted to glycerate that eventually returns to the Calvin- Benson cycle, thus reducing the accumulation of the toxic photorespiratory intermediates and enhancing carbon recycling (Fig. 2.1). We predict that, in wild-type plants, this shunt may only be activated under extremely high photorespiratory fluxes. Plants contain two GLYR proteins, the cytosolic GLYR1 and the plastid/mitochondrion- dual localized GLYR2, both of which are believed to be involved in aldehyde detoxification during abiotic stress (Allan et al., 2008; Mekonnen & Ludewig, 2016; Zarei et al., 2017). Although GLYR1 was shown to be more efficient in its glyoxylate reductase activity, it can also convert succinic semialdehyde to γ-hydroxybutyrate in γ-aminobutyrate (GABA) metabolism (Hoover et al., 2007; Zarei et al., 2017). During abiotic stress, the expression of both GLYR1 and GLYR2 is upregulated, presumably to detoxify the accumulated succinic semialdehyde in plants (Allan et al., 2008; Mekonnen & Ludewig, 2016; Zarei et al., 2017). However, the positive role of succinic semialdehyde reduction during abiotic stress contradicts the rescuing effect of glyr1 in hpr1 under high light, and GABA metabolism is not known to be directly linked to 29 photorespiration. Therefore, the succinic semialdehyde reductase activity of GLYR1 does not seem to be involved in its role related to HPR1 and CAT2. Since the knockout mutants of GLYR1 are comparable to Col-0 in plant growth and levels of the photorespiratory metabolites, this cytosolic photorespiratory shunt may not be important when the main photorespiratory pathway in the peroxisome is functional, possibly due to the predominant peroxisomal location of glyoxylate. Glyoxylate is highly reactive and can inhibit photosynthesis and other metabolic reactions, thus is toxic to the plant (Mulligan et al., 1983; Cook et al., 1985; Chastain & Ogren, 1989; Campbell & Ogren, 1990; Lu et al., 2014). There are two major glyoxylate metabolic pathways in the peroxisome to maintain its homeostasis: the main photorespiratory pathway in photosynthetic tissue and the glyoxylate cycle in seeds in which glyoxylate is produced and catabolized (Dellero et al., 2016; Pan et al., 2020). Further, the cytosolic GLYR1 and the plastidic/mitochondrial GLYR2 were believed to scavenge the glyoxylate leaked from the peroxisome (Simpson et al., 2008; Dellero et al., 2016). Since cytosolic glyoxylate may become toxic to plants, it is possible that the cytosolic shunt is activated only when the main peroxisomal pathway is defective and under conditions that need very high photorespiratory flux. We provided evidence that HPR2 is pivotal in converting cytosolic hydroxypyruvate to glycerate in the photorespiratory glyoxylate shunt, as the glyr1 hpr1 hpr2 triple mutant maintains 86% of the hydroxypyruvate content observed in hpr1 hpr2 (Fig. 2.7b). However, the minor decrease in hydroxypyruvate content, as well as the slightly bigger glyr1 hpr1 hpr2 rosette size compared to hpr1 hpr2 after 18 days of high light treatment (Fig. 2.7a) suggest the existence of other mechanisms in hydroxypyruvate reduction independent of HPR2. Although the role of HPR3 in reducing hydroxypyruvate is minor, knocking out HPR3 can further exacerbate the photorespiratory phenotypes of the hpr1 hpr2 double mutant (Timm et al., 2011). Therefore, HPR3 may be involved in maintaining this photorespiratory shunt in the cytosol when both HPR1 and HPR2 are absent. We hypothesize that there is a cytosolic aminotransferase with similar activity as the peroxisomal serine:glyoxylate aminotransferase (SGAT). Although SGAT activity was only detected in the peroxisome in previous studies (Liepman & Olsen, 2001, 2004), cytosolic activity may not be detectable unless the cytosolic route is activated. SGAT does not have any apparent homologs in the Arabidopsis genome (Liepman & Olsen, 2004), but substrate promiscuity is 30 common for aminotransferases (Koper et al., 2022). SGAT is also known as AGT1 because of its alanine:glyoxylate transferase activity; additionally, it can catalyze other amino donor:acceptor combinations such as serine:pyruvate and asparagine:glyoxylate (Liepman & Olsen, 2001; Koper et al., 2022). Human (Homo sapiens) AGT2, which localizes to mitochondria and has broad substrate specificity (Koper et al., 2022), has three homologs in Arabidopsis: AtAGT2, AtAGT3, and AtPYD4 (pyrimidine 4) (Liepman & Olsen, 2003; Koper et al., 2022). AtAGT2 was reported to localize to mitochondria and peroxisomes (Carrie et al., 2009). The enzyme activity and localization of AtAGT3 and AtPYD4 are unclear, thus may be candidates for the hypothetical cytosolic aminotransferase with SGAT activity. Some of the metabolic profiles of glycine and serine in this work have complicated patterns, possibly due to their participation in other metabolic pathways. Under ambient (21%) O2 conditions, 32% of the photorespiratory carbon is thought to leave this pathway as serine (Busch et al., 2018; Fu et al., 2023), indicating that photorespiratory serine may be strongly connected with other metabolisms. Serine has been shown to play a role in the biosynthesis of other amino acids, proteins, and lipids (Ros et al., 2014). While glycine is converted to serine in the mitochondrion during photorespiration, serine can also be converted to glycine in the cytosol and plastid (Rosa-Téllez et al., 2023). This serine-glycine interconversion is an important component in one-carbon metabolism, which is crucial for synthesizing nucleotides and methylated compounds and maintaining nutrient balance (Hanson & Roje, 2001; Rosa-Téllez et al., 2023). Furthermore, it has been reported that excess glycine from photorespiration is used for protein synthesis (Cegelski & Schaefer, 2005) and glutathione accumulation (Noctor et al., 1999). Finally, glycine and serine were found in the vacuole (Riens et al., 1991; Fürtauer et al., 2019), which may represent an inactive pool or have a slow response to environmental changes. In addition to alternative routes and photorespiratory shunts, post-translational modification of photorespiratory proteins seems to play an important regulatory role in photorespiration. For example, Arabidopsis phosphorylation‐mimetic mutant of serine hydroxymethyltransferase 1 (SHMT1) exhibited compromised performance under salt or drought stress (Liu et al., 2019), indicating that phosphorylation of photorespiratory components is important in stress response. Although the specific function of these modifications needs further investigation, phosphorylation also plays important roles in maintaining enzymatic activity and cofactor binding in GOXs (Jossier et al., 2020) and HPR1 (Liu et al., 2020), and thioredoxins in 31 mitochondria can modify the redox status of the L-protein of glycine decarboxylase complex (GDC) (Reinholdt et al., 2019; da Fonseca-Pereira et al., 2020). Furthermore, many potential sites for phosphorylation, ubiquitination, acetylation, and redox modifications have been identified in photorespiratory enzymes (Hodges, 2022; Aroca et al., 2023), suggesting the possible involvement of additional post-translational modifications in regulating photorespiration. For example, HPR2 may be regulated by post-translational modifications to increase its enzymatic activity when the cytosolic shunt is activated. Our work has provided evidence for the important role of an alternative photorespiratory route in the cytosol under high light conditions when the major photorespiratory pathway is defective, supporting the metabolic flexibility of photorespiration and substantiating the contribution of photorespiration to stress response. Further investigations are needed to obtain a more comprehensive understanding of the regulation of photorespiration, which may ultimately help to generate new crop varieties with high productivity without compromising their stress tolerance. 2.4 Methods 2.4.1 Plant materials and growth conditions Wild-type and mutant lines of Arabidopsis thaliana used in the study are all from the ecotype Col-0. T-DNA insertion mutants were obtained from the Arabidopsis Biological Resource Center (ABRC, Ohio State University, USA) and confirmed by PCR-based genotyping. Previously characterized mutants are hpr1-1 (SALK_067724) (Timm et al., 2008), cat2-1 (SALK_076998) (Queval et al., 2007), plgg1-1 (SALK_053469) (Pick et al., 2013), and glyr1-1 (SALK_057410) (Zarei et al., 2017). Previously uncharacterized mutants used in this study are glyr1-2 (SALK_202680C), glyr1-3 (SALK_203580C), and hpr2-3 (SALK_105876). Primers used for genotyping are listed in Table A1. Arabidopsis seeds were sown on plates containing half-strength Linsmaier and Skoog basal salt (1/2 LS, Caisson Labs), 1% sucrose and 0.8% agar (Phytoblend, Caisson Labs). After seed stratification in the dark at 4 °C for 3 to 7 days, plates were placed in the Percival Intellus Environmental Controller under normal lights (~100 μmol m-2 s-1 white light), 21 °C, and 12h/12h light/dark cycle. At ~1.5 weeks old, seedlings were transplanted into the soil and moved to the growth chambers with the same growth conditions. For high light treatment, 2-week-old plants were transferred to a chamber with ~700 μmol m-2 s-1 white light, and the same 32 temperature and photoperiod as those for NL growth. Plants grown under high CO2 were directly sown in the soil and grown under 2,000 ppm CO2, with the same settings for other parameters. For the glyoxylate and serine feeding experiments, glyoxylate or serine was filter- sterilized and added to the autoclaved medium to a final concentration of 0.4 mM and 0.1 mM, respectively. Plates were placed in the normal light Percival and high light growth chamber, respectively, at the same time. 2.4.2 Suppressor screening and mapping Seeds of hpr1-1 were treated with ethyl methanesulfonate (EMS) to induce random mutations. M1 plants were self-pollinated and M2 seeds were harvested. The M2 generation was grown under high light for suppressor screening, and individuals that grew bigger or greener than hpr1-1 were selected. Candidates with persistent suppressor phenotypes in the M3 generation were backcrossed to hpr1-1. Plants displaying the suppression phenotypes in the BC1F2 generation were used for additional backcrosses. In the BC3F2 generation of the shpr7 suppressor, 80 individuals with the suppression phenotypes were selected. Genomic DNA from these plants was extracted by Wizard Genomic DNA Purification Kit (Promega) and pooled together as one sample, which was used for DNBSEQ PE150 whole genome sequencing with 80X coverage. To minimize the influence of unrelated background mutations, 29 individuals from hpr1-1 were also sequenced with 30X coverage. Sequencing and bioinformatic analysis were performed by BGI Genomics (https://www.bgi.com). Selfed progenies from BC3F2 plants exhibiting suppression phenotypes were used for follow-up experiments. 2.4.3 Generation of transgenic lines and RT-PCR The 35S::FLAG-GLYR1 construct was generated by Gateway cloning according to manufacturer’s instructions (ThermoFisher). Briefly, the coding sequence (CDS) of GLYR1 was amplified from cDNA and cloned into the entry vector pDONR207 through the BP reaction. Next, GLYR1 was inserted into the destination vector pEarleyGate202 through the LR reaction. Entry and expression constructs were confirmed by Sanger sequencing. The construct was transformed into Agrobacterium tumefaciens strain GV3101 by electroporation, and Arabidopsis lines were transformed by Agrobacteria using floral dipping. Positive transformants were selected on 1/2 LS plates containing 1% sucrose and 10 μg/ml glufosinate ammonium. 33 For RT-PCR analysis, RNA was extracted from the leaf tissue using NucleoSpin RNA Plant kit (MACHEREY-NAGEL), then reverse transcribed using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems). The CDS of the target genes were PCR- amplified from cDNA using the GoTaq Green Master Mix (Promega) and gene-specific primers (Table A1). 2.4.4 Protein preparation and immunoblot analysis Whole rosettes from 4-week-old plants, which had been grown for 2 weeks under normal light followed by 2-week growth under high light, were collected and ground in liquid nitrogen, after which ~60 mg of the powder was homogenized with 500 μl Extraction Buffer [150 mM Tris-HCl (pH 6.8), 7.5% β-mercaptoethanol, 3% sodium dodecyl sulfate, and half a cOmplete mini protease inhibitor tablet (Roche)]. Samples were centrifuged at 17,000 xg at 4 °C for 10 min, and the supernatants were transferred to new tubes. A previously described method was used for immunoblotting (Lavell et al., 2021). Specifically, protein samples were combined with 4X Laemmli Buffer and boiled at 95 °C for 5 min. 40 μl samples were loaded onto a 4-20% Mini-PROTEAN TGX pre-cast gel (Bio-Rad) with the Dual Color Protein Ladder (Bio-Rad). After separation on the gel, proteins were transferred to a nitrocellulose membrane using Power Blotter XL System (Invitrogen) at 25V for 7min, which was then blocked with 5% milk in TBST for 1 h. The anti-GLYR1 antibody was developed by PhytoAB Inc. (CA, USA), based on the antigen peptide PAFPLKHQQKDMRLA with an additional C on C-terminus for conjugation. The membranes were incubated in 1:1000 rabbit anti-GLYR1 antiserum in the blocking buffer overnight at 4 °C. The membranes were washed multiple times with TBST, then incubated in 1:10,000 goat anti-rabbit IgG HRP secondary antibody (PhytoAB Inc.) in TBST for 2 h, washed again with TBST, and then developed with the SuperSignal West Pico PLUS Chemiluminescent Substrate kit (Thermo Scientific). A replicated protein gel with the same loading and running conditions was used for Coomassie staining. 2.4.5 Metabolite extraction and quantification Leaf samples were collected in the late afternoon, after ~10 h light treatment. Two to three well-expanded leaves with similar age were taken from plants with similar morphologies. In the cases where the mutants show different morphologies, the whole rosettes were taken. Plant tissue was frozen and ground in liquid nitrogen. 34 Metabolite extraction was performed using the methanol/chloroform/water method and quantification was conducted by Gas chromatography–mass spectrometry (GC-MS), as described previously (Fu et al., 2023). Specifically, 500 μl chloroform/methanol (3:7, v/v) was added to the tissue powder, followed by incubation at -20 °C for 2 h with occasional shaking. Then the internal standard adonitol and 400 μl water were added to the sample, followed by vigorously mixing and centrifugation at 4 °C, 12000 xg for 10 min. The upper phase (methanol in water) was taken, dried in a lyophilizer and stored in -80°C. Before analysis, samples were derivatized by adding 20 mg/ml methoxyamine hydrochloride dissolved in dry pyridine and incubated at room temperature overnight. Next, the reaction mixture was silylated by N, O-Bis (trimethylsilyl) trifluoroacetamide with 1% trimethylchlorosilane at 60°C overnight. The trimethylsilyl (TMS) derivatives were analyzed by an Agilent 7890A GC system/5975C inert XL Mass Selective Detector with a 30 m VF-5ms column (Agilent). 1 μl of each sample was injected into the 230°C inlet, where splitless, 10:1 split, or 20:1 split mode was chosen depending on the levels of the metabolites in plant samples. The oven temperature gradient is as follows: 40°C for 1 min, increased to 80°C in 1 min, then further increased by 10°C/min to 240°C followed by 20°C /min to 320°C, and 5 min holding at 320°C. Scan mode of the MS was used to monitor ions with mass to charge ratio (m/z) between 50–600. Metabolites were identified by m/z values, using retention time in comparison with authentic standards or the NIST Mass Spectral Library. Software MassLynx (Waters) was used for peak extraction and integration. Metabolites were quantified against the internal standard. 