CHLOROPLAST LIPID METABOLISM IN THE CONTEXT OF PLANT GROWTH AND DEVELOPMENT By Ron Cook A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Biochemistry and Molecular Biology – Doctor of Philosophy 2023 ABSTRACT For over two billion years, most life on earth has depended on oxygenic photosynthesis for fuel and sustenance. In plants, the descendants of ancient cyanobacteria operate as subcellular photosynthetic organelles, the chloroplasts, where an extensive membrane infrastructure converts light into high-energy chemical bonds. Chloroplast membranes are distinctive in that their lipid components primarily rely on sugars as head groups, as opposed to phosphate-based moieties. Plant membrane metabolism is therefore highly geared towards the conversion of de novo-synthesized phospholipids into chloroplast galactolipids, and in Arabidopsis thaliana, portions of these pathways operate in parallel at the chloroplast and the endoplasmic reticulum (ER). Here, I present novel insights into the roles of chloroplast-associated lipid phosphate phosphatases LPPγ, LPPε1, and LPPε2, which dephosphorylate phosphatidic acid (PA) to make diacylglycerol (DAG), the substrate for galactosylation reactions. LPPγ and LPPε1 were determined to act on ER-assembled PA, with their catalytic activity at the chloroplast outer envelope membrane. All three chloroplast LPPs appeared uninvolved in the dephosphorylation of chloroplast-derived PA, despite localization of LPPε2 to the interior chloroplast membranes. Growth inhibition in lppγ lppε1 double mutant plants implicated PA pools at the outer envelope membrane as affecting developmental regulation, thus linking LPPγ or LPPε1 to plant growth and development. The connection between chloroplast lipid metabolism and plant growth regulation was also exploited in a suppressor screen using a transgenic Arabidopsis line, in which overexpression of the plastid lipase-encoding gene PLIP3 leads to accumulation of the defense hormone jasmonic acid (JA). These PLIP3-OX lines exhibit unique JA-induced morphological phenotypes, and suppression of these phenotypes was targeted in the screen. One mutant, sup72, had a point mutation in KEEP ON GOING (KEG) which co-segregated with the suppression phenotype. KEG is known to have a repressive role in abscisic acid (ABA) signaling, and its apparent effects on JA signaling in sup72 indicate it may also facilitate coordination of the ABA and JA pathways. In another mutant, sup11, PLIP3-OX suppression was caused by a nonsense mutation in CDK8, linking the gene product to activation of JA-responsive transcription. Overall, these Arabidopsis lines with distorted chloroplast lipid pathways provide greater insight into the nuances of metabolism and lipid trafficking, as well as connections to broader elements of plant growth and development. ACKNOWLEDGEMENTS The work presented here was possible thanks to the support of many peers, coworkers, staff, family, and friends. As my degree-granting program, the Biochemistry and Molecular Biology (BMB) department provided a helpful framework for courses, teaching, and research, as well as administrative support, particularly from Jessica Lawrence. As the BMB graduate program director, David Arnosti was very accommodating, and also a pleasure to interact with. As a student at the MSU-DOE Plant Research Laboratory (PRL), the communication and responsiveness of the PRL staff to any of my needs was always superb, whether it was from the administrative staff, growth chamber facility, plant transformation facility, instrument repairs, or computer support. I am also grateful to PRL funding for financial support, in addition to that of the BioMolecular Science Gateway program, the MSU Plant Science Fellowship, and Plant Biotechnology for Health and Sustainability fellowship, which also partially funded a valuable 8-week industry internship. The collaborative environment of the PRL facilitated discussion with researchers in other labs, and such interactions with Deepak Bhandari and Nate Havko were particularly useful in the context of this work. Leah Johnson from the Gregg Howe lab walked me through the preparation of genomic sequencing samples, and the analysis of the resulting data. I also collaborated with the David Kramer lab to obtain the photosynthetic data presented here, which was generated by Jeff Cruz. Contributions by John Froehlich were absolutely instrumental, not just thanks to his execution of the chloroplast import experiments, but also because of his continuous guidance on my work with chloroplasts and protein biochemistry. Naturally, I have to thank all of the Benning lab members with whom I’ve worked over the past six years, and who are too numerous to mention all by name. In particular, my technical training with lipid analysis mostly came from Patrick Horn and Anastasiya Lavell, and many of the other skills and ideas applied here were developed with help from Carrie Hiser, Yang-Tsung Lin, and Yang Xu. Of the undergraduate students, Yash Manne was an exceptional peer who had an essential role in helping us to analyze genomic sequencing results, applying skills that he had developed as part of his data science degree. Likewise, undergraduate Ilayda Korkmaz was very quick to learn and operate independently in her ongoing pursuit to measure lipid phosphatase activity. Lab work in the Benning lab was efficient, and entertaining, thanks to Linda Danhof, who assisted with iv Arabidopsis crossing, mutagenesis, and transformation, as well as innumerable aspects of lab management. I also appreciate the oversight and input from my guidance committee members Gregg Howe, Hideki Takahashi, Tom Sharkey, and Yair Shachar-Hill. Personal friends and family provided fresh perspectives and suggestions throughout, with my parents, Orna and Boaz, always curious about research developments. Finally, I would like to thank Christoph Benning for his mentorship, patience, generosity, and genuine devotion to the success and well-being of his students. v TABLE OF CONTENTS CHAPTER 1: Chloroplast membrane lipid metabolism and connections to signaling ....................... 1 Introduction ................................................................................................................................... 2 Chloroplast membrane lipid metabolism................................................................................... 2 The roles of phosphatidic acid in the chloroplast ..................................................................... 5 Chloroplast lipids and jasmonic acid signaling .......................................................................... 7 Summary of research aims........................................................................................................... 8 REFERENCES .............................................................................................................................................. 11 CHAPTER 2: Metabolic and developmental roles of chloroplast phosphatidate phosphatases ... 17 Abstract ......................................................................................................................................... 18 Introduction ................................................................................................................................. 18 Results ........................................................................................................................................... 20 Discussion ..................................................................................................................................... 25 Methods........................................................................................................................................ 29 REFERENCES .............................................................................................................................................. 50 CHAPTER 3: A suppressor screen targeting novel components of OPDA conversion to jasmonic acid ............................................................................................................................................................. 55 Abstract ......................................................................................................................................... 56 Introduction ................................................................................................................................. 56 Results ........................................................................................................................................... 58 Discussion ..................................................................................................................................... 61 Methods........................................................................................................................................ 63 REFERENCES .............................................................................................................................................. 78 CHAPTER 4: Analysis, conclusions, and perspectives .......................................................................... 80 Introduction ................................................................................................................................. 81 Chloroplast LPPs and PA ............................................................................................................. 81 PLIP3-OX suppressor screen ...................................................................................................... 86 Conclusion .................................................................................................................................... 90 REFERENCES .............................................................................................................................................. 91 vi CHAPTER 1: Chloroplast membrane lipid metabolism and connections to signaling Major components of this chapter, including Figures 1.1 and 1.2, have been published in Cook et al. 2021 [1]. I wrote part 2 of the review, with some contribution to parts 1 and 4. 1 Introduction Plants are the basis for much of Earth’s multicellular life, owing to their capacity for using light energy and water to reduce CO2 into various organic molecules. While chloroplasts are primarily associated with photosynthesis in plants, these organelles host additional metabolic networks that are essential for robust cellular constitution and physiology. In particular, chloroplasts have a central role in glycerolipid metabolism, which serves both to maintain functional membranes in fluctuating environments, and as a chassis for broader sensing, signaling, and response mechanisms to external stimuli. Chloroplast membrane lipid metabolism Chloroplast membranes have evolved to accommodate an extensive photosynthetic apparatus, while maintaining minimal dependence on limiting nutrients. While phosphorus is a component of most lipids in virtually all other biological membranes, within chloroplasts it exists in less than half of envelope membrane lipids and less than 15% of thylakoid membrane lipids [2, 3]. Instead, these membranes are primarily composed of galactolipids, which are entirely derived from photosynthetic products made of carbon, oxygen, and hydrogen. In addition, sulfolipids are present as an alternative to phosphorus-based anionic membrane lipids. Plant chloroplast lipid metabolism is also closely linked to that of the ER, as the acyl components of ER-assembled lipids are synthesized and exported by chloroplasts, and ER-assembled lipids are often imported back into chloroplasts. Glycerolipid precursors and chloroplast fatty acid export Nearly all plant lipid biosynthesis begins with fatty acid (FA) biosynthesis in the chloroplast stroma by a Type II FA synthase similar to that of prokaryotes [4]. These FAs have various metabolic fates, including cuticular hydrocarbons, sphingolipids, and hormones, but the majority of FAs are esterified to glycerol to form glycerolipids. In the plastid pathway of glycerolipid biosynthesis, the acyltransferase ATS1 transfers 18:1 acyl groups from acyl-acyl carrier protein (acyl-ACP) to the sn- 1 position of glycerol 3-phosphate [5, 6]. ATS2 then transfers an additional acyl group from acyl- ACP to the sn-2 position, producing phosphatidic acid (PA) at the inner leaflet of the chloroplast inner envelope membrane (IEM) [7]. Because ATS2 is specific to 16:0 acyl-ACP, lipids with a 16- 2 carbon moiety at the sn-2 position can be identified as originating from plastid-synthesized PA [8]. The plastid pathway for membrane lipid biosynthesis is also referred to as the “prokaryotic” pathway, although its enzyme components actually have eukaryotic origins [9]. Fatty acids destined for the ER are released from ACP in the stroma by IEM-associated thioesterases, exported, and activated by acyl-CoA synthetases associated with the outer envelope membrane (OEM) [10, 11]. Acyl-CoAs are used for PA biosynthesis in the ER just as acyl- ACPs are used in plastid PA biosynthesis, with one key difference in substrate specificity: the ER acyltransferase which acylates the sn-2 position is specific to 18-carbon substrates [12]. This allows for lipids with 18-carbon chains at the sn-2 position to be identified as derivatives of ER- synthesized PA, or the “eukaryotic” pathway. Chloroplast studies have determined that a dedicated FA export machinery is required to account for observed FA transport rates, but the proteins and mechanisms involved remain largely unknown [13]. One export component in the IEM, FAX1, has been shown to contribute to efficient FA transport [14]. However, null fax1 mutants do maintain substantial FA export, indicating that supplementary or partially redundant export factors likely coexist with FAX1. Subsequent research led to the discovery of FAX2, FAX3, and FAX4, with FAX2/4 involved in plastid FA export in seeds, and FAX3 acting in partial redundancy with FAX1 in vegetative tissues [15-17]. While the FAX proteins may account for FA transport across the IEM, FA transfer across the intermembrane space would likely require mediation, as would FA flipping across the OEM for carboxyl exposure to cytosolic acyl-CoA synthetases. Discovery of novel FA export components was attempted through a suppressor screen, described in chapter 3. Chloroplast galactolipids The two primary glycerolipid constituents of chloroplast membranes are monogalactosyldiacylglycerol (MGDG) and digalactosyldiacylglycerol (DGDG) [3]. In some plants, including Arabidopsis, tomato, tobacco, and spinach, the sn-2 position of MGDG may contain either 16:3 or 18:3 acyl moieties, meaning that both plastid- and ER-assembled PA is directed towards MGDG biosynthesis. Such plants are referred to as 16:3 plants. In contrast, 18:3 plants, 3 which include legumes and monocots, only have 18:3 acyl groups at the sn-2 position of MGDG, indicating that MGDG is exclusively derived from ER-synthesized PA [18]. Due to the popularity of spinach and Arabidopsis in plant basic research, galactolipid metabolism is better characterized in 16:3 plants than in 18:3 plants. In 16:3 plants, bulk MGDG synthesis under nutrient replete conditions is observed at the IEM, and requires diacylglycerol (DAG) and UDP-galactose as substrates [19]. This reaction is catalyzed by the monogalactosyldiacylglycerol synthase (MGD1), which is associated with the outer leaflet of the chloroplast IEM in 16:3 plants [20-22]. 16:3 plants also exhibit PA phosphatase (PAP) activity primarily associated with the IEM, which presumably provides MGD1 with DAG substrate [19]. ER-derived MGDG is synthesized from precursors imported to the IEM by the TGD complex, although it is still unclear whether PA or DAG is the imported species [23]. On the other hand, pea chloroplasts exhibit substantial UDP:DAG galactosyltransferase activity in the OEM, which may explain the predominance of ER- derived galactolipids in 18:3 plants [24]. 18:3 plants also have far lower PAP activity in chloroplast envelopes, which is localized to the IEM [8, 25, 26]. Therefore, MGDG in 18:3 plants is possibly synthesized at the OEM from ER-derived DAG, while MGDG biosynthesis in 16:3 plants occurs at the IEM from a mixture of plastid-derived PA and ER-derived DAG or PA. DGDG biosynthesis by a UDP-galactose:MGDG galactosyltransferase (DGD1) was initially identified in pea chloroplast envelopes [27]. The dgd1 mutant was subsequently isolated in Arabidopsis, and the enzyme DGD1 was localized to the OEM and determined to require MGDG and UDP-galactose as substrates, likely at the cytosolic side of the membrane [28-31]. Despite equivalent concentrations of plastid-derived MGDG in the OEM and the IEM, DGDG has very low amounts of 16:3 acyl groups, indicating that DGD1 specifically galactosylates ER-derived MGDG [32]. This could be due to substrate preference, or to a low abundance of 16:3 MGDG at the outer leaflet of the OEM. DGD1 also contains an N-terminal domain that has been implicated in lipid transfer between the envelope membranes [32]. In 16:3 plants, MGD1 and DGD1 are the primary catalysts for galactolipid biosynthesis in the absence of environmental stress. However, in response to changing biotic and abiotic factors, other enzymes are synthesized or activated which redirect chloroplast lipid metabolism from this 4 baseline. In particular, phosphate depletion induces expression of genes encoding extraplastidic phospholipases and PAPs, as well as OEM-localized galactosyltransferases, which work together to convert extraplastidic phospholipids into galactolipids [1] (Fig. 1.1). Chloroplast anionic lipids In the chloroplast, the anionic membrane lipids phosphatidylglycerol (PG) and sulfoquinovosyldiacylglycerol (SQDG) are both synthesized at the IEM. PG is the only major phospholipid component of the IEM and thylakoid membranes, and its biosynthesis begins with the activation of plastid-synthesized PA to CDP-DAG [33]. PG phosphate synthase then exchanges the activated head group for glycerol 3-phosphate, producing PG phosphate [34-36], which is subsequently dephosphorylated by PG phosphate phosphatase, generating PG [37]. For SQDG biosynthesis, a UDP-sulfoquinovose precursor is produced from UDP-glucose and sulfite by SQD1 in the chloroplast stroma [38, 39]. SQD2 then synthesizes SQDG from the UDP-sulfoquinovose and DAG at the IEM [40, 41]. During phosphate deprivation, the SQDG biosynthetic pathways are upregulated, and the majority of chloroplast PG is replaced with SQDG [38, 40] (Fig. 1.1). The roles of phosphatidic acid in the chloroplast Although PA is the precursor for all other chloroplast glycerolipids, its low abundance means that quantification of chloroplast PA is difficult [42]. However, studies on PA-protein interactions and transgenic plants with alterations to PA metabolism do provide some preliminary insights into the role of PA, beyond its existence as a lipid precursor. PA interactions with proteins of lipid metabolism Several major proteins involved in chloroplast lipid metabolism are known to specifically bind PA (Fig. 1.2). MGD1 has been shown to require allosteric activation by PA and PG in order to synthesize MGDG from DAG and UDP-galactose [43]. Because DAG is itself an inhibitor of PAP activity [44], PA activation of MGD1 presumably maintains a consistent proportion in the activities of PAP and MGD1. This balance would prevent an excess accumulation of either PA or DAG in the IEM. Based on these discoveries, PA appears to have a typical role in allosteric activation of a metabolic pathway by the initial precursor. In addition, PA may promote MGDG export to the 5 OEM for subsequent DGDG biosynthesis: The N-terminal extension of DGD1 binds specifically to PA, potentially leading to PA-mediated membrane fusion that facilitates galactolipid transfer between the envelope membranes [32]. PA may also be a substrate or a regulator in the import of ER lipids to the IEM in 16:3 plants, a process that is mediated by the TGD complex [45]. The subunit TGD2 is anchored in the IEM by its N-terminus, while its C-terminus binds specifically to PA; however, the functional role of this interaction is unclear [46]. In addition, the OEM-localized TDG4 protein involved in the import of ER lipid precursors also specifically binds PA, and its PA binding site is required for activity [47- 49]. Thylakoid membrane biosynthesis may also be regulated by PA. CHLOROPLAST SEC14-LIKE1 protein (CPSFL1), which is required for vesicle formation at the IEM and thylakoid membrane biogenesis, has a specific binding site for PA and traffics phosphoinositides to membranes enriched in PA [50, 51] (Fig. 1.2). While the specific roles of plastid phosphoinositides are not fully elucidated, they are known to be involved in development and signaling processes within chloroplasts through interactions with proteins such as WKS1, VIPP1 and VIPP2 [52]. Effects of modifying chloroplast PA metabolism To better understand the potential regulatory and metabolic roles of PA, rerouting of lipid precursors to PA biosynthesis was carried out in 16:3 plants by targeting DAG Kinase (DAGK) to specific plastid compartments. In tobacco, introduction of a bacterial DAGK fused to the N- terminus of the small subunit of rubisco introduced DAGK activity to the chloroplast stroma-facing membranes, although the exact location was not determined. This resulted in accumulation of ER-derived PA, and subsequently ER-derived PG, in the chloroplast. These transgenic plants exhibited stunted growth, a substantial reduction in chloroplast lipids relative to ER lipids, and a smaller proportion of plastid-derived lipids within the chloroplast [53]. It remains puzzling as to why redirecting both plastid- and ER-derived DAG into PA synthesis at the stromal side of the chloroplast envelope would result in a disproportionate decrease of prokaryotic galactolipids. A similar study in Arabidopsis targeted DAGK to chloroplast membrane leaflets facing the stroma, intermembrane space, or cytosol [54]. Surprisingly, DAGK targeted to stroma-facing membrane 6 leaflets did not result in the phenotype witnessed in tobacco, and plant growth and membrane lipid composition was largely unaffected. Further analysis revealed that the majority of DAGK- derived PA in this case was being degraded by phospholipase A activity, preventing a significant increase in PA accumulation. Therefore, excess PA at the IEM inner leaflet is likely responsible for the phenotypes of tobacco lines in which DAGK is targeted to this membrane. In the same Arabidopsis study [54], it was also discovered that DAGK targeted to the intermembrane space of the chloroplast resulted in an increased rate of PA accumulation and stunted plant growth. Taken together, these results suggest that excess PA in the IEM has a negative impact on the development of 16:3 plants. Lipid phosphate phosphatases hypothesized to catalyze PAP activity in chloroplast lipid metabolism have been identified as LPPγ, LPPε1, and LPPε2 [55]. These were shown to associate