Glutathione and glutaredoxins, major regulators of root system architecture Jose Abraham Trujillo Hernandez

To cite this version:

Jose Abraham Trujillo Hernandez. Glutathione and glutaredoxins, major regulators of root system architecture. Agricultural sciences. Université de Perpignan, 2019. English. ￿NNT : 2019PERP0017￿. ￿tel-02336714￿

HAL Id: tel-02336714 https://tel.archives-ouvertes.fr/tel-02336714 Submitted on 29 Oct 2019

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Délivré par UNIVERSITE DE PERPIGNAN VIA DOMITIA

Préparée au sein de l’école doctorale Et de l’unité de recherche “ Énergie et Environment ” ED305

Spécialité : Biologie

Présentée par José Abraham Trujillo-Hernández

Le glutathion et les glutaredoxines, des régulateurs majeurs de l'architecture du système racinaire

Soutenue le 18 Juin devant le jury composé de

Mme Christine FOYER, Professeur, University of Rapporteur Leeds M. Benjamin PÉRET, Directeur de Recherche, Rapporteur CNRS M. Julio SAEZ, Directeur de Recherche, Examinateur Université de Perpignan Via Domitia Mme Sandra BENSMIHEN, Chargé de Recherche, Examinateur CNRS M. Pierre FRENDO , Professeur, Institut Sophia Examinateur Agrobiotech M. Jean-Philippe REICHHELD, Directeur de CoDirecteur Recherche, Université de Perpignan Via Domitia de these M. Christophe BELIN, Maître de Conférences, CoDirecteur Université de Perpignan Via Domitia de these

This thesis is dedicated to my lovely wife Dulce, whose wisdom and kindness give me the energy of a thousand suns...

Acknowledgements I would like to say thank you to my parents Jose and Luz Maria that even if they were not physically present during my Ph.D., their moral support was critical to reaching my goals. Studying a Ph.D. is a difficult task because it requires discipline, intelligence, and talent, but is easier when you have good professors to show you the way. For this reason, I want to say thanks to Jean-Philippe and Christophe Belin for not giving up on me and show me instead of how to do science. To all the people that I had the opportunity to meet in Perpignan, especially Myriam, Sandrine, Ben, Beren, Anaid, Meli, Aleix, Caro, el Grec, Victor, Alejandra, Jean-Baptiste, Annia, Hadjer, Saionara, Zac, and Rafa, for taking the time to know me better, and share with me so many moments. Eventhough some of them were not as happy ones as we could wish, it made stronger our friendship... I wish you all the best. To all my current and former colleagues from the LGDP which are very nice people and excellent scientists, I wish you success in their professional and personal projects to come. I would also like to say an especially thank you to my friends Salvatore, Alheli, Charlotte, Hernandez, Celso, Panpan, Ariadna, Moaine, and Olka that made my daily work more pleasant with their occurrences and good talks. I want to say also thank you to the all people of the University of Perpignan for allowing me to perform my studies in this institution, in which the people from the LGDP have an especially mention, particularly Claire, Veronique, and Laëtitia. Without their assistance, experience and knowledge, I could not have made the same advances in my work. I would like to say thanks as well to the members of the Ph.D. committee, Lucia STRADER, Cécile BOUSQUET-ANTONELLI, and Laurent LAPLAZE, who were actively involved in my formation through interesting discussion and advice during these three years of my Ph.D. I would like to say a special thanks to Lucia Strader and Jerry Cohen for hosting me in their laboratories in the United States for almost three weeks, they are incredible as people as well as scientists. Moreover, I want to say thanks for having been key elements of my formation. Finally I want to thank the Mexican government and CONACYT who gave me all the economic support that I needed to obtain a doctoral degree.

LIST OF ABBREVIATIONS

GSH Reduced glutathione GSSG Oxidized glutathione IBA Indole-3-butyric acid IAA Indole-3-acetic acid ABA Abscisic acid GRXS17 Glutaredoxin S17 ROXY19 Glutaredoxin cc-type BSO Buthionine sulphoximine NPA 1-N-Naphthylphthalamic acid NOA 1-naphthoxyacetic acid cad2 Cadmium-sensitive 2, a weak allele of GSH1 pad2 Phytoalexin-deficient 2, a Weak Allele of GSH1 zir1 Zinc tolerance induced by iron 1 , a weak allele of GSH1 rml1 Root meristemless1, a weak allele of GSH1 rax1 Regulator of apx2 1-1, a weak allele of GSH1 cat2 Mutant of the catalase 2 gene LR Lateral root LRP Lateral root primordia ARFs Auxin response factors PHT1.4 Phosphate transporter 1;4 SPX1 SPX domain-containing protein 1 RNS1 Ribonuclease 1 PIN PIN-FORMED proteins ABC ATP Binding Cassette proteins AUX Auxin resistant LAX Auxin resistant like AUX1 PILS Pin like transporters TIR1/AFB Transport inhibitor response1/Auxin-related F-Box proteins PDR8 Pleiotropic drug resistance8/ABCG36 PDR9 Pleiotropic drug resistance9/ABCG37 IBR1 Indole-3-butyric acid response 1 IBR3 Indole-3-butyric acid response 3 IBR10 Indole-3-butyric acid response 10 ECH2 Enoyl-CoA hydratase 2 AIM1 Abnormal inflorescence meristem PED1 Peroxisomal 3-ketoacyl-CoA thiolase 3 UGT74D1 UDP-glucosyl transferase 74D1 UGT74E2 UDP-glucosyl transferase 74E2 NTRA NADPH-dependent thioredoxin reductase A NTRB NADPH-dependent thioredoxin reductase B DR5 Synthetic auxin response element AuxRE Auxin response element ROS Reactive oxygen species PTS1 Peroxisomal targeting signal 1 PTS2 Peroxisomal targeting signal 2 NADPH Nicotinamide adenine dinucleotide phosphate GST Glutathione s-transferase GUS β-glucuronidase gene GFP Green fluorescent protein PGPR Plant growth promoting rhizobacteria RH Root hair PR Primary root PXA1 Peroxisomal ABC-transporter1 GSH1 γ-glutamylcysteine synthetase GSH2 Glutathione synthase PEX5 Peroxin 5 PEX7 Peroxin XPP Xylem pools pericycle TRX Thioredoxin NTS NADPH-dependent thioredoxin reductase system NGS NADPH-dependent glutathione/glutaredoxin system EMS Ethyl methanesulfonate NGM Next Generation Mapping GATK Genome Analysis Toolkit NGS Next-Generation Sequencing technologies EIL-1 Ethylene-insensitive3-like 1 HAM1 Hairy meristem 1 CRK8 Cysteine-rich RECEPTOR-like kinase 8

Table of Contents I GENERAL INTRODUCTION...... 1 General Introduction...... 3 I.1 ROOT SYSTEM...... 3 I.1.1 Origin of the root...... 3 I.1.1.1 A brief history of land plants...... 3 I.1.1.2 Root evolution...... 5 I.1.1.3 Root system variation in Angiosperms...... 9 I.1.2 Root system architecture in Arabidopsis as a model of study...... 11 I.1.2.1 The primary root anatomy...... 11 I.1.2.2 Root hair development...... 12 I.1.2.3 Lateral root formation...... 15 I.1.3 Root system responses to environmental conditions...... 18 I.1.3.1 Effect of abiotic stress conditions in the root architecture...... 18 I.1.3.2 Root architecture responses to biotic stress...... 20 I.2 AUXIN CONTROL OF ROOT DEVELOPMENT...... 21 I.2.1 General considerations...... 21 I.2.2 Auxin synthesis...... 23 I.2.3 Auxin transport...... 27 I.2.4 Auxin signaling pathways...... 31 I.2.5 Roles of auxin in root development and plasticity...... 33 I.2.6 Tools for study auxin in Arabidopsis...... 35 I.3 INDOLE-3-BUTYRIC ACID...... 37 I.3.1 Generalities...... 37 I.3.2 Peroxisomal conversion of IBA to IAA...... 38 I.3.2.1 Peroxisome functions in plant development...... 38 I.3.2.2 IBA derived IAA...... 41 I.3.3 Indole butyric acid conjugation and transport...... 43 I.3.4 Indole-3-butyric acid functions...... 44 I.4 REACTIVE OXYGEN SPECIES...... 47 I.4.1 ROS generation...... 47 I.4.2 ROS as signaling molecule in root development...... 51 I.4.3 Scavenging of reactive oxygen species...... 53 I.5 THIOLS HOMEOSTASIS...... 55 I.5.1 Thiol modifications...... 55 I.5.2 Thioredoxin system and glutathione...... 57 I.6 GLUTATHIONE...... 59 I.6.1 Glutathione synthesis and functions...... 59 I.6.2 Glutathione distribution...... 63 I.6.3 Glutathionylation...... 65 I.6.4 Glutathione-S-transferases...... 66 I.6.5 Glutaredoxins system in the control of the root development...... 67 I.6.6 Transcriptomic control by glutathione...... 69 I.7 REACTIVE OXYGEN SPECIES AND AUXINS...... 71 OBJECTIVES OF THIS STUDY...... 73 II RESULTS...... 75 II.1 CHAPTER 1 ROOT SYSTEM RESPONSES TO IBA ARE CONTROLLED BY GLUTATHIONE...... 77 PERSPECTIVES...... 120 II.2 CHAPTER 2: CONTRIBUTION OF GLUTAREDOXINS IN LATERAL ROOT DEVELOPMENT...... 123 PERSPECTIVES...... 152 II.3 CHAPTER 3: IDENTIFICATION OF THE CAUSAL MUTATION IN THE SUPPRESSOR VARIANTS OF THE grxs17 MUTANT...... 155 II.3.1 INTRODUCTION...... 155 II.3.2 RESULTS...... 159 II.3.2.1 Suppressor lines of the grxs17 mutant...... 159 II.3.2.2 NGM and SHOREmap failed to identify the causal mutation in the suppressor variants...... 161 II.3.2.3 Identification of the causal mutation using Samtools/SnpEff and R...... 165 II.3.3 DISCUSSION...... 169 III GENERAL DISCUSSION...... 193 III.1 Glutathione as a mediator of root development...... 195 III.1.1 IBA control by glutathione...... 195 III.1.2 IBA and glutathione involved in abiotic stress response...... 199 III.2 Roles of glutaredoxin in the root system...... 201 III.3 Identification of the causal mutation in grxs17 suppressor lines...... 203 III.4 Conclusions...... 204 IV MATERIAL AND METHODS...... 205 IV.1 CHAPTER 1...... 207 Plant material and growth conditions...... 207 Phenotypic analyses of the root system...... 207 Glutathione measurements...... 208 Confocal analyses...... 208 GUS staining...... 209 Quantitative real-time PCR (qPCR)...... 209 Statistics analysis...... 210 IV.2 CHAPTER 2...... 210 Plant material and growth conditions...... 210 Phenotypic analyses of root system...... 211 Isolation of suppressor mutants of the grxs17 phenotype...... 211 Immunodetection of GRXS17...... 211 IV.3 CHAPTER 3...... 212 Whole-genome sequencing...... 212 Bioinformatic analysis...... 212 V REFERENCES...... 215 VI ANNEXES...... 245 VI.1 Annex I: Thiol Based Redox Signaling in Plant Nucleus...... 246 VI.2 Annex II: On the Elaborate Network of Thioredoxins in Higher Plants...... 256

I GENERAL INTRODUCTION

1 2 General Introduction

This manuscript aims to describe the research that I have done during my Ph.D., which was focused on deciphering the roles of IBA and glutathione in the control of the root development. For this reason, the following general introduction covers several important aspects of root development as well as IBA and glutathione functions.

I.1 ROOT SYSTEM

I.1.1 Origin of the root

I.1.1.1 A brief history of land plants

A long time ago during the Cambrian–Early Ordovocian period (515 to 470 mya), plants left the prehistoric water to inhabit the crust-forming microorganisms soil, through the evolution of a variety of new adaptive organs and physiological systems, in which bryophytes were the first plants to colonize land (Figure I.1) (Puttick et al., 2018b; Salamon et al., 2018; Shekhar et al., 2018; Szövényi et al., 2019). Indeed, cryptospores (ancient spores) were found in Argentina and Saudi-Arabia that were dated to the middle Ordovician, suggesting the supercontinent Gondwana was the place in which land plants first appeared (Rubinstein et al., 2010; Salamon et al., 2018). The establishment of the ancient events and ancestors, that led to the evolution from algae to land plants have been a difficult task to achieve, due to the lack of fossil evidence, the limited information from the phylogenetic analyses, and the impossibility for experimental validation. However, it is well accepted in the scientific community, that the acquisition to perform photosynthesis by eukaryotic cells around 700-1500 mya, marked the first event in the evolution of land plants (Leliaert et al., 2011).

3 Figure I.1. Key events in the evolution of land plants. According with fossil data and phylogenetic studies, the evolution of Chlorophyta lineage that appeared during the Tonian period to land plants was a slow process that took millions of years, and consisted in the acquisition of new evolutive adaptation, such as the occurrence of a vascular system, new organs like roots that increase the ability of the absorption of nutrients and attachment to the soil, as well as physiological systems that allowed the diversification and survived of terrestrial plants to the different ground such as seeds and flowers. This figure is adapted from Shekhar et al 2018.