2.4.5 Photosynthetic measurement Using a previously reported method (Li et al., 2019a), the Dynamic Environment Photosynthesis Imager (DEPI) (Cruz et al., 2016) was used to measure photosystem II quantum efficiency (ΦII) of two-week-old seedlings grown in soil under regular normal light growth conditions as mentioned above. ΦII was calculated from (Fm′-Fs)/Fm′, where Fs is the chlorophyll fluorescence emission from light-adapted leaf at steady state and Fm’ is the maximum fluorescence from light-adapted leaf during a saturating pulse of light. Custom software (Visual Phenomics) was used to process the fluorescence images (Tessmer et al., 2013), and heatmaps were generated with OLIVER (https://caapp- msu.bitbucket.io/projects/oliver/index.html). 35 2.4.6 Tobacco infiltration and confocal microscopy (performed by Dr. Amanda Koenig and Joy Li) GLYR1 was N-terminally tagged with eYFP using the Gateway LR Clonase II Enzyme mix (ThermoFisher) by combining entry vector pENTR223-GLYR1 (ABRC, G82382) and pEarleyGate104, according to the manufacturer’s instructions. The expression vector (eYFP- GLYR1) was confirmed by Sanger sequencing and transformed into GV3101 Agrobacterium tumefaciens cells by electroporation. 5 ml Agrobacteria containing the peroxisome marker (mScarlet-SRL in pGWB2)(Koenig et al., 2023) and eYFP-GLYR1 were grown in Luria broth (LB, Rif25/Gent10/Kan50) with 225 rpm shaking at 28 °C overnight. Separate flasks containing 25 ml LB (Rif25/Gent10/Kan50) with 100 μM acetosyringone were inoculated with 1 ml of the overnight cultures and grown at 28 °C to OD600 between 0.6 and 0.8. The cells were harvested by centrifugation at 3000 xg for 10 min and then resuspended in MMA infiltration buffer (10 mM MES pH 5.7, 10 mM MgCl2, 100 μM acetosyringone) to adjust the OD600 to 0.6. The cell resuspensions were then incubated at room temperature (~22 °C) for 1 h. The peroxisome marker and eYFP-GLYR1 cultures were combined in 1:1 volume and infiltrated into 6-week-old Nicotiana tabacum leaves with a 1 ml needleless syringe. Tobacco plants were recovered in a 12h/12h photoperiod chamber for 2 days before imaging. Infiltrated tobacco leaves were imaged with an Olympus Fluoview 1000 spectral-based confocal laser scanning microscope with a 40X oil objective (NA: 1.30). Images were captured using the following excitation and emission parameters: eYFP-GLYR1 (ex: 515 nm, em: 530- 560 nm), mScarlet-SRL (ex: 559 nm, em: 580-615 nm), and chlorophyll autofluorescence (ex: 515 nm, em: 655-755 nm). 2.4.7 Hydroxypyruvate reductase (HPR) activity assay Using a previously reported method (Gregory et al., 2023), the maximum activity of HPR in plants was determined by NADH or NADPH oxidation. In brief, 50-150 mg leaf samples were frozen in liquid nitrogen, and ground with 1 ml Extraction Buffer on ice using 2 ml glass-to-glass homogenizer. After centrifugation, the supernatant was used for quantifying protein concentration and measuring enzymatic activity. Protein concentration was measured using Quick Start Bradford 1x Dye Reagent (BIO-RAD) and standards of bovine serum albumin (BIO- RAD), according to the manufacturer’s instructions. For enzyme activity, a total of 200 μl 36 reaction in 96-well microplate were measured. 4-8 μl crude protein extract with 188-192 μl reaction buffer were used for NADH-dependent enzyme activity, while 16-20 μl crude protein extract with 176-180 μl reaction buffer were used for NADH-dependent enzyme activity. The reaction was initiated by adding 4 μl 25 mM Na beta‐hydroxypyruvate, and the absorbance at 340 nm was monitored for 5 min. The decrease in absorbance per min and the extinction coefficient of NAD(P)H (6.22 mM-1 cm-1) were used to calculate HPR activity in plants, which was then normalized to the protein content in the samples. 2.4.8 Statistics and plots Data analysis is performed using Microsoft Excel and Rstudio. Student’s unpaired two- tailed t-test was used for pairwise comparison. One-way ANOVA with Tukey’s HSD test was used for multi-comparison. Boxplot center lines show the medians, and the lower and upper hinges correspond to the first and third quartiles. Whiskers extend from the hinges to the largest or smallest values that are no further than 1.5 times of the interquartile range. Data points are represented as individual dots. 37 Figures Figure 2.1. The established photorespiration pathway and the cytosolic glyoxylate shunt proposed in this study. Photorespiration involves a series of reactions in the chloroplast, peroxisome, mitochondrion, and cytosol. We propose that defective GLYR1 allows the overly accumulated free glyoxylate in the cytosol to react with serine, catalyzed by an unknown aminotransferase. The hydroxypyruvate produced can be further converted through HPR2 to glycerate, which re-enters the chloroplast. Abbreviations: 2-OG, 2-oxoglutarate; 2-PG, 2- phosphoglycolate; 3-PGA, 3-phosphoglycerate; BASS6, bile acid sodium symporter 6; CAT, catalase; GDC, glycine decarboxylase complex; GGAT, glutamate:glyoxylate aminotransferase; GLYK, glycerate kinase; GLYR1, glyoxylate reductase 1; GOX, glycolate oxidase; HPR, hydroxypyruvate reductase; PGLP, 2-PG phosphatase; PLGG1, plastidial glycolate/glycerate transporter 1; Rubisco, RuBP carboxylase/oxygenase; RuBP, ribulose-1,5-bisphosphate; SGAT, serine:glyoxylate aminotransferase; SHMT, serine hydroxymethyltransferase. Created with BioRender.com. 38 Figure 2.2. Loss of function of GLYR1 partially rescues the growth phenotypes of hpr1. (a) Plants grown for 2 weeks (2w) under normal light (NL, 100 µmol m-2 s-1) followed by 2 or 2.5 weeks of growth under high light (HL, 700 µmol m-2s-1), or constantly grown under normal light for 4.5 weeks. Scale bars = 5 cm. (b) Radius measurements of the total rosette of 4-week-old plants grown under 2w normal light + 2w high light. Different letters indicate statistically significant differences (p<0.05), which were determined by One-way ANOVA with Tukey’s HSD test. Biological replicates: n=10 for Col-0, n=11 for glyr1-1 and glyr1-1 hpr1-1, n=13 for hpr1-1, and n=8 for shpr7. (c) Schematic depiction of the GLYR1 gene and positions of the mutations in various alleles. (d) Immunoblot analysis of the GLYR1 protein from different genotypes. The GLYR1 protein was detected by an anti-GLYR1 peptide antibody (top). Rubisco stained by Coomassie Blue in the SDS-PAGE gel was used as a loading control (bottom). (e) Overexpressing GLYR1 reverts the suppression phenotype back to the hpr1 mutant phenotype. Plants were grown under 2-week normal light followed by 2.5-week high light. Scale bars = 3 cm. 39 Figure 2.3. Profiling of stable photorespiratory metabolites in 4-week-old high light-treated plants. Plants were grown under 2-week normal light followed by 2-week high light. Different letters indicate statistically significant differences (p<0.05), which were determined by One-way ANOVA with Tukey’s HSD test. Biological replicates: n=6. 40 Figure 2.4. Analysis of the impact of defective GLYR1 on other photorespiratory mutants and GLYR1 protein localization. (a) Plants grown under 2-week normal light followed by 2- week high light. Scale bars = 3 cm. (b) Heatmap of photosystem II quantum efficiency for 2- week-old plants under dynamic and high light conditions. Values for the mutants were normalized to that of the wild-type Col-0. Biological replicates: n=12 for Col-0, cat2-1, glyr1-1 cat2-1, glyr1-1 hpr1-1, and plgg1-1, n=11 for glyr1-1 and hpr1-1, and n=13 for plgg1-1. (c) Maximum intensity Z-projection of confocal images spanning 30 µm of tobacco leaf tissue co- expressing eYFP-GLYR1 (cyan) and mScarlet-I-SRL (peroxisomes, red). Chloroplast signals are from chlorophyll autofluorescence (yellow). Scale bar = 10 μm. Images in (c) were generated by Amanda Koenig. 41 Figure 2.5. Profiling of transitional photorespiratory metabolites in plants transferred to the photorespiratory environment. Plants were grown under 3 weeks of high CO2 and normal light, and then transferred to ambient CO2 and high light before lights were turned on. Leaf tissue was sampled after ~10 h. Different letters indicate statistically significant differences (p<0.05), which were determined by One-way ANOVA with Tukey’s HSD test. Biological replicates: n=6. 42 Figure 2.6. Plant growth after feeding with glyoxylate. Growth phenotype (a) and fresh weight (b) of 12-day-old seedlings on plates with or without glyoxylate under normal light (NL) or high light (HL) are recorded. The p values determined by Student’s unpaired two-tailed t-test were labeled on the plot. Seedlings grown on the same plate were treated as a biological replicate. Biological replicates: n=4. Scale bars = 3 cm. 43 Figure 2.7. The role of glyr1 is partially dependent on HPR2. (a) Plants grown under 3 weeks of high CO2 and normal light before being transferred to ambient CO2 and high light on Day 0. Scale bars = 5 cm. (b) Glycolate and hydroxypyruvate levels at the transitional stage. Plants were grown under 3 weeks of high CO2 and normal light, and then transferred to ambient CO2 and high light before the lights were on. Leaf tissue was sampled after ~10 h. Different letters indicate statistically significant differences (p<0.05), which were determined by One-way ANOVA with Tukey’s HSD test. Biological replicates: n=5. 44 CHAPTER 3. Defective GLYR1 largely reverts the broad transcriptional reprogramming of the hpr1 mutant under high light conditions 3.1 Introduction In Chapter 2, under high light, deficiency in GLYR1 was shown to partially rescue the phenotypes of the photorespiration mutants hpr1 and cat2 but not those of plgg1. Further genetic, physiological and metabolic analyses supported a novel photorespiratory glyoxylate shunt in the cytosol that allows more photorespiratory carbon to be recycled back to the Calvin-Benson cycle. However, it is also important to obtain a comprehensive view of how defects in GLYR1 impact the transcriptome and cell functions globally. To this end, I performed RNA-seq analysis to analyze the transcriptome of Col-0, hpr1-1, glyr1-1, glyr1-1 hpr1-1, and shpr7. Since transcriptomic data obtained from plants under prolonged photorespiratory conditions may not reflect transcriptional changes directly caused by the imposed conditions, we chose to focus on the effects of defective GLYR1 under short-term photorespiratory conditions to reduce secondary effects. Plants were grown for 3 weeks under normal light and high CO2 (2,000 ppm CO2), where photorespiration is largely inhibited. At the end of the dark period, plants were transferred to ambient air with high light conditions to induce photorespiration. Leaf samples harvested after 3 h and 10 h of the treatment were used for RNA- seq. 3.2 Results 3.2.1 Defective GLYR1 has little influence on the transcriptome of Col-0 but extensive impact on that of hpr1 Principal component analysis (PCA) was first employed to visualize the variations in the RNA-seq data (Fig. 3.1). At both 3 h and 10 h, Col-0 and glyr1-1, and glyr1-1 hpr1-1 and shpr7, were respectively clustered together, with both clusters clearly separated from hpr1-1. This clustering pattern is consistent with the growth phenotypes of these lines, validating the high quality of the RNA-seq data and clear, reproducible transcriptomic response of the mutants. Analysis of differentially expressed genes (DEGs) revealed 10,139 and 16,104 DEGs at 3 h and 10 h respectively, between hpr1-1 and Col-0, demonstrating the remarkable transcriptional reprogramming in hpr1-1 (Fig. 3.2). Interestingly, while glyr1-1 itself had relatively few DEGs  Some results in this chapter have been submitted for publication. The analyses of RNA-seq data were mostly performed by Nicholas Panchy. 45 (59 at 3 h and 12 at 10 h), this mutation dramatically altered gene expression in the hpr1-1 background. At 10 h in particular, the number of DEGs in glyr1-1 hpr1-1 (15,955) or shpr7 (15,795) compared to hpr1-1 was very close to that in hpr1-1 compared to Col-0, highlighting the strong impact of GLYR1 loss of/reduced function in hpr1-1. Current understanding of the potential function of GLYR1 under stress conditions is mainly from its enzymatic activities as GLYR and succinic semialdehyde reductase, with little known about the downstream components. To elucidate the genes and pathways affected by the loss of function of GLYR1, I chose the top 5 DEGs, which have large fold changes (>2 fold) and small p-values, in glyr1-1 compared to Col-0 from both time points. Gene expression data in all genotype comparisons at both 3 and 10 h (Table 3.1, first 5 genes) and the gene descriptions (Table 3.2, first 5 genes) of these DEGs were collected. Interestingly, the expression patterns of these 5 DEGs in shpr7 vs. hpr1 or glyr1 hpr1 vs. hpr1 comparisons were mostly different from glyr1-1 vs. Col-0 (Table 3.1), suggesting that these DEGs have little function in the glyr1- induced transcriptional reprogramming in hpr1-1. For example, AT2G43820, which encodes a glycosyltransferase, was up-regulated in all pairwise comparisons at 3 h, showing a consistent effect of glyr1 in Col-0 and hpr1-1 background (Table 3.1 and 3.2). However, while this gene has a comparable expression level in glyr1-1 as in Col-0 at 10 h, glyr1 induced a repression of this gene in the hpr1-1 background. AT3G06325, another gene that is predicted to generate an antisense RNA, was up-regulated in both glyr1-1 and hpr1-1 at 10 h, but defective GLYR1 in hpr1-1 down-regulated this gene (Table 3.1 and 3.2). The functions of these 5 DEGs and their connections with GLYR1 are largely unknown (Table 3.2). Further investigations are needed to fully understand the role of GLYR1 under high light and other stress conditions. Among the top 5 DEGs, AT5G19880, a gene belonging to the peroxidase superfamily, had high and similar expressions in hpr1-1, glyr1-1 hpr1-1 and shpr7 at 3 h, which is inconsistent with the general rescuing trend of glyr1 in hpr1 (Table 3.1 and 3.2). To further determine the pattern for changes in gene expression, I chose 5 additional DEGs from the top in hpr1-1 for analysis (Table 3.1 and 3.2, last 5 genes). The expression levels of these DEGs were similar in glyr1-1 and Col-0, but their altered expressions in hpr1-1 were all at least partially reverted in glyr1-1 hpr1-1 and shpr7 (Table 3.1), which supports the strong impact of defective GLYR1 in hpr1-1. The functions of the last 4 DEGs (#6-#10) are all related to stress (Table 3.2), indicating that the stressful cellular environment in hpr1-1 can be alleviated by glyr1. The 46 expression of AT2G26400, which is predicted to encode an acireductone dioxygenase family protein (Table 3.2), was highly repressed in hpr1-1, but this repression was totally rescued in glyr1-1 hpr1-1 and shpr7 (Table 3.1), making this unknown protein a good candidate for an important player in the glyr1-induced rescue of hpr1. Overall, few DEGs were found between glyr1-1 and Col-0, but there were huge numbers of DEGs between glyr1-1 hpr1-1 or shpr7 and hpr1-1. The DEGs induced by the deficiency of GLYR1 in Col-0 do not seem to play important roles in hpr1, suggesting that glyr1 and hpr1 mutations together activate a distinct mechanism for hpr1 rescue. 