specifically with chloroplasts and appeared to catalyze PA dephosphorylation when produced heterologously in yeast. However, null mutant analyses determined that lppε1, lppε2, and the lppε1 lppε2 double mutant did not have any aberrant phenotypes, while the lppγ null mutation was presumed lethal in the respective study [55]. In addition, LPPε1 activity at the chloroplast OEM compensates for the lppα2 null mutant, which lacks an ER PA phosphatase [56]. A deeper investigation into these chloroplast LPPs, their involvement in different aspects of PA metabolism, and implications for the potential roles of chloroplast PA are discussed in chapter 2. These include potential regulatory roles, as PA is known to be involved in various signaling pathways outside of the chloroplast [57]. Chloroplast lipids and jasmonic acid signaling The broad regulatory effects of lipid metabolism in plants are not limited to PA, and other chloroplast membrane lipids are known to be involved in hormone pathways. In particular, synthesis of the defense hormone jasmonic acid (JA) utilizes chloroplast membrane lipid substrates, which allows for regulation of JA biosynthesis through changes in plastid lipid metabolism [1]. For example, the ratio of MGDG to DGDG has been shown to affect induction of JA biosynthesis, as was first witnessed in the stunted growth phenotype of the dgd1 mutant [28, 7 58, 59]. While a mechanistic understanding of this phenomenon remains elusive, there are also lipase-mediated initiations of JA biosynthesis, some of which have been well-characterized. JA biosynthesis begins in the chloroplast, with the conversion of 18:3 fatty acids to 12-oxo- phytodienoic acid (OPDA) in the stroma. OPDA is then exported from the chloroplast, and converted to JA through β-oxidation in the peroxisome and reduction in the cytosol or peroxisome [60]. Initiation of JA biosynthesis by chloroplast phospholipases A1 has been demonstrated in overexpression lines of plastid lipase genes PLIP1, PLIP2, and PLIP3 [61, 62]. The PLIP enzymes hydrolyze the sn-1 linolenoyl moieties of chloroplast membrane lipids, and subsequent FA conversion to OPDA and JA results in plants with an elevated JA concentration and JA response phenotype [61, 62]. Under native regulation, PLIP2 and PLIP3 are involved in interaction between abscisic acid (ABA) and JA signaling, as PLIP2/3 gene expression and subsequent JA production is induced by ABA, and plip triple mutants are hypersensitive to ABA during germination [62]. Because JA production requires OPDA export from chloroplasts, the JA-induced phenotypes of PLIP-overexpressing lines also relies on transport of this FA derivative. PLIP-overexpressing lines therefore provide an opportunity for suppression screening targeted at chloroplast FA export components, as a mutant deficient in OPDA export would be relieved of its JA-induced growth inhibition. A suppressor screen in the PLIP3 overexpression background, and its results, are detailed in chapter 3. Summary of research aims Chapter 2 details the investigation of chloroplast lipid phosphate phosphatases LPPγ, LPPε1, and LPPε2, their activity, locations within the chloroplast, involvement in lipid metabolic pathways, and insights into potential roles of PA or DAG in plant development. The design and implementation of the PLIP3-OX suppressor screen are described in chapter 3, along with the subsequent mapping of suppressor mutations, genetic approaches for determining causal mutations, and the candidate genes themselves. One specific suppressor mutation in the gene KEEP ON GOING (KEG), and the deeper insight it may provide into JA-ABA interactions are also addressed in chapter 3. The broader implications of all these results, along with directions for future research, are discussed in chapter 4. 8 FIGURES Figure 1.1. Chloroplast lipid metabolism as a scaffold for metabolic responses to environmental stress. In black, constitutive lipid metabolism in unstressed plants; In purple, constitutive pathways that are upregulated in response to phosphate deprivation; in red, non-constitutive pathways that are turned on during phosphate deprivation; in blue, pathways activated by freezing or dehydration stress. List of abbreviations in alphabetical order: ATS1/2, GLYCER-OL-3- PHOSPHATE ACYLTRANSFERASE 1/2; CDS4/5, CYTIDINE DIPHOSPHATE DIACYL-GLYCEROL SYNTHASE 4/5; DAG, diacylglycerol; DGD1, UDP-GALACTOSE:MGDG GALAC-TOSYLTRANSFERASE; DGD2, DIGALACTOSYLDIACYLGLYCEROL SYNTHASE 2; DGDG, digalactosyldiacylglycerol; ER, endoplasmic reticulum; MGDs, monogalactosyldiacylglycerol synthases; MGDG, monogalactosyldiacylglycerol; NPC5, NON-SPECIFIC PHOSPHOLIPASE C5; PA, phosphatidic acid; PAH1 and PAH2, PHOSPHATIDIC ACID PHOSPHOHYDROLASE1 and 2; PAP, PHOSPHATIDIC ACID PHOSPHATASE; PC, phosphatidylcholine; PG, phosphatidyl-glycerol; PGP, phosphatidylglycerol phosphate; PGPP1, PHOSPHATIDYLGLYCEROPHOS-PHATE PHOSPHATASE1; PLAM, plastid associated microsomes; PLDζ1/2, PHOSPHOLIPASES D ZETA1/2. SFR2, SENSITIVE TO FREEZING2; SQD1, UDP-sulfoquinovose synthase; SQD2, SQDG synthase; SQDG, sulfoquinovosyldiacylglycerol; TGDG, trigalactosyldiacylglycerol; UDP-Gal, uridine diphosphate galactose; UDP-Glc, uridine diphosphate glucose; UDP-SQ, uridine di-phosphate-sulfoquinovose. 9 Figure 1.2. Roles of phosphatidic acid (PA) in chloroplast lipid metabolism. Proteins colored in green have specific interactions with PA, which may serve as a regulator, substrate, or both. The potential role of PA as the substrate for lipid import into the chloroplast is represented by a dotted arrow and a question mark, as this remains uncertain. List of abbreviations in alphabetical order: CPSFL1, CHLOROPLAST SEC14-LIKE1 protein; DAG, diacylglycerol; DGD1, UDP-galactose:MGDG galactosyltransferase; DGDG, digalactosyldiacylglycerol; ER, endoplasmic reticulum; MGD1, monogalactosyldiacylglycerol synthase; MGDG, monogalactosyldiacylglycerol; PA, phosphatidic acid; PAP, phosphatidic acid phosphatase; PLAM, plastid associated micro-somes; TGD complex, trigalactosyldiacylglycerol complex. The numbers refer to the TGD1-5 proteins forming the TGD complex. 10 REFERENCES 1. Cook, R., J. Lupette, and C. Benning, The role of chloroplast membrane lipid metabolism in plant environmental responses. Cells, 2021. 10(3): p. 706. 2. Mackender, R. and R.M. Leech, The Galactolipid, Phospholipid, and Fatty Acid Composition of the Chloroplast Envelope Membranes of Vicia faba. L. Plant Physiol., 1974. 53(3): p. 496-502. 3. Block, M.A., et al., Preparation and characterization of membrane fractions enriched in outer and inner envelope membranes from spinach chloroplasts. II. Biochemical characterization. J. Biol. Chem., 1983. 258(21): p. 13281-13286. 4. Ohlrogge, J.B., D.N. Kuhn, and P. Stumpf, Subcellular localization of acyl carrier protein in leaf protoplasts of Spinacia oleracea. Proc. Natl. Acad. Sci. U.S.A., 1979. 76(3): p. 1194- 1198. 5. Bertrams, M. and E. Heinz, Positional Specificity and Fatty Acid Selectivity of Purified sn- Glycerol 3-Phosphate Acyltransferases from Chloroplasts. Plant Physiol., 1981. 68(3): p. 653-657. 6. Kunst, L. and C. Somerville, Altered regulation of lipid biosynthesis in a mutant of Arabidopsis deficient in chloroplast glycerol-3-phosphate acyltransferase activity. Proc. Natl. Acad. Sci. U.S.A., 1988. 85(12): p. 4143-4147. 7. Yu, B., et al., Loss of plastidic lysophosphatidic acid acyltransferase causes embryo- lethality in Arabidopsis. Plant Cell Physiol., 2004. 45(5): p. 503-510. 8. Frentzen, M., et al., Specificities and Selectivities of Glycerol-3-Phosphate Acyltransferase and Monoacylglycerol-3-Phosphate Acyltransferase from Pea and Spinach Chloroplasts. Eur. J. Biochem., 1983. 129(3): p. 629-636. 9. Sato, N. and K. Awai, "Prokaryotic Pathway" Is Not Prokaryotic: Noncyanobacterial Origin of the Chloroplast Lipid Biosynthetic Pathway Revealed by Comprehensive Phylogenomic Analysis. Genome Biol. Evol., 2017. 9(11): p. 3162-3178. 10. Andrews, J. and K. Keegstra, Acyl-CoA synthetase is located in the outer membrane and acyl-CoA thioesterase in the inner membrane of pea chloroplast envelopes. Plant Physiol., 1983. 72(3): p. 735-740. 11 11. Löhden, I. and M. Frentzen, Role of plastidial acyl-acyl carrier protein: glycerol 3- phosphate acyltransferase and acyl-acyl carrier protein hydrolase in channelling the acyl flux through the prokaryotic and eukaryotic pathway. Planta, 1988. 176(4): p. 506-512. 12. Hares, W. and M. Frentzen, Properties of the microsomal acyl-CoA: sn-1-acyl-glycerol-3- phosphate acyltransferase from spinach (Spinacia oleracea L.) leaves. J. Plant Physiol., 1987. 131(1-2): p. 49-59. 13. Koo, A.J., J.B. Ohlrogge, and M. Pollard, On the export of fatty acids from the chloroplast. J. Biol. Chem., 2004. 279(16): p. 16101-10. 14. Li, N., et al., FAX1, a novel membrane protein mediating plastid fatty acid export. PLoS Biol., 2015. 13(2): p. e1002053. 15. Tian, Y., et al., FAX2 mediates fatty acid export from plastids in developing Arabidopsis seeds. Plant Cell Physiol., 2019. 60(10): p. 2231-2242. 16. Li, N., et al., Two plastid fatty acid exporters contribute to seed oil accumulation in Arabidopsis. Plant Physiol., 2020. 182(4): p. 1910-1919. 17. Bugaeva, W., et al., Plastid fatty acid export (FAX) proteins in Arabidopsis thaliana-the role of FAX1 and FAX3 in growth and development. bioRxiv, 2023. DOI: https://doi.org/10.1101/2023.02.09.527856 18. Mongrand, S., et al., The C16: 3\C18: 3 fatty acid balance in photosynthetic tissues from 468 plant species. Phytochemistry, 1998. 49(4): p. 1049-1064. 19. Block, M.A., et al., The phosphatidic acid phosphatase of the chloroplast envelope is located on the inner envelope membrane. FEBS Lett., 1983. 164(1): p. 111-115. 20. Jarvis, P., et al., Galactolipid deficiency and abnormal chloroplast development in the Arabidopsis MGD synthase 1 mutant. Proc. Natl. Acad. Sci. U.S.A., 2000. 97(14): p. 8175- 8179. 21. Miège, C., et al., Biochemical and topological properties of type A MGDG synthase, a spinach chloroplast envelope enzyme catalyzing the synthesis of both prokaryotic and eukaryotic MGDG. Eur. J. Biochem., 1999. 265(3): p. 990-1001. 22. Xu, C., et al., Mutation of the TGD1 chloroplast envelope protein affects phosphatidate metabolism in Arabidopsis. Plant Cell, 2005. 17(11): p. 3094-110. 12 23. Roston, R.L., et al., TGD1,-2, and-3 proteins involved in lipid trafficking form ATP-binding cassette (ABC) transporter with multiple substrate-binding proteins. J. Biol. Chem., 2012. 287(25): p. 21406-21415. 24. Cline, K. and K. Keegstra, Galactosyltransferases Involved in Galactolipid Biosynthesis Are Located in the Outer Membrane of Pea Chloroplast Envelopes. Plant Physiol., 1983. 71(2): p. 366-372. 25. Heinz, E. and P.G. Roughan, Similarities and differences in lipid metabolism of chloroplasts isolated from 18: 3 and 16: 3 plants. Plant Physiol., 1983. 72(2): p. 273-279. 26. Andrews, J., J.B. Ohlrogge, and K. Keegstra, Final Step of Phosphatidic Acid Synthesis in Pea Chloroplasts Occurs in the Inner Envelope Membrane. Plant Physiol., 1985. 78(3): p. 459-465. 27. Siebertz, M. and E. Heinz, Galactosylation of different monogalactosyldiacylglycerols by cell-free preparations from pea leaves. Hoppe Seylers Z. Physiol. Chem., 1977. 358(1): p. 27-34. 28. Dörmann, P., et al., Isolation and characterization of an Arabidopsis mutant deficient in the thylakoid lipid digalactosyl diacylglycerol. Plant Cell, 1995. 7(11): p. 1801-1810. 29. Dörmann, P., I. Balbo, and C. Benning, Arabidopsis galactolipid biosynthesis and lipid trafficking mediated by DGD1. Science, 1999. 284(5423): p. 2181-2184. 30. Froehlich, J.E., C. Benning, and P. Dörmann, The digalactosyldiacylglycerol (DGDG) synthase DGD1 is inserted into the outer envelope membrane of chloroplasts in a manner independent of the general import pathway and does not depend on direct interaction with monogalactosyldiacylglycerol synthase for DGDG biosynthesis. J. Biol. Chem., 2001. 276(34): p. 31806-31812. 31. Kelly, A.A., J.E. Froehlich, and P. Dörmann, Disruption of the two digalactosyldiacylglycerol synthase genes DGD1 and DGD2 in Arabidopsis reveals the existence of an additional enzyme of galactolipid synthesis. Plant Cell, 2003. 15(11): p. 2694-2706. 32. Kelly, A.A., et al., Synthesis and transfer of galactolipids in the chloroplast envelope membranes of Arabidopsis thaliana. Proc. Natl. Acad. Sci. U.S.A., 2016. 113(38): p. 10714-10719. 13 33. Andrews, J. and J.B. Mudd, Phosphatidylglycerol Synthesis in Pea Chloroplasts. Plant Physiol., 1985. 79(1): p. 259-265. 34. Xu, C., et al., The pgp1 mutant locus of Arabidopsis encodes a phosphatidylglycerolphosphate synthase with impaired activity. Plant Physiol., 2002. 129(2): p. 594-604. 35. Hagio, M., et al., Phosphatidylglycerol is essential for the development of thylakoid membranes in Arabidopsis thaliana. Plant Cell Physiol., 2002. 43(12): p. 1456-1464. 36. Babiychuk, E., et al., Arabidopsis phosphatidylglycerophosphate synthase 1 is essential for chloroplast differentiation, but is dispensable for mitochondrial function. Plant J., 2003. 33(5): p. 899-909. 37. Lin, Y.C., et al., Arabidopsis phosphatidylglycerophosphate phosphatase 1 involved in phosphatidylglycerol biosynthesis and photosynthetic function. Plant J., 2016. 88(6): p. 1022-1037. 38. Essigmann, B., et al., Phosphate availability affects the thylakoid lipid composition and the expression of SQD1, a gene required for sulfolipid biosynthesis in Arabidopsis thaliana. Proc. Natl. Acad. Sci. U.S.A., 1998. 95(4): p. 1950-1955. 39. Shimojima, M. and C. Benning, Native uridine 5′-diphosphate–sulfoquinovose synthase, SQD1, from spinach purifies as a 250-kDa complex. Arch. Biochem. Biophys., 2003. 413(1): p. 123-130. 40. Yu, B., C. Xu, and C. Benning, Arabidopsis disrupted in SQD2 encoding sulfolipid synthase is impaired in phosphate-limited growth. Proc. Natl. Acad. Sci. U.S.A., 2002. 99(8): p. 5732-5737. 41. Tietje, C. and E. Heinz, Uridine-diphospho-sulfoquinovose: diacylglycerol sulfoquinovosyltransferase activity is concentrated in the inner membrane of chloroplast envelopes. Planta, 1998. 206(1): p. 72-78. 42. Dubots, E., et al., Role of phosphatidic acid in plant galactolipid synthesis. Biochimie, 2012. 94(1): p. 86-93. 43. Dubots, E., et al., Activation of the chloroplast monogalactosyldiacylglycerol synthase MGD1 by phosphatidic acid and phosphatidylglycerol. J. Biol. Chem., 2010. 285(9): p. 6003-11. 14 44. Malherbe, A., et al., Feedback inhibition of phosphatidate phosphatase from spinach chloroplast envelope membranes by diacylglycerol. J. Biol. Chem., 1992. 267(33): p. 23546-23553. 45. Awai, K., et al., A phosphatidic acid-binding protein of the chloroplast inner envelope membrane involved in lipid trafficking. Proc. Natl. Acad. Sci. U.S.A., 2006. 103(28): p. 10817-10822. 46. Lu, B. and C. Benning, A 25-amino acid sequence of the Arabidopsis TGD2 protein is sufficient for specific binding of phosphatidic acid. J. Biol. Chem., 2009. 284(26): p. 17420-17427. 47. Wang, Z., C. Xu, and C. Benning, TGD4 involved in endoplasmic reticulum-to-chloroplast lipid trafficking is a phosphatidic acid binding protein. Plant J., 2012. 70(4): p. 614-623. 48. Xu, C., et al., Lipid trafficking between the endoplasmic reticulum and the plastid in Arabidopsis requires the extraplastidic TGD4 protein. Plant Cell, 2008. 20(8): p. 2190- 2204. 49. Wang, Z., N.S. Anderson, and C. Benning, The phosphatidic acid binding site of the Arabidopsis trigalactosyldiacylglycerol 4 (TGD4) protein required for lipid import into chloroplasts. J. Biol. Chem., 2013. 288(7): p. 4763-71. 50. Hertle, A.P., et al., A Sec14 domain protein is required for photoautotrophic growth and chloroplast vesicle formation in Arabidopsis thaliana. Proc. Natl. Acad. Sci. U.S.A., 2020. 117(16): p. 9101-9111. 51. Kim, E.H., et al., Chloroplast-localized PITP7 is essential for plant growth and photosynthetic function in Arabidopsis. Physiol. Plant., 2022. 174(4): p. e13760. 52. Schroda, M., Phosphoinositides regulate chloroplast processes. Proc. Natl. Acad. Sci. U.S.A., 2020. 117(17): p. 9154-9156. 53. Fritz, M., et al., Channeling of eukaryotic diacylglycerol into the biosynthesis of plastidial phosphatidylglycerol. J. Biol. Chem., 2007. 282(7): p. 4613-25. 54. Muthan, B., et al., Probing Arabidopsis chloroplast diacylglycerol pools by selectively targeting bacterial diacylglycerol kinase to suborganellar membranes. Plant Physiol., 2013. 163(1): p. 61-74. 15 55. Nakamura, Y., M. Tsuchiya, and H. Ohta, Plastidic phosphatidic acid phosphatases identified in a distinct subfamily of lipid phosphate phosphatases with prokaryotic origin. J. Biol. Chem., 2007. 282(39): p. 29013-21. 56. Nguyen, V.C. and Y. Nakamura, Distinctly localized lipid phosphate phosphatases mediate endoplasmic reticulum glycerolipid metabolism in Arabidopsis. Plant Cell, 2023. 35(5): p. 1548-1571. 57. Kolesnikov, Y., et al., Phosphatidic acid in plant hormonal signaling: from target proteins to membrane conformations. Int. J. Mol. Sci., 2022. 23(6): p. 3227. 58. Lin, Y.T., et al., Reduced Biosynthesis of Digalactosyldiacylglycerol, a Major Chloroplast Membrane Lipid, Leads to Oxylipin Overproduction and Phloem Cap Lignification in Arabidopsis. Plant Cell, 2016. 28(1): p. 219-32. 59. Yu, C.W., Y.T. Lin, and H.m. Li, Increased ratio of galactolipid MGDG: DGDG induces jasmonic acid overproduction and changes chloroplast shape. New Phytol., 2020. 228(4): p. 1327-1335. 60. Ruan, J., et al., Jasmonic acid signaling pathway in plants. Int. J. Mol. Sci., 2019. 20(10): p. 2479. 61. Wang, K., et al., A Plastid Phosphatidylglycerol Lipase Contributes to the Export of Acyl Groups from Plastids for Seed Oil Biosynthesis. Plant Cell, 2017. 29(7): p. 1678-1696. 62. Wang, K., et al., Two Abscisic Acid-Responsive Plastid Lipase Genes Involved in Jasmonic Acid Biosynthesis in Arabidopsis thaliana. Plant Cell, 2018. 30(5): p. 1006-1022. 16 CHAPTER 2: Metabolic and developmental roles of chloroplast phosphatidate phosphatases Chloroplast import experiments were conducted by John Froehlich. Photosynthetic experiments were conducted by Jeffrey Cruz in the David Kramer lab. EMS mutagenesis of seeds was performed by Linda Danhof. 17 Abstract Galactolipids comprise the majority of chloroplast membranes in plants, and their biosynthesis requires dephosphorylation of phosphatidic acid (PA) at the chloroplast envelope membranes. In Arabidopsis, the lipid phosphate phosphatases LPPγ, LPPε1, and LPPε2 have been previously implicated in chloroplast lipid assembly, with LPPγ being essential, as null mutants were reported to exhibit embryo lethality. Here, we show that lppγ mutants are in fact viable, and that LPPγ, LPPε1, and LPPε2 do not appear to have central roles in the plastid pathway of membrane lipid biosynthesis. Redundant LPPγ and LPPε1 activity at the outer envelope membrane is important for plant development, and the respective lppγ lppε1 double mutant exhibits reduced flux through the ER pathway of galactolipid synthesis. While LPPε2 is imported and associated with interior chloroplast membranes, its role remains elusive, and does not include basal nor phosphate limitation-induced biosynthesis of glycolipids. The specific physiological roles of LPPγ, LPPε1, and LPPε2 have yet to be uncovered, as does the identity of the PA phosphatase required for plastid MGDG biosynthesis. Introduction In plants, photosynthesis begins with the capture and photochemical conversion of light energy by densely-packed thylakoid membranes, and green tissues devote the majority of glycerolipid metabolism to generating and maintaining these chloroplast membranes. The potential for nutrient limitation imposed by a large phospholipid-based system is mitigated in plants, which have instead evolved photosynthetic membranes mostly composed of glycolipids. The most abundant lipid constituent of chloroplast membranes is monogalactosyldiacylglycerol (MGDG), followed by digalactosyldiacylglycerol (DGDG). Together, these galactolipids account for more than two-thirds of chloroplast lipids [1-3]. The remainder is mainly phosphatidylglycerol (PG) and sulfoquinovosyldiacylglycerol (SQDG), the two major anionic lipids in chloroplasts. Phosphatidylcholine (PC) is also found in chloroplasts, where it is confined to the outer envelope membrane (OEM) [1, 3]. In 16:3 plants, which include Arabidopsis, tobacco, and spinach [4], two separate pathways of lipid biosynthesis converge to make MGDG: an endoplasmic reticulum (ER) pathway and a plastid 18 pathway (also referred to as the “eukaryotic” and “prokaryotic” pathways, respectively). 18:3 plants, including monocots and legumes, rely entirely on the ER pathway for galactolipid biosynthesis. In the ER pathway, fatty acid (FA) export from the chloroplast is followed by activation to acyl-CoAs and subsequent acyl transfer to glycerol 3-phosphate by ER acyltransferases [5, 6]. Acylation of the sn-2 position is carried out in the ER by an acyltransferase with specific preference for 18-carbon substrates [7]. The phosphatidic acid (PA) product, or a PA derivative, is imported by the chloroplast to the inner envelope membrane (IEM) [8, 9], where the enzyme MGD1 galactosylates PA-derived diacylglycerol (DAG) to make MGDG [10-12]. In the plastid pathway of MGDG biosynthesis, acyl transfer from acyl-acyl carrier protein (acyl- ACP) to glycerol 3-phosphate takes place at the stroma-facing leaflet of the IEM [13, 14]. This is followed by dephosphorylation by a PA phosphatase (PAP), and galactosylation of DAG by MGD1 [10-12]. In contrast to the ER pathway, the chloroplast sn-2 acyltransferase is specific to 16-carbon rather than 18-carbon substrates, allowing for MGDG synthesized through this pathway to be distinguished from the ER-derived lipid [15, 16]. Previously identified candidates for PAPs involved in MGDG biosynthesis in Arabidopsis include the cytosolic lipins PAH1 and PAH2 [17], and the lipid phosphate phosphatases LPPγ, LPPε1, and LPPε2 [18]. PAH1 and PAH2 may play a role in the ER pathway, while LPPγ, LPPε1, and LPPε2 are chloroplast-located and have enzymatic properties matching the PAP activity in chloroplast membranes. LPPε1 is associated with the OEM, where its activity has redundancy with that of the ER-located LPPα2 [19]. However, mutant studies in Arabidopsis report no phenotypic abnormalities in the lppε1 and lppε2 null mutants, the lppε1 lppε2 double mutant, or reduced- function lines of LPPγ. Because a null mutant of LPPγ was deemed unattainable at the time, LPPγ was proposed to be essential for plant viability [18]. In this study, chloroplast lpp mutants were revisited for further characterization. It was hypothesized that LPPε1 and LPPε2 may function redundantly with other lipid metabolic enzymes, or may be involved in a response pathway to environmental challenges, thereby not showing aberrant mutant phenotypes at standard growth conditions. In addition, independent lppγ mutants were pursued for a more complete characterization of LPPγ function in chloroplasts. 