4 The resulting unicellular photosynthetic organisms evolved later on glaucophytes, green and red algae, and subsequently the green algae into the Chlorophyta and the Streptophyta lineages. From the Streptophyta lineage two classes of the photosynthetic organism were derived, the Charophyte and the Embryophytes (land plants) lineages, in which the Charales, Coleochaetales, and Zygnematales are more closely related with the terrestrial plants (Figure I.1) (Sagan, 1967; Karol et al., 2001; McCourt et al., 2004; Keeling, 2010; Puttick et al., 2018a). Indeed, some Charophyte algae present several characteristics that are also shared by land plants such as plasmodesmata, apical growth, and branching. However, the evolution of algae to plant also include the appearance of new adaptive characteristics such as in the life cycle. This can be seen in the Charophyte algae, which are dominated by haploid phase (gametophyte), while the diploid phase (sporophyte) is restricted to the zygote. Land plants, in contrast, increase the diploid sporophytic phase , still a minority in the bryophytes clade, but becoming dominant with the appearance of tracheophytes (McCourt et al., 2004; Puttick et al., 2018a; Bowman et al., 2019; Szövényi et al., 2019). Besides the change in the gametophyte/sporophyte life cycle dominance, tracheophytes also evolved new tissues and organs such as the root system, that was a critical innovation for the successful colonization of the land, and for the plant diversification. Due to the importance of both physiology and molecular mechanisms that govern the root growth, to understand the study that I performed during my PhD thesis, the evolution of the root system will be explained in more detail in the next section.

I.1.1.2 Root evolution

One of the most fascinating innovations of the plant kingdom, that allowed their ability to survive without the advantage of motion in the new land, was the evolution of a specialized new organ known as “root”, that gave them the capacity to anchor to the ground, the perception of environmental cues and gravity, and allowed them to absorb nutrients and water beneath the soil (Scheres et al., 2002; Shekhar et al., 2018). Based on several studies, It was concluded that roots were a recent adaptation in terrestrial plants, and they appeared for the first time around 420 to 359 million years ago in the Devonian period, and it is proposed that the first root-like structure was a rhizoid-based rooting system (RBRS) (Kenrick and Strullu-Derrien, 2014; Shekhar et al., 2018).

5 Figure I.2. Fossil evidence of the history of root system evolution. This figure shows some of the most important and best-preserved fossil records that support the different hypotheses, regarding the evolution of the root system. (A) The rhizoid-based rooting system of Nothia aphylla from the Devonian period (420-359 mya), and its (B) reconstruction (Adapted from Kenrick et al., 2014). (C-D) Intermediated root system of Zosterophyllum shengfengense of the lower Devonian (From Kenrick et al., 2014) and finally (E) fossil record of stigmarian rootlets of the ancient lycophyta trees and the recreation of their root system adopted from Hetherington et al., 2016. Red arrows indicated either the rhizoids or the root system. Scale bar in (A) = 1 mm, (B) = 1.5 mm, (C) = 10 mm, (D) = 20mm, and in (F) = 3 m.

6 Indeed, fossil records of the Devonian prehistoric ecosystem, in the well-known Rhynie Chert in Scotland, showed that the basal part of land plants that was in contact to the soil was composed of RBRSs, that grew horizontally to the soil and performed functions similarly to true roots through the development of rhizoids (Figure I.2A and I.2B) (Edwards, 2004; Hetherington and Dolan, 2017). Rhizoids are cellular structures that are still present in both vascular and non-vascular plants, and are highly similar at the cellular level to root hairs of the model plant Arabidopsis thaliana. Their main function is to anchor primitive land plants to the soil, although they could participate in the acquisition of nutrients in a minor extent; rhizoids were also able to develop in the aerial axes of the land plants (Edwards, 2004; Menand et al., 2007; Kenrick and Strullu-Derrien, 2014; Honkanen et al., 2017). Due to the lack of well-preserved evidence, the hypothesis that roots evolved from a rhizoid-based rooting system is still debated. However, fossil records preserved from extinct species belonging to the genus Zosterophyllum (413 mya from the Devonian period of time) showed an intermediated root-like structure, that suggest that it was likely that for the evolution of true roots, there was first a change in the growth orientation from horizontal in RBRS to vertically in vascular plants accompanied with small ramifications (Figure I.2C and I.2D) (Hao et al., 2010). Thereafter, the evolution of new adaptations in land plants such as gravitropic sensing, meristem-dependent tissue development, and the appearance of the root cap distinguished true roots from RBRS (Raven and Edwards, 2017). However, the earliest evidence of true roots come from the Carboniferous period of time, and belong to species of Lycophyta, extinct trees that were more than 50 meters of height, and according to the fossil records they were attached to the soil by branching roots covered by root hairs, this early roots are known as stigmarian system, and it is believed that this system was produced by modified leaves (Figure I.2E and I.2F). The discovery of the root system in the ancient Lycophyta trees, and their further examination give the origin for an alternative hypothesis for the evolution of roots, that postulated that is likely that roots evolved from leaves instead of rhizoids (Hetherington et al., 2016; Szövényi et al., 2019). Besides the different theories about the tissue origin of roots, there is a consensus in the scientific community, that agrees that the evolution of roots took place independently in Lycophyta and Euphyllophyta lineages, and even they evolved in a different manner within the Euphyllophyta clade.

7 Figure I.3. Root system types in Angiosperms. Dicots (e.g. arabidopsis and carrot) and monocots (e.g. maize and rice) of the Angiosperms lineage present different root system architecture profiles. As it is shown in the top panel, the RSA of both classes of Angiosperm is dominated by the primary root in the early stages of development, while in monocots the dominance of the root system by the primary root is substituted by the increasing number of adventitious roots in mature plants (low panel). This figure is adapted from Bellini et al., 2014.

8 This hypothesis is supported by the great phenotypic variation from different Angiosperms root systems (Kenrick and Strullu-Derrien, 2014; Baars, 2017; Shekhar et al., 2018; Szövényi et al., 2019).

I.1.1.3 Root system variation in Angiosperms

Adaptation is one of the principal forces that drive the evolution of the root system, and the variation among close relatives within the same clade. Moreover, differences in the root system architecture can be observed in the same specie depending on the bio- composition of the soil, as well as the nutrients status in the ground, which bold the extraordinary plasticity of the roots. Overall, there are two classes of root system architecture: fibrous and taproot, that both are defined by the distribution and size of the primary root, adventitious roots, as well as the arrangement of secondary and tertiary roots (Smith and de Smet, 2012; Shekhar et al., 2018). During germination, the first root structure that appears is the radicle, from which the primary root develops. Taproot is characterized by the dominance of the primary root after germination and from which secondary lateral roots develop. This class of root system, also known as allorhizic, is presented in eudicots species of the angiosperm lineage such as Arabidopsis thaliana, and Daucus carota (carrots) (Figure I.3). While fibrous root system, also known as homorhizic root, is observed in land plants from the monocots clade. Contrary to taproots, the primary root is important during germination, thereafter post-embryonic secondary roots dominate the root system architecture, developing other root organs known as adventitious roots (growing from shoot tissues rather than the primary root, and present also in eudicots). Examples of post-embryonic adventitious roots are crowns and brace roots, that can be seen in the root system architecture of important commercial crop species of monocots like Zea mays (corn) and Oryza sativa (Figure I.3) (Bellini et al., 2014). Due to the importance of the root system architecture for the survival, and adaptation of the plant to different constraints in the soil environment, the study of the actors and mechanism that govern the root system plasticity is an important topic in plant biology. In this regard, Arabidopsis thaliana has been demonstrated to be an excellent model for the study of the root system, mostly due to the simplicity of its root system, and the broad number of techniques, that have been developed for this plant model (Shekhar et al., 2018).

9 Figure I.4. Primary root anatomy In (A) is shown the cell types and tissues that composed the root apical meristem in Arabidopsis, that is critical for the development and growth of the root system. In (B) the well known zones of the primary root, each of them being characterized by changes in cell division, development and proliferation. Scale bar = 50µM. Figure adapted from (Barrada et al., 2015).

10 For this reason, and because it is the model that I used for the study presented in this Ph.D. thesis, in the next sections, I am going to describe more in details the different tissues and zones that compose the root system, using mostly the information available from Arabidopsis thaliana.

I.1.2 Root system architecture in Arabidopsis as a model of study

I.1.2.1 The primary root anatomy

Despite the complex processes that depend on the primary root, its architecture is remarkably simple. In Arabidopsis, it is composed by a radial arrangement of seven well characterized types of tissue, that come from four types of stem cells, also known as initial cells, that are localized in the root tip, more precisely in the distal part of the Root Apical Meristem (Figure I.4A). The central part of the primary root are the vascular tissues (stele), that is responsible for the transport of metabolites, nutrients, and water across the entire plant, and they developed from one type of stem cells; this class of initials cells also give rise to the pericycle layer of the root. The cortex and the endodermis tissues come from the second class of initials, while the epidermis and the lateral root cap that are important layers for protection against mechanical damage, are formed from another type of initials cells (Scheres et al., 2002). The columella is the only tissue of the root tip that is developed from his own class of stem cells, and together with the lateral root cap form the root cap that protects the meristem of the growing root against mechanical abrasion, while ensuring other important functions including regulation of the gravitropism (Blancaflor et al., 1998). In to the middle of the four classes of initial cells, there is a small group of cells known as quiescent center (QC), that is characterized by its low mitotic activity, and is determinant for maintaining the specification of the stem cells niche (Van Den Berg et al., 1997). Besides the different classes of tissues that compose the root, this organ is divided into four distinct zones according to their cellular activity (Figure I.4B). In the distal part of the apical root, there is a part known as meristem zone (MZ), that is characterized by a high rate of mitotic cell division; after several rounds of division the daughter cells make a switch in the transition zone (TZ) from division to cellular enlargement (elongation zone; ET), the TZ in the primary root is clearly identified by observing the increased size of the 11 cortical cells, and recently this zone has been catching the attention of scientists, due to the discoveries from several studies, that showed that the TZ is a point for hormonal cross-talk and sensing endogenous and extracellular signals, that ultimately control the growth of the primary root, as well as modulate the plasticity of the root system (Barrada et al., 2015; Kong et al., 2018). In the elongation zone, the cells develop the central vacuole, accompanied by modification in the cell wall that allows the increase of the cell size, mainly in the apico-basal direction (Anderson et al., 2010). When the cell reaches a proper size, elongated cells from different tissues enter a process of specialization in the so-called differentiation zone (DZ), in which several molecular events led to the formation of new organs such as secondary roots, and the specialization of epidermal cells that are characterized by the development of root hairs, both events will be described in more in detail in the next section (Scheres et al., 2002; Kong et al., 2018).

I.1.2.2 Root hair development

The differentiaton zone is the site in which root cells acquire specialization, easily visible in the vasculature tissue, endodermis, and root epidermis. In the epidermis some cells develop a tubular extension known as root hair. Epidermal cells that developed the root hair are called trichoblasts and non-root-hair cells are called atrichoblasts (Dolan et al., 1993). The molecular mechanism that determines whether cells will become either root-hair or non-root-hair cells, highly depends on the position of the epidermal cells within the root. In Arabidopsis, cells that are in the junction of two cortical cells (“H” position) are trichoblasts, and cells that are adjacent to one cortical cell (“N” position) will be unable to develope root hair (Figure I.5A) (Balcerowicz et al., 2015; Salazar-Henao et al., 2016). In the N cells, there is an accumulation of the transcription factors WEREWOLF (WER) and MYB23 that interact with a basic Helix-Loop-Helix transcription factor GLABRA3 (GL3), and ENHANCER OF GLABRA3 (EGL3) that with TRANSPARENT TESTA GLABRA (TTG) protein, they form a system that induce the expression of the transcription factor GLABRA2 (GL2) that prevents the expression of the bHLH transcription factor ROOT HAIR DEFECTIVE 6 (RHD6).

12 The activity of RHD6 is necessary for the expression of genes involved in the activation of root hair development (Rerie et al., 1994; Lee and Schiefelbein, 1999; Walker et al., 1999; Payne et al., 2000; Esch et al., 2003; Zhang et al., 2003). Interestingly, it has been proposed that GL3/EGL3/TTG complex induces the expression of the transcription factors CAPRICE (CPC), TRIPTYCHON (TRY), and ENHANCER OF TRY AND CPC1 (ETC1), that are able to migrate to the H cells, in which they inactivate the GL3/EGL3/TTG protein complex, forming a negative loop to suppress the expression of GL2 in H cells. These mechanisms allow the transcription of genes involved in the root hair development, such as the bHLH transcription factors ROOT HAIR DEFECTIVE 6- LIKE 2 and 4 (RSL2 and RSL4), in which RSL4 has been shown to be determinant for the size and expansion of the root hairs, as well as LOTUS JAPONICUS ROOTHAIRLESS1- LIKE (LRL1 and LRL2) transcription factors, that are involved in the root hair morphogenesis during nutritional responses (Schellmann et al., 2002; Wada et al., 2002; Kirik et al., 2004; Yi et al., 2010; Bruex et al., 2012; Datta et al., 2015; Lan et al., 2015; Salazar-Henao et al., 2016). Moreover, the transcription factor TRY induces the expression of the leucine repeat receptor-like kinase SCRAMBLED (SCM), that is specifically expressed in H cells, and reinforce the inhibition of the activity of the GL3/EGL3/TTG1 system by blocking the transcription factor WER. In addition, it has been showed that the zinc-finger protein JACKDAW (JKD) acts upstream of the SCM enzyme (Figure I.5A) (Kwak and Schiefelbein, 2007, 2008; Kwak et al., 2014). The importance of the GL3/EGL3/TTG system to determine the specialization of the epidermal cells into trichoblast and atrichoblast, is demonstrated by the increased number of trichoblast cells that is observed after down-regulation of any of the members of the GL3/EGL3/TTG system (Galway et al., 1994; Bernhardt et al., 2003; Zhang et al., 2003; Balcerowicz et al., 2015). Besides the mechanism described above, downstream events such as cell polarity, and intracellular changes in trichoblast cells are determinant for the proper growth of the root hairs (Balcerowicz et al., 2015). In trichoblasts, the cellular polarity is determined by the accumulation of the protein RHO GTPase (RAC/ROP) in the plasma membrane, which correlates with the site of root hair emergence (Figure I.5B).