3.2.2 Defective GLYR1 largely reverts the broad transcriptional reprogramming in hpr1 To further determine the impact of glyr1 on the transcriptome of hpr1, the DEGs that have a more than 4-fold change in hpr1-1 compared to Col-0 at 10 h were selected to generate a heatmap (Fig. 3.3). Almost all these DEGs showed reversion to some extent in glyr1-1 hpr1-1 and shpr7; the majority of them even had comparable or nearly comparable levels of genes expression in glyr1-1 hpr1-1, shpr7 and Col-0. This data indicates that glyr1 can effectively revert the transcriptional reprogramming in hpr1. To decipher the biological pathways affected, we employed weighted gene co-expression network analysis (WGCNA) to identify co-expression modules. Using the soft threshold of 30 (Fig. 3.4a), a total of 17 co-expression modules, whose sizes ranged from 11 to 4,680 genes, were constructed (Fig. 3.4b and c, Table 3.3). Generally, the modules that were associated with 3 h were different from those for 10 h (Fig. 3.4c), indicating that the transcriptome is being remodeled along with the increased exposure to high light. Most modules were correlated with hpr1-1 but rarely with the other lines, suggesting the large rescuing effects of defective GLYR1 in hpr1-1. One exception is the lightcyan module, which had positive correlations with Col-0 and glyr1-1 at 10 h but no other lines (Fig. 3.4c). Gene Ontology (GO) enrichment analysis of this module showed enrichment of the biosynthetic process for anthocyanin-containing compounds (Fig. 3.4d), which is consistent with the reduced anthocyanin phenotype in hpr1-1, glyr1-1 hpr1- 1 and shpr7 as shown in Chapter 2. The top modules that comprise large numbers of genes, including turquoise (4,680 genes), blue (3,886 genes), brown (3,047 genes), yellow (2,310 genes), green (2,144 genes) and red (1,815 genes), were all closely associated with hpr1-1 but rarely with the other lines (Fig. 3.4c, Table 3.4). Among these, the yellow and red modules were positively and negatively 47 associated with hpr1-1, respectively, at 3 h (Fig. 3.4c). GO enrichment analysis showed that, at 3 h, stress response was activated in hpr1-1 whereas protein translation-related processes were repressed (Fig. 3.5), suggesting that the cell environment in hpr1-1 at 3 h became stressful and started to disrupt housekeeping activities. The other 4 modules were associated with hpr1-1 at 10 h, with blue and green modules positively associated and turquoise and brown modules negatively associated. The blue, green and turquoise/brown modules were enriched in protein degradation (Fig. 3.6a), nuclear proteins (Fig. 3.6b), and chloroplast-related proteins (Fig. 3.6c and d), respectively. These results suggest that the stressful cell environment in hpr1-1 may accelerate protein damage and degradation, especially for chloroplast proteins, and that the transcription machinery is highly activated to cope with this adversary. These data are also consistent with the accumulation of photorespiratory intermediates in hpr1-1 (see Chapter 2), which likely induces a stressful environment that interferes with broad cellular functions. Defective GLYR1 partially rescues this metabolic accumulation in hpr1-1, which helps to prevent the unfavorable cell environment and consequently largely reverts the transcriptional reprogramming in hpr1-1. 3.2.3 Photorespiratory and photosynthetic genes are largely suppressed in hpr1 and rescued by glyr1 at 10 h To determine the relationship between GLYR1 and photorespiration or photosynthesis, the expression data for genes in these two pathways were extracted from the RNA-seq data (Fig. 3.7). Most of the photorespiratory and photosynthetic genes showed a small shift in expression between 3 and 10 h, suggesting that photorespiration and photosynthesis are dynamic in early high light response. All lines showed similar expression patterns in these genes at 3 h, but at 10 h hpr1-1 displayed a unique pattern of most decreased expression of these genes, indicating that photorespiration and photosynthesis were generally suppressed in hpr1-1, a pattern that got reverted by glyr1. The top 4 genes in the expression heatmap of photorespiration showed a different pattern from the other genes because they were highly up-regulated in hpr1-1 at 10 h (Fig. 3.7a). None of these genes encode major components in photorespiration (Table 3.4), thus their upregulation in hpr1-1 may be due to their function in other biological pathways. To obtain more information about the photorespiratory glyoxylate shunt in the cytosol proposed in Chapter 2, gene expression data of candidate aminotransferases, AGT2, AGT3 and 48 PYD4, was gathered along with HPR2 in all genotype comparisons at both 3 and 10 h (Table 3.5). The expression of AGT2 only had small changes in the mutants, but it was repressed in hpr1-1 at 10 h and rescued by defective GLYR1, similar to that of HPR2 and most of the photorespiratory genes. In contrast, at 10 h, AGT3 and PYD4 were up-regulated in hpr1-1, which was reverted by GLYR1 defects. Among the three candidates, PYD4 is the only gene that differentially expressed in hpr1-1 at 3 h and had the biggest alternations of expression among genotypes. However, the expression patterns of these three candidate aminotransferases did not provide directive clues about their involvement in the glyoxylate shunt, considering their potential functions in other pathways and the possible regulations on the enzymatic level. Therefore, investigations on the localizations and enzymatic activities of these candidates are still the key to determining the aminotransferase involved in the cytosolic glyoxylate shunt. Taken together, the expression of genes related to photorespiration and photosynthesis was largely suppressed in hpr1 at 10 h, a pattern that was mostly rescued by glyr1. It is likely that photorespiration and photosynthesis are disrupted by the stressful environment in hpr1. 3.3 Discussions In this work, I have shown that while glyr1-1 itself did not cause many transcriptional changes in the wild-type background, defects in GLYR1 significantly rescued the hpr1-1 mutant at the transcriptome level under high light conditions. The top DEGs in glyr1-1 may be the key to understanding the function of GLYR1 in Col-0, but these genes do not seem to play important roles in the hpr1-1 background, suggesting that lacking the function of both GLYR1 and HPR1 activates a distinct mechanism which otherwise has minimal influence in the glyr1-1 single mutant. Broad and high-level transcriptional reprogramming was found in hpr1, probably because of the accumulation of photorespiratory metabolites. Due to the impaired main photorespiratory pathway in the peroxisome in hpr1, the photorespiratory intermediates gradually accumulate in the cell as plants are moved to high light conditions, causing an unfavorable environment that disrupts regular biological processes. The lack of GLYR1 in hpr1 can activate the cytosolic glyoxylate shunt proposed in Chapter 2, which helps to recycle more carbon back to the Calvin- Benson cycle and therefore partially prevents the formation of a stressful cell environment. Photorespiratory and photosynthetic genes were generally suppressed in hpr1 at 10 h, and this 49 suppression was largely rescued by glyr1. This pattern may be similar to the genes in other primary cellular processes in a stressful environment. These results support the existence of the photorespiratory glyoxylate shunt in the cytosol, but other possible mechanisms for how glyr1 rescues hpr1 also exist. The genes that are differentially expressed in hpr1-1 and fully reverted in glyr1-1 and shpr7 at 3 h may be good candidates for the key regulators in this rescue. 3.4 Methods 3.4.1 Plant growth and RNA extraction Arabidopsis lines used in this Chapter were described in Chapter 2. Seeds were directly sown to soil in pots and stratified in the dark at 4 °C for 3 to 7 days. Then the pots were moved to high CO2 (2,000 ppm) conditions, ~100 μmol m-2 s-1 white light, 21 °C, and 12h/12h light/dark cycle for growth. After growing under high CO2 for 3 weeks, plants were moved to ambient air condition at the end of the dark period, followed by 3 h or 10 h of high light treatment at ~700 μmol m-2 s-1. Five biological replicates each for Col-0, hpr1-1, glyr1-1, glyr1-1 hpr1-1, and shpr7 were used, and 2-3 well-expanded leaves of similar age were sampled on each plant. Plant tissue was frozen and ground in liquid nitrogen. Total RNA was isolated using the NucleoSpin RNA Plant kit (MACHEREY-NAGEL), after which DNA was further depleted using the TURBO DNA-free Kit (Invitrogen). 3.4.2 RNA sequencing and data analysis RNA-seq was performed by the MSU Genomics Core, who prepared mRNA libraries with the KAPA mRNA HyperPrep Kit (Roche) and pooled them together with samples from other researchers onto one S4 lane of Illumina NovaSeq 6000. The sequencing yielded 2x150 bp paired end reads with ~20M read pairs per sample. Analysis of the RNA-seq data was mainly performed by Nicholas Panchy (MSU Bioinformatics Core), who used the nf-core/rnaseq v3.10.1 pipeline (https://zenodo.org/records/7505987) built with Nextflow v22.10.4 (Di Tommaso et al., 2017) to process and quantify transcriptomic reads using the standard defaults unless otherwise specified. Briefly, Salmon v1.9.0 (Patro et al., 2017) and the fq v0.9.1 (https://github.com/stjude-rust- labs/fq) were used to sub-sample FastQ files and auto-infer read strandedness. Adapter and quality trimming were performed using Trim Galore! v0.6.7 50 (https://zenodo.org/records/5127899) with Cutadapt v3.4 (Martin, 2011). STAR v2.7.9a was used to map the raw FastQ reads to the reference genome and project the alignments onto the transcriptome (Dobin et al., 2013). The alignments were sorted and indexed using SAMtools v1.16.1 (Li et al., 2009), and downstream BAM-level transcript quantification was performed with Salmon v1.9.0 with the --seqBias --gcBias tags (Patro et al., 2017). Reads were mapped to the TAIR 10.1 version of the A. thaliana genome (GCA_000001735). Genes with missing transcript IDs were filtered from the GTF file, which is a known issue with some GTF files (see https://github.com/nf-core/rnaseq/issues/1086). DESeq2 v1.38.3 was used to perform differential expression analysis for 3-h and 10-h samples separately (Love et al., 2014). Tximport v1.26.1 was used to import transcript abundances and construct a gene-level DESeqDataSet object from Salmon quant.sf files (Soneson et al., 2016). Genes were filtered for those with a count of at least 10 in 5 samples. The fgsea v1.24.0 was used for pre-ranked gene set enrichment analysis (GSEA), and the genes were ranked by the Wald statistic (Korotkevich et al., 2021). Gene ontology enrichments were performed using the TAIR annotation file for A. thaliana downloaded from http://current.geneontology.org/products/pages/downloads.html from the 2024-04-24 release. Heatmaps of differentially expressed, photosynthetic, and photorespiration genes were generated using the pheatmap v1.0.12 package (https://github.com/raivokolde/pheatmap). Genes (rows) were clustered using row scaled, Euclidean distance and the ward.D clustering approach. Weighted Gene Co-expression Network Analysis (WGNCA), including network construction, module trait correlations, and intramodular connectivity was performed with the WCGNA v1.72-5 package (Langfelder & Horvath, 2008, 2012), incorporating samples from the 3 and 10 h samples for best results. Genes were filtered as in differential gene expression analysis and count data was normalized using the getVarianceStabilizedData function from the DEseq2 package. A soft power threshold of 30 was used by applying the elbow criterion to scale independence and mean connectivity. For module trait correlation analysis, the different combinations of time and genotype were discretized into presence (1)/ absence (0) values. Enrichment of gene ontology for WGCNA modules was done using the same sources for GO annotations, but with a hypergeometric test using the phyper function. However, fgsea v1.24.0 was also used to test the enrichment of differentially expressed genes in modules using the previous ranking approach for DEGs. 51 Figures Figure 3.1. Principal Component Analysis plots of RNA-seq samples. All 5 biological replications of each line at 3 h (a) and 10 h (b) time points were plotted. dubl, the glyr1-1 hpr1-1 double mutant. 52 Figure 3.2. Volcano plots to display differentially expressed genes (DEGs). Four pairwise comparisons: hpr1-1 vs. Col-0, glyr1-1 vs. Col-0, shpr7 vs. hpr1-1, and glyr1-1 hpr1-1 vs. hpr1- 1 at 3 h (a) and 10 h (b) are shown. Genes that have an adjusted p-value of < 0.01 are considered DEGs and shown above the gray dashed line. Blue, red, and black dots represent the down- regulated, up-regulated, and unchanged genes, respectively. 53 Figure 3.3. Heatmap of the expression of selective DEGs. DEGs in hpr1-1 at 10 h that have an absolute log2 fold ratio>2 and an adjusted p-value of < 0.01 were selected. Gene expression was normalized to have an average value of 0 and indicated by the color legend. 54 Figure 3.4. Construction of co-expression networks by weighted gene co-expression network analysis (WGCNA). (a) Plots showing the relationships between soft threshold and scale independence (left) and between soft threshold and mean connectivity (right). Soft threshold is determined by the elbow criterion. (b) A hierarchical cluster tree showing co- expression modules identified by WGCNA. Different colors represent different gene modules. The grey module includes genes that failed to be assigned. (c) Heap map of correlations between gene modules and plant genotypes at both time points. The correlation value is indicated by the color legend, and the number in parenthesis in each cell represents the adjusted p-value of the correlation. dubl, the glyr1-1 hpr1-1 double mutant. (d) Gene Ontology (GO) enrichment analysis in the lightcyan module. 55 Figure 3.4 (cont’d) 56 Figure 3.5. Gene Ontology (GO) enrichment analysis of modules closely connected to hpr1 at 3 h. The top 10 enriched terms in the yellow (a) and red (b) modules are shown. 57 Figure 3.6. Gene Ontology (GO) enrichment analysis of modules closely connected to hpr1 at 10 h. The top 10 enriched terms in the blue (a), green (b), turquoise (c) and brown (d) modules are shown. 58 Figure 3.6 (cont’d) 59 Figure 3.7. Heatmap of the expression of photorespiratory and photosynthetic genes. Expression of photorespiration (a) and photosynthesis (b) genes were normalized to have an average of 0 and indicated by the color legend. 60 Figure 3.7 (cont’d) 61 Table 3.1. Expression of selected top DEGs. 3h Tables Gene ID AT1G20390 AT5G19880 AT2G43820 AT2G05380 AT3G06325 AT2G26400 AT2G36790 AT1G05680 AT5G43450 AT2G41730 AT1G20390 AT5G19880 AT2G43820 glyr1 vs. Col-0 -2.64 19.81 2.74 / / / / / / / 6.25 E-29 4.72 E-11 1.01 E-14 AT2G05380 n.s. 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 AT3G06325 n.s. n.s. n.s. n.s. AT2G26400 n.s. 4.29 E-06 AT2G36790 n.s. 0 AT1G05680 n.s. 3.3 E-268 AT5G43450 n.s. 0 10 AT2G41730 n.s. 1.7 E-286 hpr1 vs. Col-0 shpr7 vs. hpr1 dubl vs. hpr1 log2 Fold Change 10h glyr1 vs. Col-0 hpr1 vs. Col-0 shpr7 vs. hpr1 / 6.32 8.88 4.96 7.78 / 5.42 14.19 5.51 6.38 n.s. 4.68 E-04 -1.67 -4.66 -5.17 / -5.75 / -4.2 -5.77 -3.24 -4.75 4.13 E-15 7.73 E-03 0 0 -1.74 18.71 4.73 3.24 / -18.42 7.98 13.2 7.02 8.81 7.78 E-15 2.28 E-12 4.25 E-50 1.09 E-21 / / 1.18 / / 27.11 -5.31 -4.61 -4.35 -6.42 / / 1.25 / / 25.67 -5.39 -4.71 -4.34 -6.44 padjust n.s. n.s. -2.67 / / 1.57 4.02 / / / / / 8.78 E-31 n.s. n.s. n.s. 3.46 E-04 1.48 E-04 n.s. n.s. n.s. 8.55 E-08 3.16 E-03 dubl vs. hpr1 -2.43 / -4.84 / -5.21 / -3.31 -5.19 -2.53 -3.48 1.25 E-27 n.s. 1.1 E-296 9 E-100 1.37 E-20 n.s. n.s. 4.51 E-20 8.72 E-18 1.03 E-11 4.7 E-264 1.37 E-74 1.8 E-135 9.7 E-168 1.19 E-10 2.1 E-271 1.68 E-77 5.5 E-135 3.4 E-168 n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. 9.5 E-110 2.9 E-59 1.9 E-56 7.5 E-134 2.72 E-69 1.63 E-12 1.23 E-20 3.02 E-78 3.04 E-44 2.3 E-10 4.15 E-13 3.91 E-43 62 Table 3.2. Functional description of selected top DEGs. Gene ID Gene description 1 AT1G20390 Transposable_element_gene; gypsy-like retrotransposon family. 2 AT5G19880 Peroxidase superfamily protein. 3 AT2G43820 UDP-GLUCOSYLTRANSFERASE 74F2 (UGT74F2); ARABIDOPSIS THALIANA SALICYLIC ACID GLUCOSYLTRANSFERASE 1 (ATSAGT1); UDP-GLUCOSE:SALICYLIC ACID GLUCOSYLTRANSFERASE 1 (SGT1); Encodes a nicotinate-O- glycosyltransferase. Induced by Salicylic acid; virus; fungus and bacteria. Also involved in the tryptophan synthesis pathway. Independent of NPR1 for their induction by salicylic acid. 4 AT2G05380 Glycine-rich protein 3 short isoform (GRP3S) mRNA; the mRNA is cell- to-cell mobile. 5 AT3G06325 Natural antisense transcript overlaps with AT3G41762. 6 AT2G26400 Encodes a protein predicted to belong to the acireductone dioxygenase family. 7 AT2G36790 UDP-GLUCOSYL TRANSFERASE 73C6 (UGT73C6); encodes a UDP- glucose:flavonol-3-O-glycoside-7-O-glucosyltransferase attaching a glucosyl residue to the 7-O-position of the flavonols kaempferol, quercetin and their 3-O-glycoside derivatives. 8 AT1G05680 URIDINE DIPHOSPHATE GLYCOSYLTRANSFERASE 74E2 (UGT74E2); Encodes a UDP-glucosyltransferase that acts on IBA (indole-3-butyric acid) and affects auxin homeostasis. The transcript and protein levels of this enzyme are strongly induced by H2O2 and may allow integration of ROS (reactive oxygen species) and auxin signaling. 