19 Results LPPγ null mutants are viable, and their lipid profile is unaltered Three independent lppγ mutant alleles of Arabidopsis were confirmed as homozygous using PCR: lppγ-1 (SAIL_1255_H02), lppγ-2 (SALK_055510), and lppγ-3 (SALK_048788), all of which have insertions within the coding sequence (Fig. 2.1A-B). In our hands, and in contrast to a previous report [18], the three lppγ insertional mutant alleles were viable and did not have abnormalities in growth or morphology under standard conditions (Fig. 2.1C). Subsequent experiments were conducted using lppγ-1 (hereafter lppγ). The lipid profile of lppγ is comparable to the wild type, with the relative membrane lipid composition and their acyl compositions unaffected (Fig. 2.2). LPPγ and LPPε1 have redundant roles affecting plant growth The LPP family in Arabidopsis is subcategorized into LPPα, LPPβ, LPPγ, LPPδ, and LPPε, with a prior phylogenetic analysis showing that LPPγ and LPPε share a subclade, and microscopy and fractionation assays localizing LPPγ and LPPε to the chloroplast [18-20]. To account for potential functional redundancies among the different chloroplast LPP isoforms, single mutants were crossed to generate the double mutants lppγ lppε1, lppγ lppε2, and lppε1 lppε2, as well as the triple mutant lppγ lppε1 lppε2. As previously reported, lppε1 lppε2 did not exhibit differences in growth or morphology [18], and this was also observed here for lppγ lppε2. Meanwhile, lppγ lppε1 showed a reduction in both growth rate and size at maturity (Fig. 2.3A). This phenotype of lppγ lppε1 was replicated in the triple mutant lppγ lppε1 lppε2, with no additive effect of introducing lppε2. These phenotypes show that the activities of LPPγ and LPPε1 are at least partially redundant and required for proper development under standard growth conditions, while LPPε2 activity is separate from that of LPPγ and LPPε1. Complementation studies verified the redundancy of LPPγ and LPPε1, as either gene is sufficient to reverse the phenotype of lppγ lppε1 when expressed using either native (Fig. 2.3B) or CaMV 35S (Fig. 2.3C) promoters. LPPε2 overexpression in the lppγ lppε1 background does not complement the mutant phenotype, further implicating the role of LPPε2 as discrete from LPPγ and LPPε1 (Fig. 2.3C). 20 LPP localization within the chloroplast To characterize sub-chloroplast location of the three plastid LPP isoforms, we employed chloroplast import experiments with radioactive precursor proteins. For this purpose, LPPs labeled with 3H-Leu were synthesized in vitro using wheat germ lysate and a coding sequence template, and the translation products were incubated with intact pea chloroplasts. All three LPPs appeared in pellets after chloroplast fractionation, confirming these are membrane-associated proteins (Fig. 2.4). LPPε2 was efficiently cleaved and imported to a trypsin-protected membrane, which could be either thylakoid or IEM. Meanwhile, LPPε1 import was inefficient, and LPPγ was not processed, but membrane-associated and protease-sensitive. The redundancy of LPPγ and LPPε1 is therefore likely at the chloroplast OEM, with LPPε2 unable to compensate due to its confinement to the IEM or thylakoids. Unexpectedly, the translation product of LPPε1 ran at approximately 25 kDa on polyacrylamide electrophoresis gels, despite an expected molecular weight of ~30 kDa and a high sequence similarity and equivalent length to LPPε2, which did run at the expected size. To test whether chloroplast import separates LPPε2 from the redundant activity of LPPγ and LPPε1, a truncated LPPε2 lacking 59 N-terminal residues was introduced into lppγ lppε1. The wild- type growth phenotype was restored in these transgenic lines (Fig. 2.5), confirming that the redundant LPPγ and LPPε1 activity does not occur at the IEM nor thylakoids, and that the three LPPs have equivalent enzymatic activity. A rescue was also observed with a corresponding 51- residue N-terminal truncation of LPPε1, demonstrating that this region is not necessary for proper localization of LPPε1, and is possibly the missing component of the in vitro translation product. Mutants lacking LPPγ, LPPε1, and LPPε2 have largely unaltered membrane lipid profiles in leaves and unaffected lipid fluxes in isolated chloroplasts As previously reported, lppε1 lppε2 did not show aberrations in relative quantities of membrane lipids, nor in lipid acyl composition [18]. The same was observed for lppγ lppε2. Differences in lipid composition, including PA content, were also not observed in lppγ lppε1 (Fig. 2.6A). In addition, major lipids did not have altered acyl compositions, and the lipid profile of the lppγ lppε1 lppε2 triple mutant was comparable to that of lppγ lppε1 (Fig. 2.6B). 21 To test whether any of these LPPs have a significant role in the plastid pathway of galactolipid assembly, isolated chloroplasts from the lppγ lppε1 lppε2 triple mutant were incubated with 14C- acetate, and acyl flux through chloroplast lipid pools was determined based on pulse experiments. A deficiency in the PAP activity required for plastid MGDG biosynthesis would be observed as lower rates of PA conversion to MGDG [21]. Meanwhile, a decrease in flux caused by lower plastid PA production would be observed as higher relative PC accumulation, as plastid- synthesized FA transfer to PC continues to occur while PA and its derivatives accumulate more slowly [22]. Here, this is seen in isolated chloroplasts from the ats1-1 mutant, which is deficient in plastid PA biosynthesis (Fig. 2.6C) [14]. Surprisingly, chloroplasts from lppγ lppε1 lppε2 did not show slower conversion of PA to MGDG than wild-type chloroplasts, nor was there a relative decrease in labeling of PA, MGDG, or PG compared to PC (Fig. 2.6C). Therefore, the basal lipid biosynthetic pathways within the chloroplast do not appear to be dependent on the chloroplast LPPs. LPPγ and LPPε1 contribute to the ER pathway of galactolipid biosynthesis The role of chloroplast LPPs on the ER pathway contribution to galactolipid biosynthesis was tested by 14C-acetate pulse-chase analysis of polar lipids in whole leaves. During the chase, both lppγ lppε1 and lppγ lppε1 lppε2 exhibited slower conversion of PC to MGDG than wild type, with no additive effect by the LPPε2 deletion (Fig. 2.7). LPPγ and LPPε1 activity therefore facilitate MGDG biosynthesis from ER-derived phospholipids, while LPPε2 does not participate in either of the two galactolipid biosynthetic pathways. Crosses of lppγ lppε1 to tgd and rbl10 mutants have additive phenotypes, while crossing to ats1 results in severely reduced fitness In order to better contextualize the roles of LPPγ and LPPε1 in overall lipid metabolism, the double mutant was crossed to various characterized lipid mutants. Among these, tgd1-1 and tgd4-1 are deficient in lipid import from the ER pathway, rbl10-1 has decreased PAP activity in the plastid pathway, and ats1-1 is severely reduced in stromal PA biosynthesis [9, 14, 21, 23-25]. The triple mutants lppγ lppε1 tgd1-1, lppγ lppε1 tgd4-1, and lppγ lppε1 rbl10-1 did not exhibit recovery nor 22 exacerbation of the lppγ lppε1 growth defect, with lppγ lppε1 tgd1-1 leaves additionally showing the pale color of tgd1-1. The lipid profile of tgd1-1 is changed by genetic stacking of lppγ lppε1, with the triple mutant showing a small relative decrease in DGDG and a restoration of fully desaturated 16-carbon chains on MGDG, while the acyl profile of DGDG in lppγ lppε1 tgd1-1 remains identical to that of tgd1-1 (Fig. 2.8A). The lipid profile lppγ lppε1 rbl10-1 does not differ from that of rbl10-1, retaining the specific decrease in 16:3 acyl groups on MGDG (Fig. 2.8B). While lppγ ats1-1 plants were obtained and appear normal, the triple mutant was not successfully generated after crossing of lppγ ats1-1 to lppγ lppε1. 192 seeds were sown from the lppγ ats1-1 x lppγ lppε1 F2 generation, of which approximately 12 (6.25%) would be expected to be triple mutants. Instead, only 171 seeds germinated, and none were determined to be lppγ lppε1 ats1- 1 triple mutants. Genetic linkage would not account for this result, as ATS1 is on chromosome 1, LPPγ on chromosome 5, and LPPε1 on chromosome 3. The ungerminated seeds represent ~11% of the F2 segregating population, and may include triple mutants that have a severe or complete decline in viability. Such effects on fitness have been previously witnessed in crosses of ats1-1 to mutants disrupted in the ER pathway [9, 24, 26]. Light sensitivity is not the primary cause of growth inhibition in lppγ lppε1 In addition to its slow growth and stunted appearance, lppγ lppε1 also visibly accumulates anthocyanins in leaves under standard growth conditions. This is particularly noticeable at the phase separation step of lipid extraction (Fig. 2.9A). Anthocyanin accumulation is associated with a wide range of stresses, including a photoprotective role in excess light. Chlorophyll fluorescence measurements showed lower photosystem II efficiency (φII) in the double mutant, which was mostly accounted for by higher energy-dependent quenching (qE) (Fig. 2.9B). To test whether excess light leads to a stress-induced growth inhibition in lppγ lppε1, plants were grown at a reduced light intensity of 35 μmol photons m-2 s-1. These plants remained stunted, and their lipid profile was also unchanged, indicating that light sensitivity is not a major contributor to the growth phenotype of lppγ lppε1 (Fig. 2.9C). Likewise, increasing the ambient CO2 concentration to 1800 ppm did not alleviate the growth phenotype of the double mutant (Fig. 2.9D). 23 Salicylic acid signaling is not induced in lppγ lppε1 As discussed in chapters 1 and 3, chloroplast lipid mutants exist that are known to have hormone- driven changes to morphology. Among these, the dgd1 mutant and PLIP overexpression lines have severe growth defects resulting from the constitutive production of jasmonic acid (JA) [27- 30]. The phenotype lppγ lppε1 does not resemble that of these constitutive JA mutants, as leaf and petiole elongation in lppγ lppε1 is not disproportionately stunted, anthocyanins are not distributed in vascular tissues in the absence of light stress, and mutants have shown susceptibility to fungus gnats in the growth chambers. While these phenotypes differ from those of constitutive JA lines, we considered they may resemble previous descriptions of the constitutive salicylic acid (SA) mutant cpr1-1, among others [31, 32]. Stunted growth in these mutants is known to be suppressed at elevated temperature [32], so the morphology of lppγ lppε1 was compared to that of cpr1-1 at 22°C and 28°C. In our comparison, lppγ lppε1 did not have a strong resemblance to cpr1-1 at standard temperature (Fig. 2.10A), and growth was not rescued at elevated temperature (Fig. 2.10B). Furthermore, constitutive SA was tested in lppγ lppε1 by probing for the response factor pathogenesis response 1 (PR1) using Western blotting, and PR1 accumulation was not observed in lppγ lppε1 in the absence of applied stresses (Fig. 2.10C). The SA biosynthesis mutant sid2-2 was also included as a negative control [33]. A screen in the lppγ lppε1 background has yielded preliminary suppressor mutants In order to identify genes potentially responsible for the stunted growth of lppγ lppε1, the double mutant was mutagenized with EMS, and M2 plants subjected to a visual suppressor screen. Two mutants out of ~800 screened on soil, designated sup3 and sup25, had strong suppression phenotypes and were chosen for backcrossing. Additional suppressors sup30 and sup32 were selected among ~4500 that were initially screened on agar plates, of which 24 had been transferred to soil for a secondary visual screen. These suppressor mutants are shown in Fig. 2.11. After mutant backcrossing to lppγ lppε1, the F2 segregating populations will be sequenced and causative mutations identified using the strategies applied in chapter 3. 24 Chloroplast-imported LPPs are not required for lipid changes under phosphate deprivation, excess light, or low temperature While the phospholipid content in chloroplasts is already relatively low in nutrient-replete conditions, it is further decreased when SQDG is substituted for PG under phosphate deprivation [34]. Because lppε2 has a wild-type phenotype under standard conditions, it is possible that LPPε2 instead plays a role in responding to environmental changes. Its import into the chloroplast and its PA phosphatase activity would suggest it may be important for the increase in SQDG under phosphate deprivation. Because the import assay also showed limited LPPε1 import, the single mutants lppε1 and lppε2 were examined under phosphate deprivation, along with the lppε1 lppε2 double mutant. While root growth was slightly slower in the double mutant, this difference was not exacerbated on media with a 95% reduction in phosphate content (Fig. 2.12A). In addition, lppε1 lppε2 did not appear to be impaired in its ability to accumulate SQDG in place of PG (Fig. 2.12B). Other conditions in which a chloroplast PAP may be applied include transitions to excess light or decreased temperature, as these changes may require phosphatase-dependent lipid turnover or remodeling. To test this, three-week old plants grown at standard conditions (22°C, 120 µmol m- 2 s-1 photons) were transferred to either low-temperature (10°C) or high-light (270 µmol m-2 s-1 photons) chambers. After one week at these conditions, no visible difference was seen between lppε mutants and wild-type plants (Fig. 2.13A). Likewise, no differential effects on lipid profiles were observed between Col-0 and lppε1 lppε2 at reduced temperature or increased light (Fig. 2.13B). Discussion LPPγ, LPPε1, and LPPε2 do not have major roles in the plastid pathway of galactolipid biosynthesis PA biosynthesis and conversion to MGDG in isolated lppγ lppε1 lppε2 chloroplasts is equivalent to that of wild-type chloroplasts. In contrast, a mutant lacking the chloroplast rhomboid-like protein RBL10 has been shown to be deficient in the conversion of chloroplast PA to MGDG, indicating that PA dephosphorylation specific to the plastid pathway partially depends on functional RBL10 [21]. Because the conversion of plastid PA to MGDG is decreased in rbl10, but 25 not in lppγ lppε1 lppε2, these LPPs are not the RBL10-dependent PAPs involved in plastid galactolipid biosynthesis. Therefore, LPPγ, LPPε1, or LPPε2 serve other metabolic or physiological roles, while the identity of the plastid pathway PAP is still unknown (Fig. 2.14). Because the rbl10- 1 mutant phenotype does resemble that of an expected deficiency in plastid pathway PAP activity, additional investigation into the role of RBL10 in the plastid may reveal its identity. LPPγ, LPPε1, and LPPε2 have the same catalytic activity, with LPPγ and LPPε1 located at the chloroplast OEM, and LPPε2 within the IEM or thylakoids. With chloroplast import assays showing an efficient import of LPPε2 into the trypsin-protected internal compartments of chloroplasts, and little to no equivalent import of LPPγ nor LPPε1, it is likely that LPPε2 is the only characterized PA phosphatase exclusively located at internal chloroplast membranes (Fig. 2.14). LPPε1 may be dual-localized to both envelope membranes, as import assays have shown inefficient chloroplast import to trypsin-protected compartments. Published protease assays on chloroplasts from epitope-tagged LPPε1 lines in Arabidopsis have shown OEM association [19]. LPPγ is likely exclusively at the OEM, as it has redundant activity with LPPε1, no apparent processing or import, and has been shown to be chloroplast-associated [18]. The complementation of the lppγ lppε1 phenotype by an N-terminal truncation of LPPε2 confirms the equivalent enzymatic activity, previously shown to be PAP activity [18], among the three enzymes. LPPε2 may associate with either thylakoids or the IEM. Because the bulk of chloroplast membrane lipid biosynthesis takes place in the envelope membranes, the location of LPPε2 in thylakoids would explain why lppε2 mutants are unaffected in lipid phenotypes. In such case, LPPε2 may be involved in turnover of thylakoid lipids, although it is likely not involved in mild transitions to low temperature or high light. On the other hand, if LPPε2 is IEM-associated, it would not be the primary PAP in the plastid pathway of galactolipid or sulfolipid biosynthesis, as mutants can accumulate these glycolipids without hinderance under standard and phosphate-limited growth conditions. 26 LPPγ and LPPε1 contribute to the ER pathway of galactolipid biosynthesis With LPPγ and LPPε1 located in the OEM, we would expect that if they are involved in one of the galactolipid biosynthetic pathways, it would be the ER pathway (Fig. 2.14). Indeed, based on the radiolabeling results in whole leaves and chloroplasts of lppγ lppε1, the double mutant appears to be specifically impaired in the ER pathway of MGDG biosynthesis. In addition, the failure to obtain homozygous lppγ lppε1 ats1-1 triple mutants, with ats1-1 being deficient in the plastid pathway, is also reported for known ER pathway mutants tgd1-1, tgd4-1, and tgd5-1 [9, 24, 26]. This is in contrast to the viability of lppγ lppε1 tgd1-1 and lppγ lppε1 tgd4-1 triple mutants, which remain deficient in just the ER pathway, and in line with a disruption of both galactolipid biosynthetic pathways in lppγ lppε1 ats1-1. Activity of the desaturase FAD5 may be altered in the tgd1-1 mutant The differences between tgd1-1 and lppγ lppε1 tgd1-1 in DGDG content and MGDG acyl composition may both derive from a decrease in FAD5 activity. In the tgd1-1 mutant, the majority of acyl chains on the sn-2 position of DGDG are 16:0, as opposed to 18:3 in the wild type [23]. Likewise, the sn-2 position of MGDG in tgd1-1 is not only depleted of 18-carbon chains, but also enriched in saturated 16-carbon moieties, which is seen affecting the total acyl composition of tgd1-1 in Fig. 2.8A. This would suggest that DGD1 cannot utilize plastid-derived MGDG as a substrate if the sn-2 position is desaturated, and in tgd1-1 this desaturation is downregulated to provide compensatory substrate for DGD1. In lppγ lppε1 tgd1-1, desaturation of MGDG sn-2 chains is restored and 16:3 levels are recovered, which deprives DGD1 of substrate and results in decreased DGDG content. The desaturase FAD5 is responsible for the initial desaturation of sn-2 chains on chloroplast MGDG [35, 36]. FAD5 expression or activity may therefore be downregulated in tgd1-1, and repression of FAD5 is reversed in lppγ lppε1 tgd1-1. PA at the intermembrane-facing leaflet of the OEM may negatively regulate plant growth in lppγ lppε1 Despite evidence for the involvement of LPPγ and LPPε1 in the ER pathway of galactolipid biosynthesis, the acyl profile of MGDG in lppγ lppε1 strongly resembles that of the wild type. This would suggest that Arabidopsis is somehow able to compensate for the decreased lipid flux from 27 the ER in the double mutant. Meanwhile, the tgd1-1 mutant shows a similar pattern of lipid fluxes in whole leaves labeled using 14C-acetate, but is also affected in the acyl profile of MGDG and DGDG, and accumulates high levels of PA [9, 23]. In addition, unlike lppγ lppε1, it has pale leaves, likely due to an overall decrease in thylakoid membrane content. Based on these data, it appears that the ER pathway is more severely affected in tgd1-1 than in lppγ lppε1. However, in tgd1-1, as well as similar mutants affecting import complex components TGD2-5, plant growth is not inhibited as it is in lppγ lppε1 [24, 26, 37, 38]. The growth deficiency in lppγ lppε1 is therefore likely distinct from the broad effects on lipid flux from the ER to the chloroplast, and may rather result specifically from PA accumulation in a particular membrane compartment. In a previous Arabidopsis study, a bacterial DAG kinase was targeted to stromal-facing, intermembrane-facing, or cytosolic-facing leaflets of the chloroplast membranes. Among these, only the lines in which DAG kinase was introduced to the intermembrane space appeared stunted in growth [39]. As in lppγ lppε1, the membrane lipid composition appeared unaffected despite the growth inhibition. Given these results, and because LPPγ is exclusive to the OEM, it can be reasoned that the phenotype of lppγ lppε1 is a specific result of PA accumulation at the intermembrane-facing leaflet of the OEM (Fig. 2.14). This PA pool would be negligible relative to total PA content, resulting in the apparent lack of difference in PA during quantification. There is currently no insight into the means by which PA at the inner leaflet of the OEM affects plant development in lppγ lppε1. It appears as though the hormone pathways of JA and SA are not involved, based on morphological, physiological, and biochemical observations. The retention of growth inhibition at low light and high CO2 suggests that it is not a direct result of constraints or inhibition of photosynthesis. The severity of growth inhibition in lppγ lppε1 may itself be the reason for the double mutant’s light sensitivity, with quenching mechanisms activated to prevent photosynthetic sink limitations from leading to oxidative stress. In this absence of obvious leads, the mutant screen for suppressors of lppγ lppε1 will provide opportunities for elucidating the roles of PA and chloroplast LPPs in the regulation of plant development. 28 LPPε2 has an unknown role in the interior chloroplast membranes The role of LPPε2 is yet more elusive, with no discernable phenotypic differences in lppε1 or lppε2 mutants at standard, phosphate-limited, high-light, or low-temperature conditions. The enzyme may therefore be involved in responses to environmental conditions that have not yet been tested, with possible roles in lipid signaling, remodeling, or degradation. LPPε2 may have redundancies with LPPε1, depending on their specific locations at the interior membranes of the chloroplast. Therefore, future studies introducing various environmental challenges would be appropriate using the lppε1 lppε2 double mutant. LPPs and PA in the chloroplast Our continued investigation of chloroplast LPPs has confirmed that LPPγ and LPPε1 are involved in basal lipid metabolism, but as part of the ER-derived galactolipid biosynthesis pathway and not the plastid pathway. The stunted growth of lppγ lppε1 implicates PA as a potential regulatory molecule at the site of LPPγ and LPPε1 activity, and consequently these enzymes’ function may not be limited to just metabolism. LPPε2 is imported into the inner envelope or thylakoids, but like the other LPPs, does not significantly contribute to plastid-derived galactolipid biosynthesis. Its role has yet to be discovered, as does the identity of the plastid pathway PA phosphatase. Methods Plants Insertional mutant lines of Arabidopsis thaliana were ordered from the Arabidopsis Biological Resource Center (ABRC). The T-DNA lines with insertions in LPPγ were SAIL_1255_H02, SALK_055510, and SALK_048788, corresponding to lppγ-1, lppγ-2, and lppγ-3 [40-42]. The lppε1 and lppε2 mutants refer to T-DNA lines SALK_000157 and SALK_055964, respectively. cpr1-1 seeds were obtained from the Sarah Lebeis lab, and sid2-2 seeds from the Federica Brandizzi lab. ats1-1, tgd1-1, tgd4-1, and rbl10-1 mutant seeds were available in the Benning lab stock. Double and triple mutants were generated by crossing, which entailed removal of petals and stamens from unopened flowers, cross-pollination of the emasculated flowers by the male, and protection of pollinated pistils using plastic cling film for ~1 week, until silique formation [43]. 