13 Figure I.5. Root hair development. (A) Genetic control of the root hair development in H and N cells known as trichoblast and atrichoblast respectively. The specialization of the epidermal cells relies on its position with respect the adjacent cortical cells, and the genetic control that determine the activity and expression of genes involved in the development of the root hair, such as GL2 (Figure adapted from Salazar-Henao et al 2016). In part (B) is illustrated the accumulation of the AtRop proteins (Green) that define the site in the plasma membrane in which the root hair will grow. Scale bar = 25 µm (Adapted from Molendijk et al 2001). Finally, in (C) is shown the pH distribution within the development of the root hair (Adapted from Bibikova et al., 1998).

14 Indeed, overexpression of the ROP2 GTPase gene led to the formation of more than one root hair in a single trichoblast cell in Arabidopsis, which demonstrated the critical role of the proper level of ROP2 in specific plasma membrane sites for the normal root hair development (Molendijk et al., 2001; Jones et al., 2002). Moreover, other members of the RHO GTPase family protein have been shown to be involved in the establishment of the cell polarity, such as ROP4 and ROP6, that are also localized in the plasma membrane during the root hair development. Overexpression of both ROP4 and ROP6, led to a mislocalization of the cell polarity that modified the normal gradient of Ca2+ that is required for root hair elongation (Schiefelbein et al., 1992; Molendijk et al., 2001). Besides the polarization of the trichoblast cells that is highly dependent of members of the ROP family, there are also changes in the distribution of the pH within the cell. It has been shown by fluorescent histochemical analyses, that the cell wall of trichoblast cells are more acid than the cytosol (Figure I.5C). Those changes in the subcellular pH distribution have been demonstrated to be critical for the normal progression of the cell outgrowth of trichoblast (Bibikova et al., 1998; Carol and Dolan, 2002; Balcerowicz et al., 2015). Besides epidermal cells specialization, in the differentiation zone also takes place the development of post-embryonic organs known as lateral roots.

I.1.2.3 Lateral root formation

Lateral roots (LR) are new organs that grow from pericycle cells adjacent to xylem cells, that have acquired the status of founder cellsvia oscillating endogenous developmental signals and environmental conditions. Upon initiating signals, the founder cells start to divide and produce a Lateral Root Primordium (LRP) that will pass eight developmental stages in the differentiation zone to become a mature organ, as illustrated in Figure I.6A (Dubrovsky et al., 2000; Vilches-Barro and Maizel, 2015). In the first stage, founder cells enter to a round of anticlinal division to produce one cellular layer, subsequently, they divide periclinally to form a two-layer of cells in stage II. Additional divisions take place in stage III and IV to produce four layers of cells, that in stage V through expansion and division acquires a dome-shaped structure that crosses from the endodermis to the cortex layer of the primary root. In the stage VI the LRP starts to resemble the meristem of the primary root that in stage VII reaches the epidermis.

15 Figure I.6. Lateral root formation. (A) Different stages of the LR development that is characterized by the periclinal and anticlinal division of xylem pole pericycle cells, that will cross several root layers until emerging from the epidermis (adapted from Casimiro et al 2003). (B) The activity of the gene expression marker DR5-luciferase that indicate the localization of the oscillation zone and pre-branch sites (Adapted from Moreno-Risueno et al 2010). (C) Schematic representation of the oscillation zone and the early stages of lateral root development (Adapted from Van Norman et al., 2013).

16 Finally, the last stage VIII also called “emergence” is the moment in which the LRP emerges from the epidermis of the primary root, giving rise to the new lateral root (Malamy and Benfey, 1997; Casimiro et al., 2003). While the formation of the lateral root primordium takes place in the differentiation zone, the specification for the xylem pole pericycle cells to become founder cells occurs early in the transition and elongation zone of the root, in which it has been described oscillation of gene expression every six hours (oscillation zone, OZ). This process is known as “priming” and is essential for the subsequent determination of pre-branch sites, that mark which cells will become founder cells and form lateral root primordia (Figure I.6B and I.6C) (Moreno-Risueno et al., 2010). Indeed, mutations in some of the genes that are expressed in the OZ, particularly transcription factors such as MADS-box proteins, AUXIN RESPONSE FACTORs (ARFs), and NAC showed impairments in the number of both lateral root and pre-branch sites, which demonstrated the importance of the early activation of specific genes in the OZ for the LR development. However, it has been shown that not all pericycle cells that experience this oscillatory gene activation are destined to become founder cells, which suggest that additional downstream mechanisms are involved in this process (Moreno-Risueno et al., 2010; Van Norman et al., 2013). Experiments performed in the model plant Arabidopsis thaliana have been unraveling several aspects of the modulation of the LR formation. Genetic and molecular analyses have demonstrated that during lateral root initiation, the activation of the AUXIN/ INDOLE-3-ACETIC ACID (AUX/IAA) negative regulators of the transcription factors AUXIN RESPONSE FACTORS (ARFs) are critical for the asymmetric division of pericycle cells (De Smet et al., 2010; Jung and McCouch, 2013). Indeed, the slr-1 mutant of the gene that encodes the AUX/IAA14 was unable to form lateral roots due to the arrest in pericycle cell division. Its activity is required for the expression of BODENLOS (BDL)/IAA12-MONOPTEROS (MP)/ ARF5, that together with other ARF are determinant for the proper lateral root initiation process (Fukaki et al., 2002; De Smet et al., 2010). Besides that modulation of LR formation by transcription factors, it has also been demonstrated by genetic experiments, that the down-regulation of the KIP-RELATED PROTEIN2 ( KRP2 ) gene that encodes a cyclin-dependent kinases (CDKs) inhibitor, it is critical for the early events of lateral root formation, by controlling the transition from G1 to S phase of the pericycle cell cycle, which is determinant for the initiation of the lateral

17 root primordia (Himanen et al., 2002; Casimiro et al., 2003). Until now, I have been describing the anatomy of the primary root, the specification of the root hairs, and the key mechanism that govern the development of lateral roots, which is highly important to understand the magnificent plasticity of the root system, critical for responses to changes in environmental constraints. It is well known that nutrient status, biotic and abiotic stresses induce changes in the size of the primary root, as well as the number and length of post-embryonic new organs and specialized cells (such as in the case of root hairs). For this reason, the next section will aim to give some examples of biotic and abiotic factors that induce changes in the root system architecture.

I.1.3 Root system responses to environmental conditions

I.1.3.1 Effect of abiotic stress conditions in the root architecture

Lateral root formation and root hair growth are modulated in response to different environmental factors. For example, during abiotic stress conditions such as drought, changes in temperature or excess of salt in the rhizosphere (defined as the surrounding soil area that interacts with the root system), there is a decrease in water availability known as osmotic stress, which induces molecular modifications that alter the root system architecture. This process is modulated by the plant hormone abscisic acid (ABA). Abscisic acid is well known for its roles in controlling plant germination and aperture of the stomatal pore, and as a master regulator of stress-responsive genes (Finkelstein et al., 2002; Nemhauser et al., 2006; Xiong et al., 2006; Danquah et al., 2014; Kim et al., 2016). In the root system, exogenous application of elevated concentration of ABA have a negative effect on the growth of the primary root, lateral roots, and root hairs length, an effect also observed during osmotic stress conditions (Schnall and Quatrano, 1992; Duan et al., 2013; Zhao et al., 2014; Rymen et al., 2017). Moreover, exogenous application of ABA reduced the number of lateral roots, an effect that is suppressed in the mutant of Flowering time control protein (FCA), that has been proposed to directly bind ABA (Razem et al., 2006). The negative effect of ABA in the number of lateral roots has been shown to occur after the emergence of the LRP, and just before the activation of the meristem of the mature lateral root, which it is not dependent on the Arabidopsis ABA- insensitive (ABI) genes (De Smet et al., 2003, 2006). Several studies focusing on the ABA 18 response of lateral roots, demonstrated that the effect of the plant hormone in the lateral roots is more complex than previously thought. Roots under stress conditions and with exogenous ABA showed the normal arrest in the elongation of the primary root and lateral roots. However, authors also showed a subsequent recovery phase in the growth of both root organs. These changes in the kinetics of the growth suggest an evolutionary adaptive response to overpass osmotic stress, and at least in the lateral roots, it has been shown to be mediated by the ABA receptor PYRABACTIN RESISTANCE 1-LIKE PROTEIN (PYL8) (Geng et al., 2013; Zhao et al., 2014). While water deficit can cause a substantial change in the root architecture, lacking nutrients also alter the growth of the root system. Nitrogen is a macro-element that is uptake by the roots in the form of nitrate and is essential for normal growth of the plant. Indeed, variations in the concentration of nitrogen in the culture medium alter the plasticity of the root system (Kant et al., 2011). In the absence of nitrogen, Arabidopsis plants present a decrease in the primary root growth and lateral root density, but conversely increases the rate of lateral root elongation (Remans et al., 2006; Krouk et al., 2010). Root responses to N-deprivation seem to be mediated by the nitrogen transport NTR1.1, a plasma membrane protein that is involved in sensing nitrogen availability, and there are strong evidence of the role of NTR1.1 in the control of the auxin transport, that is required for normal response to N-deprivation (Remans et al., 2006; Krouk et al., 2010). In contrast to the effect of nitrogen availability in the LR number, phosphorus- deprivation induces lateral root density. However, as in N-deprivation event, the absence of phosphorous produces an increase in the elongation of the lateral roots and increments the root hair length (Bates and Lynch, 1996; Linkohr et al., 2002; Shin et al., 2005). It is still unknown which is the molecular mechanism that is involved in monitoring the extracellular and intracellular phosphorous levels. However, it is well known that during phosphorous deprivation, there is an induction of what is known as phosphate starvation- inducible genes (PSI), that encodes phosphorous influx transporters, signaling proteins, and Pi-releasing enzymes that increase the uptake of phosphorous, such as PHT1.4, SPX1, and RNS1 (Niu et al., 2013; López-Arredondo et al., 2014; Lan et al., 2015; Puga et al., 2017; Ham et al., 2018). The induction of PSI is mediated by the transcription factor PHOSPHATE STARVATION RESPONSE 1 (PHR1) and its activity is regulated by the members of the family of proteins SPX-domain, that are also involved in the inositol

19 polyphosphates (InsP) sensing (Puga et al., 2014; Wild et al., 2016). Genes involved in the control of the root hair elongation during phosphorous deprivation responses include the UBIQUITIN-SPECIFIC PROTEASE14 (UBP14/PER1), the homeodomain-containing protein ALFIN-LIKE6 (AL6/PER2), the ACTIN-RELATED PROTEIN6 (ARP6) involved in chromatin remodeling, and the OVARIAN TUMOR DOMAIN protein (OTU5). Evidence of the roles of those genes in the root hair response during phosphorous deprivation is shown in the corresponding mutants that are impaired in the root hair elongation under phosphorous starvation conditions (Li et al., 2010; Smith et al., 2010; Chandrika et al., 2013; Suen et al., 2018). Besides the effect of the abiotic stress conditions such as water deficit and nutrients availability in the root architecture, biotic factors like microorganisms are also able to induce changes in the root system.

I.1.3.2 Root architecture responses to biotic stress

Underground, there is a broad number and types of microorganisms that habit the rhizospheric area and establish communication with the plant. It has been mentioned that there are around 106–109 bacteria, and 105–106 fungi per gram of soil that compete for carbon-based metabolites derived from the root (Doornbos et al., 2012; Chuberre et al., 2018). The effect of microorganisms on the root system depends on the species that establish the interaction with the root. Microorganisms such as Fusarium oxysporum, and Ralstonia solanacearum are well-known to cause severe soilborne-diseases on important crops species (Haas and Défago, 2005). The early response of the root system to a pathogenic infection is characterized by the arrest of the primary root elongation, which can be seen after the inoculation with the flagellar protein derived from Pseudomonas aeruginosa (Flg22) (Millet et al., 2010; Pečenková et al., 2017). Intriguingly, soilborne- diseases can be naturally controlled by the microbiome status of the soil in what is known as disease-suppressive soil effect, which consists in the hard competition between some species of microorganisms for the uptake of nutrients produced by the plants in the rhizosphere (Berendsen et al., 2012; Schlatter et al., 2017). This phenomenon is exploited in several laboratories for the development of natural strategies to suppress pathogenic microorganisms (Haas and Défago, 2005), in particularly the use of mutualistic interactions with a class of bacteria known as “Plant Growth-Promoting Bacteria” (PGPB). This group is composed of different orders of bacterial species and can establish

20 mutualistic interactions with the plants in the rhizosphere, which positively affects the capacity of the root to explore the soil and uptake nutrients (Glick, 2012). The mechanisms by which PGPB induce positive changes in the root system, rely on either the modification of nutrients sources in the rhizosphere or by altering the levels of several plant hormones (Ma et al., 2002; Glick, 2012; Poitout et al., 2017). Bacterial species such as Rhizobium spp. are able to secrete nitrogenases that improve the fixation of nitrogen, as well as organic acids to increase the uptake of phosphorus (Yanni et al., 2001). Rhizobium spp. also induce changes in the root system by altering the levels of ethylene through the synthesis of rhizobitoxine, which inhibits the ACC synthase involved in the synthesis of ethylene (Yuhashi et al., 2000). Other strains of Rhizobium produce ACC deaminase that catalyzes the conversion of 1-aminocyclopropane-1-carboxylate (ACC) to ammonium and α–ketobutyrate (Ma et al., 2002). The reduction of the plant hormone ethylene by different Rhizobium strains induces changes in root system, that result in the expansion of the surface area to acquire nutrients from the soil. Some Plant Growth-Promoting Bacteria such as Pseudomonas aeruginosa, Klebsiella sp., Rhizobium sp., Mesorhizobium sp., among others, have also the capacity to secrete plant hormones that alter the root system, especially the endogenous auxin indole-3-acetic acid (IAA) a major player in the control of the root system (Ahemad and Kibret, 2014). Because auxins are the major morphogenetic hormones in plants, the next section is dedicated to review what is known about the metabolism and functions of auxinic compounds.