9 AT5G43450 Encodes a protein whose sequence is similar to ACC oxidase. 10 AT2G41730 HRG1; H2O2 response gene; sensor/responder of H2O2; involved in maintaining embryonic root meristem activity. Expression in rosette leaves is activated by high concentration of boron. 63 Table 3.3. Number of genes in each co-expression module determined by WGCNA. moduleColors nGenes black blue brown cyan green greenyellow grey lightcyan magenta midnightblue pink purple red salmon tan turquoise yellow 148 3886 3047 36 2144 90 843 11 121 17 127 91 1815 38 76 4680 2310 Table 3.4. The first 4 photorespiratory genes in the expression heatmap shown in Fig. 3.7a. Gene ID Gene name AT1G36370 MORE SULPHUR ACCUMULATION1 (MSA1); SERINE HYDROXYMETHYLTRANSFERASE 7 (SHM7) AT4G13890 SERINE HYDROXYMETHYLTRANSFERASE 5 (SHM5); EMBRYO SAC DEVELOPMENT ARREST 36 (EDA36); EMBRYO SAC DEVELOPMENT ARREST 37 (EDA37) AT3G17240 LIPOAMIDE DEHYDROGENASE 2 (mtLPD2) AT1G22020 SERINE HYDROXYMETHYLTRANSFERASE 6 (SHM6) 64 Table 3.5. Expression of candidates for the hypothetical cytosolic aminotransferase proposed in Chapter 2. Gene ID Gene name AT1G79870 HPR2 AT4G39660 AGT2 AT2G38400 AGT3 AT3G08860 PYD4 3h glyr1 vs. Col-0 / / / / hpr1 vs. Col-0 / / / 3.87 shpr7 vs. hpr1 / / / / 10h dubl vs. hpr1 / / / / glyr1 vs. Col-0 / / / / hpr1 vs. Col-0 -1.6 -0.78 1.44 4.24 shpr7 vs. hpr1 1.41 0.61 -1 -7.97 dubl vs. hpr1 1.5 0.71 -1.09 -4.49 65 CHAPTER 4. The role of photorespiration in plant immunity 4.1 Introduction Plants have developed a sophisticated immune system during their co-evolution with pathogens in nature, which include two interacting and connected layers: pattern-triggered immunity (PTI) and effector-triggered immunity (ETI) (Dodds & Rathjen, 2010; Yuan et al., 2021). PTI is triggered by the elicitors from pathogens, like the peptide flg22, which comes from the conserved domain of the bacterial flagellin (Yu et al., 2017). During PTI, intracellular signaling, transcriptional reprogramming, and other physiological responses such as reactive oxygen species (ROS) burst, callose deposition, and biosynthesis of phytohormones salicylic acid (SA) and jasmonate (JA), limit pathogen growth (Yu et al., 2017). ETI is activated by the recognition of virulent effectors that are secreted from pathogens into the plant cell (Cui et al., 2015). ETI responses are similar to PTI but stronger, and often lead to local programmed cell death called hypersensitive response (HR) (Cui et al., 2015). As introduced in detail in Chapter 1, evidence for the role of photorespiration in plant immunity is emerging. Studies of the roles of photorespiration in plant-pathogen interaction have focused on H2O2 since photorespiration is considered as a major source of H2O2 in photosynthetic cells (Foyer et al., 2009). In the photorespiratory pathway, glycolate is converted to glyoxylate by glycolate oxidases (GOXs) in the peroxisome, producing H2O2 that is then scavenged by catalases (CATs). Defects in GOXs generally induce depressed immune response and susceptible disease phenotypes in Arabidopsis, tobacco and tomato (Rojas et al., 2012; Ahammed et al., 2018), suggesting that the H2O2 produced by GOXs is important to immunity. In cat mutants, H2O2 accumulation along with SA-dependent defense phenotypes was found under normal conditions without pathogen infection (Takahashi et al., 1997; Chamnongpol et al., 1998; Mittler et al., 1999; Chaouch et al., 2010; Chaouch & Noctor, 2010), and Arabidopsis CAT2 has been shown to coordinate defense signaling (Giri et al., 2017; Yuan et al., 2017; Lv et al., 2019), highlighting the crucial role of CAT-mediated H2O2 in immune response. It has also been reported that CAT and GOX can act together to regulate H2O2 homeostasis in defense response (Zhang et al., 2016; Williams et al., 2018). Additionally, the glycine decarboxylase complex (GDC) (Navarre & Wolpert, 1995; Yao et al., 2002; Cristina Palmieri et al., 2010; Gilbert & Wolpert, 2013), serine hydroxymethyltransferase (SHMT) (Fu et al., 2022), and 66 glutamate:glyoxylate aminotransferase (GGAT) (González‐lópez et al., 2021) have been shown to involve in ROS homeostasis in immunity. Photorespiratory metabolites also contribute to plant immunity. A soybean cytosolic SHMT, which has impaired tetrahydrofolate (THF) binding, was identified to confer resistance to the soybean cyst nematode (Liu et al., 2012; Kandoth et al., 2017; Korasick et al., 2020), suggesting the connection between folate metabolism and immunity. In soybean, HPR was also found to interact with P34, the receptor for the P. syringae elicitor syringolide, and applying glycerate and 3-PGA to the plant was able to restrain syringolide-induced HR (Okinaka et al., 2002). Additionally, photorespiration-associated amino acids have been shown to induce disease resistance in plants (Kadotani et al., 2016; Yang et al., 2017; Toyota et al., 2018). Other photorespiratory enzymes, including serine:glyoxylate aminotransferase (SGAT) (Taler et al., 2004; Ahammed et al., 2018) and glycerate kinase (GLYK) (Gao et al., 2020), were also reported to play roles in plant-pathogen interaction, but the underlying mechanisms are unclear. Although studies have shown that photorespiration impacts plant immune response through multiple processes, our current understanding of the underlying mechanism is still largely fragmentary. To further investigate the interaction between the photorespiratory pathway and plant immune system, I employed two strategies using the model plant Arabidopsis thaliana and the bacterial pathogen Pseudomonas syringae pv. tomato strain DC3000 (Pst DC3000). To test the hypothesis that photorespiration plays a positive role in plant immunity, I first assessed disease phenotypes of photorespiratory mutants and found that the peroxisomal photorespiratory enzyme HPR1 and the chloroplastic glycolate/glycerate transporter PLGG1 contribute to plant immune response via the photorespiratory pathway. Additionally, I quantified photorespiration under pathogen infection using the gas exchange experiment, but did not find evidence for the upregulation of photorespiration that had been reported by previous studies. 4.2 Results 4.2.1 Null mutants of HPR1 and PLGG1 show defects in immune response To understand whether components of the photorespiratory pathway influence the warfare between the plant and pathogen, I tested a collection of Arabidopsis photorespiratory mutants, including those deficient in enzymes or transporters. Among the mutants, hpr1 and plgg1 consistently showed increased disease susceptibility. 67 Consistent with previous reports of the growth phenotypes of photorespiratory mutants (Timm et al., 2008; Yang et al., 2012), both hpr1-1 and plgg1-1 plants grown in ambient air exhibited small rosettes; in addition, plgg1-1 also had lesions on leaves (Fig. 4.1a). When infiltrated with Pst DC3000, at 2 dpi (2 days post infiltration), both knockout lines of HPR1 (hpr1-1 and hpr1-2) and the null PLGG1 allele showed a higher level of bacterial growth in plants compared with the wild-type (Fig. 4.1b). Consistently, the infected leaves of the hpr1 mutants also exhibited more severe water-soaking symptoms (Fig. 4.1c), the symptoms that describe the pathogen-driven establishment of an aqueous apoplastic environment favoring infection (Xin et al., 2016). Therefore, defective HPR1 or PLGG1 seems to compromise plant immune response. To get a better understanding of the role of HPR1 and PLGG1 in plant immunity, the two main layers of the immune system, PTI and ETI, were tested in the mutants. PTI was induced by pre-treatment of 100 nM of flg22 followed by Pst DC3000 infiltration into the plants 22-24 h after flg22 inoculation, called the flg22 protection assay. At 2 dpi, the hpr1-1 and plgg1-1 mutants with flg22 pretreatment showed significantly higher levels of bacterial populations and more obvious water-soaking symptoms than Col-0 in the same group (Fig. 4.1d and e), indicating that the PTI response in hpr1-1 and plgg1-1 is compromised. To induce the ETI response, Pst DC3000 (avrPpt2), an avirulent strain carrying the effector protein AvrRpt2, was inoculated into the plants. AvrRpt2-induced ETI requires the activation of the receptor protein RPS2 (Spoel & Dong, 2012), whose null mutant rps2-101c (Mackey et al., 2003) exhibited severe disease symptoms such as massive bacterial growth, necrosis and chlorosis in the leaves at 3 dpi (Fig. 4.1f and g). The disease symptoms in hpr1-1 and plgg1-1 were weaker than those in rps2-101c but apparently stronger than in Col-0 (Fig. 4.1f and g), suggesting impaired ETI response in these two photorespiratory mutants. Other than bacterial growth inside the plants, more aspects of the PTI response were investigated in hpr1-1 and plgg1-1. In response to flg22 treatment, comparable apoplastic ROS burst, a hallmark of early PTI response (Yu et al., 2017), was observed in hpr1-1, plgg1-1 and Col-0 (Fig. 4.2a), suggesting that HPR1 or PLGG1 defect does not impact apoplastic ROS burst. Late PTI response, including callose deposition as a physical barrier against pathogens (Yu et al., 2017) and expression of the SA signaling marker gene PR1 (Pieterse et al., 2012), were also tested. Callose deposition was induced by flg22 application in both Col-0 and plgg1-1 plants at 8 68 h post infiltration (8 hpi), but with clearly fewer induced callose deposits in plgg1-1 (Fig. 4.2b and c). Although flg22 triggered PR1 expression in all three lines at 24 hpi, hpr1-1 and plgg1-1 displayed much reduced upregulation compared with Col-0 (Fig. 4.2d). The weaker callose deposition and PR1 gene expression in hpr1-1 and plgg1-1 are consistent with their susceptible disease phenotypes in the flg22 protection assay (Fig. 4.1d and e), indicating compromised PTI in hpr1-1 and plgg1-1. Taken together, results from bacterial growth, callose deposition and PR1 gene expression analyses support the conclusion that defects in HPR1 and PLGG1 impair both PTI and ETI responses. 4.2.2 The growth and disease phenotypes of hpr1 and plgg1 can be largely rescued by high CO2 conditions To determine whether the function of HPR1 and PLGG1 in plant immunity is dependent on photorespiration specifically, and not a pleiotropic effect, mutant plants were grown under high CO2 (2,000 ppm), where photorespiration is largely inhibited. As expected, the hpr1-1 and plgg1-1 mutants grown under high CO2 had similar morphologies as Col-0 (Fig. 4.3a). To rule out the direct influence of high CO2 on pathogens, Pst DC3000 and Pst DC3000 (avrPpt2) were grown on plates under ambient air or high CO2. Results showed that these two strains had comparable growth irrespective of the environmental CO2 concentrations (Fig. 4.3b). The flg22 protection assay and Pst DC3000 (avrPpt2) infiltration were performed again on high CO2-grown mutants to evaluate the function of HPR1 and PLGG1 in both PTI and ETI responses. With flg22 or mock treatments, the hpr1-1 and plgg1-1 mutants grown under high CO2 exhibited comparable bacterial growth and water-soaking symptoms as Col-0 at 2 dpi (Fig. 4.4a-d). The null allele of RPS2 grown under high CO2 still showed susceptibility to Pst DC3000 (avrRpt2) at 3 dpi, but the disease phenotypes in hpr1-1 and plgg1-1 were similar to those in Col-0 (Fig. 4.4e and f). These results revealed that the compromised immune response in hpr1-1 and plgg1-1 can be largely rescued by high CO2, prompting the conclusion that the effects of HPR1 and PLGG1 in plant immunity are dependent on the photorespiratory pathway. 4.2.3 Photorespiration rate is unchanged during plant interaction with Pst DC3000 and flg22 in Arabidopsis As described in Chapter 1, various parameters, including the difference of net CO2 assimilation rate between 2% and 21% O2, photorespiratory CO2 compensation point (Γ*) and 69 the ratio of glycine to serine (Gly/Ser), have been used to estimate photorespiration rate. Using these methods, previous studies reported increased photorespiration rate in tomato upon Pst DC3000 infection (Ahammed et al., 2018), in banana seedlings inoculated with Fusarium oxysporum f. sp. cucumerinum (FOC) (Dong et al., 2016), and in nitrate-induced FOC-resistant cucumber plants (Sun et al., 2021), suggesting that photorespiration may be actively regulated by plants as a strategy in defense. To further determine how photorespiration is regulated during defense, I performed gas exchange experiments to test whether an increased photorespiration rate can be seen in the Arabidopsis-Pst DC3000 model system used in my studies. First, I used a moderate concentration of Pst DC3000 (~1 × 106 CFU ml−1) to inoculate the plant and measured photorespiration rate in plants before and at 14-18 hpi of Pst DC3000 infiltration. Pst DC3000-treated plants had similar rubisco oxygenation rates (vo) to mock-treated plants, suggesting that the photorespiration rate does not change under this condition (Fig. 4.5a). Photosynthesis-related parameters, including net CO2 assimilation rate (A), rubisco carboxylation rate (vc) and quantum efficiency of photosystem II (ΦII), were also found to be comparable between the two groups of plants (Fig. 4.5a). Next, I reasoned that the potential regulation of photorespiration may be restricted to a certain stage during plant-pathogen interaction instead of throughout the whole process, and therefore used a higher concentration of Pst DC3000 (~1 × 107 CFU ml−1) and a late time point (23-25 hpi). Plants after Pst DC3000 infiltration showed a much lower vo than those treated with mock (Fig. 4.5b), which was opposite to what I had expected. A, vc and ΦII were also found to decrease in Pst DC3000-treated plants at this time point, raising the possibility of cell damage that disrupts normal biological processes including photorespiration. To avoid the difficulties in finding a perfect time point after Pst DC3000 infiltration, I switched to flg22 treatment to obtain a strong immune response without cell death. However, no obvious changes in vo, A, vc or ΦII were found at 20-24 hpi with a high concentration of flg22 (500 nM) (Fig. 4.6), suggesting that neither photorespiration nor photosynthesis is influenced by PTI response at this time point. Overall, using Arabidopsis with Pst DC3000 or flg22 treatment, I have not been able to reproduce the increased photorespiration rate reported previously. More investigations are needed to reach a clear conclusion on if and how photorespiration changes during plant-pathogen interaction. 