29 EMS mutagenesis Approximately 13,000 PLIP3-OX seeds were incubated in 0.1% Tween®20 (Sigma-Aldrich) for 15 minutes in a tube rotator, after which seeds were allowed to settle, and the solution was removed. 0.2% Ethyl methanesulfonate (EMS) in water was added, and seeds were incubated overnight (~16 hrs) in tube rotator. Seeds were washed 7 times with water, incubated in water for 2 hours, and washed with water one more time. Growth conditions Unless otherwise stated, plants were grown in SUREMIX™ Professional All-Purpose Perlite Mix (Michigan Grower Products, Inc.) at a temperature of 22°C, a long-day 16/8-hr light/dark cycle, and a light intensity of approximately 120 µmol m-2 s-1. Plants for phosphate response assays were grown on vertical plates on medium containing half-strength Murashige and Skoog (MS), 1% sucrose, 1% Phytoblend™ agar (Caisson labs), and buffered with 2.5 mM MES at pH 5.7 [44]. For low-phosphate plates, MS was replaced with a mixture of 5% MS and 95% phosphate-free MS. Plants used for chloroplast isolation were grown on horizontal plates containing full-strength MS, 1% sucrose, and 0.8% Phytoblend™ agar, buffered with 2.5 mM MES at pH 5.7. Plate-grown plants were all incubated at 22°C, a long-day 16/8-hr light/dark cycle, and a light intensity of approximately 120 µmol m-2 s-1. Constructs and transformation For overexpression lines, coding sequences were amplified using cDNA templates synthesized from Arabidopsis leaf mRNAs. These were cloned into pENTR using the pENTR™ /D-TOPO® kit (ThermoFisher), and recombined to pH35GY expression vectors [45] using Gateway™ LR Clonase™ II (ThermoFisher). Stop codons were excluded to generate C-terminal YFP fusions. Constructs for N-terminal truncations were made using the Q5® Site-Directed Mutagenesis Kit (New England Biolabs). For constructs using native promoters, target genes were amplified together with ~2 kb upstream regions from Arabidopsis gDNA using primers that include AscI and SalI restriction sites. These were then integrated into pCAMBIA1390 vectors using restriction digests and ligation. Primer sequences are listed and defined in Table 2.1. Vectors were used to transform Agrobacterium tumefaciens strain GV3101, and plants were transformed via vacuum- 30 mediated floral dip [46]. For the in vitro translation step of the chloroplast import experiment, LPP sequences from pENTR were recombined with pDEST14 expression vectors. Lipid profiling Tissue was harvested from 4-week-old soil-grown plants, and lipids were extracted, separated, and analyzed as described [47]. In brief, lipids were extracted from leaves in 20:10:1 methanol:chloroform:formic acid solution, after which a half-volume of 0.2 M phosphoric acid, 1 M KCl solution was added and vortexed. After phase separation, the bottom organic phase was transferred to a new tube, dried with pure nitrogen, resuspended in chloroform, and loaded onto an ammonium sulfate-impregnated thin layer chromatography (TLC) plate. TLC was run using a mobile phase of 91:30:7.5 acetone:toluene:water. PA was separated using 2-dimensional TLC on unimpregnated plates, with a mobile phase of 65:25:4 chloroform:methanol:water for the first dimension, and 50:20:10:10:5 chloroform:acetone:methanol:acetic acid:water for the second [9]. Lipids were briefly stained with iodine for identification, then scraped into glass tubes and derivatized to fatty acid methyl esters (FAMEs) by adding 1 M methanolic HCl for a 25-minute incubation at 80°C. Equal volumes of 0.9% aqueous sodium chloride and hexane were then added, and phases were separated after vortexing. The FAME-containing hexane phase was transferred to a new tube, dried under N2 gas, and resuspended with hexane. FAMEs were identified and quantified using gas chromatography-flame ionization detection. Chloroplast isolation from Arabidopsis Intact chloroplasts were isolated as described in [48]. In brief, two-week old plants grown on horizontal plates were harvested in the morning using a razor blade, and suspended in ice-cold chloroplast buffer (330 mM sorbitol, 50 mM HEPES, 1 mM MgCl2, 1M MnCl2, 2 mM EDTA, 0.1% BSA, KOH to pH 7.3). In the buffer, leaves were promptly cut with scissors and then homogenized for 10-30 s using a T25 digital ULTRA-TURRAX® homogenizer (IKA) with a 15-mm probe at 8,000 rpm. Homogenate was filtered through two layers of Miracloth (Millipore), pelleted at 700 x g for 5 min, and resuspended in buffer, which was loaded onto tubes containing a step gradient. The step gradient consisted of 85% Percoll® (Sigma) at the bottom, upon which 40% Percoll® was gently loaded (Percoll® solutions were prepared with the same solute composition as the 31 chloroplast buffer). After centrifugation at 2000 x g for 10 min with no brake, intact chloroplasts were transferred from the interphase to a new tube, washed once with buffer, pelleted, and resuspended in a small volume of buffer. To determine µg/mL chlorophyll, absorbance at 652 nm was measured in 80% acetone, and multiplied by an extinction coefficient of 28 [49]. 14C-acetate labeling 14C-acetate labeling was carried out for intact chloroplasts at 100 µg/mL chlorophyll in chloroplast buffer (330 mM sorbitol, 50 mM HEPES, 1 mM MgCl2, 1M MnCl1, 2 mM EDTA, 0.1% BSA, KOH to pH 7.3) with 0.6 mM UDP-galactose. 10 µCi/mL 14C-acetate (American Radiolabeled Chemicals, Inc., Catalog No. ARC 0173) was added in the dark and on ice. Samples were transferred to a 24- well plate on a shaker, at room temperature and with an LED light source of ~100 µmol m-2 s-1. 14C-acetate labeling of leaves was carried out in 10 cm petri plates, with leaves floating on 25 mM MES pH 5.7 buffer. The 1-hr pulse and first wash included 0.001% Tween®20, and the pulse included 1 µCi/mL 14C-acetate. After the pulse, leaves were washed once with the same buffer, containing no radioactivity. Equivalent MES buffer excluding the Tween was used for the second and third washes, and as the chase incubation buffer. Leaf samples also used an LED light source of ~100 µmol m-2 s-1. For both leaves and isolated chloroplasts, samples were harvested at the designated timepoints, and polar lipids were extracted and separated using TLC as previously described. Plates were imaged using phosphor screens, and radioactivity was measured for scraped lipids by addition of 4a20™ counting cocktail (Research Products International) and quantification using a liquid scintillation counter. Photosynthesis measurements Photosynthetic measurements were obtained from 4-week-old plants grown in soil at a temperature of 22°C, a long-day 16/8-hr light/dark cycle, and a light intensity of approximately 120 µmol m-2 s-1. Chlorophyll fluorescence images were recorded using dynamic environmental phenotype imager (DEPI) chambers as previously described [50], but with a series of constant actinic intensities. Briefly, plants were dark-adapted for 30 min, and relative yields of chlorophyll fluorescence were estimated for the fully dark-adapted state (F0) and during an intense flash (~0.3 s, 10,000 μmol m-2 s-1) to saturate photosystem II (PSII) photochemistry. A series of actinic 32 light intensities (50, 100, 200, 300, 400, 500, 600 μmol photons m-2 s-1) were then applied in sequence. After five minutes of illumination at each intensity, relative chlorophyll fluorescence yields were in the steady state (Fs) and during saturation flashes (Fm’). Photosynthetic parameters were calculated for photosystem II efficiency (φII) and energy-dependent quenching (qE) as previously described [50, 51]. Protein extraction, SDS-PAGE, and immunoblotting ~50-100 mg of leaf tissue was harvested into 2 mL tubes containing 5-10 zirconia/silica beads (2.3 mm, BioSpec Products, Cat. No. 11079125z) and frozen in liquid nitrogen. Cells were broken using bead-beater at 30 Hz for 3 min. 150 µL loading buffer (4% sodium dodecyl sulfate (SDS), 20% glycerol, 10% β-mercaptoethanol, 0.125 M Tris-HCl, pH 6.8) containing plant protease inhibitor cocktail (Sigma-Aldrich Cat. No. P9599) was added to the sample. The sample was incubated at room temperature for 15 minutes with occasional vortexing, then centrifuged at 15,000 x g for 2 minutes. Supernatant was transferred to new tube, allowed to incubate at room temperature for 15 minutes, then loaded onto Bio-Rad 4-20% Mini-PROTEAN® TGX™ precast polyacrylamide gels (Cat. No. 4561094). Gels were run at 150 V for 45-60 minutes, using Tris-glycine-SDS running buffer (25 mM Tris, 192 mM glycine, 0.1% SDS), and transferred to a PVDF membrane at 100 V for 80 minutes using Tris-glycine transfer buffer (25 mM Tris, 192 mM glycine, 20% methanol), with chilling and stirring. Membranes were washed in PBS-T (137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM KH2PO4, 0.04% Tween®20), and blocked using PBS-T with 5% nonfat dry milk. Membranes were incubated with antibodies for 1 hr, then washed 4 times for 5 minutes with PBS- T. The primary antibody used was the rabbit PR-1 antibody from Agrisera (Cat. No. AS10 687), and the secondary antibody was the HRP-conjugated donkey anti-rabbit IgG, from Agrisera (Cat No. AS10 845). An undergraduate student expressed concern for the fate of the donkey. Chemiluminescence was detected using Bio-Rad Clarity Western ECL Substrate (Cat. No. 1705061). Chloroplast import assays Chloroplast import assays were carried out as previously described in [52]. In summary, chloroplasts were extracted from 8- to 12-day-old pea seedlings, isolated by centrifugation using 33 a 40% Percoll cushion, and resuspended in import buffer (IB; 330 mM sorbitol, 50 mM HEPES- KOH, pH 8.0) at 1 mg/mL chlorophyll. Separately, pDEST14 vectors with LPP or control sequences were translated using the Promega TNT® Coupled Wheat-germ Lysate System, with the addition of 0.05 mCi 3H-leucine in each 50 µL reaction. After translation, the radiolabeled reaction product was diluted with an equal volume of 2x IB containing 50 mM unlabeled L-leucine. Equal volumes of 1 mg/mL chloroplasts in IB and 12 mM Mg-ATP in IB were added to the translation product (final Mg-ATP concentration of 4 mM), and the mixture was incubated at room temperature and ambient room light (~10 µmol m-2 s-1) for 30 min. The sample was then divided in half, one of which was incubated with trypsin (final conc. 0.1 mg/mL) for 20 minutes on ice, and then quenched with trypsin inhibitor (final conc. 0.2 mg/mL). The other half was the negative control. After protease treatment, chloroplasts were recovered by centrifugation using a 40% percoll cushion, resuspended in lysis buffer (25 mM HEPES-KOH, 4.0 mM MgCl2, pH 8.0), and centrifuged to fractionate into total soluble (S) and total membranes (P). Fractions were analyzed using SDS- PAGE, followed by fluorography and exposure to X-ray film. 34 FIGURES AND TABLES Figure 2.1. Insertional sites and genotyping primers (A), PCR genotyping confirmations (B), and growth phenotypes (C) of the three independent lppγ mutants. Pictured plants are 4 weeks old. L, left primer paired with right primer; B, insertion border primer paired with right primer; for lppγ-1: L, 1+2; B, 11+2; for lppγ-2: L, 3+4; B, 12+4; for lppγ-3: L, 5+6; B, 12+6; see Table 2.1 for primer descriptions and sequences. 35 Figure 2.2. Relative lipid content and acyl composition of major membrane lipids in lppγ. MGDG, monogalactosyldiacylglycerol; DGDG, digalactosyldiacylglycerol; PG, phosphatidylglycerol; SQDG, sulfoquinovosyldiacylglycerol; PE, phosphatidylethanolamine; PI, phosphatidylinositol; PC, phosphatidylcholine. Three biological replicates; bars indicate standard deviation. 36 Figure 2.3. (A) Morphology of lpp double mutants and triple mutant. Complementation of lppγ lppε1 by LPP genes driven by native (B) or overexpression (C) promoters. 5-week-old plants, grown in a 12/12 hr light/dark cycle. 37 Figure 2.4. Import of 3H-labeled in vitro translation products in isolated pea chloroplasts. Arc6 and FtsH12 are IEM-localized controls, with Arc6 containing a domain that extends into the intermembrane space, which upon digestion results in a smaller protein indicated by the asterisk. TP, translation product; P, pellet from fractionated chloroplasts; S, supernatant from fractionated chloroplasts; pr, protein prior to cleavage of transit peptide; m, mature protein; IE, inner envelope-localized. This experiment was performed by John Froehlich. 38 Figure 2.5. Rescue of lppγ lppε1 phenotype with N-terminal truncations of LPPε1 and LPPε2. Genes are expressed under 35S CaMV promoter. 6-week-old plants, grown in a 12/12 hr light/dark cycle. 39 Figure 2.6. (A) Lipid compositions and (B) lipid acyl compositions of lppγ lppε1 and lppγ lppε1 lppε2. (C) Relative incorporation of radioactivity into membrane lipids in isolated chloroplasts fed with 14C-acetate. MGDG, monogalactosyldiacylglycerol; DGDG, digalactosyldiacylglycerol; PG, phosphatidylglycerol; SQDG, sulfoquinovosyldiacylglycerol; PE, phosphatidylethanolamine; PI, phosphatidylinositol; PC, phosphatidylcholine. Three biological replicates; bars indicate standard deviation. 40 Figure 2.7. Relative incorporation of radioactivity to membrane lipids in leaves after feeding with 14C-acetate. x-axis represents time after removal of radioactive acetate. MGDG, monogalactosyldiacylglycerol; PG, phosphatidylglycerol; DGDG, digalactosyldiacylglycerol; SQDG, sulfoquinovosyldiacylglycerol; PI, phosphatidylinositol; PE, phosphatidylethanolamine; PC, phosphatidylcholine. Three biological replicates; bars indicate standard deviation. 41 Figure 2.8. Relative lipid content and galactolipid acyl compositions of (A) lppγ lppε1 tgd1-1 and (B) lppγ lppε1 rbl10-1. MGDG, monogalactosyldiacylglycerol; PG, phosphatidylglycerol; DGDG, digalactosyldiacylglycerol; SQDG, sulfoquinovosyldiacylglycerol; PI, phosphatidylinositol; PE, phosphatidylethanolamine; PC, phosphatidylcholine. Three biological replicates; bars indicate standard deviation. 42 Figure 2.9. (A) Anthocyanin in aqueous phase in extract from Col-0 vs lppγ lppε1. (B) Photosystem II efficiency (φII) and energy-dependent quenching (qE) in Col-0 vs lppγ lppε1. (C) Growth of Col- 0 vs lppγ lppε1 under low light. (D) Growth of Col-0 vs lppγ lppε1 at elevated CO2. Three biological replicates; bars indicate standard deviation. 43 Figure 2.10. Growth of lppγ lppε1 compared with constitutive SA mutant cpr1-1 or SA-deficient mutant sid2-2 at (A) 22°C and (B) 28°C. Plants are approximately 4 weeks old. (C) Probing of the SA response factor PR1 in lppγ lppε1 compared with wild-type and SA mutant controls. 44 Figure 2.11. Suppressor mutants of lppγ lppε1. Plants in the upper panel are approximately 4 weeks old, plants in lower panel are 5-8 weeks old. 45 Figure 2.12. Effect of low phosphate on growth (A) and lipid profile (B) of lppε mutants. HP, 1x phosphate; LP, 0.05x phosphate; MGDG, monogalactosyldiacylglycerol; DGDG, digalactosyldiacylglycerol; PG, phosphatidylglycerol; SQDG, sulfoquinovosyldiacylglycerol; PE, phosphatidylethanolamine; PI, phosphatidylinositol; PC, phosphatidylcholine. Three biological replicates; bars indicate standard deviation. 46 Figure 2.13. (A) Size and morphology of lppε mutants after one week at high light or low temperature. (B) Lipid composition and acyl compositions of major chloroplast lipids in lppε mutants at high light or low temperature. Std, standard light and temperature (22°C, 120 µmol m-2 s-1 photons); HL, high light (270 µmol m-2 s-1 photons); cold, 10°C. Three biological replicates; bars indicate standard deviation. 47 Figure 2.14. Model illustrating the localizations and activities of chloroplast LPPs, denoted by their Greek letters. Dotted lines represent uncertain localization. The unknown RBL10-dependent PA phosphatase is represented by a white box and question mark. Bolded “PA” in purple represents PA at the inner leaflet of the outer envelope, which may have an inhibitory effect on plant growth. ACP, acyl carrier protein; CoA, coenzyme A; DAG, diacylglycerol; ER, endoplasmic reticulum; FAS, fatty acid synthesis; IEM, chloroplast inner envelope membrane; MGDG, monogalactosyldiacylglycerol; OEM, chloroplast outer envelope membrane; PA, phosphatidic acid; PC, phosphatidylcholine; Thy, thylakoid membranes. 48 No. Primer Sequence Description 1 CCATTGAAGAAGCTTGAGCAC lppγ-1 genotyping left primer 2 ACCAACTCGCACCAACAATAC lppγ-1 genotyping right primer 3 ATGGAATCTCCCATCTCCTTG lppγ-2 genotyping left primer 4 CACTTTTCCGTCACTTTCTCG lppγ-2 genotyping right primer 5 CCATTGAAGAAGCTTGAGCAC lppγ-3 genotyping left primer 6 ACCAACTCGCACCAACAATAC lppγ-3 genotyping right primer 7 TCTGTTGATACCAGAGGTGGC lppε1 genotyping left primer 8 GGATCGATTTCGAATTCTGCT lppε1 genotyping right primer 9 CAAGAGAGTTCCAAGTTACA lppε2 genotyping left primer 10 GAATCGTTTGATTTGACTTATAG lppε2 genotyping right primer 11 GCCTTTTCAGAAATGGATAAATAGCCTTGCTTCC LB1 genotyping border primer, paired with right primer for lppγ-1 12 ATTTTGCCGATTTCGGAAC LBb1.3 genotyping border primer, paired with right primers of lppγ-2, lppγ-3, lppε1, and lppε2 13 CACCATGGACCTAATACCTCAGC LPPγ for cloning to pENTR, forward 14 ATCAGATTTAGCAGAATCCATATC LPPγ, no stop, reverse 15 TTAATCAGATTTAGCAGAATCC LPPγ, stop, reverse 16 CACCATGGCAGCGTCGTCTTC LPPε1 for cloning to pENTR, forward 17 TCTCTCGTCTTTGAACCAGTT LPPε1, no stop, reverse 18 TTATCTCTCGTCTTTGAACCAG LPPε1, stop, reverse 19 CACCATGGCAGCGTCATCATCTTC LPPε2 for cloning to pENTR, forward 20 TCTGTCATCTTTAAACCAGTTAAG LPPε2, no stop, reverse 21 TCATCTGTCATCTTTAAACCAG LPPε2, stop, reverse 22 CTGAAGGCGCGCcaaaattgaacaatagaatatc LPPγ 2 kb upstream forward, with AscI, 23 GTCAAGTCGACttgcattgaagcttgttcctag LPPγ genomic reverse, with SalI 24 CTGAAGGCGCGCCaaaatcaaccaaaaaacctaaacc LPPε1 2 kb upstream forward, with AscI, 25 GTCAAGTCGACttgattttaacaaacgggctctg LPPε1 genomic reverse, with SalI 26 CTGAAGGCGCGCCataacgatatctcggccggac LPPε2 2 kb upstream forward, with AscI, 27 GTCAAGTCGActatggatgtatggttgatctgc LPPε2 genomic reverse, with SalI 28 ATGACCGTTAAAAGATTCTCTAG LPPε1 alternative start for 51-residue N-terminal truncation, for use with NEB Q5 SDM kit, forward 29 ATGGCCGATTTGGTTAAAACCAATG LPPε2 alternative start for 59-residue N-terminal truncation, for use with NEB Q5 SDM kit, forward 30 GGTGAAGGGGGCGGCCGC pENTR/D-TOPO immediate upstream, for use with NEB Q5 SDM kit, reverse Table 2.1. Primers used for cloning of LPP genes, and genotyping of lpp mutants. 49 REFERENCES 1. Block, M.A., et al., Preparation and characterization of membrane fractions enriched in outer and inner envelope membranes from spinach chloroplasts. II. Biochemical characterization. J. Biol. Chem., 1983. 258(21): p. 13281-13286. 2. Mackender, R. and R.M. Leech, The Galactolipid, Phospholipid, and Fatty Acid Composition of the Chloroplast Envelope Membranes of Vicia faba. L. Plant Physiol., 1974. 53(3): p. 496-502. 3. Dörmann, P. and C. Benning, Galactolipids rule in seed plants. Trends Plant Sci., 2002. 7(3): p. 112-118. 4. Mongrand, S., et al., The C16: 3\C18: 3 fatty acid balance in photosynthetic tissues from 468 plant species. Phytochemistry, 1998. 49(4): p. 1049-1064. 5. Andrews, J. and K. Keegstra, Acyl-CoA synthetase is located in the outer membrane and acyl-CoA thioesterase in the inner membrane of pea chloroplast envelopes. Plant Physiol., 1983. 72(3): p. 735-740. 6. Löhden, I. and M. Frentzen, Role of plastidial acyl-acyl carrier protein: glycerol 3- phosphate acyltransferase and acyl-acyl carrier protein hydrolase in channelling the acyl flux through the prokaryotic and eukaryotic pathway. Planta, 1988. 176(4): p. 506-512. 7. Hares, W. and M. Frentzen, Properties of the microsomal acyl-CoA: sn-1-acyl-glycerol-3- phosphate acyltransferase from spinach (Spinacia oleracea L.) leaves. J. Plant Physiol., 1987. 131(1-2): p. 49-59. 8. Roston, R.L., et al., TGD1,-2, and-3 proteins involved in lipid trafficking form ATP-binding cassette (ABC) transporter with multiple substrate-binding proteins. J. Biol. Chem., 2012. 287(25): p. 21406-21415. 9. Xu, C., et al., Mutation of the TGD1 chloroplast envelope protein affects phosphatidate metabolism in Arabidopsis. Plant Cell, 2005. 17(11): p. 3094-110. 10. Jarvis, P., et al., Galactolipid deficiency and abnormal chloroplast development in the Arabidopsis MGD synthase 1 mutant. Proc. Natl. Acad. Sci. U.S.A., 2000. 97(14): p. 8175- 8179. 11. Miège, C., et al., Biochemical and topological properties of type A MGDG synthase, a spinach chloroplast envelope enzyme catalyzing the synthesis of both prokaryotic and eukaryotic MGDG. Eur. J. Biochem., 1999. 265(3): p. 990-1001. 50 12. Shimojima, M., et al., Cloning of the gene for monogalactosyldiacylglycerol synthase and its evolutionary origin. Proc. Natl. Acad. Sci. U.S.A., 1997. 94(1): p. 333-337. 13. Bertrams, M. and E. Heinz, Positional Specificity and Fatty Acid Selectivity of Purified sn- Glycerol 3-Phosphate Acyltransferases from Chloroplasts. Plant Physiol., 1981. 68(3): p. 653-657. 14. Kunst, L. and C. Somerville, Altered regulation of lipid biosynthesis in a mutant of Arabidopsis deficient in chloroplast glycerol-3-phosphate acyltransferase activity. Proc. Natl. Acad. Sci. U.S.A., 1988. 85(12): p. 4143-4147. 15. Yu, B., et al., Loss of plastidic lysophosphatidic acid acyltransferase causes embryo- lethality in Arabidopsis. Plant Cell Physiol., 2004. 45(5): p. 503-510. 16. Frentzen, M., et al., Specificities and Selectivities of Glycerol-3-Phosphate Acyltransferase and Monoacylglycerol-3-Phosphate Acyltransferase from Pea and Spinach Chloroplasts. Eur. J. Biochem., 1983. 129(3): p. 629-636. 17. Nakamura, Y., et al., Arabidopsis lipins mediate eukaryotic pathway of lipid metabolism and cope critically with phosphate starvation. Proc. Natl. Acad. Sci. U.S.A., 2009. 106(49): p. 20978-20983. 18. Nakamura, Y., M. Tsuchiya, and H. Ohta, Plastidic phosphatidic acid phosphatases identified in a distinct subfamily of lipid phosphate phosphatases with prokaryotic origin. J. Biol. Chem., 2007. 282(39): p. 29013-21. 19. Nguyen, V.C. and Y. Nakamura, Distinctly localized lipid phosphate phosphatases mediate endoplasmic reticulum glycerolipid metabolism in Arabidopsis. Plant Cell, 2023. 35(5): p. 1548-1571. 20. Sato, N. and K. Awai, "Prokaryotic Pathway" Is Not Prokaryotic: Noncyanobacterial Origin of the Chloroplast Lipid Biosynthetic Pathway Revealed by Comprehensive Phylogenomic Analysis. Genome Biol. Evol., 2017. 9(11): p. 3162-3178. 21. Lavell, A., et al., A predicted plastid rhomboid protease affects phosphatidic acid metabolism in Arabidopsis thaliana. Plant J., 2019. 99(5): p. 978-987. 22. Karki, N., B.S. Johnson, and P.D. Bates, Metabolically distinct pools of phosphatidylcholine are involved in trafficking of fatty acids out of and into the chloroplast for membrane production. Plant Cell, 2019. 31(11): p. 2768-2788. 51 23. Xu, C., et al., A permease-like protein involved in ER to thylakoid lipid transfer in Arabidopsis. EMBO J., 2003. 