I.2 AUXIN CONTROL OF ROOT DEVELOPMENT

I.2.1 General considerations

The term “auxin” comes from the Greek “auxein” that means “to grow” and is used to describe a class of plant hormones, that is involved in almost every pathway of the control of plant development (Enders and Strader, 2015). Early insights in the concept of auxin can be seen in the book “The Power of Movement in Plants” on chapter XI by Darwin and Darwin, in which they hypothesized that should exist an “influence” that is transported from the tip to the basal part of the plant, that causes the gravitropic effect observed in experiments with roots (Darwin and Darwin, 1881). 21 Figure I.7. Classical auxin assays. (A) Avena test developed by Wen 1928, adopted from Enders and Strader 2015. (B and C) Modern assay for testing auxin effect and functions. This assay consists in placing the plants in medium containing a desired concentration of the auxin compound. After a few periods of incubation, it is highly evident the auxin affects the root system, which consists in the arrest on the primary root growth, an increase in the number of lateral roots and an increment in the root hair length. Arrows indicate the local effect in the root system, such as primary root elongation, lateral root, and root hair lengths.

22 However auxin (named heteroauxin) was first isolated in 1934 from humane urine by Kögl and Kostermans, and only in 1941 from higher plants by Haagen-Smit and coworkers (Kögl and Kostermaas, 1934; Haagen-Smit et al., 1941). Contributions to the discovery of auxin compounds were made by the introgression of the “Avena test” developed by Went in 1928, which consists in the removal of the tip from coleoptiles of Avena sativa plants, and the placement of an agar cube containing the compound desired to be tested in one side of the tip. The auxin activity is therefore measured by the grade of curvature of the coleoptile in response to the compound (Went 1928) (Figure I.7A). The Avena test has now been replaced by more simple assays, that consist in the root system response to the auxin-related compounds (Figure I.7B and I.7C). Based on genetic and chemical strategies, it was possible to identify several molecules with auxinic effects in higher plants, such as Indole-3-acetic acid (IAA), 4-chloroindole-3-acetic acid (4-Cl-IAA), phenylacetic acid (PAA), indole-3-butyric acid (IBA), and indole-3-propionic acid (IprA), which can be found in different amounts in plants depending on the specie (Simon and Petrášek, 2011; Enders and Strader, 2015). The most abundant auxin is indole- 3-acetic acid (IAA), and it has been demonstrated to be a versatile molecule that controls cell division, cell elongation and cell differentiation. Due to the plethora of functions that depend on auxin, its homeostasis and distribution are tightly regulated via complex networks of pathways, in which there are still several aspects not yet well understood (Teale et al., 2006; Strader and Zhao, 2016).

I.2.2 Auxin synthesis

Indole-3-acetic acid is de novo synthesized by two main pathways, the tryptophan- dependent pathway and the one independent on tryptophan. In the Trp-dependent way, the production of the indole ring of the amino acid relies on the synthesis of chorismate from the so-called Shikimate pathway in plastids (Figure I.8) (Tzin and Galili, 2010; Mano and Nemoto, 2012; Ljung, 2013). Afterward, four precursors are produced in the cytosol derived from tryptophan: the indole-3-pyruvic acid (IPA), indole-3-acetamide (IAM), tryptamine (TAM), and the indole-3-acetaldoxime (IAOx), that by subsequent reactions are converted to IAA (Ljung, 2013; Enders and Strader, 2015).

23 Figure I.8. IAA synthesis by Tryptophan dependent pathways. IAA can be synthesized from four precursors derived from tryptophane: IPA, IAM, TAM, and IAOX. Several enzymes involved in the different pathways have been discovered and characterized. However, there are still several aspects that remain to be elucidated (?) (Adapted from Mano and Nemoto 2012).

24 Tryptophan aminotransferases TAA1, TAR1, and TAR2, catalyze the conversion of tryptophan to IPA which is then converted to IAA by the YUCCA (YUC) family of enzymes (Tao et al., 2008; Yamada et al., 2009; Zhou et al., 2011). The synthesis of IAA from IPA precursor is considered the main route in the synthesis of auxin in higher plants, and it has been shown that de novo synthesis of IAA from IPA is highly important for plant development (Ljung, 2013). Indeed, mutants impaired in the enzymes involved in the IPA pathway showed drastic phenotypes in Arabidopsis seedlings (Yamada et al., 2009; Zhou et al., 2011). Mutants plants of the TAA1 gene resulted in the impairment of the primary root growth, and defects in root gravitropism (Zhou et al., 2011). Moreover, mutants plants of the homologs of TAA1, produced less lateral roots and small root hairs, confirming the important roles of the IPA pathway in root development (Yamada et al., 2009). Less is known about the synthesis of IAA from the other precursors derived from tryptophan. It has been proposed that tryptamine is produced by the activity of tryptophan decarboxylases and YUC enzymes, while IAOx, presumably existing only in the Brassicaceae clade, and it is formed by the activity of CYP79B2 and CYP79B3, both are cytochrome P450 enzymes (Quittenden et al., 2009; Ljung, 2013). IAOx can derive as well the precursor IAM, which is converted to IAA by IAM hydrolases and the precursor Indole-3-acetonitrile (IAN) that is converted to IAA by Nitrilases (NITs) (Nemoto et al., 2009). Intriguingly, the IAOx biosynthetic pathway also produces camalexin (CAM) and indole glucosinolates (Igs), which are important during defense responses to plant pathogens (Bender and Celenza, 2009; Mano and Nemoto, 2012; Ljung, 2013). Besides the de novo synthesis of IAA from tryptophan, auxin can be synthesized by pathways involving auxin conjugates with amino acids and sugars in reactions catalyzed by several hydrolase enzymes (Ludwig-Müller, 2011; Ljung, 2013). Indeed, it has been shown by genetic studies that amidohydrolases mutants ilr1-1 and iar3-2 are resistant to IAA- conjugates in the primary root elongation, suggesting the involvement of those enzymes in the synthesis of IAA from auxin conjugates (LeClere et al., 2002). Moreover, genes encoding ILR1, IAR3, and additional amidohydrolases, such as ILL1 and ILL2, are expressed differentially within the plant, which suggests that the synthesis of IAA from auxin-conjugates is important for the spatial control of auxin activities (Rampey et al., 2004).

25 Figure I.9. Auxin transport in Arabidopsis root. The auxin distribution within the root is mediated by the family of transporters PIN, AUX, and ABC which are localized in different parts within the root tissue, such as in the meristem and lateral root primordia to ensure the proper auxin maxima required for root development. PINs are well known to be efflux auxin carriers while AUX proteins are involved in the influx of auxin. Some ABCs have dual activity depending on the cellular environment. Figure adapted from Petrášek and Friml 2009.

26 The enzymatic reactions by which IAA is conjugated with different compounds, involve the activity of UDP-glucose transferases and amino acid conjugate synthetases (Ludwig-Müller, 2011). Members of the GH3 family of amino acid conjugate synthetases have been shown to be able to catalyze the conjugation of IAA with amide groups (Ludwig-Müller, 2011). It has been demonstrated that glucosyltransferase enzyme UGT84B1 is involved in the specific ester conjugation of IAA, by catalyzing the covalent link of the auxin with the hydroxyl group of a sugar moiety, which negatively controls the auxin activity of IAA (Jackson et al., 2002b,a). Indeed, ectopic expression of UGT84B1 gene produced an increase in IAA-glucose levels in Arabidopsis, and the transgenic lines are resistant to exogenous application of auxin (Jackson et al., 2002b,a). It has been proposed that conjugation of auxin with amides and sugars is a mechanism to control auxin activity (LeClere et al., 2002; Staswick, 2009; Ludwig-Müller, 2011). Indeed, plants treated with exogenous auxin conjugates, such as IAA-Trp suppress the root responses to the active auxin IAA, suggesting that auxin-conjugates exert a negative control of auxin activity (Staswick, 2009). Moreover, it has been suggested as well that specific conjugation of IAA with the amino acids aspartate, and glutamate are molecular marks for the auxin degradation (Ludwig-Müller, 2011). Besides auxin synthesis from different pathways, the auxin maxima required for the specific and normal auxin activity depend on a fine-tuned distribution of auxin in different tissues by specific auxin carriers, which represent another level of auxin homeostasis control.

I.2.3 Auxin transport

IAA is so far the most abundant type of endogenous auxin present in terrestrial plants. It is involved in a broad number of physiological and developmental processes. Specific functions of IAA rely on the activity of specifically localized carriers that are able to control the flux of auxin through the entire plant. PIN-FORMED (PIN) proteins and members of the ATP Binding Cassette (ABC) family have been demonstrated to control the auxin efflux, while the uptake of auxin is facilitated by the AUXIN RESISTANT1/LIKE AUX1 (AUX1/LAX) family of transporters (Teale et al., 2006; Strader and Bartel, 2011; Enders and Strader, 2015). As shown in figure I.9, PIN, ABC, and AUX proteins are specifically localized in different tissues in the meristem, which 27 suggests a complex regulation of auxin distribution in the root system (Petrášek and Friml, 2009). The family of PIN-FORMED carriers localized in cells and are divided into two classes. PIN1, PIN2, PIN3, PIN4, and PIN7 are considered to belong to the plasma membrane-localized “long PIN class” and are often polarly localized (Gälweiler et al., 1998). PIN5, PIN6, and PIN8 are considered as “short PIN class”, and are localized in the endoplasmic reticulum (ER) (Enders and Strader, 2015). PIN1 is mostly expressed in vascular tissue, but is also detected at a lower level in the epidermis and cortical cells (Blilou et al., 2005). Loss-of-function of PIN1 in the well- known pin-formed mutant, produces the strongest phenotype related with the auxin transport, that consists of the loss of apical meristem dominance, impairment of the development of the vascular tissue, gravitropic growth defects, and inflorescence abnormalities (Okada et al., 1991; Gälweiler et al., 1998). PIN2 has been demonstrated to be critical for the auxin transport that takes place in the root tip. It is localized in the cortical cells, in the root epidermis, and lateral root cap. Null-mutant of PIN2 decrease the growth of the primary root, display root gravitropic defects and show a decreased in the meristem size (Müller et al., 1998; Blilou et al., 2005). The plasma membrane PIN3 is expressed in several parts of the root, such as in the columella, pericycle, and vasculature tissues. Null alleles of the gene that encodes PIN3 show defects in the elongation of the primary root, as well as gravitropic and phototropic impairments in the apical growth (Friml et al., 2002). PIN4 is located around the quiescent center and in provascular cells, while PIN7 is located in the columella and also in the provascular root cells (Blilou et al., 2005). Single mutants pin4 and pin7 did not show impairment in the primary root growth as in the case of other pin mutants, but they show impairment in cell division of the quiescent center, columella, and lateral root cap (Blilou et al., 2005). While single pin mutants showed specific functions in the root development, it was demonstrated that there is a redundancy among the members of the PIN family. Indeed, in the triple mutant pin3 pin4 pin7, PIN1 is accumulated in the lateral and basal part of the endodermis, and PIN2 in the provascular cells. PIN4 showed an increase of expression in the columella cells in pin3 mutant and in the lateral root cap in the double mutant pin3 pin7 (Blilou et al., 2005). Moreover, the strongest phenotypes that were observed involved combinations with the mutant pin2, which demonstrated the critical roles that PIN2 play in root development

28 (Blilou et al., 2005). Apart of the well-known eight PIN carriers that are in Arabidopsis, the group of professor Kleine-Vehn identified new members of the family of auxin intracellular transporters, based in the conserved structure of the PIN proteins this new class of auxin carriers is known as PIN-like transporters (PILS) (Barbez et al., 2012). This new family of transporters is composed of 7 members that are differentially expressed across the plant. PILS5 has been shown to be expressed in the transition zone of the root and together with PILS2, it has been demonstrated their roles in lateral root development. Single mutants pils2 and pils5 resulted in an increased in lateral root density, a phenotype that is stronger in the double mutant pil2pils5 (Barbez et al., 2012). AUX1, LAX1, LAX2, and LAX3 are members of the (AUX1/LAX) family of transporters that have been shown to be involved in the influx of auxin, which is important during root development (Petrášek and Friml, 2009; Enders and Strader, 2015). Indeed, AUX1 carrier is localized in the root apex in the columella, lateral root cap, and vasculature tissues in the upper side of the plasma membrane opposite to PIN carriers such as PIN1. The asymmetric localization of AUX1 with respect PIN carriers facilitate the auxin transport, which is demonstrated in the aux1 mutant, that is impaired in the acropetal IAA transport (Swarup et al., 2001). Moreover, confocal analysis of the AUX1 expression revealed that it is also expressed in the differentiation zone of the root, in which it is critical for the root hair development, this is demonstrated in the mutant aux1 that showed a decreased in the length of the root hair compared to wild type plants (Jones et al., 2009). Phenotypic studies in the aux1 mutant showed also impairments in the oriented growth of the primary root, and a decrease in the number of mature lateral roots and lateral root primordia. The impairment in the lateral roots in aux1 is also present in the loss-of- function lax3, and shows an additive effect in the double mutant lax3 aux1 (Swarup et al., 2001, 2008). (Swarup et al., 2001, 2008). It was showed that LAX3 is expressed in the vasculature tissues and in cells adjacent to lateral root primordia, which give strong evidence for the critical role of the proper auxin transport for the development of the lateral root (Swarup et al., 2008). Besides PIN and AUX/LAX family of auxin carriers, ABCB1, ABCB4, ABCB19, and ABCB21 transporters of the ABC family are able to facilitate the non-polar auxin efflux.

29 Figure I.10. Auxin signaling mechanisms. (A) Simplification of the molecular events that are triggered by auxin signaling depending on the TIR1 and Aux/IAA repressor proteins. (B) Auxin induces cell-cycle genes by targeting S-Phase Kinase-Associated Protein 2 (ASKP2) (C) The suggested auxin signaling pathway depending on ABP1. Adapted from Powers and Strader 2016.