70 4.3 Discussion and future directions To elucidate the role of photorespiration in immunity, I tested the disease phenotypes of photorespiratory mutants and measured photorespiration rate on plants with pathogen treatments. I found that defects in Arabidopsis HPR1 and PLGG1 proteins compromise the immune response in both PTI and ETI, and the susceptible phenotypes shown in the hpr1-1 and plgg1-1 mutants can be reverted by high CO2. These findings provide additional evidence for the important role of photorespiration in defense response. Based on these results and the literature on these two photorespiratory proteins, I hypothesize that the connections between HPR1/PLGG1 and defense response might be changes in ROS homeostasis or photorespiratory intermediates. As discussed in detail in Chapter 1, in photosynthetic cells, photorespiration is a major source of H2O2 (Foyer et al., 2009), a crucial signaling molecule during plant-pathogen interaction. Current evidence shows that the regulation of photorespiratory ROS is not limited to the H2O2-producing enzyme GOX and the H2O2-scavenging enzyme CAT, but involves other photorespiratory enzymes such as GDC, SHMT1 and GGAT1 (see Chapter 1 for details). Therefore, although neither HPR1 nor PLGG1 directly participates in producing or scavenging H2O2, the disruptions of photorespiratory metabolism in hpr1-1 and plgg1-1 may still influence ROS homeostasis. Although hpr1-1 and plgg1-1 had similar apoplastic ROS burst as Col-0, HPR1, which is in the peroxisomal matrix, and PLGG1, which is on the chloroplast inner envelope, are more likely to influence the homeostasis of intracellular, instead of extracellular, ROS (Fig. 4.2a). To test if hpr1-1 and plgg1-1 have decreased levels of intracellular ROS, compromised ROS signaling, or disordered ROS response, I have generated transgenic lines of these two mutants expressing the peroxisomal H2O2 reporter HyPer (Costa et al., 2010) to monitor the dynamics of ROS inside peroxisomes, and started to test the expressions of ROS- responsive genes such as OXI1 (oxidative signal-inducible 1) (Rentel et al., 2004; Petersen et al., 2009), GSTU24 (glutathione S-transferase tau 24) and APX1 (ascorbate peroxidase 1) (Chaouch & Noctor, 2010), during defense. As detailed in Chapter 1, photorespiratory metabolites have been shown to play a role in immunity. The deficiency of either HPR1 or PLGG1 was found in previous publications (Timm et al., 2008; Pick et al., 2013) and my work (Chapter 2) to cause increases in photorespiratory intermediates, including glycolate, glyoxylate, glycine, serine, hydroxypyruvate, and glycerate, under ambient air conditions. One or more of these accumulated photorespiratory intermediates 71 may be involved in the susceptible disease phenotypes in hpr1-1 and plgg1-1. Since the application of glycerate inhibits HR in soybean (Okinaka et al., 2002), glycerate may be a good candidate. To test this hypothesis, photorespiratory metabolites in the pathogen-treated mutants and Col-0 can be measured by GC-MS to identify candidate metabolites whose level changes correlate to immune response. These metabolites together with pathogens can then be applied to plants to validate their function. I also measured photorespiration rate in Arabidopsis treated with Pst DC3000 or flg22, but failed to confirm the previously reported up-regulation of photorespiration. The low- throughput of gas exchange measurements by Li-COR might be a reason, as plants usually have big individual variations and it is difficult to capture small changes in photorespiration or investigate a series of time points with the current techniques. It is also possible that the regulation of photorespiration is highly dynamic or restricted to specific stages of the immune response, even specific plant-pathogen systems. Lastly, photorespiration is not a closed cycle (Fu et al., 2023) The measured vo only represents the rate of the initial step of photorespiration but not necessarily the whole process. If the activity of a photorespiratory enzyme or the concentration of a photorespiratory metabolite is regulated during plant-pathogen interaction after the initial stage, it may not alter vo. In summary, both the photorespiratory pathway and the immune response are complex processes. Large-scale and systematic approaches involving simultaneous measurements of photorespiration and immune response may be required, in order to obtain a comprehensive view of the interplay between these two systems under different conditions. 4.4 Methods 4.4.1 Plant and pathogen materials and growth conditions Wild-type and mutant lines of Arabidopsis thaliana used in the study are all from the ecotype Col-0. Mutants hpr1-1 (SALK_067724) and plgg1-1 (SALK_053469) were the same lines as those used in Chapter 2. The hpr1-2 mutant (SALK_143584) was characterized previously (Timm et al., 2008) and used in a previous study in my lab (Li et al., 2019a). The rps2-101c mutant was characterized before (Mindrinos et al., 1994) and was kindly provided by Dr. Brad Day. 72 Arabidopsis seeds were directly sown in the soil in pots. After seed stratification in the dark at 4 °C for 3 to 7 days, pots were moved to growth chambers with ~100 μmol m-2 s-1 white light, 21 °C, and 12 h/12 h light/dark cycle for plant growth. Pseudomonas syringae pv. tomato strain DC3000 (Pst DC3000), stain Pst DC3000 (avrPpt2) and peptide flg22 were kindly provided by Drs. Sheng Yang He and Brad Day. Pst DC3000 and Pst DC3000 (avrPpt2) were grown on Modified Luria-Bertani medium (LM, 10.0 g Bacto tryptone, 6.0 g Bacto yeast extract, 1.5 g K2HPO4, 0.6 g NaCl, 0.4 g MgSO4·7H2O per liter H2O) at 28-30 °C with Rif50 and Rif50/Kan50 respectively. 4.4.2 Pathogen infection assays Bacterial infiltration in Arabidopsis was performed using a previous protocol (Yao et al., 2013). Briefly, 5-week-old Arabidopsis plants were syringe-infiltrated with a bacterial suspension [~1 × 105 colony-forming units (CFU) ml−1 Pst DC3000 or ~ 1 × 106 CFU ml−1 Pst DC3000 (avrRpt2) in 0.25 mM MgCl2 solution]. Plants were dried in the air in the growth chamber and covered with domes to keep high humidity. Two or 3 days after inoculation, i.e., 2 dpi for Pst DC3000 and 3 dpi for Pst DC3000 (avrPpt2), plant tissue was collected and ground, and bacterial populations were determined by serial dilutions. For flg22 protection assays, plant leaves were syringe-infiltrated with 100 nM flg22 or 0.1% DMSO (mock). After 22–24 h, plants in the growth chamber without domes were infiltrated again with a bacterial suspension [~ 1 × 106 CFU ml−1 Pst DC3000 in 0.25 mM MgCl2 solution]. Bacterial populations were quantified at 2 dpi. 4.4.3 Apoplastic ROS burst assay As described previously (Zhang et al., 2019), 4 mm leaf discs were taken from ~5-week- old plants and kept floating on water (adaxial side up) in 96-well plates overnight. Then water was removed and replaced with 100 μl immune-eliciting solution [34 μg ml-1 luminol (Sigma- Aldrich), 20 μg ml-1 horseradish peroxidase (Sigma-Aldrich) and 100 nM flg22 (0.1% DMSO for mock) in water]. Luminescence was measured at 470nm with a SpectraMax L microplate reader (Molecular Devices). 4.4.4 Quantification of callose deposits Five-week-old plants were infiltrated with 100 nM flg22 or 0.1% DMSO on three leaves for each plant. After 8 h, leaves were sampled and stained using a previous protocol (Bach-Pages & Preston, 2018) with modifications. Briefly, leaves were soaked in 100% ethanol overnight to 73 clear chlorophyll, then fixed with acetic acid/ethanol solution (1:3, v/v) for 2 h. Next, samples were incubated sequentially in 75% ethanol, 50% ethanol and 150 mM K2HPO4 (pH 9.5) for 15 min at each step, then stained with 0.01% aniline blue in 150 mM K2HPO4 (pH 9.5) solution at 4 °C overnight. Leaves were mounted in 50% glycerol on glass slides and observed with the Axio Imager M1 epi-fluorescence microscope (ZEISS), using the DAPI filter. The number of callose deposits was counted with ImageJ (https://imagej.net/ij/). 4.4.5 RNA extraction and qRT-PCR Five-week-old plants were infiltrated with 100 nM flg22 or 0.1% DMSO. After ~24 h, leaf tissue was sampled and ground in liquid nitrogen. RNA was extracted using the NucleoSpin RNA Plant kit (MACHEREY-NAGEL), then reverse transcribed using the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems). Fast SYBR Green Master Mix (Applied Biosystems) was used with a 7500 Fast Real-Time PCR System (Applied Biosystems) for real- time PCR. The primer pairs are 5’-GGCTAACTACAACTACGCTG-3’ and 5’- TCTCGTTCACATAATTCCCAC-3’ for PR1; and 5’-GGTTACAAGACAAGGTTCACTC-3’ and 5’-CATTCAGGACCAAACTCTTCAG-3’ for the internal control gene PP2AA3. Data of each gene was normalized to its averaged gene expression in mock-treated Col-0. 4.4.6 Gas exchange measurements Mature, fully-expanded leaves from 5- to 5.5-week-old plants were used in gas exchange measurements with LI-6800 (LI-COR Biosciences, Lincoln, Nebraska, USA). During measurements, chamber temperature was maintained at ~25°C and vapor pressure deficit was controlled at 1-1.5 kPa H2O. Each leaf was measured under a series of light intensities: 800, 100, 50, 45, 40, 35, 30, 25 μmol m-2 s-1. The same leaf was measured twice, before and after pathogen treatments. Treatments used and corresponding time points are: Pst DC3000 (~1 × 106 CFU ml−1) at 14-18 hpi, Pst DC3000 (~1 × 107 CFU ml−1) at 23-25 hpi, 500 nM flg22 treatment at 20- 24 hpi, and 0.25 mM MgCl2 solution or 0.1% DMSO as mock treatments. Net CO2 assimilation rate (A) and quantum efficiency of photosystem II (ΦII) were directly measured with LI-6800, and rubisco carboxylate rate (vc) and oxygenation rate (vo) were estimated based on a previous protocol (Gregory et al., 2023). In brief, the estimation was based on vc = (A + RL) / (1- Γ*/Cc) and vo = (vc – A - RL) / 0.5, where the partial pressure of chloroplastic CO2 (Cc) was calculated using Cc = Ci - (A / gm). Among these parameters, the partial pressure of intercellular CO2 (Ci) was provided by LI-6800, non‐photorespiratory CO2 release in the light 74 (RL) was determined using the common intersection method (Walker et al., 2016a). Finally, the partial pressure of CO2 in the chloroplast at the photorespiratory compensation point (Γ*) and mesophyll conductance to CO2 (gm) were assumed to be 4.88 Pa and 2.23 μmol m−2 s−1 Pa−1, respectively, based on previous measurements (Bao et al., 2021). 4.4.7 Statistics and plots Data analysis and plots were conducted as described in Chapter 2. Student’s paired two- tailed t-test was used for pairwise comparison. One-way or Two-way ANOVA with Tukey’s HSD test was used for multi-comparison. 75 Figures Figure 4.1. Null mutants of HPR1 and PLGG1 show defects in growth and immune response. 76 Figure 4.1 (cont’d) (a) Five-week-old plants grown under ambient CO2. Red arrows indicate the yellow lesions in the older leaves of plgg1-1. (b, c) Quantification of bacterial populations inside the leaves (b) and water-soaking symptoms on the leaves (c) at 2 d post infiltration (2 dpi) with Pseudomonas syringae pv. tomato DC3000 (Pst DC3000) at ~ 1 × 105 CFU ml−1. (d, e) Quantification of bacterial populations inside the leaves (d) and water-soaking symptoms on the leaves (e) that were pretreated with 100 nM flg22 and then infiltrated 24 h later with Pst DC3000 ~1 × 106 CFU ml−1. Plant leaves were sampled at 2 dpi. (f, g) Quantification of bacterial populations inside the plant leaves (f) and disease symptoms on the leaves (g) at 3 dpi with Pst DC3000 (avrRpt2) at ~ 1 × 106 CFU ml−1. Five-week-old plants were used in all experiments. Different letters indicate statistically significant differences (p<0.05), which were determined by One-way (b, f) or Two- way (d) ANOVA with Tukey’s HSD test. Biological replicates: n=4. 77 Figure 4.2. PTI response of hpr1 and plgg1 in apoplastic ROS burst, callose deposition, and expression of the PR1 gene. 78 Figure 4.2 (cont’d) (a) Apoplastic ROS burst induced by 100 nM flg22. 0.1% DMSO was used as the mock treatment. Results represent the mean ± SD. Biological replicates: n=6 for mock-treated plants, and n=5 for flg22-treated plants. (b) Callose deposits induced by 100 nM flg22 at 8 h post infiltration (hpi). Leaves were stained with 0.1% aniline blue and observed under an epifluorescence microscope using a DAPI filter. Scale bar = 100 μm. (c) Callose deposits quantified using ImageJ. Biological replicates: n=4. (d) Expression of PR1 induced by 100 nM flg22 at 24 hpi. PP2AA3 was used as the internal control and data was normalized to the average of PR1 gene expression in mock-treated Col-0. Biological replicates: n=4. Five-week-old plants were used in all experiments. Different letters in (c) and (d) indicate statistically significant differences (p<0.05), which were determined by Two-way ANOVA with Tukey’s HSD test. 79 Figure 4.3. Growth phenotypes of plants and pathogens under high CO2 conditions. (a) Five-week-old plants grown under 2,000 ppm of CO2. (b) Pst DC3000 and Pst DC3000 (avrRpt2) grown in high CO2 (2,000 ppm, HC) or ambient air (AC) growth chambers for 4 or 3 days. 80 Figure 4.4. Disease phenotypes of hpr1 and plgg1 are largely rescued by high CO2 conditions. 81 Figure 4.4 (cont’d) (a, b) Quantification of bacterial population inside the leaves (a) and water-soaking symptoms on the leaves (b) of Col-0 and hpr1-1 plants that were pretreated with 100 nM flg22, and then infiltrated 24 h later with Pst DC3000 at ~1 × 106 CFU ml−1. Plant leaves were sampled at 2 dpi. (c, d) The level of bacterial populations inside the leaves (c) and water-soaking symptoms on the leaves (d) of Col-0 and plgg1-1 plants that were pretreated with 100 nM flg22, and then infiltrated 24 h later with Pst DC3000 at ~1 × 106 CFU ml−1. Plant leaves were sampled at 2 dpi. (e, f) Quantification of bacterial populations inside the leaves (e) and disease symptoms on the leaves (f) at 3 dpi with Pst DC3000 (avrRpt2) at ~ 1 × 106 CFU ml−1. Five-week-old plants were used in all experiments. Different letters indicate statistically significant differences (p<0.05), which were determined by One-way (e) or Two-way (a, c) ANOVA with Tukey’s HSD test. Biological replicates: n=4. 82 Figure 4.5. Measurement of photorespiration and photosynthesis under normal light conditions on plants treated with Pst DC3000. 83 Figure 4.5 (cont’d) (a) Rubisco oxygenation rate (vo), net CO2 assimilation rate (A), rubisco carboxylation rate (vc) and quantum efficiency of photosystem II (ΦII) of Col-0 plants before Pst DC3000 (~1 × 106 CFU ml−1) treatment and at 14-18 hpi. Biological replicates: n=4. (b) vo, A, vc and ΦII of Col-0 plants before Pst DC3000 (~1 × 107 CFU ml−1) treatment and at 23-25 hpi. Biological replicates: n=3. MgCl2 solution (0.25 mM) was used as mock. The p values were determined by Student’s paired two-tailed t-test. 84 Figure 4.6. Measurements of photorespiration and photosynthesis under normal light conditions on plants treated with flg22. vo, A, vc and ΦII of Col-0 plants before 500 nM flg22 treatment and at 20-24 hpi are shown. Biological replicates: n=4. DMSO (0.1%) was used as mock. The p values were determined by Student’s paired two-tailed t-test. 85 CHAPTER 5. Summary and future perspectives My dissertation focuses on the role of photorespiration in plants under stress conditions, high light and pathogen infection in particular. I applied various strategies to dissect the relationships between photorespiration and stress conditions and investigate the underlying mechanisms. To understand how photorespiration is modulated under high light, I performed a screening for suppressors of the photorespiratory mutant hpr1. GLYR1, which encodes a cytosolic enzyme that converts glyoxylate to glycolate, was identified as the gene carrying the causal mutation that partially reverts the mutant phenotypes of hpr1 under high light, including growth, photorespiratory metabolites, and photosynthesis. Defective GLYR1 can also rescue the phenotypes of the catalase mutant cat2 under high light, but not the mutant of the PLGG1. Further investigations with metabolic and genetic tools suggest the existence of a cytosolic glyoxylate shunt of photorespiration, where glyoxylate and serine can be converted into hydroxypyruve by an unknown aminotransferase, feeding into the HPR2-mediated reaction. The glycerate produced by HPR2 returns to chloroplast and eventually carbons are recycled to the Calvin-Benson cycle. This cytosolic glyoxylate shunt supports the flexibility of photorespiration, which may be important in plant response to stress conditions like high light when the main photorespiratory pathway is deficient as in the cases of hpr1 and cat2. To further investigate the function of GLYR1, I performed RNA-seq for Col-0, glyr1-1, hpr1-1, glyr1-1 hpr1-1 and shpr7. The glyr1 mutant only has a small number of differentially expressed genes compared to Col-0, but glyr1 hpr1 and shpr7 have a massive number of differentially expressed genes compared to hpr1. The transcriptional reprogramming in hpr1 involves broad biological processes, which can be largely reverted by the lack of a functional GLYR1. This RNA-seq data is consistent with the accumulation of photorespiratory metabolites in hpr1 and its partial rescue by glyr1, supporting the cytosolic glyoxylate shunt proposed in Chapter 2. To investigate the role of photorespiration in plant immunity, I treated photorespiratory mutants with pathogens and found two photorespiratory mutants, hpr1 and plgg1, showing susceptible phenotypes under pathogen infections. High CO2 conditions can rescue the susceptible phenotypes of these two mutants, indicating that the role of HPR1 and PLGG1 in plant immunity is dependent on photorespiration. Although further investigations are needed to 86 elucidate the underlying mechanisms, my current results suggest the positive role of photorespiration in plant immunity. My work provides evidence to the important role of photorespiration in plant stress response, a key area where our understanding is still fragmentary. Photorespiration is a complex process that involves a series of reactions with multiple enzymes and metabolites across chloroplasts, peroxisomes, mitochondria and the cytosol, therefore large-scale experiments that simultaneously assay for different aspects of the pathway during stress response are required to obtain a full understanding of the role of photorespiration in plant stress response. In addition, my work and previous work from my lab showed that HPR1 and PLGG1 play positive roles in both high light and immunity, suggesting that photorespiration may contribute to abiotic and biotic stresses using similar mechanisms. Hence, it will also be interesting to expand this line of research to compare photorespiration under different stresses. Lastly, knowledge gained on the role of photorespiration in stress response may guide future breeding and engineering efforts for the simultaneous improvement of yield and resilience in crop plants. 