22(10): p. 2370-2379. 24. Xu, C., et al., Lipid trafficking between the endoplasmic reticulum and the plastid in Arabidopsis requires the extraplastidic TGD4 protein. Plant Cell, 2008. 20(8): p. 2190- 2204. 25. Wang, Z., C. Xu, and C. Benning, TGD4 involved in endoplasmic reticulum-to-chloroplast lipid trafficking is a phosphatidic acid binding protein. Plant J., 2012. 70(4): p. 614-623. 26. Fan, J., et al., Arabidopsis TRIGALACTOSYLDIACYLGLYCEROL5 interacts with TGD1, TGD2, and TGD4 to facilitate lipid transfer from the endoplasmic reticulum to plastids. Plant Cell, 2015. 27(10): p. 2941-2955. 27. Dörmann, P., et al., Isolation and characterization of an Arabidopsis mutant deficient in the thylakoid lipid digalactosyl diacylglycerol. Plant Cell, 1995. 7(11): p. 1801-1810. 28. Lin, Y.T., et al., Reduced Biosynthesis of Digalactosyldiacylglycerol, a Major Chloroplast Membrane Lipid, Leads to Oxylipin Overproduction and Phloem Cap Lignification in Arabidopsis. Plant Cell, 2016. 28(1): p. 219-32. 29. Wang, K., et al., A Plastid Phosphatidylglycerol Lipase Contributes to the Export of Acyl Groups from Plastids for Seed Oil Biosynthesis. Plant Cell, 2017. 29(7): p. 1678-1696. 30. Wang, K., et al., Two Abscisic Acid-Responsive Plastid Lipase Genes Involved in Jasmonic Acid Biosynthesis in Arabidopsis thaliana. Plant Cell, 2018. 30(5): p. 1006-1022. 31. Bowling, S.A., et al., A mutation in Arabidopsis that leads to constitutive expression of systemic acquired resistance. Plant Cell, 1994. 6(12): p. 1845-1857. 32. Van Wersch, R., X. Li, and Y. Zhang, Mighty dwarfs: Arabidopsis autoimmune mutants and their usages in genetic dissection of plant immunity. Front. Plant Sci., 2016. 7: p. 1717. 33. Wildermuth, M.C., et al., Isochorismate synthase is required to synthesize salicylic acid for plant defence. Nature, 2001. 414(6863): p. 562-565. 34. Cook, R., J. Lupette, and C. Benning, The role of chloroplast membrane lipid metabolism in plant environmental responses. Cells, 2021. 10(3): p. 706. 52 35. Kunst, L., J. Browse, and C. Somerville, A mutant of Arabidopsis deficient in desaturation of palmitic acid in leaf lipids. Plant Physiol., 1989. 90(3): p. 943-947. 36. Heilmann, I., et al., Identification of the Arabidopsis palmitoyl- monogalactosyldiacylglycerol Δ7-desaturase gene FAD5, and effects of plastidial retargeting of Arabidopsis desaturases on the fad5 mutant phenotype. Plant Physiol., 2004. 136(4): p. 4237-4245. 37. Awai, K., et al., A phosphatidic acid-binding protein of the chloroplast inner envelope membrane involved in lipid trafficking. Proc. Natl. Acad. Sci. U.S.A., 2006. 103(28): p. 10817-10822. 38. Lu, B., et al., A Small ATPase Protein ofArabidopsis, TGD3, Involved in Chloroplast Lipid Import. J. Biol. Chem., 2007. 282(49): p. 35945-35953. 39. Muthan, B., et al., Probing Arabidopsis chloroplast diacylglycerol pools by selectively targeting bacterial diacylglycerol kinase to suborganellar membranes. Plant Physiol., 2013. 163(1): p. 61-74. 40. Sessions, A., et al., A High-Throughput Arabidopsis Reverse Genetics System. Plant Cell, 2002. 14(12): p. 2985-2994. 41. Alonso, J.M., et al., Genome-Wide Insertional Mutagenesis of Arabidopsis thaliana. Science, 2003. 301(5633): p. 653-657. 42. Alonso, J.M., et al., Genome-wide insertional mutagenesis of Arabidopsis thaliana. Science, 2003. 301(5633): p. 653-657. 43. Cook, R., How to cross Arabidopsis thaliana. 2021: YouTube. https://www.youtube.com/watch?v=0Tvkco9fa-E&ab_channel=roncook9. 44. Murashige, T. and F. Skoog, A revised medium for rapid growth and bio assays with tobacco tissue cultures. Physiol. Plant., 1962. 15(3): p. 473-497. 45. Kubo, M., et al., Transcription switches for protoxylem and metaxylem vessel formation. Genes Dev., 2005. 19(16): p. 1855-1860. 46. Logemann, E., et al., An improved method for preparing Agrobacterium cells that simplifies the Arabidopsis transformation protocol. Plant Methods, 2006. 2(1): p. 1-5. 53 47. Wang, Z. and C. Benning, Arabidopsis thaliana polar glycerolipid profiling by thin layer chromatography (TLC) coupled with gas-liquid chromatography (GLC). J. Vis. Exp., 2011(49): p. e2518. 48. Aronsson, H. and P. Jarvis, A simple method for isolating import-competent Arabidopsis chloroplasts. FEBS Lett., 2002. 529(2-3): p. 215-220. 49. Arnon, D.I., Copper enzymes in isolated chloroplasts. Polyphenoloxidase in Beta vulgaris. Plant Physiol., 1949. 24(1): p. 1-15. 50. Cruz, J.A., et al., Dynamic environmental photosynthetic imaging reveals emergent phenotypes. Cell Syst., 2016. 2(6): p. 365-377. 51. Baker, N.R. and K. Oxborough, Chlorophyll fluorescence as a probe of photosynthetic productivity, in Chlorophyll a Fluorescence: A Signature of Photosynthesis. 2004. p. 66- 79. 52. Froehlich, J., Studying Arabidopsis Envelope Protein Localization and Topology Using Thermolysin and Trypsin Proteases, in Chloroplast Research in Arabidopsis: Methods and Protocols, R.P. Jarvis, Editor. 2011, Humana Press: Totowa, NJ. p. 351-367. 54 CHAPTER 3: A suppressor screen targeting novel components of OPDA conversion to jasmonic acid EMS mutagenesis of seeds was performed by Linda Danhof. Portions of this chapter have been published in Liu et al. 2021 [1]. My contribution to the published work was to optimize the growth and screening protocol, and provide instructional videos for students on seed sowing and plant crossing. 55 Abstract Fatty acid export from chloroplasts is basal to plant lipid metabolism, and it is a complex process due to the amphipathic nature of the mobile molecule, and the multiple organic and aqueous barriers it must cross. Here, a forward genetic screen was implemented in the background of a transgenic Arabidopsis line, PLIP3-OX, that excessively produces and exports the fatty acid derivative 12-oxo-phytodienoic acid (OPDA). Because cytosolic OPDA is efficiently converted into the defense hormone jasmonic acid (JA), PLIP3-OX plants have a distinctive JA-induced morphological phenotype that is dependent on OPDA production, export, and conversion. To identify plants with impaired OPDA export capacity, mutagenized PLIP3-OX plants were screened for suppression of the JA-induced phenotype. Two lines isolated from the screen were determined to exhibit PLIP3-OX suppression due to mutations in KEG4 and CDK8. KEG4 may be involved in abscisic acid (ABA)-JA signal coordination by stabilizing a transcriptional repressor of the JA response, while targeting activators of the ABA response for degradation; CDK8 is likely itself a transcriptional activator of JA response genes. Introduction Nearly all fatty acid (FA) biosynthesis in plants occurs in the chloroplast stroma, and FA export from the chloroplast feeds the various lipid pathways in the plant cell. Free FAs generated in the stroma must cross two envelope membranes and an aqueous intermembrane space, a process complicated by their amphipathic nature (Fig. 3.1). FA export therefore cannot depend on diffusion alone, and protein factors must be involved in transport [2]. However, because FA export is essential for plant viability, null mutants of FA export factors would be lethal and therefore difficult to identify. The chloroplast inner envelope membrane (IEM) transporter FAX1 has been characterized as a component of the FA export machinery, with fax1 null mutants retaining much of their transport capacity, and exhibiting various mild phenotypes such as reduced cuticle deposition and a higher 16:3 acyl content of MGDG [3]. It is therefore not surprising that FAX1 was identified using a reverse genetic approach, rather than through a phenotype-driven genetic screen. Subsequent research on FAX1 homologs identified IEM FA transporters FAX2-4, with FAX3 operating in 56 vegetative tissues alongside FAX1 [4-6]. While these may account for much or all of the FA transport across the IEM, factors facilitating transfer across the intermembrane space and outer envelope membrane (OEM) are still unknown. Because FA export is the basis for a metabolic network that is both extensive and essential, forward genetic approaches could miss null mutants due to lethality, while potentially overlooking reduced-function mutants with indistinct phenotypes. Chloroplast FA export is not limited to the precursors of glycerolipids and cuticular constituents, but also includes various FA-derived oxylipins which are involved in environmental responses [7]. Among these, chloroplast-derived 12-oxo-phytodienoic acid (OPDA) is exported as part of the jasmonic acid (JA) biosynthetic pathway [8]. As an FA derivative, OPDA faces the same physical constraints that require export mediation, namely a charged carboxyl end which must cross two hydrophobic membranes, and a hydrocarbon tail which must dissociate from membranes and cross the intermembrane space. While an OEM OPDA transporter, JASSY, has been identified, proteins required for OPDA transport across the IEM and intermembrane space remain undetermined [9]. 18:3 FA precursors can be directed towards OPDA production by lipase activity, which releases the linolenoyl substrates from glycerolipids [10]. Such lipases include the plastid lipases PLIP1, PLIP2, and PLIP3, which contribute to excess OPDA production and conversion to JA in PLIP-OX transgenic Arabidopsis plants (Fig. 3.2) [11, 12]. The distinctive JA-induced phenotype of these plants includes stunted growth, altered relative dimensions of leaves and petioles, and anthocyanin accumulation in vascular tissues (Fig. 3.3). The clarity of this phenotype makes it a strong background for suppressor screening, and because the phenotype depends on OPDA export, a suppressor screen may uncover mutants in OPDA or general FA transport. Therefore, a suppressor screen in the PLIP3-OX transgenic background was designed and implemented to discover new factors in OPDA or FA export, in a forward genetic approach that had previously been unfeasible. 57 Results Screen design and preparation Of the three described PLIP overexpression lines, PLIP1-OX has the mildest JA phenotype, PLIP2- OX has the most severe phenotype, which severely limits seed production, and PLIP3-OX has an intermediate phenotype, which is clearly distinguishable from the wild type while maintaining reproductive capacity [11, 12]. PLIP3-OX was therefore chosen as the background for mutant suppressor screening. PLIP3-OX seeds were previously mutagenized with ethyl methanesulfonate (EMS), grown, allowed to self-fertilize, and M2 seeds were harvested in 96 separate batches. M2 plants were screened visually for suppression. Initially, vertical plates were considered as an option for visual screening. However, PLIP3-OX plants do not exhibit a strong JA-induced phenotype in early stages of growth, rendering plate screening unfeasible (Fig. 3.4). EMS mutants were therefore grown in soil, with visual screening taking place at approximately four weeks after sowing. Primary screen The primary screen for PLIP3-OX suppressor mutants was performed according to the following visual criteria: rosette diameter relative to Col-0, anthocyanin content and distribution, ratio of leaf length to petiole length, and ratio of leaf length to leaf width. Approximate values for these parameters in Col-0, PLIP3-OX, and suppressor mutants are provided in Table 3.1. Inflorescence apical dominance, plant fertility, and leaf color were also noted, though not used as selection criteria. Plants were selected for secondary screening if a stronger resemblance to wild type than PLIP3-OX was observed in two or more of these characteristics. Selected plants that were ultimately sequenced are shown in Fig. 3.3. In total, approximately 5000 plants were screened visually, with 90 mutants selected for secondary screening. Integration of primary screen into a lab course Primary screening was also integrated into a Course-based Undergraduate Research Experience (CURE) [1]. Undergraduate students enrolled in an entry-level biology laboratory course engaged in sowing Arabidopsis seeds, maintaining plants, characterizing the phenotypes of Col-0 and 58 PLIP3-OX controls, and screening for suppressor mutants. After selecting the mutants, students used PCR analysis to confirm the presence of the PLIP3-OX transgene in the suppressor lines. Mutants were then independently analyzed according to the previously described criteria, and suppressor lines meeting these criteria were chosen for secondary screening. In total, approximately 1200 of the ~5000 visually screened plants were screened by undergraduates through CURE, yielding 17 of the 90 mutants carried into secondary screening. Secondary screen Mutants isolated from the primary screen may show suppression due to various causes, beyond the targeted defects in OPDA export. These are expected to include mutants in OPDA biosynthesis, JA perception, and downstream JA signaling. In order to mitigate these possibilities, a secondary screen was implemented in which OPDA, JA, and the JA catabolite 12-OH JA were directly quantified. Mutants lacking OPDA were discounted as OPDA biosynthetic mutants, and mutants retaining high JA were discounted as defective in JA perception or downstream signaling. Out of the 90 mutants screened, 23 were determined to have non-zero levels of OPDA, and lower levels of JA than PLIP3-OX. Results of the JA metabolite concentrations are shown in Table 3.2. The 23 suppressor mutants passing the secondary screen were designated candidates for OPDA export, appropriate for subsequent mutation mapping. Candidates were back-crossed to PLIP3-OX, and the F1 phenotype was monitored to determine whether mutations were dominant, semi-dominant, or recessive. F2 seeds were harvested separately for each F1 plant and used in subsequent F2 segregation analyses and mutation mapping. Whole genome sequencing and mutation mapping Mutants with recessive suppression alleles were prioritized for mapping. These were determined based on an unsuppressed PLIP3-OX phenotype in the F1 backcross to PLIP3-OX, and one-quarter or fewer F2 plants showing PLIP3-OX suppression. Four such mutants were selected: sup11, sup12, sup53, and sup72, as these had clear phenotypes, sufficient seeds, and recessive suppression alleles (Fig. 3.3). For each, the segregating F2 population from the PLIP3-OX backcross was grown, and gDNA extracted from individual plants showing suppression and background 59 phenotypes. The extracted gDNA was pooled based on phenotype, with each suppressor or background pool consisting of DNA from 30-200 plants. Pooled gDNA was submitted for whole genome sequencing, and results were analyzed using the SIMPLE pipeline developed for causal mutation mapping [13]. Plots generated by SIMPLE remove uncorrelated mutations, and then use LOESS smoothing when plotting the remaining mutations [13]. This provides an accessible visualization of the region containing the causal mutation, shown for each mutant in Fig. 3.5. Lists of potential causal mutations in Tables 3.3-3.6. Causal gene candidates were pursued further for the suppressor mutants sup72 and sup11. A mutation in KEG may lead to PLIP3-OX suppression in sup72 Sequencing data analysis for sup72 indicated co-segregation of PLIP3-OX suppression with a cytosine to thymine base substitution, at nucleotide position 4369 in the coding sequence of KEEP ON GOING (KEG). The substitution corresponds to a change of the histidine residue at position 1457 to tyrosine. KEG is a RING-type E3 ligase, with a negative regulatory role in the abscisic acid (ABA) and JA signaling pathways [14, 15]. It is composed of RING and ankyrin domains, a kinase domain, and a C-terminal domain of 12 HERC2-like repeats [14]. The H1457Y mutation is in the 10th HERC2-like repeat and may alter the protein-protein interactions associated with the HERC2- like domain. A previous study on the mutant keg-4 had determined that alterations to the HERC2- like domain increases the repressive role of KEG in the ABA pathway, and it is therefore likely that a similar effect is witnessed in sup72 with respect to JA signaling [16]. A mutation in CDK8 leads to PLIP3-OX suppression in sup11 Sequencing data analysis for sup11 showed co-segregation of PLIP3-OX suppression with the introduction of an early stop codon to the gene CDK8, corresponding to residue position 149. Unpublished data from the Howe lab demonstrates a similar suppression of the jazD phenotype by mutation of CDK8. The jazD phenotype resembles that of PLIP3-OX, as it lacks ten transcriptional repressors of JA response genes [17]. cdk8 suppression of jazD indicates that CDK8 has an important role in driving expression of JA response genes, and the elimination of CDK8 in sup11 likely results in the same suppression. 60 In order to test whether the mutation of CDK8 was indeed causal for PLIP3-OX phenotypic suppression, sup11 was crossed to a null cdk8 insertional mutant (SALK_138675). The F1 generation, which is heterozygous for the PLIP3-OX transgene, showed suppression for cdk8 x sup11, while the control cross of cdk8 x PLIP3-OX background retained the JA response phenotype (Fig. 3.6). These F1 results confirm that two non-functional alleles of CDK8 suppress the JA response in PLIP3-OX, while one functional CDK8 allele is sufficient to maintain the response. Crosses of insertional mutants of candidate genes to sup12 and sup53 will determine causal mutations While suppressor mutants sup12 and sup53 have been sequenced, the co-segregating mutations do not include obvious candidates for suppression. Causal mutations for suppression will be determined by crossing the suppressor mutant to insertional mutants for each of the co- segregating altered genes. The cross in which the insertional mutant is in the same gene as the causal suppressor mutation will appear suppressed in the F1 generation, while the others should retain the PLIP3-OX phenotype. Discussion Primary screen and course integration Based on the mutants identified, the visual primary screen was effective in targeting JA-related genes. The distinct phenotypes allow for rapid identification of suppressor mutants, in a manner accessible to contributors who are otherwise unexperienced with Arabidopsis. Thus, integration of the primary screen with a lab course was successful, and undergraduate students were able to identify promising suppression phenotypes. In fact, primary screening continued solely through CURE would be sufficient for providing future mutants, as the downstream backcrossing, growing, sequencing, and analysis of selected candidates represent a more significant bottleneck than the initial screen. The CURE also provided the Benning lab with opportunities to recognize and recruit capable undergraduates. From the undergraduate perspective, direct exposure to an academic research project was useful in gauging their interest in a future academic career. 61 Secondary screen The purpose of the secondary screen was to eliminate mutants deficient in OPDA biosynthesis or JA perception and downstream signaling, thereby narrowing the candidate pool to mutants deficient in the conversion of OPDA to JA. This was accomplished by measuring OPDA and JA levels in the plants and eliminating mutants that appeared to accumulate high JA or no OPDA. While the secondary screen was successful in reducing the number of candidates from 90 to 23, the results from sup11 and sup72 indicate that false positives do pass through secondary screening. Both of the presumed causative mutations in these suppressors affect expression of JA response genes, rather than JA biosynthesis. False positives are likely a result of the variability in the JA response values, which was witnessed in both wild-type and PLIP3-OX controls included in each batch of hormone quantification. While it is not ideal for such off-target mutants to pass through secondary screening, false positives may be preferable to elimination of true JA biosynthetic mutants. KEG mutation in sup72 likely results in greater JAZ12 stability and repression of the JA response The substitution in sup72 results in a single residue change from histidine to tyrosine at the 10 th HERC2-like repeat in the C-terminal domain of KEG. The full protein is composed of an N-terminal RING-HCa domain, a kinase domain, and ankyrin repeats, followed by the HERC2-like repeats, and its repressive role in the ABA pathway is essential for plant viability [14]. In the absence of ABA, KEG maintains low levels of the transcriptional activator ABSCISIC ACID INSENSITIVE 5 (ABI5) by continually ubiquitinating ABI5. In response to ABA, KEG self-ubiquitinates, allowing ABI5 to accumulate and activate the ABA response [18, 19]. Early development is therefore arrested in null keg mutants, due to excess ABA signaling, while KEG overexpression results in reduced ABA sensitivity [14, 18]. However, the keg-4 mutant, which contains a substitution in the 5th HERC2- like repeat, exhibits low ABA sensitivity rather than hypersensitivity [16]. Comparative localization of native and mutant KEG revealed that in keg-4, the protein is less strongly associated with the trans-Golgi network (TGN) and more abundant in the cytosol, indicating that the HERC2-like domain helps to sequester the protein at the TGN, where KEG is less effective in repression of the ABA response [20]. More recently, KEG has been implicated in JA signaling, as the JA response 62 repressor JAZ12 is stabilized through interaction with the HERC2-like domain of KEG, and KEG overexpression protected JAZ12 from degradation [15]. In this context, there are two likely models for KEG-mediated suppression of the JA response in sup72. In the first, the mutation in the HERC2-like domain of KEG in sup72 directly affects the interaction of KEG and JAZ12, in such a way as to increase the stability and repressive role of the JAZ repressor. While this is the most straightforward explanation, the keg-4 phenotype would suggest that a more general role of the HERC2-like domain in subcellular targeting could be affected in sup72. If the effect of an altered HERC2-like domain in sup72 is comparable to that of keg-4, the increase in cytosolic KEG would amplify its repressive role in the ABA pathway through increased degradation of ABI5, and similarly repress the JA response due to increased interactions with JAZ12 (Fig. 3.7). These models can be tested by directly studying the stability of the JAZ12- KEG4 interactions, as well as determining the sensitivity of sup72 to ABA. CDK8 may be involved in transcriptional activation of JA response genes Using a genetic approach, phenotypic suppression of PLIP3-OX was shown here to be caused by a nonsense mutation of CDK8 in sup11. The suppression is likely the result of a muted transcriptional response to JA, as opposed to decreased JA biosynthesis. CDK8 is known to be a nuclear-localized protein, involved in activation of various stress-responsive regulatory pathways [21]. In addition, unpublished data from the Howe lab shows cdk8 mutant suppression of the jazD phenotype. Because the jazD phenotype results from de-repression of JA response genes, and does not directly depend on JA biosynthesis, CDK8 likely acts as a positive regulator of these response genes. Methods Plant strains PLIP3-OX lines had been previously developed in the Benning lab and were obtained from the lab seed stocks [12]. The cdk8 insertional mutant, SALK_138675, was ordered from the Arabidopsis Biological Resource Center (ABRC). 