30 However, by genetic and auxin transport studies it was discovered that ABCB4 and ABCB21 are able to switch their activity to auxin influx when AUX proteins are not functional, or depending on cellular auxin levels (Kamimoto et al., 2012; Kubeš et al., 2012). In agreement with the involvement of the auxin transport in the gravitropic growth, disruption of the members of the ABC auxin carriers induces gravitropic phenotype in roots (Noh et al., 2003; Lewis et al., 2007; Mravec et al., 2008). ABCB4 has a strong expression in the elongation zone of the primary root and in lateral roots, and null mutations in the mutants atpgp4 demonstrated the important roles of ABCB4 in lateral root development, due to the increase in lateral root number and root hair length presented in atpgp4 compared to wild-type plants (Santelia et al., 2005). The auxin import activity in the root system together with the efflux carriers are critical for the formation of lateral roots, root hairs, and the gravitropic root growth, which bold the importance of the correct auxin distribution for normal root development (Bennett et al., 1996; Teale et al., 2006; Swarup et al., 2008; Jones et al., 2009).

I.2.4 Auxin signaling pathways

The molecular mechanism by which auxin drives and regulates most aspects of plant development depends on the SCF (SKP, CULLIN, F-BOX), TIR1/AFB (TRANSPORT INHIBITOR RESPONSE1/AUXIN–RELATED F-BOX PROTEINS), and Aux/IAA auxin receptor system (Figure I.10A) (Strader and Zhao, 2016). In the SCFTIR1/AFB-Aux/IAA system, ARF (Auxin Response Factors) transcription factors are inactivated by heterodimerization with the Aux/IAA repressor proteins. However, active IAA acts as a molecular glue, allowing the interaction of Aux/IAA proteins with the SCFTIR1/AFB ubiquitin E3-ligase complex. This process induces the polyubiquitination of the AUX/IAA proteins by the SCFTIR1/AFB system, which results in the Aux/IAA degradation by the proteasome. The ARF transcription factors released from Aux/IAA bind to DNA sequences, the auxin response elements (AuxREs), in the promoter of target genes. This triggers the transcription of auxin inducible genes, that initiate molecular changes to control a broad number of physiological pathways (Jing et al., 2015; Yu et al., 2015; Li et al., 2016). In Arabidopsis plants, there are 6 TIR1/AFBs and 29 Aux/IAA proteins which show differential auxin affinity, depending on the combination of the members, as well as the class of auxin. Indeed, in vitro analyses have shown that PAA has less affinity than IAA 31 and 4-Cl-IAA for TIR1/AFBs receptor (Calderón Villalobos et al., 2012; Shimizu-Mitao and Kakimoto, 2014). Auxin response factors join to specific DNA sequences with canonical signatures (TGTCTC or TGTCGG), core of the AuxREs, in the promoters of target genes. There are 22 ARF in Arabidopsis in which ARF5 to ARF8 and ARF19 are gene transcriptional activators while the rest of the ARF are transcriptional repressors (Li et al., 2016). However, it has been proposed that some ARF could play a dual function depending on the molecular circumstances. Recently, studies focusing on downstream events in the auxin response have revealed a pathway independent of Aux/IAA degradation to control auxin inducible gene expression. This pathway involves the activity of INDOLE-3-BUTYRIC ACID RESPONSE 5 (IBR5) and MITOGEN-ACTIVATED PROTEIN KINASE 12 (MPK12), however the mechanism of this control remains to be elucidated (Powers and Strader, 2016). Targets of the ARFs include Aux/IAA, SAUR, GH3, and ASL2/LOB families of genes, that are involved in a variety of physiological and molecular functions that rely on the auxin signaling pathway (Ulmasov et al., 1997; Okushima et al., 2005; Guilfoyle and Hagen, 2007; Boer et al., 2014; Liao et al., 2015). For example, SAUR proteins have been implicated in the inhibition of PP2C.D phosphatases, that ultimately lead to the modulation of the H+-ATPase enzyme that is important for cell expansion (Spartz et al., 2014; Powers and Strader, 2016). Besides the auxin signaling mechanism dependent on TIR and Aux/IAA protein complex, it has been shown that auxin also join to the F-box protein SKP2 (Jurado et al., 2010). Similarly to TIR1 signaling process, auxin induces the proteasomal degradation of DPB-E2FC complex allowing the expression of genes involved in the cell cycle progression (Figure I.10B). Mutation of the SKP2 gene led to an increased content of E2FC and DPB proteins and impair cell division, while overexpression of SKP2 gene increased the number of lateral root primordia (Del Pozo et al., 2002; Jurado et al., 2010; Powers and Strader, 2016). Another target of auxin is the controversial ABP1 (Auxin Binding Protein 1), identified many years ago as a protein able to bind IAA (Batt and Wilkins, 1976; Ray, 1977; Napier, 1995). It has been suggested that auxin induces the interaction of ABP1 with the TRANSMEMBRANE KINASE RECEPTOR (TMK1). This would lead to the modulation of auxin transport by relocation of PIN1 carrier, as well as the remodeling of cytoskeleton by the control of microtubule and F-actin polymerization,

32 in a process depending on RHO-LIKE GTPASE and CRIB motif-containing proteins (RICs) (Figure I.10C) (Feng and Kim, 2015). Conditional inactivation of ABP1 by cellular immunization shown that ABP1 has roles in cell cycle and shoot development (Braun et al., 2008). However, the ABP1 mechanism of auxin signaling has been a subject to debate, due to scientific reports that claim that phenotypes of the mutants used in initial studies were due to co-segregating mutations. This conclusion is supported by the phenotypic characterization of new alleles of the ABP1 gene produced by CRISPR technology and T- DNA insertions, which did not display the phenotypes presented in past studies (Gao et al., 2015; Michalko et al., 2016). Further studies have to be done to determine whether or not, ABP1 is involved in the auxin signaling mechanism and moreover elucidate new aspects of the auxin pathway that remain elusive in the control of the root development.

I.2.5 Roles of auxin in root development and plasticity

The synthesis of auxin by different pathways and the refined auxin distribution mediated by specific auxin carriers facilitates the control of the spatiotemporal auxin activity, which is important for the regulation of root architecture. Indeed, auxin distribution mediated by polar and non-polar auxin transport has been shown to be important for root development, which is demonstrated by the phenotype observed in mutants of the PIN and AUX1/LAX family of carriers, such as pin1, pin2, pin3 and aux1 that show a decrease in the primary root elongation, and lateral root number together with a decreased in the length of the root hairs and impairment in the gravitropic growth (Blilou et al., 2005; Grieneisen et al., 2007). Besides the auxin transport, the localized synthesis of auxin has been shown to be important in plant development. Indeed, grafting experiments demonstrated that local synthesis of IAA from IPA pathway to generate an auxin maximum is critical for root meristem maintenance, as well as flower development (Brumos et al., 2018; Morffy and Strader, 2018). Intriguingly, the level of auxin in the root tissue have been correlated with the expression of the transcription factors known as PLETHORA (PLT), which together with SCARECROW (SCR) and the plant-specific transcription factor TEOSINTE- BRANCHED CYCLOIDEA PCNA (TCP) are critical for the meristem maintenance and cell differentiation (Overvoorde et al., 2010; Shimotohno et al., 2018). In addition, overexpression of CYP79B2 gene involved in the formation of IAA from indole-3- 33 acetaldoxime precursor result in an increase in the length of the hypocotyls, as well as an increment in post-embryonic adventitious roots in the presence of the amino acid tryptophan, that emphasizes the role of de novo auxin synthesis in cell expansion and differentiation. It can also be easily seen in the lateral root formation, root hairs, and primary root development in response to exogenous application of endogenous auxin, and the root architecture phenotype in mutants impaired in the auxin pathway (Zhao et al., 2002; Blilou et al., 2005; Strader et al., 2010, 2011; Mano and Nemoto, 2012). During lateral root development, the specific localization of auxin maxima in founder cells is required for the successful initiation of lateral root primordia formation, that relies on the activation of SOLITARY-ROOT/INDOLE-3-ACETIC ACID (IAA)14- AUXIN RESPONSE FACTOR (ARF)7-ARF19 and the sequential activity of the module BODENLOS/IAA12-MONOPTEROS/ARF5, which allow the first asymmetrical division during LRP development (Okushima et al., 2005; Benková and Bielach, 2010; De Smet et al., 2010). Afterward, an auxin gradient with a maximum in the tip is rapidly set up in the LRP in a process that depends on the auxin carrier PIN1, which is required for the activity of the auxin response factor GNOM (guanine-nucleotide exchange factor) (Kleine-Vehn et al., 2008). Besides lateral root formation auxins are also determinant for root hair development since it participates in the differentiation of epidermal cells, polarity and specification site in the epidermal cell (Salazar-Henao et al., 2016). Intriguingly, auxin flow required for root hair development seems to be supplied by atrichoblast cells to root hair cells in a process mediated by AUX1 auxin carrier (Jones et al., 2009). Moreover, the aux1 mutant is defective in the root hair response during phosphorous deprivation, which is complement by the expression of AUX1 in the root cap and epidermal cells, demonstrating that localized auxin transport is critical to supply appropriate auxin flux for the normal response during phosphorous deprivation (Parry, 2018; Bhosale et al., 2018). In root hair cells, auxin triggers the expression of several genes mainly via the action of ARF7 and ARF19 (Bargmann et al., 2013). Nevertheless, double mutant arf7 arf19 was not perturbed in root hair formation, but was more affected in lateral root development, which suggests a possible redundancy among other ARFs for the root hair development (Okushima et al., 2005). Recently, it has been shown that RSL4, a critical transcription factor involved in the outgrowth of trichoblast, is regulated by auxin activity and it has been proposed that the

34 RSL4 control by auxin is mediated by the ARF5, ARF7, and ARF8 transcription factors (Mangano et al., 2017). Overall, in recent years genetic and molecular studies have clearly shown the pivotal roles for auxin during root development. However, there are still a plethora of molecular events and functions in which auxin is involved and remain to be elucidated. In this regard, the development of auxin research tools is an active subject, and because I used some of these tools in my work, I will briefly describe them in the next part.

I.2.6 Tools for study auxin in Arabidopsis

Advances in understanding the functions that rely on auxin have been increased by the development of fluorescent, and histochemical reporters that facilitate auxin research. In this regard, the pioneering work by the group of professor Guilfoyle in the development of the well-known DR5 reporter established a new area in the auxin research field. DR5 auxin signaling marker consists of a synthetic promoter, with 7 to 9 repetitions of the canonical TGTCTC motif that can be fused with GUS, GFP, and Luciferase reporter genes (Figure I.11) (Ulmasov et al., 1997; Friml, 2003; Moreno-Risueno et al., 2010). However, the discovery of the TGTCGG as a new conserved binding site for some ARFs, in the promoter of auxin-responsive genes, allowed the development of a novel auxin signaling reporter known as DR5v2 (Figure I.11). This tool combines on the same transgene the two binding sites, each fused with a specific fluorescent protein. This new reporter demonstrated to be an excellent tool for the analysis of cellular processes driven by auxin, due to of its capacity to reveal more auxin signaling events than the original DR5, such as in embryogenesis, cotyledons, and vasculature tissues, as well as in the root system (Liao et al., 2015). Another auxin reporter frequently used to study auxin functions and responses is the DII-VENUS system that is based in the auxin signaling mechanism. DII- Venus consists in the translational fusion between the auxin-binding domain of the Aux/IAA proteins (DII domain), and the VENUS fluorescent protein, expressed under the control of any constitutive promoter (CaMV35S or UBQ10). This new auxin marker has the advantage over DR5 and DR5v2 to allow a more direct approximation of auxin distribution, since it relies on auxin perception rather than auxin signaling output (Brunoud et al., 2012).

35 Figure I.11. Auxin reported lines. DR5 is a transcriptional marker for auxin activity, that consists in the canonical TGTCTC sequences presented in the promoters of the auxin-inducible genes fused with protein reporters such as GUS and GFP. DR5v2 is a new version of the DR5, that in addition to TGTCTC fused with GFP DR5v2, also hold an extra TGTCGG conserved sequences fused with tomato fluorescent protein, which improve the observation of cellular sites in which auxin is activated. Finally, the new reporter line R2D2 that is based in the DII domain of the Aux/IAA proteins, which is degraded by the proteasome in the presence of auxin, allowing the visualization of the auxin signaling in real time. QC: quiescent center.

36 The DII-Venus was improved in a new ratiometric version (Ratiometric version of two DIIs = R2D2) developed by Weijers and coworkers, that consists in having in the same transgene the usual DII-VENUS and a mutated version of the DII domain fused with the red fluorescent protein, which gives the advantage of having an internal control for quantitative analysis (Figure I.11) (Liao et al., 2015). Another recent tool designed for the study of auxin distribution and transport are the fluorescent auxin analogs that conjugate natural and synthetic auxins with 7-nitro-2,1,3- benzoxadiazole (NBD), which allows mimicking the normal distribution of exogenous application of auxin and observe it in real- time (Hayashi et al., 2014). The development of fluorescent and histochemical probes are powerful tools for unraveling several molecular and cellular aspects of auxin metabolism that until now have been elusive for plant biologists, including aspects in the auxin synthesis and functions. In this regard and because is one of the main subjects of this work, the next chapter focuses on another endogenous auxin known as IBA, which is another way of IAA synthesis . While little is known about its metabolism and its regulation, it has been shown to be important for several physiological aspects of the root system.