87 REFERENCES Ahammed GJ, Li X, Zhang G, Zhang H, Shi J, Pan C, Yu J, Shi K. 2018. Tomato photorespiratory glycolate-oxidase-derived H2O2 production contributes to basal defence against Pseudomonas syringae. Plant Cell and Environment 41: 1126–1138. Allan WL, Simpson JP, Clark SM, Shelp BJ. 2008. γ-Hydroxybutyrate accumulation in Arabidopsis and tobacco plants is a general response to abiotic stress: Putative regulation by redox balance and glyoxylate reductase isoforms. Journal of Experimental Botany 59: 2555– 2564. Aroca A, García-Díaz I, García-Calderón M, Gotor C, Márquez AJ, Betti M. 2023. Photorespiration: regulation and new insights on the potential role of persulfidation. Journal of Experimental Botany 74: 6023–6039. Bach-Pages M, Preston GM. 2018. Methods to quantify biotic-induced stress in plants. In: Medina C, López-Baena FJ, eds. Host-Pathogen Interactions: Methods and Protocols. New York, NY: Springer New York, 241–255. Bao H, Morency M, Rianti W, Saeheng S, Roje S, Weber APM, Walker BJ. 2021. Catalase protects against nonenzymatic decarboxylations during photorespiration in Arabidopsis thaliana. Plant Direct 5: 1–13. Bassi R, Dall’osto L. 2021. Dissipation of light energy absorbed in excess: the molecular Mmechanisms. Annu. Rev. Plant Biol. 2021 72: 47–76. Benjamin I, Kenigsbuch D, Galperin M, Abrameto JA, Cohen Y. 2009. Cisgenic melons over expressing glyoxylate-aminotransferase are resistant to downy mildew. European Journal of Plant Pathology 125: 355–365. Betti M, Bauwe H, Busch FA, Fernie AR, Keech O, Levey M, Ort DR, Parry MAJ, Sage R, Timm S, et al. 2016. Manipulating photorespiration to increase plant productivity: recent advances and perspectives for crop improvement. Journal of Experimental Botany 67: 2977– 2988. Brikis CJ, Zarei A, Trobacher CP, Deell JR, Akama K, Mullen RT, Bozzo GG, Shelp BJ. 2017. Ancient plant glyoxylate/succinic semialdehyde reductases: GLYR1s are cytosolic, whereas GLYR2s are localized to both mitochondria and plastids. Frontiers in Plant Science 8. Busch FA, Sage RF, Farquhar GD. 2018. Plants increase CO2 uptake by assimilating nitrogen via the photorespiratory pathway. Nature Plants 4: 46–54. Camejo D, Guzmán-Cedeño Á, Moreno A. 2016. Reactive oxygen species, essential molecules, during plant-pathogen interactions. Plant Physiology and Biochemistry 103: 10–23. Campbell WJ, Ogren WL. 1990. Glyoxylate inhibition of ribulosebisphosphate carboxylase/oxygenase activation in intact, lysed, and reconstituted chloroplasts. Photosynthesis Research 23: 257–268. Carrie C, Kühn K, Murcha MW, Duncan O, Small ID, O’Toole N, Whelan J. 2009. Approaches to defining dual-targeted proteins in Arabidopsis. Plant Journal 57: 1128–1139. Cegelski L, Schaefer J. 2005. Glycine metabolism in intact leaves by in vivo 13C and 15N labeling. Journal of Biological Chemistry 280: 39238–39245. 88 Chamnongpol S, Willekens H, Moeder W, Langebartels C, Sandermann H, Van Montagu M, Inzé D, Van Camp W. 1998. Defense activation and enhanced pathogen tolerance induced by H2O2 in transgenic tobacco. Proceedings of the National Academy of Sciences of the United States of America 95: 5818–5823. Chaouch S, Noctor G. 2010. Myo-inositol abolishes salicylic acid-dependent cell death and pathogen defence responses triggered by peroxisomal hydrogen peroxide. New Phytologist 188: 711–718. Chaouch S, Queval G, Vanderauwera S, Mhamdi A, Vandorpe M, Langlois-Meurinne M, Van Breusegem F, Saindrenan P, Noctor G. 2010. Peroxisomal hydrogen peroxide is coupled to biotic defense responses by ISOCHORISMATE SYNTHASE1 in a daylength-Related manner. Plant Physiology 153: 1692–1705. Chastain CJ, Ogren WL. 1989. Glyoxylate inhibition of ribulosebisphosphate carboxylase/oxygenase activation state in vivo. Plant and Cell Physiology 30: 937–944. Chaux F, Peltier G, Johnson X. 2015. A security network in PSI photoprotection: Regulation of photosynthetic control, NPQ and O2 photoreduction by cyclic electron flow. Frontiers in Plant Science 6. Chen Z, Silva H, Klessig DF. 1993. Active oxygen species in the induction of plant systemic acquired resistance by salicylic acid. Science 262: 1883–1886. Chern M, Bai W, Chen X, Canlas PE, Ronald PC. 2013. Reduced expression of glycolate oxidase leads to enhanced disease resistance in rice. PeerJ 2013: 1–16. Ching SLK, Gidda SK, Rochon A, van Cauwenberghe OR, Shelp BJ, Mullen RT. 2012. Glyoxylate reductase isoform 1 is localized in the cytosol and not peroxisomes in plant cells. Journal of Integrative Plant Biology 54: 152–168. Conrath U, Chen Z, Ricigliano JR, Klessig DF. 1995. Two inducers of plant defense responses, 2,6-dichloroisonicotinic acid and salicylic acid, inhibit catalase activity in tobacco. Proceedings of the National Academy of Sciences of the United States of America 92: 7143– 7147. Cook CM, Mulligan RM, Tolbert NE. 1985. Inhibition and stimulation of ribulose-1,5- bisphosphate carboxylase/oxygenase by glyoxylate. Archives of Biochemistry and Biophysics 240: 392–401. Costa A, Drago I, Behera S, Zottini M, Pizzo P, Schroeder JI, Pozzan T, Schiavo F Lo. 2010. H2O2 in plant peroxisomes: An in vivo analysis uncovers a Ca2+-dependent scavenging system. Plant Journal 62: 760–772. Cousins AB, Pracharoenwattana I, Zhou W, Smith SM, Badger MR. 2008. Peroxisomal malate dehydrogenase is not essential for photorespiration in arabidopsis but its absence causes an increase in the stoichiometry of photorespiratory CO2 release. Plant Physiology 148: 786– 795. Cousins AB, Walker BJ, Pracharoenwattana I, Smith SM, Badger MR. 2011. Peroxisomal hydroxypyruvate reductase is not essential for photorespiration in Arabidopsis but its absence causes an increase in the stoichiometry of photorespiratory CO 2 release. Photosynthesis Research 108: 91–100. 89 Crisp PA, Ganguly DR, Smith AB, Murray KD, Estavillo GM, Searle I, Ford E, Bogdanović O, Lister R, Borevitz JO, et al. 2017. Rapid recovery gene downregulation during excess-light stress and recovery in Arabidopsis. The Plant Cell 29: 1836–1863. Cristina Palmieri M, Lindermayr C, Bauwe H, Steinhauser C, Durner J. 2010. Regulation of plant glycine decarboxylase by S-nitrosylation and glutathionylation. Plant Physiology 152: 1514–1528. Cruz JA, Savage LJ, Zegarac R, Hall CC, Satoh-Cruz M, Davis GA, Kovac WK, Chen J, Kramer DM. 2016. Dynamic environmental photosynthetic imaging reveals emergent phenotypes. Cell Systems 2: 365–377. Cui H, Tsuda K, Parker JE. 2015. Effector-triggered immunity: From pathogen perception to robust defense. Annual Review of Plant Biology 66: 487–511. Dellero Y, Jossier M, Schmitz J, Maurino VG, Hodges M. 2016. Photorespiratory glycolate- glyoxylate metabolism. Journal of Experimental Botany 67: 3041–3052. Deslandes L, Rivas S. 2012. Catch me if you can: Bacterial effectors and plant targets. Trends in Plant Science 17: 644–655. Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, Batut P, Chaisson M, Gingeras TR. 2013. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29: 15–21. Dodds PN, Rathjen JP. 2010. Plant immunity: towards an integrated view of plant–pathogen interactions. Nature Reviews Genetics 11: 539–548. Dong X, Wang M, Ling N, Shen Q, Guo S. 2016. Potential role of photosynthesis-related factors in banana metabolism and defense against Fusarium oxysporum f. sp. cubense. Environmental and Experimental Botany 129: 4–12. Eisenhut M, Bräutigam A, Timm S, Florian A, Tohge T, Fernie AR, Bauwe H, Weber APM. 2017. Photorespiration is crucial for dynamic response of photosynthetic metabolism and stomatal movement to altered CO2 availability. Molecular Plant 10: 47–61. Eisenhut M, Roell MS, Weber APM. 2019. Mechanistic understanding of photorespiration paves the way to a new green revolution. New Phytologist 223: 1762–1769. Esser C, Kuhn A, Groth G, Lercher MJ, Maurino VG. 2014. Plant and animal glycolate oxidases have a common eukaryotic ancestor and convergently duplicated to evolve long-chain 2-hydroxy acid oxidases. Molecular Biology and Evolution 31: 1089–1101. da Fonseca-Pereira P, Souza PVL, Hou LY, Schwab S, Geigenberger P, Nunes-Nesi A, Timm S, Fernie AR, Thormählen I, Araújo WL, et al. 2020. Thioredoxin h2 contributes to the redox regulation of mitochondrial photorespiratory metabolism. Plant Cell and Environment 43: 188–208. Foyer CH. 2018. Reactive oxygen species, oxidative signaling and the regulation of photosynthesis. Environmental and Experimental Botany 154: 134–142. Foyer CH, Bloom AJ, Queval G, Noctor G. 2009. Photorespiratory metabolism: Genes, mutants, energetics, and redox signaling. Annual Review of Plant Biology 60: 455–484. Foyer CH, Hanke G. 2022. ROS production and signalling in chloroplasts: cornerstones and evolving concepts. The Plant Journal 111: 642–661. 90 Fu X, Gregory LM, Weise SE, Walker BJ. 2023. Integrated flux and pool size analysis in plant central metabolism reveals unique roles of glycine and serine during photorespiration. Nature Plants 9: 169–178. Fu X, Walker BJ. 2022. Dynamic response of photorespiration in fluctuating light environments. Journal of Experimental Botany. Fu S, Wang K, Ma T, Liang Y, Ma Z, Wu J, Xu Y, Zhou X. 2022. An evolutionarily conserved C4HC3-type E3 ligase regulates plant broad-spectrum resistance against pathogens. Plant Cell 34: 1822–1843. Fürtauer L, Küstner L, Weckwerth W, Heyer AG, Nägele T. 2019. Resolving subcellular plant metabolism. The Plant Journal 100: 438–455. Gao C, Xu H, Huang J, Sun B, Zhang F, Savage Z, Duggan C, Yan T, Wu CH, Wang Y, et al. 2020. Pathogen manipulation of chloroplast function triggers a light-dependent immune recognition. Proceedings of the National Academy of Sciences of the United States of America 117: 9613–9620. Gilbert BM, Wolpert TJ. 2013. Characterization of the LOV1 -Mediated, Victorin-Induced, Cell-Death Response with Virus-Induced Gene Silencing. Molecular Plant-Microbe Interactions 26: 903–917. Giraldo – González JJ, de Souza Carvalho FM, Ferro JA, Herai RH, Chaves Bedoya G, Rodas Mendoza EF. 2021. Transcriptional changes involved in kumquat (Fortunella spp) defense response to Xanthomonas citri subsp. citri in early stages of infection. Physiological and Molecular Plant Pathology 116. Giri MK, Singh N, Banday ZZ, Singh V, Ram H, Singh D, Chattopadhyay S, Nandi AK. 2017. GBF1 differentially regulates CAT2 and PAD4 transcription to promote pathogen defense in Arabidopsis thaliana. Plant Journal 91: 802–815. Givan C V., Tsutakawa S, Hodgson JM, David N, Randall DD. 1988. Glyoxylate reductase activity in pea leaf protoplasts nucleotide specificity and subcellular location. Journal of Plant Physiology 132: 593–599. González‐lópez MDC, Jijón‐moreno S, Dautt‐castro M, Ovando‐vázquez C, Ziv T, Horwitz BA, Casas‐flores S. 2021. Secretome analysis of arabidopsis–trichoderma atroviride interaction unveils new roles for the plant glutamate:Glyoxylate aminotransferase ggat1 in plant growth induced by the fungus and resistance against botrytis cinerea. International Journal of Molecular Sciences 22: 1–28. Gregory LM, Roze L V., Walker BJ. 2023. Increased activity of core photorespiratory enzymes and CO2 transfer conductances are associated with higher and more optimal photosynthetic rates under elevated temperatures in the extremophile Rhazya stricta. Plant Cell and Environment 46: 3704–3720. Hanson AD, Roje S. 2001. One-carbon metabolism in higher plants. Annual Review of Plant Physiology and Plant Molecular Biology 52: 119–137. He L, Jin P, Chen X, Zhang TY, Zhong KL, Liu P, Chen JP, Yang J. 2021. Comparative proteomic analysis of Nicotiana benthamiana plants under Chinese wheat mosaic virus infection. BMC Plant Biology 21: 1–17. 91 Hodges M. 2022. Photorespiration and improving photosynthesis. In: Lüttge U, Cánovas FM, Risueño M-C, Leuschner C, Pretzsch H, eds. Progress in Botany Vol. 84. Cham: Springer Nature Switzerland, 171–219. Hoover GJ, Van Cauwenberghe OR, Breitkreuz KE, Clark SM, Merrill AR, Shelp BJ. 2007. Characteristics of an Arabidopsis glyoxylate reductase: General biochemical properties and substrate specificity for the recombinant protein, and developmental expression and implications for glyoxylate and succinic semialdehyde metabolism in planta. Canadian Journal of Botany 85: 883–895. Huang W, Hu H, Zhang SB. 2015. Photorespiration plays an important role in the regulation of photosynthetic electron flow under fluctuating light in tobacco plants grown under full sunlight. Frontiers in Plant Science 6. Huang W, Yang YJ, Wang JH, Hu H. 2019a. Photorespiration is the major alternative electron sink under high light in alpine evergreen sclerophyllous Rhododendron species. Plant Science 289: 110275. Huang W, Zhang SB, Hu H. 2014. Sun leaves up-regulate the photorespiratory pathway to maintain a high rate of CO2 assimilation in tobacco. Frontiers in Plant Science 5. Huang J, Zhao X, Chory J. 2019b. The Arabidopsis transcriptome responds specifically and dynamically to high light stress. Cell Reports 29: 4186-4199.e3. Jossier M, Liu Y, Massot S, Hodges M. 2020. Enzymatic properties of recombinant phosphomimetic photorespiratory glycolate oxidases from Arabidopsis thaliana and Zea mays. Plants 9. Kadotani N, Akagi A, Takatsuji H, Miwa T, Igarashi D. 2016. Exogenous proteinogenic amino acids induce systemic resistance in rice. BMC Plant Biology 16: 1–10. Kalapos B, Juhász C, Balogh E, Kocsy G, Tóbiás I, Gullner G. 2021. Transcriptome profiling of pepper leaves by RNA-Seq during an incompatible and a compatible pepper-tobamovirus interaction. Scientific Reports 11: 1–19. Kandoth PK, Liu S, Prenger E, Ludwig A, Lakhssassi N, Heinz R, Zhou Z, Howland A, Gunther J, Eidson S, et al. 2017. Systematic mutagenesis of serine hydroxymethyltransferase reveals an essential role in nematode resistance. Plant Physiology 175: 1370–1380. Kaur N, Li J, Hu J. 2013. Peroxisomes and Photomorphogenesis. In: del Río LA, ed. Subcellular Biochemistry. Peroxisomes and their Key Role in Cellular Signaling and Metabolism. Dordrecht: Springer Netherlands, 195–211. Kleine T, Kindgren P, Benedict C, Hendrickson L, Strand A. 2007. Genome-wide gene expression analysis reveals a critical role for CRYPTOCHROME1 in the response of Arabidopsis to high irradiance. Plant Physiology 144: 1391–1406. Koenig AM, Liu B, Hu J. 2023. Visualizing the dynamics of plant energy organelles. Biochemical Society Transactions 51: 2029–2040. Koper K, Han S-W, Casas Pastor D, Yoshikuni Y, Maeda HA. 2022. Evolutionary origin and functional diversification of aminotransferases. Journal of Biological Chemistry 298: 102122. 