63 EMS mutagenesis Approximately 13,000 PLIP3-OX seeds were incubated in 0.1% Tween©20 (Sigma-Aldrich) for 15 minutes in a tube rotator, after which seeds were allowed to settle, and the solution was removed. 0.2% Ethyl methanesulfonate (EMS) in water was added, and seeds were incubated overnight (~16hrs) in tube rotator. Seeds were washed 7 times with water, incubated in water for 2 hours, and washed one more time. Plant growth conditions Plants were grown in SUREMIX™ Professional All-Purpose Perlite Mix (Michigan Grower Products, Inc.) at 22°C and a light intensity of approximately 120 µmol m-2 s-1, under a 16/8-hr light/dark cycle in a growth chamber. Plants grown in the classroom for the CURE-based primary screen had a less stable environment as they were grown on open, lighted racks, and may have experienced some deviations from this regime. Hormone quantification Fresh plant tissue was harvested, flash-frozen, ground, and incubated in extraction buffer (80:20 methanol:water, 0.1% formic acid, 100 mg/L butylated hydroxytoluene) containing internal standard (100 nM abscisic acid-d6) for 24 hr at 4°C. Samples were analyzed using liquid chromatography/mass spectrometry as described [11]. Genome sequencing and analysis Genomic DNA was isolated from segregating F2 plants using the Promega Wizard® Genomic DNA Purification Kit, and quantified using a Qubit fluorometer. Equal amounts of DNA from 30-200 plants were pooled based on phenotype, and then sent to BGI genomics for paired-end 150 Illumina sequencing, or the BGI DNBseq™ platform. At least 10 Gb was sequenced for each sample. Sequencing data was processed using the SIMPLE pipeline [13]. Modifications to the SIMPLE pipeline were made by undergraduate student Yash Manne, for compatibility with data arriving from the DNBseq™ platform. 64 FIGURES AND TABLES Figure 3.1. Fatty acid (FA) export from the chloroplast requires an amphipathic molecule to cross both hydrophobic membranes and an aqueous intermembrane space. This process requires facultative protein factors in order to be thermodynamically favorable. ACP, acyl carrier protein; CoA, coenzyme A; IEM, chloroplast inner envelope membrane; LACS, long-chain acyl-CoA synthetase; OEM, chloroplast outer envelope membrane; TE, thioesterase. 65 Figure 3.2. Schematic of jasmonic acid (JA) biosynthesis in plants. The red arrow represents OPDA export from the chloroplast, the primary target of the PLIP3-OX suppressor screen. MGDG, monogalactosyldiacylglycerol; 12-OPDA, 12-oxo-phytodienoic acid; PG, phosphatidylglycerol; PLIPs, plastid lipases. 66 Figure 3.3. PLIP3-OX suppressor mutants selected for genomic sequencing. These M2 plants were backcrossed to PLIP3-OX, and the F2 generation was sequenced. 67 Figure 3.4. 8-day-old plants grown on vertical plates, as a test for a plate-based primary screen. The JA-induced PLIP3-OX phenotype does not appear in young seedlings. 68 Figure 3.5. LOESS plots generated from the SIMPLE pipeline, showing chromosomal regions in which mutations co-segregate with suppression phenotypes in sup11, sup12, sup53, and sup72. 69 Figure 3.6. F1 generation of sup11 crossed to cdk8, in which the JA-induced PLIP3-OX phenotype remains suppressed. In contrast, cdk8 crossed to PLIP3-OX retains its JA-induced phenotype. Two independent F1 plants are shown for each cross. 70 Figure 3.7. A model for PLIP3-OX suppression in sup72. A substitution in the HERC2-like domain weakens KEG association with the TGN, and the resulting higher KEG concentration at the cytosol increases stabilization of JAZ12 and subsequent repression of JA response genes. 71 relative rosette inflorescence apical leaf length: leaf length: batch date batch # mutant # diameter anthocyanin dominance fertility color petiole length leaf width notes N/A N/A col-0 1.00 strong AD 2 3.00 N/A N/A PLIP3-OX 0.20 weak AD 10 1.25 08.17.2018 #1.02 1 0.75 wt distribution near-wt AD wt 3 1.50 08.17.2018 #1.05 2 0.30 wt distribution weak AD weak wt ? 2.00 attacked by herbivores, indistinguishible from mutant #3 08.17.2018 #1.05 3 0.30 wt distribution weak AD weak wt ? 2.00 attacked by herbivores, indistinguishible from mutant #2 08.17.2018 #1.08 4 0.50 less than wt weak-intermediate AD slight pale 2 1.50 bolted very early (~week2-3), and senesced early (rosette already completeley senesced at week 6). Had to harvest 08.17.2018 #1.09 5 0.40 less than wt weak AD yes pale ? 2.00 cauline leaves for hormone analysis Seems infertile, filament doesn't elongate. Could be a JA- #8.01 6 no deficient phenotype #8.03 7 yes #8.04 8 yes #8.07 9 yes 09.20.2018 #1.11 10 0.67 low strong AD yes normal 3 1.75 slow to grow, slow to bolt. Only started bolting after week 7, low -> high at and rosette had many more leaves. Only started bolting in 09.20.2018 #2.01 11 0.40 later stage strong AD no normal -> dark 1.5 1.50 week 7 09.20.2018 #2.02 12 0.80 low strong AD yes normal 3 1.75 09.20.2018 #2.02 13 0.50 low moderate AD yes pale 3 1.40 09.20.2018 #2.04 14 0.60 low moderate AD yes pale 3 1.25 appears to retain JA phenotype with the exception of growth 09.20.2018 #2.05 15 1.00 high strong AD yes dark 2 2.00 inhibition. Update: growth more stunted in later weeks (~5-8) 09.20.2018 #2.06 16 0.50 low yes slight pale 3 3.00 09.20.2018 #2.06 17 0.40 low yes slight pale 2 2.00 10.24.2018 #2.08 18 0.33 low strong AD yes normal 2 1.50 10.24.2018 #2.09 19 0.50 moderate normal 3 3.00 slightly late to bolt 10.24.2018 #2.09 20 0.30 moderate strong AD normal 1.5 1.70 flat leaves, low trichome density 10.24.2018 #2.09 21 0.40 high slight dark 3 2.00 slow to bolt 10.24.2018 #2.10 22 0.30 normal immature? 10.24.2018 #2.10 23 0.15 high? normal 5 1.25 JA not measured; Plant had strong JA phenotype 12/12/2018 10.24.2018 #2.11 24 0.40 low moderate AD normal 2 3.00 10.24.2018 #2.11 25 0.40 low moderate AD yes nomal 3 2.00 10.24.2018 #2.11 26 0.70 low strong AD normal 2.5 2.50 10.24.2018 #3.01 27 0.15 moderate slight pale 4 1.25 JA not measured; Plant had strong JA phenotype 12/12/2018 10.24.2018 #3.01 28 0.30 low normal 2 2.00 10.24.2018 #3.01 29 0.25 low moderate AD normal 1 2.00 spindly 10.24.2018 #3.01 30 0.40 high moderate AD yes slight pale? 3 2.00 10.24.2018 #3.01 31 0.45 low moderate AD yes slight dark 1.5 2.00 folded leaves JA not measured; Apparent recovery from JA phenotype: back- 10.24.2018 #3.01 32 0.20 low pale 2 1.50 cross necessary. very pale plant, seed was in PLIP3-OX control vial, JA not 11.12.2018 unknown 33 0.60 low very pale 4 2.00 measured 12.11.2018 #3.03 34 0.20 moderate normal 2 1.33 12.11.2018 #3.03 35 0.20 moderate normal 2 1.33 12.11.2018 #3.04 36 0.20 low moderate AD normal 2 1.33 12.11.2018 #3.06 37 0.25 low moderate AD normal 2 1.50 crumpled, bushy leaves, looks like butter lettuce strange, smooth, few leaves, some circular others very skinny. 12.11.2018 #3.07 38 0.20 low normal 1.5 1.00 Early senescence 12.11.2018 #3.07 39 0.25 moderate strong AD normal 2 2.00 spiky leaves 01.08.2019 #7.09 40 not backcrossed - too old 01.08.2019 #7.11 41 no sterile 01.08.2019 #8.08 42 01.08.2019 #7.04 43 01.08.2019 #8.08 44 01.08.2019 #8.08 45 not backcrossed 01.08.2019 #7.02 46 no sterile 01.08.2019 #7.02 47 not backcrossed 01.08.2019 #7.05 48 no hormone measurments - sample lost 01.08.2019 #7.05 49 promising 01.08.2019 #7.05 50 too small for JA meas., too old to backcross 01.08.2019 #7.01 51 too young to backcross 01.08.2019 #7.01 52 too young to backcross 01.21.2019 #3.08 53 01.21.2019 #3.08 54 01.21.2019 #3.08 55 01.21.2019 #3.09 56 01.21.2019 #3.09 57 01.21.2019 #3.09 58 01.21.2019 #3.09 59 01.21.2019 #3.09 60 01.21.2019 #3.10 61 01.21.2019 #3.10 62 01.21.2019 #3.10 63 01.21.2019 #3.10 64 01.21.2019 #3.10 65 01.21.2019 #3.10 66 01.21.2019 #3.11 67 01.21.2019 #3.11 68 01.21.2019 #3.12 69 01.21.2019 #3.12 70 01.21.2019 #3.12 71 01.21.2019 #4.01 72 01.21.2019 #4.01 73 leaves are completely vertical, no apparent bolt for flowers as 02.24.2019 #4.02 74 0.40 low normal 1.5 4.00 of 03.27.2019. crumpled leaf appearance 04.08.19: 02.24.2019 #4.02 75 0.70 moderate unclear normal 3 2.00 02.24.2019 #4.02 76 0.20 moderate dark 1 1.00 02.24.2019 #4.03 77 0.35 low slight pale 2 2.00 02.24.2019 #4.03 78 0.35 moderate slight pale 2 1.50 02.24.2019 #4.03 79 0.35 low slight pale 2 1.25 02.24.2019 #4.03 80 0.35 moderate slight pale 1 1.30 04.08.19: appears 02.24.2019 #4.04 81 0.80 low strong AD sterile pale 2 2.00 pale. 04.08.2019: similar phenotype to #83 (coi1?) 02.24.2019 #4.04 82 0.40 low slight pale 2 1.80 04.08.19: appears 02.24.2019 #4.04 83 0.50 low sterile normal 2 3.00 04.08.2019: similar phenotype to #81 (coi1?) 02.24.2019 #4.04 84 0.40 moderate slight pale 2 2.00 02.24.2019 #4.06 85 0.25 moderate normal 2 1.60 02.24.2019 #4.06 86 0.25 low slight pale 2 2.00 02.24.2019 #4.06 87 0.30 moderate slight pale 6 3.00 02.24.2019 #4.07 88 0.35 low slight pale 3 1.50 02.24.2019 #4.07 89 0.35 low normal 3 2.00 02.24.2019 #4.07 90 0.60 low strong AD fertile slight pale 2 2.00 Table 3.1. Phenotypic data collected on mutants selected from visual primary screen. 72 raw total nmol per g harvest date harvest mass (g) OPDA (nM) JA (nM) JA-Ile (nM) 12-OH JA (response) OPDA (pmol) JA (pmol) JA-Ile (pmol) 12-OH JA (response) OPDA JA JA-Ile 12-OH JA (response) wt 0.085 72.03874 0.2924 0.2827 57.630992 0.23392 0 0.22616 0.678 0.003 0.000 0.003 PLIP3-OX 0.05 55.55396 1.89235 4.5056 44.443168 1.51388 0 3.60448 0.889 0.030 0.000 0.072 #1 0.08 86.99594 3.2662 0.12569 5.4923 69.597 2.613 0.101 4.394 0.870 0.033 0.001 0.055 #2 0.03 71.21643 11.95392 15.1647 56.973 9.563 0.000 12.132 1.899 0.319 0.000 0.404 #3 0.044 45.98224 4.54166 26.445 36.786 3.633 0.000 21.156 0.836 0.083 0.000 0.481 #4 0.055 82.98276 5.5287 0.1147 1.4263 66.386 4.423 0.092 1.141 1.207 0.080 0.002 0.021 #5 0.041 36.98417 3.20263 1.6751 29.587 2.562 0.000 1.340 0.722 0.062 0.000 0.033 10.22.2018 wt 0.0487 6.7083 0.22438 0.0381 4.02498 0.134628 0 0.02286 0.083 0.003 0.000 0.000 10.22.2018 PLIP3-OX 0.0464 8.7123 3.70176 0.1847 10.1446 5.22738 2.221056 0.11082 6.08676 0.113 0.048 0.002 0.131 10.22.2018 #6 0.0125 0.4378 0.10353 0.0001 0.1378 0.26268 0.062118 0.00006 0.08268 0.021 0.005 0.000 0.007 10.22.2018 #7 0.032 1.9508 0.1212 0.1103 0.2292 2.34096 0.14544 0.13236 0.27504 0.073 0.005 0.004 0.009 10.22.2018 #8 0.0435 3.8457 2.06935 0.3455 0.1715 2.30742 1.24161 0.2073 0.1029 0.053 0.029 0.005 0.002 10.22.2018 #9 0.011 2.3625 0.11746 0.0045 2.329 1.4175 0.070476 0.0027 1.3974 0.129 0.006 0.000 0.127 10.22.2018 wt 0.0548 6.4949 0.08971 0.0025 0.0392 3.89694 0.053826 0.0015 0.02352 0.071 0.001 0.000 0.000 10.22.2018 PLIP3-OX 0.039 29.1363 1.87841 0.3664 45.8129 17.48178 1.127046 0.21984 27.48774 0.448 0.029 0.006 0.705 10.22.2018 #10 0.0463 3.315 0.26712 0.0835 0.2209 1.989 0.160272 0.0501 0.13254 0.043 0.003 0.001 0.003 10.22.2018 #11 0.0464 2.7488 0.14357 0.0078 0.0569 1.64928 0.086142 0.00468 0.03414 0.036 0.002 0.000 0.001 10.22.2018 #12 0.0445 3.8768 0.06514 0.0004 0.0349 2.32608 0.039084 0.00024 0.02094 0.052 0.001 0.000 0.000 10.22.2018 #13 0.0335 0.0903 0.0131 0.0012 0.0475 0.05418 0.00786 0.00072 0.0285 0.002 0.000 0.000 0.001 10.22.2018 #14 0.0406 1.5502 0.37421 0.2387 0.1477 0.93012 0.224526 0.14322 0.08862 0.023 0.006 0.004 0.002 10.22.2018 #15 0.0592 11.7861 0.44529 0.1627 2.7687 7.07166 0.267174 0.09762 1.66122 0.119 0.005 0.002 0.028 10.22.2018 #16 0.0423 3.3474 2.36828 0.2818 5.5637 2.00844 1.420968 0.16908 3.33822 0.047 0.034 0.004 0.079 10.22.2018 #17 0.0251 0.8033 0.02793 0.022 0.2805 0.48198 0.016758 0.0132 0.1683 0.019 0.001 0.001 0.007 12.04.2018 wt #1 0.0654 17.9026 3.20226 0.0068 1.3193 10.74156 1.921356 0.00408 0.79158 0.164 0.029 0.000 0.012 12.04.2018 wt #2 0.0541 4.4491 0.19189 0.9067 2.66946 0.115134 0 0.54402 0.049 0.002 0.000 0.010 12.04.2018 PLIP3-OX #1 0.0434 2.6228 2.3626 0.0088 2.1371 1.57368 1.41756 0.00528 1.28226 0.036 0.033 0.000 0.030 12.04.2018 PLIP3-OX #2 0.0231 2.4752 2.19751 0.0067 0.4715 1.48512 1.318506 0.00402 0.2829 0.064 0.057 0.000 0.012 12.04.2018 #18 0.0542 1.2916 0.77275 0.0028 0.4847 0.77496 0.46365 0.00168 0.29082 0.014 0.009 0.000 0.005 12.04.2018 #19 0.0497 4.2542 1.93718 0.0202 0.8285 2.55252 1.162308 0.01212 0.4971 0.051 0.023 0.000 0.010 12.04.2018 #20 0.0274 6.7092 1.21785 0.5166 4.02552 0.73071 0 0.30996 0.147 0.027 0.000 0.011 12.04.2018 #21 0.0513 4.1468 5.70458 0.0031 4.2144 2.48808 3.422748 0.00186 2.52864 0.049 0.067 0.000 0.049 12.04.2018 #22 0.0081 0.1112 0.57461 0.0022 0.0624 0.06672 0.344766 0.00132 0.03744 0.008 0.043 0.000 0.005 #23 0 0 0 0 #DIV/0! #DIV/0! #DIV/0! #DIV/0! 12.04.2018 #24 0.0451 0.0002 0.1894 0 0 0.00012 0.11364 0.000 0.000 0.000 0.003 12.04.2018 #25 0.0526 5.945 1.93372 0.0018 1.2677 3.567 1.160232 0.00108 0.76062 0.068 0.022 0.000 0.014 12.04.2018 #26 0.0394 3.8877 0.6738 0.0151 0.3326 2.33262 0.40428 0.00906 0.19956 0.059 0.010 0.000 0.005 #27 0 0 0 0 #DIV/0! #DIV/0! #DIV/0! #DIV/0! 12.04.2018 #28 0.0221 0.9739 0.36067 0.0041 0.0687 0.58434 0.216402 0.00246 0.04122 0.026 0.010 0.000 0.002 12.04.2018 #29 0.0167 0.3745 0.79272 0.0064 0.4379 0.2247 0.475632 0.00384 0.26274 0.013 0.028 0.000 0.016 12.04.2018 #30 0.0354 1.9402 2.71679 0.0634 1.4812 1.16412 1.630074 0.03804 0.88872 0.033 0.046 0.001 0.025 12.04.2018 #31 0.0489 4.4346 1.0784 0.1155 0.6942 2.66076 0.64704 0.0693 0.41652 0.054 0.013 0.001 0.009 #32 01.16.2019 wt #1 0.0475 112.90915 5.36159 0.1624 67.74549 3.216954 0 0.09744 1.426 0.068 0.000 0.002 01.16.2019 wt #2 0.0437 85.57158 7.03778 0.1864 51.342948 4.222668 0 0.11184 1.175 0.097 0.000 0.003 01.16.2019 PLIP3-OX #1 0.0258 98.04684 22.71287 0.0709 9.4297 58.828104 13.62772 0.04254 5.65782 2.280 0.528 0.002 0.219 01.16.2019 PLIP3-OX #2 0.0263 121.65531 60.86976 0.1904 32.6805 72.993186 36.52186 0.11424 19.6083 2.775 1.389 0.004 0.746 01.16.2019 #34 0.0295 119.75275 53.54001 0.1288 11.493 71.85165 32.12401 0.07728 6.8958 2.436 1.089 0.003 0.234 01.16.2019 #35 0.0322 39.81831 86.36901 0.096 1.736 23.890986 51.82141 0.0576 1.0416 0.742 1.609 0.002 0.032 01.16.2019 #36 0.032 58.09106 24.98605 0.0224 0.6661 34.854636 14.99163 0.01344 0.39966 1.089 0.468 0.000 0.012 01.16.2019 #37 0.0349 25.24707 13.36276 0.2349 0.061 15.148242 8.017656 0.14094 0.0366 0.434 0.230 0.004 0.001 01.16.2019 #38 0.0133 74.26895 34.82444 0.0577 9.6309 44.56137 20.89466 0.03462 5.77854 3.350 1.571 0.003 0.434 01.16.2019 #39 0.0485 74.82114 44.68281 0.2691 0.8187 44.892684 26.80969 0.16146 0.49122 0.926 0.553 0.003 0.010 02.15.2019 7.02 col-0 0.0364 195.79588 11.41731 0.0017 0.2282 117.477528 6.850386 0.00102 0.13692 3.227 0.188 0.000 0.004 02.15.2019 7.05 col-0 0.052 170.56108 3.81673 0.0034 0.013 102.336648 2.290038 0.00204 0.0078 1.968 0.044 0.000 0.000 02.15.2019 7.11 col-0 0.0302 110.3321 72.44077 1.3701 0.0961 66.19926 43.46446 0.82206 0.05766 2.192 1.439 0.027 0.002 02.15.2019 8.08 col-0 0.097 197.22041 12.21968 0.1039 0.1817 118.332246 7.331808 0.06234 0.10902 1.220 0.076 0.001 0.001 02.15.2019 7.02 PLIP3 0.0259 1281.42606 103.7661 0.2331 1.8158 768.855636 62.25968 0.13986 1.08948 29.686 2.404 0.005 0.042 02.15.2019 7.05 PLIP3 0.0299 695.74883 31.59615 0.05 10.6858 417.449298 18.95769 0.03 6.41148 13.962 0.634 0.001 0.214 02.15.2019 7.11 PLIP3 0.037 721.29385 29.99369 0.0396 8.8643 432.77631 17.99621 0.02376 5.31858 11.697 0.486 0.001 0.144 02.15.2019 8.08 PLIP3 0.0539 579.26309 96.22145 0.2413 11.3042 347.557854 57.73287 0.14478 6.78252 6.448 1.071 0.003 0.126 02.15.2019 #40 0.0345 95.4666 3.6102 0.0047 0.2765 57.27996 2.16612 0.00282 0.1659 1.660 0.063 0.000 0.005 02.15.2019 #41 0.05 1.44622 10.56269 0.0062 0.0662 0.867732 6.337614 0.00372 0.03972 0.017 0.127 0.000 0.001 02.15.2019 #42 0.0533 171.27429 60.49172 0.0824 7.8148 102.764574 36.29503 0.04944 4.68888 1.928 0.681 0.001 0.088 02.15.2019 #43 0.0385 38.73274 25.56334 0.0567 0.0159 23.239644 15.338 0.03402 0.00954 0.604 0.398 0.001 0.000 02.15.2019 #44 0.0183 49.69856 22.49813 0.1367 0.4806 29.819136 13.49888 0.08202 0.28836 1.629 0.738 0.004 0.016 02.15.2019 #45 0.02 14.01517 7.85793 0.0497 0.2807 8.409102 4.714758 0.02982 0.16842 0.420 0.236 0.001 0.008 02.15.2019 #46 0.047 1.1468 1.64607 0.0025 0.2523 0.68808 0.987642 0.0015 0.15138 0.015 0.021 0.000 0.003 02.15.2019 #47 0.0266 228.87213 140.6353 0.2343 5.3846 137.323278 84.3812 0.14058 3.23076 5.163 3.172 0.005 0.121 02.15.2019 #49 0.0375 318.36933 22.16193 0.0374 191.021598 13.29716 0.02244 0 5.094 0.355 0.001 0.000 02.15.2019 #51 0.0217 216.08026 84.97542 1.0423 0.2518 129.648156 50.98525 0.62538 0.15108 5.975 2.350 0.029 0.007 02.15.2019 #52 0.0204 210.45005 9.02369 0.0468 0.0624 126.27003 5.414214 0.02808 0.03744 6.190 0.265 0.001 0.002 02.20.2019 col-0 0.0483 226.37421 86.05322 0.5806 0.158 135.824526 51.63193 0.34836 0.0948 2.812 1.069 0.007 0.002 02.20.2019 col-0 0.0417 185.84344 9.89529 0.0209 0.126 111.506064 5.937174 0.01254 0.0756 2.674 0.142 0.000 0.002 02.20.2019 PLIP3-OX 0.0244 141.77249 119.5056 0.6842 39.5023 85.063494 71.70335 0.41052 23.70138 3.486 2.939 0.017 0.971 02.20.2019 PLIP3-OX 0.0168 61.05935 11.9301 0.0402 19.836 36.63561 7.15806 0.02412 11.9016 2.181 0.426 0.001 0.708 02.19.2019 #48 0.0167 158.39526 5.11149 0.1555 0.7013 95.037156 3.066894 0.0933 0.42078 5.691 0.184 0.006 0.025 02.19.2019 #50 0.015 26.37566 3.76101 0.0168 0.3355 15.825396 2.256606 0.01008 0.2013 1.055 0.150 0.001 0.013 02.20.2019 #53 0.0174 50.23476 7.7665 0.0094 0.0446 30.140856 4.6599 0.00564 0.02676 1.732 0.268 0.000 0.002 02.20.2019 #54 0.0296 95.25968 44.95067 0.1433 3.9033 57.155808 26.9704 0.08598 2.34198 1.931 0.911 0.003 0.079 02.20.2019 #55 0.001 10.02346 86.03486 0.0202 2.0665 6.014076 51.62092 0.01212 1.2399 6.014 51.621 0.012 1.240 02.20.2019 #56 0.0201 10.91416 14.27315 0.0491 0.1131 6.548496 8.56389 0.02946 0.06786 0.326 0.426 0.001 0.003 02.20.2019 #57 0.0192 29.99098 16.38774 0.015 3.6354 17.994588 9.832644 0.009 2.18124 0.937 0.512 0.000 0.114 02.20.2019 #58 0.0144 24.53178 4.77357 0.0009 2.3749 14.719068 2.864142 0.00054 1.42494 1.022 0.199 0.000 0.099 02.20.2019 #59 0.0102 22.90658 15.85929 0.0427 1.6005 13.743948 9.515574 0.02562 0.9603 1.347 0.933 0.003 0.094 02.20.2019 #60 0.0086 15.58063 6.64687 0.0208 0.6123 9.348378 3.988122 0.01248 0.36738 1.087 0.464 0.001 0.043 02.20.2019 #61 0.0247 14.71152 6.33016 0.0038 0.8354 8.826912 3.798096 0.00228 0.50124 0.357 0.154 0.000 0.020 02.20.2019 #62 0.0419 47.70498 108.3485 1.4091 0.1304 28.622988 65.00911 0.84546 0.07824 0.683 1.552 0.020 0.002 02.20.2019 #63 0.0095 8.07935 4.47188 0.0264 1.7166 4.84761 2.683128 0.01584 1.02996 0.510 0.282 0.002 0.108 02.20.2019 #64 0.0066 3.97369 5.83019 0.0009 0.2511 2.384214 3.498114 0.00054 0.15066 0.361 0.530 0.000 0.023 02.20.2019 #65 0.0185 12.79721 16.17166 0.0387 0.0375 7.678326 9.702996 0.02322 0.0225 0.415 0.524 0.001 0.001 02.20.2019 #66 0.0378 98.37841 221.5578 0.3563 14.2387 59.027046 132.9347 0.21378 8.54322 1.562 3.517 0.006 0.226 02.20.2019 #67 0.026 43.83424 0.85212 0.0001 0.1616 26.300544 0.511272 0.00006 0.09696 1.012 0.020 0.000 0.004 02.20.2019 #68 0.022 30.82893 18.02843 0.023 0.6749 18.497358 10.81706 0.0138 0.40494 0.841 0.492 0.001 0.018 02.20.2019 #69 0.0383 55.90096 80.16986 0.5281 0.1369 33.540576 48.10192 0.31686 0.08214 0.876 1.256 0.008 0.002 02.20.2019 #70 0.0257 17.80209 22.44234 0.0679 0.6237 10.681254 13.4654 0.04074 0.37422 0.416 0.524 0.002 0.015 02.20.2019 #71 0.0326 121.20451 30.66939 0.0853 3.1245 72.722706 18.40163 0.05118 1.8747 2.231 0.564 0.002 0.058 02.20.2019 #72 0.0367 56.49604 13.26511 0.0034 0.123 33.897624 7.959066 0.00204 0.0738 0.924 0.217 0.000 0.002 02.20.2019 #73 0.0301 54.19367 9.39152 0.0384 0.0552 32.516202 5.634912 0.02304 0.03312 1.080 0.187 0.001 0.001 04.03.2019 col-0 #1 0.0676 393.96 36.87 0.041 3.0506 236.376 22.122 0.0246 1.83036 3.497 0.327 0.000 0.027 04.03.2019 col-0 #2 0.0604 742.87 40.35 0.002 0.0615 445.722 24.21 0.0012 0.0369 7.380 0.401 0.000 0.001 04.03.2019 PLIP3-OX #1 0.0406 165.77 118.71 0.4093 20.3691 99.462 71.226 0.24558 12.22146 2.450 1.754 0.006 0.301 04.03.2019 PLIP3-OX #2 0.0692 383.63 101.38 0.6558 12.0709 230.178 60.828 0.39348 7.24254 3.326 0.879 0.006 0.105 04.03.2019 #74 0.0333 46.56 12.86 0.0432 2.8522 27.936 7.716 0.02592 1.71132 0.839 0.232 0.001 0.051 04.03.2019 #75 0.0371 283.17 33.86 0.0564 7.9707 169.902 20.316 0.03384 4.78242 4.580 0.548 0.001 0.129 04.03.2019 #76 0.0172 22.75 62.24 0.3109 0.6754 13.65 37.344 0.18654 0.40524 0.794 2.171 0.011 0.024 04.03.2019 #77 0.0483 68.17 508.19 2.2286 0.3205 40.902 304.914 1.33716 0.1923 0.847 6.313 0.028 0.004 04.03.2019 #78 0.039 71.38 389.48 0.4468 44.3897 42.828 233.688 0.26808 26.63382 1.098 5.992 0.007 0.683 04.03.2019 #79 0.041 59.6 52.58 0.2281 0.6124 35.76 31.548 0.13686 0.36744 0.872 0.769 0.003 0.009 04.03.2019 #80 0.0318 103.9 389.91 0.7571 47.722 62.34 233.946 0.45426 28.6332 1.960 7.357 0.014 0.900 04.03.2019 #81 0.0724 10.1 10.21 0.0621 0.1091 6.06 6.126 0.03726 0.06546 0.084 0.085 0.001 0.001 04.03.2019 #82 0.0426 95.27 104.57 0.3654 1.7009 57.162 62.742 0.21924 1.02054 1.342 1.473 0.005 0.024 04.03.2019 #83 0.0422 7.08 23.5 0.4786 0.1953 4.248 14.1 0.28716 0.11718 0.101 0.334 0.007 0.003 04.03.2019 #84 0.0483 62.02 90.9 0.1665 6.8208 37.212 54.54 0.0999 4.09248 0.770 1.129 0.002 0.085 04.03.2019 #85 0.024 76.25 46.77 0.1564 20.2348 45.75 28.062 0.09384 12.14088 1.906 1.169 0.004 0.506 04.03.2019 #86 0.0432 80.79 64.77 0.2539 1.5846 48.474 38.862 0.15234 0.95076 1.122 0.900 0.004 0.022 04.03.2019 #87 0.0406 84.62 112.39 0.449 2.5267 50.772 67.434 0.2694 1.51602 1.251 1.661 0.007 0.037 04.03.2019 #88 0.0461 68.78 144.3 0.6818 0.3611 41.268 86.58 0.40908 0.21666 0.895 1.878 0.009 0.005 04.03.2019 #89 0.0527 310.63 81.48 0.1828 0.0977 186.378 48.888 0.10968 0.05862 3.537 0.928 0.002 0.001 04.03.2019 #90 0.0626 187.02 34.72 0.247 0.1475 112.212 20.832 0.1482 0.0885 1.793 0.333 0.002 0.001 Table 3.2. Hormone data collected from plants selected in primary screen. Light green highlights represent plants determined to meet the criteria for passing the hormone-based secondary screen. 