I.3 INDOLE-3-BUTYRIC ACID

I.3.1 Generalities

The auxin indole-3-butyric acid has been recognized as an endogenous auxin by several research groups. However, for many years its nature was the subject of intense debate, due to the difficulties that represent its detection (Normanly, 2010; Novák et al., 2012; Frick and Strader, 2018). The evidence that support the classification of IBA as natural plant hormone come from studies that were able to detect IBA in different species, such as Daucus carota, Medicago truncatula, Nicotiana tabacum, Arabidopsis thaliana, and Zea mays. Moreover, the identification of several enzymes that specifically target IBA gave stronger evidence for the existence of IBA as an endogenous auxin (Sutter and Cohen, 1992; Barkawi et al., 2007; Campanella et al., 2007; Strader et al., 2010; Liu et al., 2012; Frick and Strader, 2018). Indole-3-butyric acid was first described in 1935 by Zimmerman and Wilcoxon as a synthetic strong rooting compound, and due to its stability 37 and activity, it is currently used in commercial plant media such as in Rootone® (Epstein and Ludwig-Müller, 1993; Zimmerman and Wilcoxon, 1993; Frick and Strader, 2018). IBA is structurally highly similar to IAA, differing only by the presence of two extra carbon atoms in the side chain of the IBA structure. Nevertheless and perhaps one of the most important differences between IAA and IBA is their capacity to join to the TIR1 complex to trigger auxin signaling responses. Indeed, by computational and surface plasmon resonance analyses, it was shown that while IAA and IPA are able to bind to TIR1, IBA is not capable to adopt the correct molecular orientation to join successfully to the receptor complex (Uzunova et al., 2016). While the aspects regarding its synthesis remain elusive, the conversion of IBA to IAA have been detected in several plants, includin Zea mays (Ludwig-Muller and Epstein, 1991; Epstein and Ludwig-Müller, 1993; Ludwig-Müller, 2007). Genetic and biochemical analyses elucidated that in fact, IBA acts as a precursor for the synthesis of IAA independent of usual de novo synthesis pathways. Feeding assays with labeled IBA and genetic experiments validated this hypothesis (Zolman et al., 2000, 2007, 2008; Strader et al., 2010; Strader and Bartel, 2011). After years of debate, it is now generally accepted that IBA functions fully depend on its conversion to IAA, although we cannot exclude that IBA acts as its own, in very specific contexts or in some species (Frick and Strader, 2018). Important advances allowed to decipher the mechanism and major players in the IBA conversion to IAA as well as the physiological meaning of the IBA-derived-IAA. In this regard, the discovery that IBA-to-IAA conversion takes place in the peroxisome represents a great step towards the understanding of the IBA roles in plant development (Zolman et al., 2000, 2001; Frick and Strader, 2018). Because the peroxisome is the subcellular site of the synthesis of IAA from IBA, the next part will explain briefly the functions in which the peroxisome is involved, followed by the description of the peroxisomal conversion of IBA- to-IAA, as well as the functions of IBA.

I.3.2 Peroxisomal conversion of IBA to IAA

I.3.2.1 Peroxisome functions in plant development

Peroxisomes are small and mobile cellular compartments enclosed by a single membrane (around 0.2–1 µm in diameter), and, as its name suggests, one of its functions is 38 to host metabolic oxidative reactions, and to detoxify the hydrogen peroxide resulting by- product (Purdue and Lazarow, 2001; Pracharoenwattana and Smith, 2008). (Purdue and Lazarow, 2001; Pracharoenwattana and Smith, 2008). Peroxisomal maintenance and functions highly depend on a complex system composed by the peroxin (PEX) family of proteins, that is composed of 22 members in Arabidopsis (Hayashi and Nishimura, 2006). The knowledge of the specific functions and control of the peroxins is still under investigation. However, genetic evidence demonstrated that PEX5 and PEX7 play pivotal roles in peroxisome functions (Hu et al., 2012; Kao et al., 2018). PEX5 and PEX7 are peroxins that form part of the complex system involved in the transport of proteins with target signal sequences known as PTS1 and PTS2, that are characterized by holding the aminoacid sequence “SKL” and “R[L/I]X5HL” respectively. PEX5 is necessary to import peroxisomal proteins with both PTS1 and PTS2 signals, while PEX7 is only involved in the uptake of peroxisomal proteins with PTS2. The activity of peroxins are critical for the metabolic reactions that take place in the peroxisomes, such as fatty acid beta-oxidation, photorespiration, and synthesis of hormones and compounds (Hayashi and Nishimura, 2006; Kao et al., 2018). Fatty Acid beta-oxidation is one of the well-known metabolic pathways that take place in almost any type of peroxisome for energy production (Wanders and Waterham, 2006). Overall, the sequential reactions include first an activation of the fatty acid in the cytosol by acyl-CoA synthetase, then in the peroxisome the acyl-CoA ester is dehydrogenated to enoyl-CoA, which is hydrated to form 3-OH-acyl-CoA. Thereafter, a new dehydrogenase reaction takes place to produce 3-keto-acyl-CoA. This is the substrate for the final thiolase enzyme to form acetyl-CoA and acyl-CoA, and eventually propionyl- CoA depending on the fatty acid (Poirier et al., 2006; Wanders and Waterham, 2006). Peroxisomes are also the site of different functions that rely on the type of tissue and organism, for example in methylotrophs yeast strains they are important for methanol metabolism and assimilation (van der Klei et al., 2006), synthesis of cholesterol, and ether phospholipids such as plasmalogens in humans (Purdue and Lazarow, 2001; Wanders and Waterham, 2006), glycolysis in trypanosomes, in which they are named glycosomes (Pracharoenwattana and Smith, 2008). In plants, there are involved in the glyoxylate cycle, photorespiration in photosynthetic tissue, and synthesis of plant hormones such as jasmonates and IAA (Pracharoenwattana and Smith, 2008; Hu et al., 2012).

39 Peroxisomes that perform glyoxylate reactions are named glyoxysomes and perform reactions similar to the tricarboxylic acid (TCA) cycle. Glyoxylate cycle is performed by the enzymes citrate synthase (CSY), aconitase (ACO), isocitrate lyase (ICL), malate synthase (MLS) and malate dehydrogenase (MDH) (Kunze et al., 2006). The function of the glyoxylate cycle is to generate sugars from lipids to supply seedlings with energy during germination, and early steps of plant development when photosynthesis is limited. In agreement with these critical functions of glyoxysomes, mutant plants in the genes that encode glyoxylate enzymes, display severe developmental defects and problems in seed germination (Kunze et al., 2006; Pracharoenwattana and Smith, 2008). Besides glyoxylate cycle, peroxisomes are also the place for one part of a complex process so-called photorespiration that includes reactions in the cytosol, mitochondria, and chloroplasts (Hu et al., 2012). Overall, photorespiration occurred when oxygen is fixed instead of CO2 by ribulose 1,5-bisphosphate (RuBP) carboxylase (Rubisco), leading to the formation of glycolate, that is converted to glycine in the peroxisome (Bauwe et al., 2010; Hu et al., 2012). The complex photorespiration pathway is highly important for maintaining the Calvin cycle functional when oxygen is present. However, it is hypothesized that photorespiration is an ancient mechanism that is still present in modern angiosperms, especially in C3 plants, and is subject for research to improve commercial crops (Bauwe et al., 2010; Timm et al., 2016). Peroxisomes are also the place for the synthesis of a few plant hormones, such as Jasmonates and IAA. Jasmonates are key regulators in some plant developmental process and biotic stress responses (Dar et al., 2015; Wasternack and Song, 2017). They are synthesized by a process involving three subcellular compartments: the mitochondria, peroxisome, and cytosol. In the chloroplast polyunsaturated fatty acids are converted to the Jasmonate precursor 12-oxo-phytodienoic acid (OPDA). Thereafter, OPDA is imported into the peroxisome by the multifunctional enzyme transporter PXA1, in which it is converted to jasmonic acid by three series of β-oxidation reactions that involve OPR3, OPCL1, ACX1, ACX5, AIM1, and PED1 enzymes. Jasmonate is then activated in the cytosol by JAR1 enzyme to produce JA-Ile, that is the active hormone, playing critical roles in plant development. Indeed, seeds from peroxisomal mutants involved in jasmonates synthesis are not viable, suggesting that OPDA pathway is highly important for seed development (Hu et al., 2012).

40 Besides jasmonates synthesis, peroxisomes also synthesize IAA from the auxin precursor IBA. Because most of the research presented in this work is focused on this peroxisomal pathway, the next section is dedicated to what is known about IBA as endogenous auxin and its functions in plant development.

I.3.2.2 IBA derived IAA

Indole-3-butyric acid is considered to be an endogenous inactive auxin that requires conversion to IAA by serial β-peroxisomal-like reactions, (Figure I.12A) (Xuan et al., 2015; Frick and Strader, 2018). Several studies have focused in discovering the peroxisomal enzymes that are involved in this conversion. In a pioneer work by the groups of Bonnie Bartel and Bethany Zolman, several mutants involved in the synthesis of IBA- derived-IAA were isolated by testing a pool of Arabidopsis mutants produced by ethyl methanesulfonate (EMS), that are resistant to IBA but are normally sensitive to IAA (Zolman et al., 2000). Later the mapping and identification of the causal mutations led to a proposed model for the conversion of IBA to IAA in the peroxisome. It has been suggested that IBA is transported into the peroxisome by the PEROXISOMAL TRANSPORTER1 (PXA1/CTS/ABCD1), which import a broad number of metabolic compounds (Zolman et al., 2001). Indeed, loss-of-function of the PXA1 transporter-encoding gene shows resistance to the exogenous application of IBA, and is also impaired in germination, which depends on the peroxisomal fatty acids β-oxidation (Zolman et al., 2001; Frick and Strader, 2018). Based on a genetic analysis it has been proposed that IBA is converted to IAA by the activity of the peroxisomal enzymes acyl-CoA dehydrogenase/oxidase-like IBR3 (Zolman et al., 2007), enoyl-CoA hydratase IBR10 (Zolman et al., 2008), enoyl-CoA hydratase ENOYL-COA HYDRATASE2 (ECH2) (Strader et al., 2011), the dehydrogenase/reductase INDOLE- 3-BUTYRIC ACID RESPONSE1 (IBR1) (Zolman et al., 2008), and the PED1 3-ketoacyl-CoA thiolase (Zolman et al., 2000). While genetic evidence indicate that IBR3, IBR10 and IBR1 seem to be specific to the IBA conversion, PED1 has been demonstrated to act in several pathways, such as fatty acid β-oxidation, synthesis of jasmonate, and the conversion of IBA to IAA (Hayashi, 1998; Zolman et al., 2000).

41 Figure I.12. Peroxisomal conversion of IBA. In (A) is shown the β-oxidation reactions that take place in the peroxisome to produce the active IAA, and enzymes involved in the IBA metabolism. (B) The conversion of IBA to IAA that is localized in the root cap has been shown to be an important source of auxin for the root development. (C) Strong evidence of the critical roles that IBA perform in the root development can be observed clearly in the mutants of the IBA conversion, such as ibr1ibr3ibr10 and ech2.

42 The identification of the intermediates of the enzymatic reactions converting IBA to IAA remains to be elucidated. The IBA-to-IAA conversion takes place in several root tissues, such as in the lateral root primordia and in the root cap (Figure I.12B, I.12C), in which localized IBA-derived-IAA is critical for the LR pre-branch sites and the development of the lateral root primordia (Strader et al., 2011; De Rybel et al., 2012; Xuan et al., 2015). The regulation of the IBA pathway is poorly understood, however, it has been proposed that modulation of the IBA conversion to IAA includes the transport of the IBA to other parts of the plant, and the enzymatic modification of Indole-3-butyric acid.

I.3.3 Indole butyric acid conjugation and transport

Little is known about the molecular pathways that control the rate of the peroxisomal conversion of IBA to IAA. However, it has been proposed that IBA-derived IAA can be controlled by the transport of the IBA to other tissues, and by chemical modifications in its structure. In Arabidopsis, the PLEIOTROPIC DRUG RESISTANCE8 (ABCG36/PDR8) and the ABCG37 (PDR9), two members of the family of membrane ATP Binding Cassette (ABC) transporters, have been proposed to be IBA carriers that efflux the inactive auxin to other tissues (Figure I.12A). Mutants abcg36 and abcg37 hyper-accumulate [3H]IBA, but not [3H]IAA, and are hypersensitive to exogenous IBA but not IAA (Strader and Bartel, 2009; Ruzicka et al., 2010). In Arabidopsis, both ABCG36 and ABCG37 are localized in the root cap and in the epidermis of the root tip, which could suggest that those IBA carriers are important to secrete the IBA to the rhizosphere. The family of ABC transporters is present in all classes of living organisms, and in the Arabidopsis thaliana genome there are 130 genes that are predicted to encode ABC proteins and 43 encode type G ABC transporters (ABCG) (Hwang et al., 2016). For this reason, it could be possible that there are additional members of this family able to transport IBA in a localized or systemic manner. Besides the control of IBA homeostasis by some members of the ABCG subclass of carriers, IBA can be regulated by post-translational modification mediated by the family of UDP-glycosyltransferases (UGT) and amino acid conjugate synthetases (Figure I.12A) (Frick and Strader, 2018). Among the 120 members of the UGT enzymes in Arabidopsis, 43 UGT74E2 and UGT75D1 are glycosyltransferases that have been proposed to structurally modify IBA. Ectopically expressing of UGT75D1 and UGT74E2 increased the level of IBA-glc and affects shoot development and abiotic stress responses (Li et al., 2001; Tognetti et al., 2010; Zhang et al., 2016). However, as in the case of the IBA transport, the big family of UGT in plants suggests that it is possible that other UGT members can control of IBA homeostasis via conjugation. Little is known about the conjugation of IBA with amino acids, however, it has been shown that the enzymes belong to the amino acid synthetase family GH3, such as GH3.4, GH3.5, GH3.6, GH3.17, and GH3.15 were able to catalyze the amino acid conjugation to IBA as well as other auxins. Nevertheless, further research has to be done in order to identify conjugation enzymes that react with IBA in a specific manner as well as their roles in the IBA metabolism (Staswick et al., 2005; Frick and Strader, 2018). Due to the absence of tools to study IBA, such as the lack of reporter lines of its synthesis, transport, and localized conversion in planta, there are many aspects of its homeostasis that remain to be elucidated. However, several studies have shown that IBA plays pivotal functions in plant development. In the next section, I present what we know about the roles that IBA perform during plant development and especially in the root system.