92 Korasick DA, Kandoth PK, Tanner JJ, Mitchum MG, Beamer LJ. 2020. Impaired folate binding of serine hydroxymethyltransferase 8 from soybean underlies resistance to the soybean cyst nematode. Journal of Biological Chemistry 295: 3708–3718. Korotkevich G, Sukhov V, Budin N, Shpak B, Artyomov MN, Sergushichev A. 2021. Fast gene set enrichment analysis. bioRxiv: 060012. Krieger-Liszkay A, Shimakawa G. 2022. Regulation of the generation of reactive oxygen species during photosynthetic electron transport. Biochemical Society Transactions 50: 1025– 1034. Lakhssassi N, Patil G, Piya S, Zhou Z, Baharlouei A, Kassem MA, Lightfoot DA, Hewezi T, Barakat A, Nguyen HT, et al. 2019. Genome reorganization of the GmSHMT gene family in soybean showed a lack of functional redundancy in resistance to soybean cyst nematode. Scientific Reports 9: 1–16. Langfelder P, Horvath S. 2008. WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics 9: 559. Langfelder P, Horvath S. 2012. Fast R Functions for Robust Correlations and Hierarchical Clustering. Journal of Statistical Software 46: 1–17. Launay A, Jolivet S, Clément G, Zarattini M, Dellero Y, Le Hir R, Jossier M, Hodges M, Expert D, Fagard M. 2022. DspA/E-Triggered Non-Host Resistance against E. amylovora Depends on the Arabidopsis GLYCOLATE OXIDASE 2 Gene. International Journal of Molecular Sciences 23: 4224. Lavell A, Smith M, Xu Y, Froehlich JE, De La Mora C, Benning C. 2021. Proteins associated with the Arabidopsis thaliana plastid rhomboid‐like protein RBL10. The Plant Journal 108: 1332–1345. Laxa M. 2017. Regulatory cis-elements are located in accessible promoter regions of the CAT2 promoter and affect activating histone modifications in Arabidopsis thaliana. Plant Molecular Biology 93: 49–60. Laxa M, Fromm S. 2018. Co-expression and regulation of photorespiratory genes in Arabidopsis thaliana: A bioinformatic approach. Current Plant Biology 14: 2–18. Laxa M, Müller K, Lange N, Doering L, Pruscha JT, Peterhänsel C. 2016. The 5′UTR Intron of Arabidopsis GGT1 Aminotransferase Enhances Promoter Activity by Recruiting RNA Polymerase II. Plant Physiology 172: 313–327. Lefevere H, Bauters L, Gheysen G. 2020. Salicylic acid biosynthesis in plants. Frontiers in Plant Science 11: 1–7. Levey M, Timm S, Mettler-Altmann T, Borghi GL, Koczor M, Arrivault S, Weber APM, Bauwe H, Gowik U, Westhoff P. 2019. Efficient 2-phosphoglycolate degradation is required to maintain carbon assimilation and allocation in the C4 plant Flaveria bidentis. Journal of Experimental Botany 70: 575–587. Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, Marth G, Abecasis G, Durbin R, Subgroup 1000 Genome Project Data Processing. 2009. The sequence alignment/map format and SAMtools. Bioinformatics 25: 2078–2079. 93 Li J, Hu J. 2015. Using co-expression analysis and stress-based screens to uncover Arabidopsis peroxisomal proteins involved in drought response (D Bassham, Ed.). PLOS ONE 10: e0137762. Li M, Kim C. 2022. Chloroplast ROS and stress signaling. Plant Communications 3: 100264. Li J, Tietz S, Cruz JA, Strand DD, Xu Y, Chen J, Kramer DM, Hu J. 2019a. Photometric screens identified Arabidopsis peroxisome proteins that impact photosynthesis under dynamic light conditions. The Plant Journal 97: 460–474. Li J, Weraduwage SM, Preiser AL, Tietz S, Weise SE, Strand DD, Froehlich JE, Kramer DM, Hu J, Sharkey TD. 2019b. A cytosolic bypass and g6p shunt in plants lacking peroxisomal hydroxypyruvate reductase1. Plant Physiology 180: 783–792. Liepman AH, Olsen LJ. 2001. Peroxisomal alanine: Glyoxylate aminotransferase (AGT1) is a photorespiratory enzyme with multiple substrates in Arabidopsis thaliana. Plant Journal 25: 487–498. Liepman AH, Olsen LJ. 2003. Alanine aminotransferase homologs catalyze the glutamate:Glyoxylate aminotransferase reaction in peroxisomes of Arabidopsis. Plant Physiology 131: 215–227. Liepman AH, Olsen LJ. 2004. Genomic analysis of aminotransferases in Arabidopsis thaliana. Critical Reviews in Plant Sciences 23: 73–89. Liu Y, Guérard F, Hodges M, Jossier M. 2020. Phosphomimetic T335D mutation of Hydroxypyruvate Reductase 1 modifies cofactor specificity and impacts Arabidopsis growth in air. Plant Physiology 183: 194–205. Liu S, Kandoth PK, Warren SD, Yeckel G, Heinz R, Alden J, Yang C, Jamai A, El-Mellouki T, Juvale PS, et al. 2012. A soybean cyst nematode resistance gene points to a new mechanism of plant resistance to pathogens. Nature 492: 256–260. Liu Y, Mauve C, Lamothe-Sibold M, Guérard F, Glab N, Hodges M, Jossier M. 2019. Photorespiratory serine hydroxymethyltransferase 1 activity impacts abiotic stress tolerance and stomatal closure. Plant Cell and Environment 42: 2567–2583. Love MI, Huber W, Anders S. 2014. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biology 15: 550. Lu Y, Li Y, Yang Q, Zhang Z, Chen Y, Zhang S, Peng XX. 2014. Suppression of glycolate oxidase causes glyoxylate accumulation that inhibits photosynthesis through deactivating Rubisco in rice. Physiologia Plantarum 150: 463–476. Lutziger I, Oliver DJ. 2001. Characterization of Two cDNAs Encoding Mitochondrial Lipoamide Dehydrogenase from Arabidopsis 1. Plant Physiology 127: 615–623. Lv T, Li X, Fan T, Luo H, Xie C, Zhou Y, Tian CE. 2019. The calmodulin-binding protein IQM1 interacts with CATALASE2 to affect pathogen defense. Plant Physiology 181: 1314– 1327. Ma F, Jazmin LJ, Young JD, Allen DK. 2014. Isotopically nonstationary 13C flux analysis of changes in Arabidopsis thaliana leaf metabolism due to high light acclimation. Proceedings of the National Academy of Sciences of the United States of America 111: 16967–16972. 94 Ma Y, Ma X, Gao X, Wu W, Zhou B. 2021. Light induced regulation pathway of anthocyanin biosynthesis in plants. International Journal of Molecular Sciences 22. Ma H, Sheng C, Qiao L, Zhao H, Niu D. 2020. A comparative proteomic approach to identify defence-related proteins between resistant and susceptible rice cultivars challenged with the fungal pathogen Rhizoctonia solani. Plant Growth Regulation 90: 73–88. Mackey D, Belkhadir Y, Alonso JM, Ecker JR, Dangl JL. 2003. Arabidopsis RIN4 is a target of the type III virulence effector AvrRpt2 and modulates RPS2-mediated resistance. Cell 112: 379–389. Mano S, Hayashi M, Nishimura M. 1999. Light regulates alternative splicing of hydroxypyruvate reductase in pumpkin. Plant Journal 17: 309–320. Manohar M, Tian M, Moreau M, Park SW, Choi HW, Fei Z, Friso G, Asif M, Manosalva P, Von Dahl CC, et al. 2015. Identification of multiple salicylic acid-binding proteins using two high throughput screens. Frontiers in Plant Science 5: 1–14. Martin M. 2011. Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet.journal 17: 10. Mathioudakis MM, Veiga RSL, Canto T, Medina V, Mossialos D, Makris AM, Livieratos I. 2013. Pepino mosaic virus triple gene block protein 1 (TGBp1) interacts with and increases tomato catalase 1 activity to enhance virus accumulation. Molecular Plant Pathology 14: 589– 601. Mekonnen DW, Ludewig F. 2016. Phenotypic and chemotypic studies using Arabidopsis and yeast reveal that GHB converts to SSA and induce toxicity. Plant Molecular Biology 91: 429– 440. Messant M, Timm S, Fantuzzi A, Weckwerth W, Bauwe H, Rutherford AW, Krieger- Liszkay A. 2018. Glycolate induces redox tuning of photosystem II in vivo: Study of a photorespiration mutant. Plant Physiology 177: 1277–1285. Mindrinos M, Katagiri F, Yu G-L, Ausubel FM. 1994. The A. thaliana disease resistance gene RPS2 encodes a protein containing a nucleotide-binding site and leucine-rich repeats. Cell 78: 1089–1099. Mishra Y, Johansson Jänkänpää H, Kiss AZ, Funk C, Schröder WP, Jansson S. 2012. Arabidopsis plants grown in the field and climate chambers significantly differ in leaf morphology and photosystem components. BMC Plant Biology 12. Misra BB, de Armas E, Chen S. 2016. Differential metabolomic responses of PAMP-triggered immunity and effector-triggered immunity in Arabidopsis suspension cells. Metabolomics 12: 1– 15. Mitsuya Y, Takahashi Y, Berberich T, Miyazaki A, Matsumura H, Takahashi H, Terauchi R, Kusano T. 2009. Spermine signaling plays a significant role in the defense response of Arabidopsis thaliana to cucumber mosaic virus. Journal of Plant Physiology 166: 626–643. Mittler R, Herr EH, Orvar BL, Van Camp W, Willekens H, Inzé D, Ellis BE. 1999. Transgenic tobacco plants with reduced capability to detoxify reactive oxygen intermediates are hyperresponsive to pathogen infection. Proceedings of the National Academy of Sciences of the United States of America 96: 14165–14170. 95 Moreno JI, Martín R, Castresana C. 2005. Arabidopsis SHMT1, a serine hydroxymethyltransferase that functions in the photorespiratory pathway influences resistance to biotic and abiotic stress. Plant Journal 41: 451–463. Mulligan RM, Wilson B, Tolbert NE. 1983. Effects of glyoxylate on photosynthesis by intact chloroplasts. Plant Physiology 72: 415–419. Muraoka H, Tang Y, Terashima I, Koizumi H, Washitani I. 2000. Contributions of diffusional limitation, photoinhibition and photorespiration to midday depression of photosynthesis in Arisaema heterophyllum in natural high light. Plant, Cell and Environment 23: 235–250. Murota K, Shimura H, Takeshita M, Masuta C. 2017. Interaction between Cucumber mosaic virus 2b protein and plant catalase induces a specific necrosis in association with proteasome activity. Plant Cell Reports 36: 37–47. Navarre DA, Wolpert TJ. 1995. Inhibition of the glycine decarboxylase multienzyme complex by the host-selective toxin victorin. The Plant Cell 7: 463–471. Ngou BPM, Ahn HK, Ding P, Jones JDG. 2021. Mutual potentiation of plant immunity by cell- surface and intracellular receptors. Nature 592: 110–115. Noctor G, Arisi A-CM, Jouanin L, Foyer CH. 1999. Photorespiratory glycine enhances glutathione accumulation in both the chloroplastic and cytosolic compartments. Journal of Experimental Botany 50: 1157–1167. Okinaka Y, Yang C-H, Herman E, Kinney A, Keen NT. 2002. The P34 syringolide elicitor receptor interacts with a soybean photorespiration enzyme, NADH-dependent hydroxypyruvate reductase. Molecular Plant-Microbe Interactions® 15: 1213–1218. Ort DR. 2001. When there is too much light. Plant Physiology 125: 29–32. Ort DR, Baker NR. 2002. A photoprotective role for O2 as an alternative electron sink in photosynthesis? Current Opinion in Plant Biology 5: 193–198. Osei-Bonsu I, McClain AM, Walker BJ, Sharkey TD, Kramer DM. 2021. The roles of photorespiration and alternative electron acceptors in the responses of photosynthesis to elevated temperatures in cowpea. Plant, Cell & Environment 44: 2290–2307. Pan R, Liu J, Wang S, Hu J. 2020. Peroxisomes: versatile organelles with diverse roles in plants. New Phytologist 225: 1410–1427. Patro R, Duggal G, Love MI, Irizarry RA, Kingsford C. 2017. Salmon provides fast and bias- aware quantification of transcript expression. Nature Methods 14: 417–419. Pérez-Delgado CM, García-Calderón M, Sánchez DH, Udvardi MK, Kopka J, Márquez AJ, Betti M. 2013. Transcriptomic and metabolic changes associated with photorespiratory ammonium accumulation in the model legume Lotus japonicus. Plant Physiology 162: 1834– 1848. Petersen LN, Ingle RA, Knight MR, Denby KJ. 2009. OXI1 protein kinase is required for plant immunity against Pseudomonas syringae in arabidopsis. Journal of Experimental Botany 60: 3727–3735. Pick TR, Bräutigam A, Schulz MA, Obata T, Fernie AR, Weber APM. 2013. PLGG1, a plastidic glycolate glycerate transporter, is required for photorespiration and defines a unique 96 class of metabolite transporters. Proceedings of the National Academy of Sciences of the United States of America 110: 3185–3190. Pieterse CMJ, Van Der Does D, Zamioudis C, Leon-Reyes A, Van Wees SCM. 2012. Hormonal modulation of plant immunity. Annual Review of Cell and Developmental Biology 28: 489–521. Pracharoenwattana I, Zhou W, Smith SM. 2010. Fatty acid beta-oxidation in germinating Arabidopsis seeds is supported by peroxisomal hydroxypyruvate reductase when malate dehydrogenase is absent. Plant Molecular Biology 72: 101–109. Pruitt RN, Locci F, Wanke F, Zhang L, Saile SC, Joe A, Karelina D, Hua C, Fröhlich K, Wan WL, et al. 2021. The EDS1–PAD4–ADR1 node mediates Arabidopsis pattern-triggered immunity. Nature 598: 495–499. Queval G, Issakidis-Bourguet E, Hoeberichts FA, Vandorpe M, Gakière B, Vanacker H, Miginiac-Maslow M, Van Breusegem F, Noctor G. 2007. Conditional oxidative stress responses in the Arabidopsis photorespiratory mutant cat2 demonstrate that redox state is a key modulator of daylength-dependent gene expression, and define photoperiod as a crucial factor in the regulation of H2O2-induced cell death. Plant Journal 52: 640–657. Reinholdt O, Schwab S, Zhang Y, Reichheld J-P, Fernie AR, Hagemann M, Timm S. 2019. Redox-Regulation of Photorespiration through Mitochondrial Thioredoxin o1. Plant Physiology 181: 442–457. Rentel MC, Lecourieux D, Ouaked F, Usher SL, Petersen L, Okamoto H, Knight H, Peck SC, Grierson CS, Hirt H, et al. 2004. OXI1 kinase is necessary for oxidative burst-mediated signalling in Arabidopsis. Nature 427: 858–861. Reumann S, Weber APM. 2006. Plant peroxisomes respire in the light: Some gaps of the photorespiratory C2 cycle have become filled—Others remain. Biochimica et Biophysica Acta (BBA) - Molecular Cell Research 1763: 1496–1510. Riens B, Lohaus G, Heineke D, Heldt HW. 1991. Amino acid and sucrose content determined in the cytosolic, chloroplastic, and vacuolar compartments and in the phloem sap of spinach leaves. Plant Physiology 97: 227–233. Roach T, Krieger-Liszkay A. 2014. Regulation of photosynthetic electron transport and photoinhibition. Current Protein & Peptide Science 15: 351–362. Rojas CM, Mysore KS. 2012. Glycolate oxidase is an alternative source for H2O2 production during plant defense responses and functions independently from NADPH oxidase. Plant Signaling and Behavior 7: 752–755. Rojas CM, Senthil-Kumar M, Wang K, Ryu C-M, Kaundal A, Mysore KS. 2012. Glycolate oxidase modulates reactive oxygen species-mediated signal transduction during Nnonhost resistance in Nicotiana benthamiana and Arabidopsis. The Plant Cell 24: 336–352. Ros R, Muñoz-Bertomeu J, Krueger S. 2014. Serine in plants: Biosynthesis, metabolism, and functions. Trends in Plant Science 19: 564–569. Rosa-Téllez S, Alcántara-Enguídanos A, Martínez-Seidel F, Casatejada-Anchel R, Saeheng S, Bailes CL, Erban A, Barbosa-Medeiros D, Alepúz P, Matus JT, et al. 2023. The serine– 97 glycine–one-carbon metabolic network orchestrates changes in nitrogen and sulfur metabolism and shapes plant development. The Plant Cell: 1–23. Rossel JB, Wilson IW, Pogson BJ. 2002. Global changes in gene expression in response to high light in Arabidopsis. Plant Physiology 130: 1109–1120. Ruban A V. 2016. Nonphotochemical chlorophyll fluorescence quenching: Mechanism and effectiveness in protecting plants from photodamage. Plant Physiology 170: 1903–1916. Saji S, Bathula S, Kubo A, Tamaoki M, Aono M, Sano T, Tobe K, Timm S, Bauwe H, Nakajima N, et al. 2017. Ozone-sensitive arabidopsis mutants with deficiencies in photorespiratory enzymes. Plant and Cell Physiology 58: 914–924. Sánchez-Casas P, Klessig DF. 1994. A salicylic acid-binding activity and a salicylic acid- inhibitable catalase activity are present in a variety of plant species. Plant Physiology 106: 1675– 1679. Segarra G, Casanova E, Bellido D, Odena MA, Oliveira E, Trillas I. 2007. Proteome, salicylic acid, and jasmonic acid changes in cucumber plants inoculated with Trichoderma asperellum strain T34. Proteomics 7: 3943–3952. Selinski J, Scheibe R. 2019. Malate valves: old shuttles with new perspectives. Plant Biology 21: 21–30. Sharma N, Nagar S, Thakur M, Suriyakumar P, Kataria S, Shanker AK, Landi M, Anand A. 2023. Photosystems under high light stress: throwing light on mechanism and adaptation. Photosynthetica 61: 250–263. Shi X, Bloom A. 2021. Photorespiration: the futile cycle? Plants 10: 908. Shi Y, Ke X, Yang X, Liu Y, Hou X. 2022a. Plants response to light stress. Journal of Genetics and Genomics 49: 735–747. Shi Q, Sun H, Timm S, Zhang S, Huang W. 2022b. Photorespiration alleviates photoinhibition of Photosystem I under fluctuating light in tomato. Plants 11. Shikanai T. 2007. Cyclic Electron Transport Around Photosystem I: Genetic Approaches. Annual Review of Plant Biology 58: 199–217. Silva BN, Picanço BBM, Martins SCV, Rodrigues FA. 2023. Impairment of the photorespiratory pathway on tomato leaves during the infection process of Septoria lycopersici. Physiological and Molecular Plant Pathology 125. Simpson JP, Di Leo R, Dhanoa PK, Allan WL, Makhmoudova A, Clark SM, Hoover GJ, Mullen RT, Shelp BJ. 2008. Identification and characterization of a plastid-localized Arabidopsis glyoxylate reductase isoform: Comparison with a cytosolic isoform and implications for cellular redox homeostasis and aldehyde detoxification. Journal of Experimental Botany 59: 2545–2554. Smith K, Strand DD, Kramer DM, Walker BJ. 2023. The role of photorespiration in preventing feedback regulation via ATP synthase in Nicotiana tabacum. Plant Cell and Environment. Soneson C, Love MI, Robinson MD. 2016. Differential analyses for RNA-seq: transcript-level estimates improve gene-level inferences. F1000Research 2016 4:1521 4: 1521. 