73 chr pos ref alt mutation_effect gene At_num CDS_change protein_change EMS_mut.ref EMS_mut.alt EMS_wt.ref EMS_wt.alt notes insertional lines homozygous apparently fine (Rohr et al 2019, plant phys), and 5 22400218 C T stop_gained TIG AT5G55220 1366C>T Gln456* 4 63 42 29 homozygous line SALK_089907C apparently available SALK_089907C vln5 SAIL_512_F03 and GABI_225F09 homozygous mutants 5 23213868 C T missense_variant VLN5 AT5G57320 721C>T Pro241Ser 1 57 35 12 were fertile (zhang 2010) CS863116 (homoSAIL) 5 24883408 C T missense_variant&splice_region_variant ML1 AT5G61960 76G>A Glu26Lys 2 59 46 24 mutant viable and fertile according to Anderson 2005 SALK_015088C 5 24914071 C T stop_gained ARF2 AT5G62000 2146C>T Gln716* 3 58 40 18 early flowers infertile, later flowers fertile (Okushima 2005) CS24602 mutation looks less likely to cause problems (Ala->Val). No publications on this gene specifically. Predictions: Plant invertase/pectin methylesterase inhibitor superfamily protein; plastid-localized; there is line SALK_134445C, which is 5 25037961 C T missense_variant AT5G62350 AT5G62350 458C>T Ala153Val 0 56 38 21 homozygous SALK_134445C hard to find papers on this, because it is used in many studies as a housekeeping gene for looking at expression. "Phenotype curated by ABRC: Left-handed helical growth in the root and other rapidly elongating organs. Strong severity of defects in cell expansion, cytokinesis, and vascular development." This does 5 25182210 C T missense_variant TUBB2 AT5G62690 563C>T Ser188Phe 0 44 42 17 not resemble my mutant CS68678 homozygous sterile, both male and female "We examined the reproductive development of these mutants and found that blap75 sterility is due to abortion of male and female gametophytes (data not shown)" (Chelysheva 2008). Sterility detailed more in Bonnet 2013. This would not explain rescue by MeJA application. However, stop codon is in the essential DUF domain (Bonnet 2013), so this mutant is expected to be sterile SALK_093589 5 25444029 C T stop_gained RMI1 AT5G63540 345G>A Trp115* 0 65 46 23 here, so I don't know what to think SALK_093519 AKA HEN3. no mention of infertility in Wang 2004, although shorter siliques were described for mutant. Floral development 5 25464611 C T stop_gained CDKE-1 AT5G63610 447G>A Trp149* 1 59 39 21 gene. Homozygous line SALK_072781C exists SALK_072781C Mg transporter essential for pollen development (Chen 2009). Homozygous lethal. Very possible that R->K change in residue 5 25807197 C T missense_variant MRS2-2 AT5G64560 1103G>A Arg368Lys 0 49 48 17 368 out of 394 does not cause loss of function SALK_030174 (het) Acyl-CoA oxidase 2; used in catabolism of long-chain FAs in peroxisome, no problems germinating or setting seed for acx2-1 null mutant, and wound-induced JA response is not SALK_030578C 5 26011654 C T missense_variant ACX2 AT5G65110 601G>A Ala201Thr 2 50 42 17 compromised either (Pinfield-Wells 2005) SALK_006464-het aka Chloroplast RNA Editing Factor 7 (CREF7). Mutant strain SALK_078415 "displayed no aberrant visible phenotype" (Yagi 5 26552651 C T missense_variant PCMP-H61 AT5G66520 773C>T Ala258Val 2 64 40 27 2013) SALK_078415C 519-residue "transmembrane protein". SALK_016436C (homozygous) exists, but this insertion is ~100bp upstream of start codon. High-throughput phenotyping shows reduced tolerance of cold and oxidative stress for SALK_016436C, and no SALK_016436C change for heat, osmotic, NaCl, ABA, or hypoxia stress tolerance WiscDsLoxHs015_03G 5 26689989 C T missense_variant AT5G66820 AT5G66820 439C>T Pro147Ser 1 54 40 23 (Luhua 2013) (CS901367) Mn transporter? Vacuole localized. nramp4-1 null mutant 5 26863128 C T missense_variant NRAMP4 AT5G67330 1087C>T Leu363Phe 2 55 34 18 "displays no obvious phenotype" (Lanquar 2005) CS859760 5 23247163 C T missense_variant VIN3 AT5G57380 1238G>A Gly413Glu 5 56 33 32 CS875198 het 5 23905083 C T missense_variant AT5G59250 AT5G59250 592C>T Leu198Phe 6 67 51 21 HP59; PSUT; PLASTIDIC SUGAR TRANSPORTER CS67300 CS416473 (5'utr, not 5 23925623 C T synonymous_variant LTP4 AT5G59310 150G>A Pro50Pro 5 60 36 16 homo) 5 25166065 C T missense_variant NPF2.11 AT5G62680 1216G>A Gly406Arg 3 64 40 26 GLUCOSINOLATE TRANSPORTER-2 SALK_052811C CS863998 (WISCDSLOX297300_11N 5 26474855 C T missense_variant AT5G66270 AT5G66270 170G>A Arg57Gln 10 48 51 24 Zinc finger C-x8-C-x5-C-x3-H type family protein ) SALK_127997C, 5 26947794 C T missense_variant AT5G67550 AT5G67550 826G>A Glu276Lys 4 47 39 14 transmembrane protein SALK_052167 (het) Table 3.3. Outputs from SIMPLE following genomic sequencing of sup11, showing possible causative mutations that co-segregate with suppression phenotype. A list of insertional mutants was also included here, for crossing to sup11 to identify the causal suppressor mutation. 74 chr pos ref alt mutation_effect gene At_num CDS_change protein_change EMS_mut.ref EMS_mut.alt EMS_wt.ref EMS_wt.alt 1 17046048 C T upstream_gene_variant AT1G45100 AT1G45100 -4583C>T 0 1 8 5 1 26401359 G A missense_variant GATL9 AT1G70090 433G>A Val145Ile 2 55 41 25 1 26901261 G A missense_variant CEL3 AT1G71380 415C>T Pro139Ser 2 26 32 19 1 28464675 G A missense_variant CLV1 AT1G75820 1978C>T Arg660Cys 5 56 41 25 1 29034238 G A missense_variant AT1G77280 AT1G77280 322C>T Leu108Phe 3 32 47 21 1 29071522 G A missense_variant AT1G77350 AT1G77350 346G>A Glu116Lys 5 47 46 27 1 29108381 C T missense_variant AT1G77460 AT1G77460 3541C>T Leu1181Phe 2 50 42 17 1 29751567 C T stop_gained AT1G79090 AT1G79090 366G>A Trp122* 1 47 44 18 1 30012292 C T missense_variant DTA4 AT1G79760 239C>T Thr80Ile 1 27 38 19 1 30043481 C T missense_variant ROPGEF12 AT1G79860 529G>A Asp177Asn 0 47 31 26 1 30244604 C T missense_variant VQ11 AT1G80450 85G>A Val29Ile 2 38 30 16 Table 3.4. Outputs from SIMPLE following genomic sequencing of sup12, showing possible causative mutations that co-segregate with suppression phenotype. 75 chr pos ref alt mutation_effect gene At_num CDS_change protein_change EMS_mut.ref EMS_mut.alt EMS_wt.ref EMS_wt.alt ratio 2 1343761 C T splice_region_variant&intron_variant AT2G04060 AT2G04060 463+7G>A 2 34 40 22 0.589606 2 1600559 C T downstream_gene_variant AT2G04570 AT2G04570 *4430C>T 1 37 28 15 0.624847 2 2895400 C T missense_variant HEN2 AT2G06990 266C>T Ser89Phe 0 37 29 25 0.537037 2 4539413 C T upstream_gene_variant AT2G11405 AT2G11405 -3762C>T 0 43 28 19 0.595745 2 368009 C T missense_variant AHK4 AT2G01830 8G>A Arg3Lys 9 41 34 20 0.44963 2 466905 C T upstream_gene_variant AT2G01990 AT2G01990 -408G>A 4 40 31 21 0.505245 2 935961 C T missense_variant AT2G03110 AT2G03110 448C>T Pro150Ser 2 25 31 13 0.630471 2 2342483 GA G upstream_gene_variant AT2G06020 AT2G06020 -51delA 10 7 20 0 0.411765 2 2992124 C T upstream_gene_variant AT2G07210 AT2G07210 n.-22G>A 0 36 22 21 0.511628 2 3161788 C T non_coding_exon_variant AT2G07550 AT2G07550 n.3722G>A 0 32 29 11 0.725 2 3232139 C T upstream_gene_variant AT2G07660 AT2G07660 n.-439G>A 0 40 26 20 0.565217 2 3434738 T G missense_variant AT2G07721 AT2G07721 143A>C Glu48Ala 2 3 7 0 0.6 2 4595941 G A upstream_gene_variant AT2G11465 AT2G11465 n.-507G>A 21 26 39 0 0.553191 2 4989494 C T upstream_gene_variant AT2G12380 AT2G12380 n.-136G>A 1 35 24 22 0.493961 2 6341031 G A upstream_gene_variant AT2G14780 AT2G14780 n.-334C>T 13 17 51 0 0.566667 2 6589845 C T missense_variant AT2G15180 AT2G15180 323G>A Gly108Glu 13 24 41 0 0.648649 2 7394275 C T missense_variant AT2G17010 AT2G17010 1309G>A Asp437Asn 22 14 43 0 0.388889 2 7406858 C T upstream_gene_variant anac036 AT2G17040 -265C>T 20 31 76 0 0.607843 2 7411519 C T missense_variant AT2G17050 AT2G17050 3595G>A Val1199Met 17 15 58 0 0.46875 2 14770384 C T missense_variant AT2G35050 AT2G35050 677C>T Pro226Leu 6 20 38 19 0.435897 1 11267234 G A missense_variant SGR2 AT1G31480 634G>A Ala212Thr 26 18 49 0 0.409091 2 368009 C T missense_variant AHK4 AT2G01830 8G>A Arg3Lys 9 41 34 20 0.44963 2 935961 C T missense_variant AT2G03110 AT2G03110 448C>T Pro150Ser 2 25 31 13 0.630471 2 2895400 C T missense_variant HEN2 AT2G06990 266C>T Ser89Phe 0 37 29 25 0.537037 2 6589845 C T missense_variant AT2G15180 AT2G15180 323G>A Gly108Glu 13 24 41 0 0.648649 2 7411519 C T missense_variant AT2G17050 AT2G17050 3595G>A Val1199Met 17 15 58 0 0.46875 2 13754645 C T missense_variant GLR3.7 AT2G32400 1240G>A Val414Ile 9 26 37 19 0.403571 2 14770384 C T missense_variant AT2G35050 AT2G35050 677C>T Pro226Leu 6 20 38 19 0.435897 4 10665994 C T missense_variant AT4G19570 AT4G19570 479C>T Ala160Val 28 22 47 0 0.44 4 14063045 C T missense_variant AT4G28450 AT4G28450 732G>A Met244Ile 22 18 49 0 0.45 4 15648149 C T missense_variant CYP95 AT4G32420 2126G>A Arg709Lys 20 20 44 0 0.5 Table 3.5. Outputs from SIMPLE following genomic sequencing of sup53, showing possible causative mutations that co-segregate with suppression phenotype. 76 change mut.ref mut.alt wt.ref wt.alt ratio notes AT5G01010 retinal-binding protein Leu112Leu 20 37 64 11 0.502456 ARABIDOPSIS THALIANA MYB DOMAIN PROTEIN 3R5, ATMYB3R5, MYB mutant in 2017 Chen Nature paper is SALK_031972. MYB3R5 is transcriptional repressor. AT5G02320 DOMAIN PROTEIN 3R-5, MYB3 -4181G>A (upstream variant) 16 27 56 3 0.57706 Myb3r5 mutant continues to grow during zeocin treatment (inducer of double-stranded breaks) AT5G02502 Oligosaccaryltransferase; OST4B -1G>A (splice region variant) 17 28 34 6 0.472222 part of protein glycosylation complex? Encodes KEEP ON GOING (KEG), a RING E3 ligase involved in abscisic acid signaling. KEG is essential for Arabidopsis growth and development. ABA promotes KEG degradation via the ubiquitin dependent 26S proteasome KEG is essential for development past the seedling stage. Full protein is 1625 residues. Mutation AT5G13530 pathway 4369C>T His1457Tyr 11 38 43 4 0.690404 H1457Y is in Herc2-like repeat. AT5G12850 TANDEM ZINC FINGER 8, TZF8 185C>T Ser62Phe 15 46 54 8 0.625066 excluded from nucleus (Koroleva 2004 plant journal); otherwise not much info AT5G10220 ANN6, ANNAT6, ANNEXIN 6, ANNEXIN ARABIDOPSIS THALIANA 6 823G>A Glu275Lys 15 39 47 15 0.480287 annexins are Ca-dependent membrane binding proteins involved in signaling AT5G06830 hypothetical protein -416G>A 20 46 50 10 0.530303 no info in literature AT5G03370 acylphosphatase family -4271G>A 15 31 52 9 0.526372 don't know AT5G03340 ATCDC48C, CELL DIVISION CYCLE 48C -1646G>A 27 41 51 7 0.482252 "Critical Roles in Cell Division, Expansion, and Differentiation" 2008 Park Plant Physiology AT5G02290 NAK, PBL11, PBS1-LIKE 11 1006G>A Asp336Asn 27 40 54 10 0.440765 kinase, involved in signaling AT5G01950 Leucine-rich repeat protein kinase family protein 1642+3G>A 20 39 60 18 0.430248 no info in literature AT5G01100 FRB1, FRIABLE 1 430G>A Gly144Arg 26 30 59 10 0.390787 important for cell adhesion (fucosyltranferase?). Mutants have very crumpled appearance. AT5G17090 Cystatin/monellin superfamily protein 322G>A Glu108Lys 15 28 50 11 0.470835 no info in literature AT5G26010 -873C>T 15 45 41 10 0.553922 AT2G33470 GLTP1 -2993_-2992insT 10 5 22 0 0.333333 glycosphingolipid transfer protein AT5G05480 1136G>A Gly379Glu 20 29 41 9 0.411837 AT5G05560 3133-21C>T 13 39 47 16 0.496032 AT5G06150 532G>A Ala178Thr 17 35 38 14 0.403846 AT5G07740 *278G>A 13 27 32 7 0.495513 AT5G12980 NOT9B 550G>A Glu184Lys 19 25 53 10 0.409452 negative regulation of translation? AT5G13020 *90G>A 20 38 39 12 0.419878 AT5G13050 -3740G>A 9 26 32 11 0.487043 -2932_-2931delAT 24 8 27 0 0.25 AT5G13260 -2589_-2588delAT 9 13 14 3 0.414439 AT5G13470 *4686C>G 22 45 48 14 0.445835 AT5G13480 905G>A Ser302Asn 11 22 30 14 0.348485 AT5G15050 AtGlcAT14B 744G>A Trp248* 15 30 41 14 0.412121 glucuronosyltransferase, see Dilokpmiol 2014 Plant Signaling and Behavior AT5G15060 Lateral organ boundaries (LOB) family protein 343G>A Val115Ile 11 27 31 8 0.505398 LOB domain is associated with transcription factors AT5G17710 *68G>A 13 45 35 13 0.505029 AT5G59620 n.-2330G>T 3 5 14 1 0.558333 Table 3.6. Outputs from SIMPLE following genomic sequencing of sup72, showing possible causative mutations that co-segregate with suppression phenotype. 77 REFERENCES 1. Liu, J., et al., Connecting research and teaching introductory cell and molecular biology using an Arabidopsis mutant screen. Biochem. Mol. Biol. Educ., 2021. 49(6): p. 926-934. 2. Koo, A.J., J.B. Ohlrogge, and M. Pollard, On the export of fatty acids from the chloroplast. J. Biol. Chem., 2004. 279(16): p. 16101-10. 3. Li, N., et al., FAX1, a novel membrane protein mediating plastid fatty acid export. PLoS Biol., 2015. 13(2): p. e1002053. 4. Tian, Y., et al., FAX2 mediates fatty acid export from plastids in developing Arabidopsis seeds. Plant Cell Physiol., 2019. 60(10): p. 2231-2242. 5. Li, N., et al., Two plastid fatty acid exporters contribute to seed oil accumulation in Arabidopsis. Plant Physiol., 2020. 182(4): p. 1910-1919. 6. Bugaeva, W., et al., Plastid fatty acid export (FAX) proteins in Arabidopsis thaliana-the role of FAX1 and FAX3 in growth and development. bioRxiv, 2023. DOI: https://doi.org/10.1101/2023.02.09.527856 7. Cook, R., J. Lupette, and C. Benning, The role of chloroplast membrane lipid metabolism in plant environmental responses. Cells, 2021. 10(3): p. 706. 8. Ruan, J., et al., Jasmonic acid signaling pathway in plants. Int. J. Mol. Sci., 2019. 20(10): p. 2479. 9. Guan, L., et al., JASSY, a chloroplast outer membrane protein required for jasmonate biosynthesis. Proc. Natl. Acad. Sci. U.S.A., 2019. 116(21): p. 10568-10575. 10. Ellinger, D., et al., DONGLE and DEFECTIVE IN ANTHER DEHISCENCE1 lipases are not essential for wound- and pathogen-induced jasmonate biosynthesis: redundant lipases contribute to jasmonate formation. Plant Physiol., 2010. 153(1): p. 114-27. 11. Wang, K., et al., A Plastid Phosphatidylglycerol Lipase Contributes to the Export of Acyl Groups from Plastids for Seed Oil Biosynthesis. Plant Cell, 2017. 29(7): p. 1678-1696. 12. Wang, K., et al., Two Abscisic Acid-Responsive Plastid Lipase Genes Involved in Jasmonic Acid Biosynthesis in Arabidopsis thaliana. Plant Cell, 2018. 30(5): p. 1006-1022. 78 13. Wachsman, G., et al., A SIMPLE pipeline for mapping point mutations. Plant Physiol., 2017. 174(3): p. 1307-1313. 14. Stone, S.L., et al., KEEP ON GOING, a RING E3 ligase essential for Arabidopsis growth and development, is involved in abscisic acid signaling. Plant Cell, 2006. 18(12): p. 3415- 3428. 15. Pauwels, L., et al., The ring e3 ligase keep on going modulates jasmonate zim-domain12 stability. Plant Physiol., 2015. 169(2): p. 1405-1417. 16. Wawrzynska, A., et al., Powdery Mildew Resistance Conferred by Loss of the ENHANCED DISEASE RESISTANCE1 Protein Kinase Is Suppressed by a Missense Mutation in KEEP ON GOING, a Regulator of Abscisic Acid Signaling Plant Physiol., 2008. 148(3): p. 1510-1522. 17. Guo, Q., et al., JAZ repressors of metabolic defense promote growth and reproductive fitness in Arabidopsis. Proc. Natl. Acad. Sci. U.S.A., 2018. 115(45): p. E10768-E10777. 18. Liu, H. and S.L. Stone, Abscisic acid increases Arabidopsis ABI5 transcription factor levels by promoting KEG E3 ligase self-ubiquitination and proteasomal degradation. Plant Cell, 2010. 22(8): p. 2630-2641. 19. Liu, H. and S.L. Stone, Cytoplasmic degradation of the Arabidopsis transcription factor abscisic acid insensitive 5 is mediated by the RING-type E3 ligase KEEP ON GOING. J. Biol. Chem., 2013. 288(28): p. 20267-20279. 20. Gu, Y. and R.W. Innes, The KEEP ON GOING protein of Arabidopsis recruits the ENHANCED DISEASE RESISTANCE1 protein to trans-Golgi network/early endosome vesicles. Plant Physiol., 2011. 155(4): p. 1827-1838. 21. Ng, S., et al., Cyclin-dependent kinase E1 (CDKE1) provides a cellular switch in plants between growth and stress responses. J. Biol. Chem., 2013. 288(5): p. 3449-3459. 79 CHAPTER 4: Analysis, conclusions, and perspectives 80 Introduction A better understanding of key stages in plant lipid biosynthesis was sought by focusing on two fundamental yet incompletely elucidated processes in Arabidopsis: phosphatidic acid (PA) metabolism and fatty acid (FA) export in the chloroplast. A deeper study of chloroplast lipid phosphate phosphatases (LPPs) revealed that LPPγ and LPPε1 are involved in the processing of ER-derived PA, and that the enzyme responsible for plastid-derived PA dephosphorylation remains unknown. Meanwhile, a suppressor screen in the PLIP3-OX background targeted mutants with a decreased capacity for converting the plastid-derived FA derivative 12-oxo-phytodienoic acid (OPDA) to jasmonic acid (JA) in the cytosol, with the goal of attaining mutants deficient in OPDA export from chloroplasts. While such mutants have yet to be identified, mutations in KEG and CDK8 resulted in phenotypic suppression of PLIP3-OX, which should provide further insight into the gene products’ roles in the JA-responsive transcriptomic network. Chloroplast LPPs and PA Localizations and redundancies of LPPγ, LPPε1, LPPε2, and LPPα2 As discussed in chapter 2, in the context of published works [1-3], our chloroplast import data, complementation tests, fatty acid radiolabeling data, and redundancy between LPPγ and LPPε1 all indicate localization of LPPγ and LPPε1 to the chloroplast outer envelope and LPPε2 localization to the inner envelope or thylakoids. Weak import of LPPε1 was observed in the import assay presented in chapter 2, which may point to dual-localization of LPPε1. Similarly, closer observation of Figure 11A in Nguyen 2023 shows a faint band of Venus-tagged LPPε1 at lower molecular weight, which is digested by trypsin and not thermolysin [2]. This may be the small portion of LPPε1-Ven that is fully processed and located at the inner envelope, with the Venus reporter exposed to trypsin. If LPPε1 is indeed also present at the inner envelope, it may have functional redundancy with LPPε2, which would be distinct from its redundant activity with LPPγ. Interestingly, LPPε1 has also been shown to act redundantly with the ER-localized LPPα2, a phenomenon that was rationalized by associating LPPε1 activity in the chloroplast outer envelope with sites of lipid exchange with the ER [2]. This model was supported by confocal imaging data showing fluorescent-tagged LPPε1 signals emanating from subdomains of the outer envelope in 81 close proximity to the ER [2]. However, it should be noted that LPPγ was still present in the unviable lppε1 lppα2 mutant, and so was not able to compensate for the absence of LPPε1. LPPε1 activity at the outer envelope is therefore only partially redundant with that of LPPγ, likely as a result of differential distribution of these enzymes, with just LPPε1 substantially present at contact sites with the ER. Metabolic roles of LPPγ and LPPε1, and implications for the ER pathway of galactolipid metabolism While it has been demonstrated that redundant PA phosphatase (PAP) activity of LPPγ and LPPε1 contributes to the ER pathway of galactolipid biosynthesis, it was also noted that the decreased flux through the pathway in lppγ lppε1 is relatively mild: it is not sufficient to affect the acyl compositions of major galactolipids, and was only directly discernable in pulse-chase 14C-labeling experiments of fatty acids. The decreased viability of lppγ lppε1 ats1-1 also supports involvement of LPPγ and LPPε1 in the ER pathway, as the stymied supply of lipids from the plastid pathway compounds the negative effects of a disrupted ER pathway [4-6]. The modest metabolic phenotype of lppγ lppε1 indicates that alternative sources of ER-derived DAG exist. Soluble PA phosphatases PAH1 and PAH2 are known to be involved in the ER pathway, and their presence in the cytosol could provide them access to PA at the chloroplast outer envelope [7]. The pah1 pah2 double mutant has a comparable lipid phenotype to lppγ lppε1 with respect to galactolipid acyl composition and ER pathway fluxes, although plant development is not stunted. A different study reported PAH1 and PAH2 localization to the ER, which would implicate DAG as a mobile lipid between the ER and outer envelope, possibly in addition to PA [8]. However, it is possible that overexpression led to mis-localization of PAH1 and PAH2 in this study, as the phenomenon of over-produced outer envelope proteins accumulating in the ER has been shown for TGD4 [5, 9]. In either case, in lppγ lppε1, PAH1 and PAH2 continue to provide ER- derived DAG substrates for galactolipid biosynthesis from PA located at either the ER or cytosolic leaflet of the outer envelope. Pursuit of a quadruple lppγ lppε1 pah1 pah2 mutant could be informative in determining whether additional DAG-producing enzymes exist in the ER pathway. 82 If LPPγ, LPPε1, PAH1, and PAH2 are indeed together the main sources of DAG in the ER pathway, it would follow that the TGD complex imports DAG, rather than PA, into the inner envelope. Consequently, PA dephosphorylation at the inner envelope would be exclusive to the plastid pathway. This could explain why the rbl10 mutant is deficient in PA dephosphorylation in the plastid pathway, despite retaining high PAP activity in mixed envelopes isolated from chloroplasts [10]. If the four aforementioned PAPs are not the primary sources of DAG, and PA is the imported lipid species, it would mean that an unidentified, RBL10-independent enzyme dephosphorylates ER- derived PA at the inner envelope. As previously hypothesized in Lavell 2019, this PAP could be active at the intermembrane-facing leaflet of the inner envelope, and RBL10 may directly or indirectly facilitate plastid PA flipping to this leaflet for subsequent dephosphorylation [10]. Alternatively, there could be a separate RBL10-dependent PAP that acts on PA at the stromal- facing leaflet. Regarding the TGD complex, if PA is imported, binding of PA by TGD2 and TGD4 could be explained simply as substrate binding to subunits of the import complex. A different explanation for PA binding would be required in the case of a DAG substrate, which introduces the possibility of allosteric regulation by PA. Regulatory roles of LPPγ, LPPε1, and chloroplast PA The lack of growth inhibition in pah1 pah2, despite its similar lipid phenotype to lppγ lppε1, supports the hypothesis in which LPPγ and LPPε1 draw from a distinct PA pool, which is at the inner leaflet of the chloroplast outer envelope membrane. TGD4 is present in this membrane, and binds PA at its cytosolic-facing N-terminal domain [9, 11]. We hypothesized that if TGD4 is responsible for PA transfer across the outer envelope, then crossing of tgd4-1 to lppγ lppε1 would suppress growth inhibition caused by PA at the inner leaflet. However, the lppγ lppε1 tgd4-1 triple mutants remained small, indicating that either PA import is not sufficiently hindered in tgd4-1, that TGD4 binds PA only as part of the transfer from the ER to the outer leaflet, or that PA binding is regulatory and not a substrate interaction. As an allosteric regulator, one would expect PA binding to activate TGD4, as PA is known to accumulate outside of the plastid in mutants deficient 83 in TGD4, as well as ER pathway proteins TGD1, PAH1, and PAH2 [4, 7, 9]. Crossing of lppγ lppε1 to the more severe mutant alleles tgd4-2 or tgd4-3, and checking for phenotypic suppression, may provide confirmation on whether TGD4 supplies PA to the inner leaflet of the outer envelope. PA is also known to be bound to TGD2 at its C-terminal domain, which is thought to extend into the intermembrane space while the N terminus is anchored at the chloroplast inner envelope [12, 13]. The PA bound near the C-terminus is presumably at the inner leaflet of the outer envelope membrane, belonging to the same PA pool utilized by LPPγ and LPPε1. If DAG is the substrate of TGD2, and PA binds allosterically, PA may have a regulatory role as speculated for binding to TGD4. The presumed increase in this PA pool in lppγ lppε1 may affect the function of the TGD complex. Any such effect is unlikely to be the reason for growth inhibition in lppγ lppε1, as severe disruption of the complex in tgd1-1 does not result in the phenotype, nor does crossing tgd1-1 to lppγ lppε1 suppress it. The growth inhibition caused by excess PA at the inner leaflet of the outer envelope membrane is therefore distinct from its association with