I.3.4 Indole-3-butyric acid functions

The discovery of several peroxisomal enzymes that are involved in the conversion of IBA to IAA, represents a big step towards the understanding of the specific roles that indole-3-butyric acid plays during plant development and responses to stress conditions. Indeed, triple mutant ibr1ibr3ibr10 shows high impairment in the elongation of the root hairs length, and a significant decrease in the elongation of the hypocotyl, a phenotype that is also shown in the loss-of-function of the gene ECH2 (Strader et al., 2011). Moreover, the quadruple mutant ech2ibr1ibr3ibr10 shows defects in lateral root development, and reduction of the meristematic zone, which perturbs the growth of the primary root and decreases cotyledons size when are compared to wild-type plants, which is mainly due to the decrease in cell number and size (Strader et al., 2011; Katano et al., 2016). The phenotypes of decreasing IBA-derived-IAA confirmed the important roles of this biosynthetic pathway in plant development, as well as its strong contribution to the IAA 44 pool, since ech2ibr1ibr3ibr10 displays decreased level of endogenous IAA and decreased auxin signaling (Strader et al., 2011; Frick and Strader, 2018). Additional strong evidence of the role of IBA-derived-IAA in the control of root development come from studies made by the groups of Tom Beeckman and Lucia Strader (Strader et al., 2011; De Rybel et al., 2012; Xuan et al., 2015). By an elegant set of genetic experiments with Arabidopsis, it was demonstrated that the conversion of IBA to IAA that takes place in the root cap is essential for the oscillatory determination of the pre-branch sites required for the formation of the lateral roots. Moreover, mutants with low conversion of IBA to IAA show a significant decrease in the auxin signaling in lateral root primordia visualized with the auxin marker DR5-GUS, which is not rescued by the exogenous application of indole-3-acetic acid, suggesting that IBA-derived-IAA is determinant to enhance the auxin effect necessary during the development of the lateral root primordia (Strader et al., 2011). In agreement with the involvement of the IBA pathway in plant development, a mutation in the two characterized IBA transporters genes ABCG36 and ABCG37, showed increase size in the root hairs length, and an increase in cotyledons size (Strader and Bartel, 2009; Ruzicka et al., 2010). This is in agreement with the phenotype observed in lines overexpressing UGT75D1 that led to a decrease in size of cotyledon pavement cells (Zhang et al., 2016). The IBA glucosyltransferases also proved to be critical in shoot development, since overexpression of UGT74E2 in Arabidopsis produced the rosette more compact and an impairment in petioles with dark-green leaves. This suggests that the contribution of the IAA pool from the peroxisomal conversion of IBA is determinant for the normal growth of several organs of the plant (Tognetti et al., 2010). (Tognetti et al., 2010). Besides the developmental roles of IBA-derived-IAA, in silico analysis revealed that several of the genes encoding peroxisomal enzymes are up-regulated during osmotic and salt stress, suggesting putative roles in the responses to abiotic stress conditions (Frick and Strader, 2018). To conclude, it is quite clear that IBA plays essential functions during plant development, by contributing to local IAA homeostasis. However, the importance of the IBA-dependent source of auxin, relatively to IAA de novo synthesis, is not well understood. This is summarized by the question in Simon and Petrâśek (Plant Science 2011) “Why plants need more than one type of auxin?”, which is still unsolved.

45 Figure I.13. ROS production. Reactive oxygen species are produced either as by-products of the metabolism of several subcellular compartments, such as electron transfer chains in chloroplast and mitochondria, oxidation reactions in the peroxisome, or by the enzymatic activity of

NADPH oxidases (RBOH) in the apoplast. This leads to the formation of the H 2O2, that is able to act as a second messenger to modulate gene expression, for example by altering activity of transcription factors (purple). Barrel in yellow represent receptor proteins and red barrels represent receptor systems. Adapted from Noctor et al., 2018.

46 I.4 REACTIVE OXYGEN SPECIES

.- Reactive oxygen species (ROS) such as superoxide radicals (O2 ), singlet oxygen

1 . ( O2), hydrogen peroxide (H2O2), and hydroxyl radicals (OH ) are oxygen-derived compounds, that appeared on earth more than 2.5 billions of years ago, and they have been produced in both anaerobic and aerobic organisms (Czarnocka and Karpiński, 2018). ROS are produced as by-products of metabolic pathways that are capable of oxidizing proteins and cellular complex structures, resulting in the induction of cell death, a reason why where considered mainly as toxic compounds. However, ROS act also as second messengers involved in the control of many biological processes, including response to abiotic and biotic stress and control of developmental stages (Gill and Tuteja, 2010; Zandalinas and Mittler, 2018).

I.4.1 ROS generation

In plants, reactive oxygen species are formed in different compartments and cellular structures, such as the mitochondria, chloroplast, peroxisomes, and plasma membrane (Figure I.13) (Noctor et al., 2018). Due to the production of energy that takes place in the mitochondria, it has been proposed to be the major cellular site of H2O2 production (Gill and Tuteja, 2010). Hydrogen peroxide is formed in normal conditions in the mitochondria,

.- when the electron transport chain of the complex I and III produce O2 radicals, which are dismutated by SUPEROXIDE DISMUTASES (SOD) to produce H2O2. The H2O2 formed can produce OH. by Fenton reactions, which is one of the most reactive ROS produced by the cell (Mittler, 2017). Another important organelle in the production of ROS is the chloroplast, which is the

.- site of the major production of O2 . Superoxide radicals are formed by partial reduction of oxygen during photosynthesis at the photosystem I (PSI). This superoxide can generate other ROS, such as the singlet oxygen, and the highly reactive hydroxyl radicals through the Haber-Weiss and Fenton reactions both in the presence of iron. Singlet oxygen can be produced also during photosynthesis in the photosystem II (PSII) as consequences of inappropriate dissipation of energy, which lead to the formation of chlorophyll triplet state,

3 1 that reacts with triplet oxygen ( O2) and leading to O2 formation.

47 Figure I.14. Examples of systemic and localized ROS production as a signal compound. Yellow arrows indicated the transmission of a light-dependent signal to distal leaves resulting in localized ROS production. The damage produced by different factors such as pathogen attack leads to a signal that is perceived by distal leaves, inducing ROS production to prevent a possible attack. Adapted from Noctor et al., 2018.

48 1 Accumulation of O2 have been shown to be important in signaling processes such as in stress acclimation and cell death (Wagner et al., 2004; Das and Roychoudhury, 2014; Laloi and Havaux, 2015; Czarnocka and Karpiński, 2018). As mitochondria and chloroplast, peroxisomes are also producing ROS as a by- product of their normal metabolism. Superoxide radicals are formed within the

.- peroxisome, in the matrix and in their membrane. O2 produced by the peroxisome can serve as a substrate to produce H2O2. Moreover, several metabolic pathways that take place in the peroxisome produce H2O2 as a by-product of their reactions, such as β-oxidation of fatty acids, photorespiratory enzymes from the glyoxylate cycle, and the synthesis of plant hormones (Gill and Tuteja, 2010; Qu et al., 2017). For example in photorespiration, H2O2 is produced during the conversion of glycolate to glyoxylate in a reaction involving molecular oxygen, while during the oxidation of fatty acids H2O2 is produced as by- product of reactions involving acyl-CoA oxidases, flavin oxidases, and superoxide dismutases enzymes (Das and Roychoudhury, 2014; Czarnocka and Karpiński, 2018). Besides the formation of ROS as a by-product of the metabolic process, plants generate ROS in a mechanism mediated by NADPH oxidases, which have a catalytic subunit that is homologous to mammalian Respiratory Burst Oxidases, and which are probably mainly located at the plasma membrane. In the Arabidopsis genome, there are 10 genes that encode for RESPIRATORY BURST OXIDASE HOMOLOG (RBOH) enzymes, and they shared a similar protein structure, which consists in transmembrane and oxidase domains, both situated at the C-terminal region of the protein structure. Phosphorylation and calcium-binding EF-hands domains localized in the N-terminal region are also critical for the regulation of their activity. RBOH enzymes catalyze the formation of superoxide radicals by transferring an electron to molecular oxygen (Suzuki et al., 2011). Apoplastic class III heme peroxidases are other potentially important sources of superoxide and H2O2 in plants (Cosio and Dunand, 2009). While their biochemical properties are not always elucidated, some isoforms catalyse superoxide formation from oxygen, as well as H2O2 reduction (O’Brien et al., 2012). The production of ROS is maintained at a basal level under normal growth conditions. However, its localized and systemic induction by several factors, such as in biotic and abiotic stress conditions have different roles during plant development (Figure I.14).

49 .- Figure I.15. O2 and H2O2 accumulation in different root zones. MZ = Meristem zone, EZ = Elongation zone, and DZ = differentiation zone.

50 High levels of ROS have a negative impact by reacting with biological compounds

1 that produced their oxidation and degradation. Nevertheless, ROS such as O2 and H2O2 act as a signal molecule to drive the control of physiological process during plant development (Wagner et al., 2004; Laloi and Havaux, 2015; Mittler, 2017; Mhamdi and Van Breusegem, 2018).

I.4.2 ROS as signaling molecule in root development

Among the broad number of compounds and systems involved in the control of the root growth and development, reactive oxygen species play a pivotal role. Accumulation of

.- both O2 and H2O2 seems to be important for the development of the different root zones

.- (Figure I.15). High concentration of O2 have been detected in the meristem while H2O2 in the elongation and differentiation tissues, suggesting a positive role of superoxide in the promotion of cell proliferation, while H2O2 favors the increase in size and specialization of root cells (Figure I.15) (Dunand et al., 2007). The control of cell division and cell differentiation by ROS have been shown to be dependent on the transcription factor UPBEAT1 (UPB1), which modulated the expression of peroxidase genes to control the ROS level in the root. Indeed, absent of UPBEAT1 in the upb1 mutant lead to the increase of the root meristem size and primary root length, moreover overexpression of UPB1 gene produce an increase of H2O2 that resulted in a decrease of the meristem cells of the root compared to wild-type plants, suggesting that H2O2 acts as a negative regulator of cell division. In maize, the quiescent center that has a low rate of cellular division shows a high level of ROS, which is in part due to the low level of glutathione and ascorbate that have been detected in those cells (Sanchez-Fernandez et al., 1997; Jiang et al., 2003). In agreement with the critical roles of ROS during root growth, transcriptomic studies have shown a number of ROS related genes that are differential expressed in the root zones, and induced by exogenous application of auxin (Orman-Ligeza et al., 2016; Tognetti et al., 2017; Mhamdi and Van Breusegem, 2018). The root control mediated by ROS such as

.- H2O2 and O2 seems to be linked to the plant hormone ABA that is critical during stress responses. Exogenous application of ABA decrease the growth of the primary root, reduce the number of cells in the meristematic zone, and increase the production in the root tip of ROS, an effect that is partially rescued by the addition of high concentration of glutathione, a key regulator in scavenging ROS (Zhao et al., 2014). 51 Figure I.16. Control of ROS homeostasis by plant scavenging systems. Several pathways to control ROS have been evolved in plants. Key player of the detoxification of reactive oxygen species is glutathione, that acts as an electron donor to two complex systems the GPX and the AsA-GSH cycles and ultimately leads to the reduction of H2O2 to water. ASC = ascorbate, GR = glutathione reductase, DHAR = dehydroascorbate reductase, APX = ascorbate peroxidases, CAT = catalase, GPX = glutathione peroxidases, PRX = peroxiredoxins, TRXox, oxidized thioredoxin. TRXred, reduced thioredoxin, NTR = NADPH-thioredoxin reductase, FTR = ferredoxin- thioredoxin reductase, GST = glutathione S-transferase, GSH = reduced glutathione, GSSG = glutathione disulfide. Adapted from Mhamdi et al., 2010.

52 Reactive oxygen species have been also involved during the elongation of the root hairs in the epidermis of the root, which is mediated by the activity of NADPH oxidases (Foreman et al., 2003). It has been proposed that the ROS formation by NADPH oxidases activates Ca2+ channels that induce the cellular elongation of the root hairs. Inhibition of the activity of NADPH oxidases by diphenyleneiodonium chloride or loss-of-function of RBOHC block the growth of the root hairs (Foreman et al., 2003). Formation of the post- embryonic organ is another biological process in which ROS have been involved.

.- Detection of H2O2 and O2 have been detected during lateral root primordia development, suggesting roles of ROS during early events in the formation of lateral roots. Indeed, by genetic and pharmacological experiments with the RBOH family of enzymes, it was demonstrated that generation of ROS by RBOH is required for the emergence of the lateral root primordia (Manzano et al., 2014; Orman-Ligeza et al., 2016). Besides the developmental roles of ROS in the root architecture, ROS generation seems to be important during abiotic stress conditions, such as nutrient deprivation. Indeed, under lack of phosphate there is an accumulation of intracellular Fe3+ produced by the increased activity of ferroxidases LPR1 and LPR2, that alter the balance of iron and phosphorous within the cell. This leads to the production of ROS and callose deposition within the meristematic zone, that in turn decrease the growth of the primary root and induce the elongation of the root hairs (Niu et al., 2013; Yang et al., 2014; Ham et al., 2018). Moreover, during the growth of Arabidopsis plants in phosphorus, nitrogen, and potassium medium deficiency, there is an increase of H2O2 accompanied by an increase of ROS related genes (Shin et al., 2005). Overall, negative and positive functions of reactive oxygen species in the root development are depending on the type of tissue in which is induced, the class of ROS produced, and the level of ROS that is synthesized. It is for this reason that its control by several molecular systems is highly important for the normal plant development.