98 Sørhagen K, Laxa M, Peterhänsel C, Reumann S. 2013. The emerging role of photorespiration and non-photorespiratory peroxisomal metabolism in pathogen defence (A Weber, Ed.). Plant Biology 15: 723–736. South PF, Cavanagh AP, Lopez-calcagno PE, Raines CA, Ort DR. 2018. Optimizing photorespiration for improved crop productivity. Journal of Integrative Plant Biology 60: 1217– 1230. Spoel SH, Dong X. 2012. How do plants achieve immunity? Defence without specialized immune cells. Nature Reviews Immunology 12: 89–100. Sun Y, Li P, Deng M, Shen D, Dai G, Yao N, Lu Y. 2017. The Ralstonia solanacearum effector RipAK suppresses plant hypersensitive response by inhibiting the activity of host catalases. Cellular Microbiology 19. Sun Y, Li Y, Li Y, Wang M, Mur LAJ, Shen Q, Guo S. 2021. Nitrate mediated resistance against Fusarium infection in cucumber plants acts via photorespiration. Plant Cell and Environment 44: 3412–3431. Sunil B, Saini D, Bapatla RB, Aswani V, Raghavendra AS. 2019. Photorespiration is complemented by cyclic electron flow and the alternative oxidase pathway to optimize photosynthesis and protect against abiotic stress. Photosynthesis Research 139: 67–79. Suzuki N, Devireddy AR, Inupakutika MA, Baxter A, Miller G, Song L, Shulaev E, Azad RK, Shulaev V, Mittler R. 2015. Ultra‐fast alterations in mRNA levels uncover multiple players in light stress acclimation in plants. The Plant Journal 84: 760–772. Szymańska R, Ślesak I, Orzechowska A, Kruk J. 2017. Physiological and biochemical responses to high light and temperature stress in plants. Environmental and Experimental Botany 139: 165–177. Takahashi S, Badger MR. 2011. Photoprotection in plants: a new light on photosystem II damage. Trends in Plant Science 16. Takahashi S, Bauwe H, Badger M. 2007. Impairment of the photorespiratory pathway accelerates photoinhibition of photosystem II by suppression of repair but not acceleration of damage processes in Arabidopsis. Plant Physiology 144: 487–494. Takahashi H, Chen Z, Du H, Liu Y, Klessig DF. 1997. Development of necrosis and activation of disease resistance in transgenic tobacco plants with severely reduced catalase levels. Plant Journal 11: 993–1005. Taler D, Galperin M, Benjamin I, Cohen Y, Kenigsbuch D. 2004. Plant eR genes that encode photorespiratory enzymes confer resistance against disease. Plant Cell 16: 172–184. Tessmer OL, Jiao Y, Cruz JA, Kramer DM, Chen J. 2013. Functional approach to high- throughput plant growth analysis. BMC Systems Biology 7. Timm S. 2020. The impact of photorespiration on plant primary metabolism through metabolic and redox regulation. Biochemical Society Transactions 48: 2495–2504. Timm S, Bauwe H. 2013. The variety of photorespiratory phenotypes - employing the current status for future research directions on photorespiration. Plant Biology 15: 737–747. 99 Timm S, Florian A, Fernie AR, Bauwe H. 2016. The regulatory interplay between photorespiration and photosynthesis. Journal of Experimental Botany 67: 2923–2929. Timm S, Florian A, Jahnke K, Nunes-Nesi A, Fernie AR, Bauwe H. 2011. The hydroxypyruvate-reducing system in arabidopsis: Multiple enzymes for the same end. Plant Physiology 155: 694–705. Timm S, Florian A, Wittmiß M, Jahnke K, Hagemann M, Fernie AR, Bauwe H. 2013. Serine acts as a metabolic signal for the transcriptional control of photorespiration-related genes in Arabidopsis. Plant Physiology 162: 379–389. Timm S, Hagemann M. 2020. Photorespiration-how is it regulated and how does it regulate overall plant metabolism? Journal of Experimental Botany 71: 3955–3965. Timm S, Mielewczik M, Florian A, Frankenbach S, Dreissen A, Hocken N, Fernie AR, Walter A, Bauwe H. 2012. High-to-low CO2 acclimation reveals plasticity of the photorespiratory pathway and indicates regulatory links to cellular metabolism of Arabidopsis. PLoS ONE 7. Timm S, Nunes-Nesi A, Florian A, Eisenhut M, Morgenthal K, Wirtz M, Hell R, Weckwerth W, Hagemann M, Fernie AR, et al. 2021. Metabolite profiling in Arabidopsis thaliana with moderately impaired photorespiration reveals novel metabolic links and compensatory mechanisms of photorespiration. Metabolites 2021, Vol. 11, Page 391 11: 391. Timm S, Nunes-Nesi A, Pärnik T, Morgenthal K, Wienkoop S, Keerberg O, Weckwerth W, Kleczkowski LA, Fernie AR, Bauwe H, et al. 2008. A cytosolic pathway for the conversion of hydroxypyruvate to glycerate during photorespiration in Arabidopsis. Plant Cell 20: 2848–2859. Di Tommaso P, Chatzou M, Floden EW, Barja PP, Palumbo E, Notredame C. 2017. Nextflow enables reproducible computational workflows. Nature Biotechnology 35: 316–319. Toyota M, Spencer D, Sawai-Toyota S, Jiaqi W, Zhang T, Koo AJ, Howe GA, Gilroy S. 2018. Glutamate triggers long-distance, calcium-based plant defense signaling. Science 361: 1112–1115. Voss I, Sunil B, Scheibe R, Raghavendra AS. 2013. Emerging concept for the role of photorespiration as an important part of abiotic stress response. Plant Biology 15: 713–722. Wada M. 2013. Chloroplast movement. Plant Science 210: 177–182. Walker BJ, Skabelund DC, Busch FA, Ort DR. 2016a. An improved approach for measuring the impact of multiple CO2 conductances on the apparent photorespiratory CO2 compensation point through slope-intercept regression. Plant Cell and Environment 39: 1198–1203. Walker BJ, VanLoocke A, Bernacchi CJ, Ort DR. 2016b. The Costs of Photorespiration to Food Production Now and in the Future. Annual Review of Plant Biology 67: 107–129. Wang A, Shu X, Jing X, Jiao C, Chen L, Zhang J, Ma L, Jiang Y, Yamamoto N, Li S, et al. 2021. Identification of rice (Oryza sativa L.) genes involved in sheath blight resistance via a genome-wide association study. Plant Biotechnology Journal 19: 1553–1566. Wang Z, Wang Y, Wang Y, Li H, Wen Z, Hou X. 2022. HPR1 is required for high light intensity induced photorespiration in Arabidopsis thaliana. International Journal of Molecular Sciences 23: 1–17. 100 Wasternack C, Song S. 2017. Jasmonates: Biosynthesis, metabolism, and signaling by proteins activating and repressing transcription. Journal of Experimental Botany 68: 1303–1321. Williams A, Pétriacq P, Schwarzenbacher RE, Beerling DJ, Ton J. 2018. Mechanisms of glacial-to-future atmospheric CO2 effects on plant immunity. New Phytologist 218: 752–761. Wiludda C, Schulze S, Gowik U, Engelmann S, Koczor M, Streubel M, Bauwe H, Westhoff P. 2012. Regulation of the photorespiratory GLDPA gene in C4 Flaveria : An intricate interplay of transcriptional and posttranscriptional processes. The Plant Cell 24: 137–151. Wingler A, Lea PJ, Quick WP, Leegood RC. 2000. Photorespiration: metabolic pathways and their role in stress protection. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 355: 1517–1529. Wingler A, Quick WP, Bungard RA, Bailey KJ, Lea PJ, Leegood RC. 1999. The role of photorespiration during drought stress: An analysis utilizing barley mutants with reduced activities of photorespiratory enzymes. Plant, Cell and Environment 22: 361–373. Xin XF, Nomura K, Aung K, Velásquez AC, Yao J, Boutrot F, Chang JH, Zipfel C, He SY. 2016. Bacteria establish an aqueous living space in plants crucial for virulence. Nature 539: 524– 529. Xu JJ, Fang X, Li CY, Zhao Q, Martin C, Chen XY, Yang L. 2018a. Characterization of arabidopsis thaliana hydroxyphenylpyruvate reductases in the tyrosine conversion pathway. Frontiers in Plant Science 9: 1–13. Xu YP, Yang J, Cai XZ. 2018b. Glycolate oxidase gene family in Nicotiana benthamiana: Genomewide identification and functional analyses in disease resistance. Scientific Reports 8: 1– 11. Yang Y, Jin H, Chen Y, Lin W, Wang C, Chen Z, Han N, Bian H, Zhu M, Wang J. 2012. A chloroplast envelope membrane protein containing a putative LrgB domain related to the control of bacterial death and lysis is required for chloroplast development in Arabidopsis thaliana. New Phytologist 193: 81–95. Yang J, Sun C, Fu D, Yu T. 2017. Test for L-glutamate inhibition of growth of Alternaria alternata by inducing resistance in tomato fruit. Food Chemistry 230: 145–153. Yao N, Tada Y, Sakamoto M, Nakayashiki H, Park P, Tosa Y, Mayama S. 2002. Mitochondrial oxidative burst involved in apoptotic response in oats. Plant Journal 30: 567–579. Yao J, Withers J, He SY. 2013. Pseudomonas syringae infection assays in Arabidopsis. In: Goossens A, Pauwels L, eds. Jasmonate Signaling: Methods and Protocols. Totowa, NJ: Humana Press, 63–81. Yu X, Feng B, He P, Shan L. 2017. From chaos to harmony: responses and signaling upon microbial pattern recognition. Annual Review of Phytopathology 55: 109–137. Yuan HM, Liu WC, Lu YT. 2017. CATALASE2 coordinates SA-mediated repression of both auxin accumulation and JA biosynthesis in plant defenses. Cell Host and Microbe 21: 143–155. Yuan M, Ngou BPM, Ding P, Xin XF. 2021. PTI-ETI crosstalk: an integrative view of plant immunity. Current Opinion in Plant Biology 62: 102030. 101 Yue H, Huang LP, Lu DYH, Zhang ZH, Zhang Z, Zhang DY, Zheng LM, Gao Y, Tan XQ, Zhou XG, et al. 2021. Integrated Analysis of microRNA and mRNA Transcriptome Reveals the Molecular Mechanism of Solanum lycopersicum Response to Bemisia tabaci and Tomato chlorosis virus. Frontiers in Microbiology 12: 1–17. Zabala M de T, Littlejohn G, Jayaraman S, Studholme D, Bailey T, Lawson T, Tillich M, Licht D, Bölter B, Delfino L, et al. 2015. Chloroplasts play a central role in plant defence and are targeted by pathogen effectors. Nature Plants 1: 1–14. Zarei A, Brikis CJ, Bajwa VS, Chiu GZ, Simpson JP, DeEll JR, Bozzo GG, Shelp BJ. 2017. Plant glyoxylate/succinic semialdehyde reductases: Comparative biochemical properties, function during chilling stress, and subcellular localization. Frontiers in Plant Science 8: 1–13. Zelitch I, Schultes NP, Peterson RB, Brown P, Brutnell TP. 2009. High glycolate oxidase activity is required for survival of maize in normal air. Plant Physiology 149: 195–204. Zhang M, Li Q, Liu T, Liu L, Shen D, Zhu Y, Liu P, Zhou JM, Dou D. 2015. Two cytoplasmic effectors of phytophthora sojae regulate plant cell death via interactions with plant catalases. Plant Physiology 167: 164–175. Zhang L, Paasch BC, Chen J, Day B, He SY. 2019. An important role of l-fucose biosynthesis and protein fucosylation genes in Arabidopsis immunity. New Phytologist 222: 981–994. Zhang Y, Song R, Yuan H, Li T, Wang L, Lu K, Guo J, Liu W. 2021. Overexpressing the N‐ terminus of CATALASE2 enhances plant jasmonic acid biosynthesis and resistance to necrotrophic pathogen Botrytis cinerea B05.10. Molecular Plant Pathology 22: 1226–1238. Zhang Z, Xu Y, Xie Z, Li X, He ZH, Peng XX. 2016. Association-dissociation of glycolate oxidase with catalase in rice: A potential switch to modulate intracellular H2O2 levels. Molecular Plant 9: 737–748. Zhao H, Lou Y, Sun H, Li L, Wang L, Dong L, Gao Z. 2016. Transcriptome and comparative gene expression analysis of Phyllostachys edulis in response to high light. BMC Plant Biology 16. Zhao J, Sun Q, Quentin M, Ling J, Abad P, Zhang X, Li Y, Yang Y, Favery B, Mao Z, et al. 2021. A Meloidogyne incognita C‐type lectin effector targets plant catalases to promote parasitism. New Phytologist 232: 2124–2137. Zhao J, Zhang Y, Bian X, Lei J, Sun J, Guo N, Gai J, Xing H. 2013. A comparative proteomics analysis of soybean leaves under biotic and abiotic treatments. Molecular Biology Reports 40: 1553–1562. Zheng XT, Yu ZC, Tang JW, Cai ML, Chen YL, Yang CW, Chow WS, Peng CL. 2021. The major photoprotective role of anthocyanins in leaves of Arabidopsis thaliana under long-term high light treatment: antioxidant or light attenuator? Photosynthesis Research 149: 25–40. 102 APPENDIX Figure A1. Characterizations of shpr7, T-DNA insertion lines of GLYR1, and GLYR1- overexpressing lines. (a) GLYR1 expression in shpr7 plants grown under 2 weeks of normal light followed by 4.5 weeks of high light. (b) GLYR1 expression in the T-DNA insertion lines grown under 2 weeks of normal light followed by 3 weeks of high light. The expression of UBQ10 was used as a control in (a) and (b). (c) The glyr1 single mutants and glyr1 hpr1 double mutants grown under 2 weeks of normal light followed by 2.5-week high light. (d) FLAG- GLYR1 expression in transgenic lines grown under 2 weeks of normal light followed by 2.5 weeks of high light. 103 Figure A2. Metabolic profiling of plants grown under high CO2. Plants were grown under high CO2 for 3 weeks. Different letters indicate statistically significant differences (p<0.05), which were determined by One-way ANOVA with Tukey’s HSD test. Biological replicates: n=6. 104 Figure A3. Profiling of transitional hydroxypyruve in plants transferred to the photorespiratory environment. The same data of hydroxypyruve was used as Fig. 2.5, but hpr1 and glyr1 hpr1 were removed to focus on cat2 and glyr1 cat2. Plants were grown under 3 weeks of high CO2 and normal light, and then transferred to ambient CO2 and high light before lights were turned on. Leaf tissue was sampled after ~10 h. Different letters indicate statistically significant differences (p<0.05), which were determined by One-way ANOVA with Tukey’s HSD test. Biological replicates: n=6. 105 Figure A4. Plant growth after feeding with serine. Growth phenotype (a) and fresh weight (b) of 12-day-old seedlings on plates with or without serine under normal light (NL) or high light (HL) are recorded. The p values determined by Student’s unpaired two-tailed t-test were labeled on the plot. Seedlings grown on the same plate were treated as a biological replicate. Biological replicates: n=3. Figure A5. Characterizations of hpr2-3. (a) Schematic depiction of the HPR2 gene and the position of the mutation in hpr2-3. (b) HPR2 expression in hpr2-3 in plants grown under 2 weeks of normal light followed by 2 weeks of high light. UBQ10 was used as a control. 106 Figure A6. Hydroxypyruvate reductase (HPR) activities in plants. The NADH-and NADPH- dependent HPR enzyme activities in plants grown under 3-week high CO2 and normal light followed by ~10 h ambient air and high light (a) and plants grown under 2 weeks of normal light followed by 2 weeks of high light (b) are shown. Different letters indicate statistically significant differences (p<0.05), which were determined by One-way ANOVA with Tukey’s HSD test. Biological replicates: n=3 for (a) and n=4 for (b). 107 Table A1. List of primers used in this study. Primer name Primer sequence (5'-3') SALK_LBb1.3 ATTTTGCCGATTTCGGAAC SALK_067724_LP GTTGAGTTTGGATATGGCCAC SALK_067724_RP ACCAAACATCGCGATTACAAC SALK_076998_LP ACATTTTGGAGCATTGACTGG SALK_076998_RP TCTGGTGCTCCTGTATGGAAC SALK_053469_RP GTTTTGCCATAGGCTCGGCTT SALK_053469_LP CGTCGTCGTCTCCATACCCAT SALK_057410_LP ACAATCAAAACCCAAAATCCC SALK_057410_RP AAACGATCTCTTCCCCAAGAC SALK_202680_LP CTCAGCCAATCCAAATGAGTG SALK_202680_RP CGGTGTTTTGGAGCAGATATG SALK_203580_LP GCTTGCAAAAGTTTGATCACC SALK_203580_RP GTTTGGGAATCATGGGAAAAG SALK_105876_LP CACTGGATTCCCTAAACATGC SALK_105876_RP CCCTTAGCTCCTAATGCATCC Purpose Genotyping Genotyping Genotyping Genotyping Genotyping Genotyping Genotyping Genotyping Genotyping Genotyping Genotyping Genotyping Genotyping Genotyping Genotyping GLYR1-att-F GLYR1-att-R ggggacaagtttgtacaaaaaagcaggcttcATGGAAGTAGGGTTT CTGGGT Cloning ggggaccactttgtacaagaaagctgggtcCTATTCGCGGGAGAA TTTC GLYR1-CDS-F ATGGAAGTAGGGTTTCTGGGT GLYR1-CDS-R CTATTCGCGGGAGAATTTCAC FLAG-GLYR1-F GACTACAAAGACGATGACGACAAA UBQ10-F UBQ10-R TCAATTCTCTCTACCGTGATCAAGATG GGTGTCAGAACTCTCCACCTCAAGAG HPR2-RT-F ATGGAATCAATCGGAGTCCTTATGA HPR2-RT-R CCAAATCCCAAATGTGTCACATGAC 108 Cloning RT-PCR RT-PCR RT-PCR RT-PCR RT-PCR RT-PCR RT-PCR