the TGD complex. The pathways or mechanisms by which PA affects growth from the intermembrane-facing leaflet of the chloroplast outer envelope are not known, despite being shown both in lppγ lppε1 mutants and in transgenic lines where DAG kinase is targeted to the intermembrane space [3]. Salicylic acid signaling does not appear to be involved, and lppγ lppε1 plants do not resemble the JA- induced morphologies of PLIP-OX lines or dgd1 [14-16]. Hormone profiling of lppγ lppε1 may prove to be valuable in identifying relevant signaling pathways. Moreover, the suppressor mutant screen applied to lppγ lppε1 is expected to identify novel factors affecting growth regulation in the double mutant. We may discover mutants in PA trafficking to the inner leaflet of the outer envelope, which could contribute to our understanding of membrane lipid metabolism. Other mutants may uncover components that link the altered membrane composition to broader signaling pathways, or reveal novel regulatory PA-dependent factors that would provide broader insights into plant growth regulation. These discoveries should in turn provide more context for the regulatory roles of LPPγ and LPPε1, which may act as regulators through their modulation of this PA pool. 84 The role of LPPε2 is unknown It was shown that LPPε2 is located at the chloroplast inner envelope or thylakoids, and that its catalytic activity is that of a PAP equivalent to LPPγ and LPPε1. However, an aberrant phenotype of lppε2 has not been observed, and the same is true for lppε1 lppε2, in which potential redundancy is accounted for. It is unlikely that LPPε2 would be retained during evolution in absence of a metabolic or physiological purpose, and further study would require testing of more diverse environmental conditions. As chloroplasts are central to substantial portions of both metabolism and signaling, any biotic or abiotic stresses are appropriate as challenges. In fact, unpublished preliminary data from the David Kramer lab has shown a decrease in non- photochemical quenching efficiency in lppε2 under fluctuating light conditions and elevated temperature, which may provide a direction for future research. While LPPε2 has been detected in leaf chloroplasts [1], it is also possible that its primary function is in other tissues, such as roots, flowers, or seeds. An approach targeting characterization of these tissues in lppε2 and lppε1 lppε2 may also prove fruitful. The PA phosphatase of the plastid pathway is unknown The plastid pathway is dependent on PAP activity at the chloroplast inner envelope, and appears to be largely dependent on RBL10 [10]. Acyl group radiolabeling on isolated chloroplasts from lppγ lppε1 lppε2 revealed that none of the three known chloroplast LPPs are involved in the plastid pathway. Because there is some residual plastid pathway-derived MGDG in the rbl10 mutant, it is possible that the RBL10-dependent PAP activity is actually completely abolished, and weak LPPε2 or LPPε1 activity provides minimal compensation. A cross of rbl10 to lppε2 or lppε1 lppε2 would be useful in determining whether this is the case, and thus whether the primary plastid pathway PAP is partially or entirely dependent on RBL10. One possibility that had been previously discussed is that RBL10 itself is the plastid pathway PAP. This was discounted because mixed envelopes from rbl10 retain PAP activity, and it was concluded that substrate access by the PAP was deficient in the mutant rather than the phosphatase itself [10]. However, our results show that chloroplast LPPs, which are all present in rbl10, would be expected to remain active and possibly obscure the effects of a missing plastid pathway PAP in a 85 mixed envelope assay. PAP activity assays on separated envelopes would therefore be more informative, in both various lpp mutants, rbl10, and crosses between them. These results would clarify if the inner envelope PAP activity is dependent on RBL10, and whether it overlaps with some LPP activity. However, this experiment cannot directly implicate RBL10 as a PAP, and separate PA phosphatase assays on the RBL10 protein itself would be needed to address this question. It should be noted that protease activity has not been demonstrated for Arabidopsis RBL1, RBL10, RBL11, nor RBL12, and only witnessed in RBL2 [17-21]. While this is possibly just due to unique specificities for protein substrates, the case may also be that some plant rhomboid- like enzymes hydrolyze lipids rather than proteins. Another possibility, though unlikely, is that ATS1 has dual function as an acyltransferase and PA phosphatase. This is hypothesized because the ats1-1 mutant is more deficient in plastid-derived MGDG than plastid-derived PG, which does not require PA dephosphorylation [22]. As shown in chapter 2, in ats1-1 more plastid PA is allocated to PG relative to MGDG than in Col-0. Therefore, PAP activity in the plastid pathway is lower in ats1-1, in addition to the decreased acyltransferase activity. A simple way to explain this would be that ATS1 is also the lipid phosphatase, which would not be unprecedented: Arabidopsis GPAT4 and GPAT6 have lyso-PA phosphatase activity in addition to their acyltransferase activity in the cytoplasm [23]. Finally, according to a preliminary analysis using InterPro, ATS1 and ATS2 share a similar C-terminal acyltransferase domain, while ATS1 has an additional N-terminal alpha-helical bundle, the purpose of which is not known [24]. This possibility can be addressed by in vitro testing of ATS1 for PAP activity, and its dependence on the N-terminal domain. If ATS1 is indeed the plastid pathway PAP, an explanation for the PAP dependence on RBL10 would require further investigation. PLIP3-OX suppressor screen Implications for KEG mutant suppression of PLIP3-OX In chapter 3, the candidacy of a mutation in KEG for causing phenotypic suppression of PLIP3-OX in sup72 was discussed, along with possible mechanisms of suppression. KEG is a known repressor in the abscisic acid (ABA) pathway, which targets the transcriptional activator ABI5 for degradation in the absence of ABA [25, 26]. An equivalent role in the repression of the JA response 86 is also possible, as KEG has been shown to bind and stabilize the JA response repressor JAZ12 [27]. Because KEG-mediated JAZ12 stabilization and ABI5 degradation both depend on cytosolic interactions, it is hypothesized that the mutation in sup72 leads to increased KEG presence in the cytosol, thereby dampening the JA response. This effect would resemble that of the keg-4 mutant, in which an increased presence of KEG in the cytosol attenuates ABA sensitivity [28, 29]. It is also possible that instead, the KEG mutation in sup72 directly affects its binding and stabilization of JAZ12. In KEG, both the JAZ12 interaction and the trans-golgi network (TGN) sequestration away from the cytosol are dependent on its C-terminal HERC domain [27-29]. Because the KEG H1457Y mutation in sup72 is in the HERC domain, each of the two mechanisms of PLIP3-OX phenotypic suppression is a possibility. In either case, it is likely that phenotypic suppression of PLIP3-OX by the H1457Y mutation in sup72 is not due to loss of function in KEG, but rather to a change that actually increases its repressive activity. Consequently, a traditional complementation approach in sup72 with the native KEG sequence is not expected to reverse the suppression, or prove causality. In order to prove causality, a more complex approach is necessary, particularly as the role of KEG appears to be dose-dependent. In the most direct approach, a keg null mutant would be complemented with the native or H1457Y mutant gene. The transformation needs to be performed on the heterozygote, as keg null mutants are lethal shortly after germination [30]. After these complementation lines are obtained, they would be crossed to PLIP3-OX to determine phenotypic suppression. Several approaches could be taken to assess whether it is an increase in the cytosolic presence of KEG, or a change in the KEG-JAZ12 interaction that results in PLIP3-OX suppression. Studies on the binding affinities between JAZ12 and native or mutant KEG can be carried out in heterologous systems or using purified proteins. In addition, if the mutation specifically affects the KEG-JAZ12 interaction, and not KEG localization, it would be expected that the ABA pathway would be less compromised in sup72. Therefore, ABA sensitivity assays, as well as direct studies of KEG H1457Y localization, would determine the extent to which changes in its location affect its role in the JA signaling pathway. Inversely, the role of KEG can be elucidated with a test of the robustness of the JA response in the keg-4 mutant, in which KEG is known to mislocalize to the cytosol [29]. 87 A larger question that would require further study is why the KEG interactions with transcriptional regulators in the cytosol is so significant, when these proteins are active in the nucleus. In the case of the activator ABI5, it is possible that efficient KEG-mediated degradation following translation is sufficient to out-compete nuclear import. However, for a repressor like JAZ12, it is unclear why stabilization outside of the nucleus would increase repression within the nucleus, as COI1-mediated degradation of JAZ repressors is generally attributed to the nucleus [31]. It is therefore likely that JAZ12 is also targeted by cytosolic factors for degradation, an interaction which would complicate current models of JA signaling. Such factors could be identified through further study of the JAZ12 interactome, or discovered by additional screening for new PLIP3-OX suppressors. As previously reported, JAZ12 is likely essential as a viable null mutant has not been demonstrated, and therefore this avenue for studying the repressor may be valuable [27]. Implications for CDK8 mutant suppression of PLIP3-OX As described in chapter 3, a nonsense mutation of CDK8 in sup11 was determined to suppress the JA-induced phenotype of PLIP3-OX. A concurrent suppressor screen in the jazD background carried out by the Gregg Howe group yielded an equivalent jazD suppression by a CDK8 mutant. Because jazD is deficient in transcriptional repressors of JA-responsive genes [32], it is likely that CDK8 serves as a transcriptional activator in at least some portion of the JA response. Prior literature on CDK8, also referred to as HEN3, RAO1, or CDKE1, points to a transcription-level regulatory role. The CDK8 mutant hen3-1 was characterized by its exacerbation of floral deformities in the hua1 hua2 double mutant [33]. However, hen3-1 on its own appears to have normal flowers, pointing to functional overlaps with other factors controlling floral development. JA is also implicated in floral development, specifically in the maturation of male tissues [34-36]. However, there is no direct evidence of a connection between JA-dependent signaling and CDK8 in flowers, as hen3-associated floral phenotypes result from incorrect differentiation of floral tissues early in development, rather than incomplete maturation of tissues post-differentiation. The relationship between CDK8 and mitochondrial retrograde signaling also suggests a broader role for the protein beyond JA signaling, which is likely effective on a transcriptional level due to the exclusive localization of CDK8 to the nucleus [37]. More detailed studies of the PLIP3-OX cdk8 88 or jazD cdk8 plants may provide further insights into the regulatory targets of CDK8, as some elements of the JA response may be less suppressed than others. Overall, the results highlight the complex nature of overlapping stress-responsive transcriptional networks in plants, as well as the tissue-dependent variation in the roles of their components. Effectiveness of the screening approach The suppressor screen in the PLIP3-OX background was originally intended to target mutants in chloroplast OPDA export, with a visual primary screen designed for high throughput, and a secondary screen that would eliminate mutants not impaired in the conversion of OPDA to JA. Approximately 4000 plants were screened within six months, of which 90 passed the primary screen. The visual screen was therefore effective in providing suppressor mutants for further screening or analysis in an acceptable timeframe. The secondary screen, based on measurements of JA, OPDA, and 12OH-JA, was intended to enrich for mutants impaired specifically in the conversion of OPDA to JA. Mutants lacking OPDA were excluded as likely OPDA biosynthetic mutants, and mutants retaining high JA were excluded as likely deficient in JA perception or signaling. A small number of mutants also exhibited decreased JA coupled with high levels of 12-OH JA, and these were also rejected as they are likely unimpaired in OPDA processing. Although the secondary screen was efficacious in reducing the number of mutants from 90 to 23, both of the candidate suppressor mutants appear to be impaired in JA signaling. The secondary screen was therefore prone to generating false positives, although it may still be effective in enriching for the desired mutant. One technical explanation is that the data generated from hormone quantification had wide variations in the controls, both between runs and within the same run, so the method is quite noisy. It is also possible that the decrease in JA in these mutants was real, but that it resulted from regulatory feedbacks rather than a direct obstruction of OPDA processing. Subsequent work by Yosia Mugume in the Benning lab determined that two additional mutants, sup12 and sup53, lost PLIP3 function according to their fatty acid profile. These had been selected for sequencing based on results from primary and secondary screening, and confirmation of the correct transgenic sequence. These results emphasize a critical weakness in the screening 89 approach: that fatty acid profiles were not measured in the secondary screen to ensure functional expression of PLIP3. However, the hormone profiles for sup12 and sup53 would be expected to match those of other, desired suppressors, and thus provide some support for the efficacy of the hormone measurement approach. Integration of the PLIP3-OX screen into coursework The straightforward nature of the primary screen makes it compatible with introductory-level undergraduate coursework, and it was therefore incorporated into a Course-based Undergraduate Research Experience (CURE) [38]. This collaborative effort increased primary screening by approximately 30%, as screening is primarily limited by chamber space and labor. In addition, it was effective in providing access to a candidate pool of undergraduate students, some of whom were subsequently recruited by the lab and took part in various research projects. Conclusion The frameworks, results, and open questions discussed here underscore the complexities of plant metabolism and development, and the extent to which the two are inextricable. An unexpected impairment of plant growth was observed in mutants lacking lipid phosphatases LPPγ and LPPε1, two enzymes that also contribute to basal chloroplast metabolism. Meanwhile, the only known PA phosphatase to exist exclusively at the plastid interior, LPPε2, appears to be uninvolved in the major galactolipid pathway in its compartment, and its role in metabolism or regulation has yet to be elucidated. For the PLIP3-OX screen, the link between lipid hydrolysis in the chloroplast and JA signaling was exploited to identify novel factors in FA metabolism, and instead led to discovery of a putative coordination mechanism between the JA and ABA response networks. In conclusion, chloroplast lipid metabolism is integrated with various pathways affecting plant physiology, and these newly discovered interactions present valuable inroads to the study of plant growth, development, and survival. 90 REFERENCES 1. Nakamura, Y., M. Tsuchiya, and H. Ohta, Plastidic phosphatidic acid phosphatases identified in a distinct subfamily of lipid phosphate phosphatases with prokaryotic origin. J. Biol. Chem., 2007. 282(39): p. 29013-21. 2. Nguyen, V.C. and Y. Nakamura, Distinctly localized lipid phosphate phosphatases mediate endoplasmic reticulum glycerolipid metabolism in Arabidopsis. Plant Cell, 2023. 35(5): p. 1548-1571. 3. Muthan, B., et al., Probing Arabidopsis chloroplast diacylglycerol pools by selectively targeting bacterial diacylglycerol kinase to suborganellar membranes. Plant Physiol., 2013. 163(1): p. 61-74. 4. Xu, C., et al., Mutation of the TGD1 chloroplast envelope protein affects phosphatidate metabolism in Arabidopsis. Plant Cell, 2005. 17(11): p. 3094-110. 5. Xu, C., et al., Lipid trafficking between the endoplasmic reticulum and the plastid in Arabidopsis requires the extraplastidic TGD4 protein. Plant Cell, 2008. 20(8): p. 2190- 2204. 6. Fan, J., et al., Arabidopsis TRIGALACTOSYLDIACYLGLYCEROL5 interacts with TGD1, TGD2, and TGD4 to facilitate lipid transfer from the endoplasmic reticulum to plastids. Plant Cell, 2015. 27(10): p. 2941-2955. 7. Nakamura, Y., et al., Arabidopsis lipins mediate eukaryotic pathway of lipid metabolism and cope critically with phosphate starvation. Proc. Natl. Acad. Sci. U.S.A., 2009. 106(49): p. 20978-20983. 8. Eastmond, P.J., et al., Phosphatidic acid phosphohydrolase 1 and 2 regulate phospholipid synthesis at the endoplasmic reticulum in Arabidopsis. Plant Cell, 2010. 22(8): p. 2796- 811. 9. Wang, Z., C. Xu, and C. Benning, TGD4 involved in endoplasmic reticulum-to-chloroplast lipid trafficking is a phosphatidic acid binding protein. Plant J., 2012. 70(4): p. 614-623. 10. Lavell, A., et al., A predicted plastid rhomboid protease affects phosphatidic acid metabolism in Arabidopsis thaliana. Plant J., 2019. 99(5): p. 978-987. 11. Wang, Z., N.S. Anderson, and C. Benning, The phosphatidic acid binding site of the Arabidopsis trigalactosyldiacylglycerol 4 (TGD4) protein required for lipid import into chloroplasts. J. Biol. Chem., 2013. 288(7): p. 4763-71. 91 12. Awai, K., et al., A phosphatidic acid-binding protein of the chloroplast inner envelope membrane involved in lipid trafficking. Proc. Natl. Acad. Sci. U.S.A., 2006. 103(28): p. 10817-10822. 13. Lu, B. and C. Benning, A 25-amino acid sequence of the Arabidopsis TGD2 protein is sufficient for specific binding of phosphatidic acid. J. Biol. Chem., 2009. 284(26): p. 17420-17427. 14. Wang, K., et al., A Plastid Phosphatidylglycerol Lipase Contributes to the Export of Acyl Groups from Plastids for Seed Oil Biosynthesis. Plant Cell, 2017. 29(7): p. 1678-1696. 15. Wang, K., et al., Two Abscisic Acid-Responsive Plastid Lipase Genes Involved in Jasmonic Acid Biosynthesis in Arabidopsis thaliana. Plant Cell, 2018. 30(5): p. 1006-1022. 16. Lin, Y.T., et al., Reduced Biosynthesis of Digalactosyldiacylglycerol, a Major Chloroplast Membrane Lipid, Leads to Oxylipin Overproduction and Phloem Cap Lignification in Arabidopsis. Plant Cell, 2016. 28(1): p. 219-32. 17. Kanaoka, M.M., et al., An Arabidopsis Rhomboid homolog is an intramembrane protease in plants. FEBS Lett., 2005. 579(25): p. 5723-5728. 18. Kmiec-Wisniewska, B., et al., Plant mitochondrial rhomboid, AtRBL12, has different substrate specificity from its yeast counterpart. Plant Mol. Biol., 2008. 68: p. 159-171. 19. Knopf, R.R., et al., Rhomboid proteins in the chloroplast envelope affect the level of allene oxide synthase in Arabidopsis thaliana. Plant J., 2012. 72(4): p. 559-571. 20. Thompson, E.P., S.G. Llewellyn Smith, and B.J. Glover, An Arabidopsis rhomboid protease has roles in the chloroplast and in flower development. J. Exp. Bot., 2012. 63(10): p. 3559-3570. 21. Lavell, A., et al., Proteins associated with the Arabidopsis thaliana plastid rhomboid-like protein RBL10. Plant J., 2021. 108(5): p. 1332-1345. 22. Kunst, L. and C. Somerville, Altered regulation of lipid biosynthesis in a mutant of Arabidopsis deficient in chloroplast glycerol-3-phosphate acyltransferase activity. Proc. Natl. Acad. Sci. U.S.A., 1988. 85(12): p. 4143-4147. 23. Yang, W., et al., A distinct type of glycerol-3-phosphate acyltransferase with sn-2 preference and phosphatase activity producing 2-monoacylglycerol. Proc. Natl. Acad. Sci. U.S.A., 2010. 107(26): p. 12040-12045. 92 24. Paysan-Lafosse, T., et al., InterPro in 2022. Nucleic Acids Research, 2023. 51(D1): p. D418-D427. 25. Liu, H. and S.L. Stone, Abscisic acid increases Arabidopsis ABI5 transcription factor levels by promoting KEG E3 ligase self-ubiquitination and proteasomal degradation. Plant Cell, 2010. 22(8): p. 2630-2641. 26. Liu, H. and S.L. Stone, Cytoplasmic degradation of the Arabidopsis transcription factor abscisic acid insensitive 5 is mediated by the RING-type E3 ligase KEEP ON GOING. J. Biol. Chem., 2013. 288(28): p. 20267-20279. 27. Pauwels, L., et al., The ring e3 ligase keep on going modulates jasmonate zim-domain12 stability. Plant Physiol., 2015. 169(2): p. 1405-1417. 28. Wawrzynska, A., et al., Powdery Mildew Resistance Conferred by Loss of the ENHANCED DISEASE RESISTANCE1 Protein Kinase Is Suppressed by a Missense Mutation in KEEP ON GOING, a Regulator of Abscisic Acid Signaling Plant Physiol., 2008. 148(3): p. 1510-1522. 29. Gu, Y. and R.W. Innes, The KEEP ON GOING protein of Arabidopsis recruits the ENHANCED DISEASE RESISTANCE1 protein to trans-Golgi network/early endosome vesicles. Plant Physiol., 2011. 155(4): p. 1827-1838. 30. Stone, S.L., et al., KEEP ON GOING, a RING E3 ligase essential for Arabidopsis growth and development, is involved in abscisic acid signaling. Plant Cell, 2006. 18(12): p. 3415- 3428. 31. Chini, A., M. Boter, and R. Solano, Plant oxylipins: COI1/JAZs/MYC2 as the core jasmonic acid-signalling module. FEBS J., 2009. 276(17): p. 4682-4692. 32. Guo, Q., et al., JAZ repressors of metabolic defense promote growth and reproductive fitness in Arabidopsis. Proc. Natl. Acad. Sci. U.S.A., 2018. 115(45): p. E10768-E10777. 33. Wang, W. and X. Chen, HUA ENHANCER3 reveals a role for a cyclin-dependent protein kinase in the specification of floral organ identity in Arabidopsis. Development, 2004. 131(13): p. 3147–3156. 34. McConn, M. and J. Browse, The critical requirement for linolenic acid is pollen development, not photosynthesis, in an Arabidopsis mutant. Plant Cell, 1996. 8(3): p. 403-416. 93 35. Ishiguro, S., et al., The DEFECTIVE IN ANTHER DEHISCENCE1 gene encodes a novel phospholipase A1 catalyzing the initial step of jasmonic acid biosynthesis, which synchronizes pollen maturation, anther dehiscence, and flower opening in Arabidopsis. Plant Cell, 2001. 13(10): p. 2191-2209. 36. Park, J.H., et al., A knock-out mutation in allene oxide synthase results in male sterility and defective wound signal transduction in Arabidopsis due to a block in jasmonic acid biosynthesis. Plant J., 2002. 31(1): p. 1-12. 37. Ng, S., et al., Cyclin-dependent kinase E1 (CDKE1) provides a cellular switch in plants between growth and stress responses. J. Biol. Chem., 2013. 288(5): p. 3449-3459. 38. Liu, J., et al., Connecting research and teaching introductory cell and molecular biology using an Arabidopsis mutant screen. Biochem. Mol. Biol. Educ., 2021. 49(6): p. 926-934. 94