I.4.3 Scavenging of reactive oxygen species

In plants as in other organisms, there are several compounds that have the capacity to act as antioxidant agents to maintain proper cellular levels of ROS. They are divided into enzymatic and non-enzymatic ROS scavenging molecules. Superoxide dismutases (SOD), catalases (CAT), ascorbate peroxidases (APX), and glutathione peroxidase (GPX) are part 53 of the enzymatic mechanisms to detoxify reactive oxygen species, while ascorbate (ASC), and glutathione (GSH) are members of the non-enzymatic scavenging systems (Figure I.16) (Das and Roychoudhury, 2014; Czarnocka and Karpiński, 2018). SOD catalyzes the dismutation of superoxides in different cell compartments (cytosol, chloroplasts, mitochondria, and peroxisome). These isoforms are either of prokaryotic or eukaryotic origin and use Fe, Mn, Cu or Zn as a co-factor (Noctor et al., 2018). Its expression is up-regulated under biotic and abiotic stress conditions, suggesting important roles in the control of ROS homeostasis during oxidative stress (Gupta et al., 1993; Czarnocka and Karpiński, 2018).

Catalases (CAT) dismutate H2O2 into H2O and O2. In plants, there are three isoforms and they are expressed differentially in several tissues. CAT1 is principally expressed in pollen and seeds, CAT2 in both photosynthetic and non-photosynthetic tissues, such as in roots and seeds, while CAT3 is expressed in leaves and vascular tissues (Mhamdi et al.,

2010). CAT are major actors in H2O2 metabolism in the peroxisome. Disruption of the gene that encodes to CAT enzymes showed that the reduction of the catalase activity is more affected in the mutant cat2, rather than cat1 and cat3, which produced an increase of the H2O2 and oxidized glutathione (GSSG) (Bueso et al., 2007; Mhamdi et al., 2010b). Decreased in catalase activity is also observed in the roots of the cat2 mutant, in which there is also a decreased in the number of lateral roots, plant size, and it is more sensitive to stress conditions compared to wild-type plants (Bueso et al., 2007).

Ascorbate peroxidases (APX) isoforms are major H2O2 scavenging enzymes in plants through a central role in the ascorbate/glutathione cycle known also as the Foyer-Halliwell- Asada pathway (Figure I.16) (Foyer and Halliwell, 1976; Asada, 1999). As reducer of APX, ascorbate belonging to the class of non-enzymatic compounds play a major role in this cycle (Noctor et al., 2018; Czarnocka and Karpiński, 2018). It is oxidized to monodehydroascorbate (MDHA) and dehydroascorbate (DHA), which are reduced back to ascorbate by MDA reductases (MDHAR) and DHA reductases (DHAR). The reduction of DHAR is performed by glutathione (GSH), forming oxidized glutathione, which is then reduced by the activity of the enzyme glutathione reductase (GR). Ascorbate has been shown to be critical during plant development. Mutants plants known as vitamin c defective (vtc) that are impaired in the synthesis of ascorbate shown developmental defects, such as impairment in the stomata functions and plant growth, which was demonstrated to

54 be linked to the ABA and jasmonate pathways (Chen and Gallie, 2004; Kerchev et al., 2011; Bartoli et al., 2018). The importance of ascorbate during abiotic and biotic stress conditions, have been demonstrated by the fact that lowering the level of ascorbate, resulted in a strong impairment in the response to oxidative stress conditions (Bartoli et al., 2018; Kuźniak et al., 2018). In addition to its major role in the ascorbate/glutathione cycle, glutathione is also serving as reducing power for other ROS scavenging enzymes. GPXs have the capacity to detoxify H2O2 to H2O by using glutathione as a reducing power (Bela et al., 2015). More specifically in plants, GPX isoforms are reduced by thioredoxins instead of glutathione (Iqbal et al., 2006). In Arabidopsis, there are eight isoforms of GPX, which are localized in different compartments, and their expressions are induced during biotic and abiotic stress conditions, which seems to be important during root developmental process, such as lateral root formation (Passaia et al., 2014; Bela et al., 2015). Overall, there are several players and mechanisms in plants that are responsible to maintain a proper localized and systematic level of ROS, this due to the strong effect that ROS have during plant development through the oxidation of several molecular components, such as in the case of thiol proteins.

I.5 THIOLS HOMEOSTASIS

The next part regarding the thiol modification is adapted from a paper published during my Ph.D. in which I described the functions of thioredoxins in the nucleus, to consult the complete paper please see the Annex 1 of this manuscript.

I.5.1 Thiol modifications

The chemical characteristics of the sulfur atom make Cys and Met residues major sites of oxidation within proteins (Davies, 2005). Depending on their pKa and on the pH of the medium, thiol residues are deprotonated into a thiolate residue (R-S−) which is prone to oxidation. This is leading to successive oxidations to sulfenic (R-SOH), sulfinic (R

SO2H), and sulfonic (R-SO3H) acids (Davies, 2005). Thiol groups can also form a disulfide bridge (S-S) or react with reactive nitrogen species (RNS) or oxidized glutathione (GSSG) resulting in S-nitrosylation (R-SNO) or S-glutathionylation (R-S-SG). 55 Figure I.17. H2O2-induced thiol modifications and scavenging activities.

H2O2 can react with thiol residues of proteins and produce several types of modifications. Primarily, oxidation of cysteine leads to the formation of sulfenic acid, which is an intermediate for overoxidated forms sulfinic acid and sulfonic acid. As sulfinic acid can be reduced eventually by sulfiredoxins, sulfonic acids are irreversible oxidations. Sulfenic acid can also lead to the formation of disulfide bonds, or react with NO or GSH to form S-nitrosylated and S-glutathionylated residues. Those types of modifications can be reduced back to the original state, in a process that relies on the thiol reduction systems dependent on GRX and TRX. Other types of thiol modifications, (e.g. S-sulfhydrylation) exist, but are not represented here (Adapted from Martins et al., 2018).

56 Depending on their nature, most of these thiol modifications can be reversed by dedicated thiol reduction systems (TRX, GRX, and GSNO Reductase) which exhibit disulfide bond, deglutathionylation or denitrosylation activities (Figure I.17). Thiol modifications can alter the structure and/or the activity of many proteins like transcription factors, MAP kinases, and chromatin modification proteins (Martins*, Trujillo-Hernandez* and Reichheld, 2018).

I.5.2 Thioredoxin system and glutathione

Thioredoxins are important regulators of the thiol-disulfide status (Figure I.17 and Figure I.18). In this regard, I published during my Ph.D. as a co-author a book chapter describing the functions of the thioredoxin system in plants (Thormählen et al., 2018). (see annex 2). Overall, in Arabidopsis, there are 20 genes encoding to extraplastidial thioredoxins harboring the canonical WCGPC domain. The extraplastidial thioredoxins system uses NADPH-dependent TRX reductases as reducing power known as NTR system. NTR systems are located in different subcellular compartments such as in the cytosol, mitochondria, apoplast, plasma membrane and nucleus, in which NTR plays different functions as a donor of electrons to thioredoxins, that reduced cysteines in target proteins to control several developmental pathways (Figure I.19). Indeed, double mutant of the two NTR isoforms resulted in impairment of the root growth, pollen fitness, and seed development (Reichheld et al., 2007). Several evidence are suggesting a cross-talk between Glutathione-Glutaredoxin and NTR-Thioredoxin systems in plants (Figure I.18). For example, in Arabidopsis, NADPH- dependent thioredoxin reductases (NTR) are able to act as a backup to reduce glutathione in the absence of GR enzymes (Marty et al., 2009). Moreover, it has been demonstrated that thioredoxins TRXh4 and CXXS3 uses Glutathione-Glutaredoxin system as a source of reducer power instead of NTR-Thioredoxin system in plants (Gelhaye et al., 2003). The interplay between glutathione and NTR in plant development was also demonstrated by genetic analysis of the loss-of-function of the two isoforms of NTRA and NTRB, and the weak alleles of the glutathione biosynthesis pathway. Double mutant ntrantrb crossing with cad2 shows significant impairment in the growth of the primary root, a decrease in lateral root development and in auxin transport.

57 Figure I.18. NADPH-dependent thioredoxin reductase (NTR) and Glutathione- Glutaredoxin systems. Nicotinamide adenine dinucleotide phosphate (NADPH) is the source of reducer power to feed two of the major thiol reduction systems in plants and several organisms. Both systems NTR and GRX-GSH have specific targets of proteins with cysteine residues. Crosstalks between the two reduction systems exists, in which glutathione can be the donor of electrons to both TRX and GRX, and can be reduced by TRX alternatively to GR (Adapted from Meyer et al., 2012).

58 Moreover, crossing ntrantrb with a more impaired mutant in glutathione biosynthesis the rml1, arrested severely the growth of both shoot and root organs (Reichheld et al., 2007). These experimental data bold the critical roles of thiol reduction system such as glutathione in plant development and in auxin signaling (Reichheld et al., 2007; Bashandy et al., 2010).

I.6 GLUTATHIONE

I.6.1 Glutathione synthesis and functions

Glutathione (γ-glutamyl-cysteinyl-glycine) is a tripeptide thiol compound that is present in all living kingdoms and in all subcellular compartments (Figure I.20) (Foyer and Noctor, 2011; Meyer et al., 2012; Noctor et al., 2012; Hasanuzzaman et al., 2017). It is synthesized by two ATP-dependent reactions mediated by the enzymes [γ- glutamylcysteine synthetase (γ-ECS or GSH1), and glutathione synthase (GSH-S or GSH2) (Figure I.20). In the first step, the plastidial enzyme GSH1 catalyzes the formation of γ-glutamylcysteine by the formation of a peptide bond between the amino acids glutamic acid (Glu) and the sulfur-containing amino acid cysteine (Cys). The second and last step in the formation of glutathione, consists in the catalysis of an amide bond between γ-glutamylcysteine with the amino acid glycine (Gly) in a process mediated by the activity of the dual-target plastidial and cytosolic enzyme GSH2 to finally produce glutathione (Figure I.20) (Meister, 1983; Wachter et al., 2005; Noctor et al., 2012). In Arabidopsis, loss-of-function of the enzyme GSH1 produce impairment in the development of the embryo. In contrast, the null mutant gsh2 is not perturbed in embryogenesis but seeds from gsh2 are incapable to germinate. The effects observed in the mutants gsh1 and gsh2 demonstrated the important roles of glutathione in embryogenesis and seed germination (Cairns et al., 2006; Pasternak et al., 2008). Besides the gsh1 mutant, several weak alleles of the gene GSH1 have been produced by different groups. Those mutants are able to accumulate only a proportion of the glutathione present in wild-type plants due to mutations located in the catalytic region of the enzyme (Parisy et al., 2006).

59 Figure I.19. Subcellular localization of NTR in plants. Different plant TRX isoforms (yellow) are represented in their respective subcellular localization. When established, their respective reducers are represented (blue). Some key TRX target proteins (green) as well as the respective regulated functions (red spots) are also shown. NRX = Nucleoredoxins, LOV1 = Locus orchestrating victorin effects protein, NPR1 = Non-pathogenesis-related protein expressor, Ssb = Single-stranded DNA binding protein. Adpated from Thormählen et al., 2018.

60 The root meristemless1 mutant (rml1), was isolated from an EMS-screening experiment in an attempt to find components involved in the development of the root meristem (Cheng et al., 1995). Based in Restriction Fragment Length Polymorphism technique (RFLP) and complementation tests, the causal mutation was located in chromosome 4 in the exon 7 in the gene GSH1 of the synthesis of glutathione (Vernoux et al., 2000). The root meristemless1 (rml1) mutant showed an accumulation of around ~3% of total glutathione compared to wild-type plants, which causes an impairment in the primary root growth (Vernoux et al., 2000). In another study, the mutant phytoalexin- deficient 2 (pad2-1) characterized by its sensitivity to pathogen infection due to a reduction of the camalexin content, contained a EMS point mutation that produced an amino acid replacement of serine by asparagine in exon 8 of the enzyme GSH1, which decreased the content of total glutathione to around 20% compared to wild-type plants (Glazebrook and Ausubel, 1994; Parisy et al., 2006). Mutagenesis studies by x-irradiation performed by the Cobbett laboratory isolated mutants that are sensitive to cadmium. The cadmium-sensitive (cad2-1) mutant was identified as a weak allele of the GSH1 gene due to a deletion in the exon 6 near to the catalytic region of the enzyme GSH1, this produced a reduction of total glutathione of around 30%, and demonstrated the critical roles of glutathione in the responses to stress conditions induced by heavy metals (Howden et al., 1995a,b; Cobbett et al., 1998). Another glutathione mutant isolated by EMS mutagenesis is the mutant regulator of apx2 1-1 (rax1-1) that expresses constitutively the gene that encodes to ASCORBATE PEROXIDASE2 (APX2) enzyme, which is involved in the response to photo-oxidative stress conditions. This mutant accumulated only 50% of total glutathione compared to wild-type plants and altered the transcriptional expression of defense genes (Ball et al., 2004). The mutant zinc tolerance induced by iron 1 (zir1) is another weak allele of GSH1 isolated by EMS screening. It contains a point mutation that changes glutamic acid by lysine in position 385 of the GSH1 amino acid sequence. The authors reported that zir1 mutant contains only 15% of total glutathione, which produced an impairment in the response to abiotic stress conditions (Shanmugam et al., 2012). In addition to the different functions discovery by glutathione mutants, in plants and humans glutathione have been shown to be important in the cell cycle progression to S- phase, in which a high oxidation of glutathione take place in the early G1-phase (Menon and Goswami, 2007; Vivancos et al., 2010).

61 Figure I.20. Glutathione synthesis and distribution. The first steps of the synthesis of GSH involved the formation of the γ- glutamylcysteine. This reaction is catalyzed by the chloroplastic enzyme γ-ECS, from which it can either be exported to other compartments or proceed with the second reaction mediated by GSH-S to produce glutathione. As it can be noticed from the diagram, several aspects of its transport remain to be elucidated. OPT, CLT, and MRP family of carriers are determinant for the distribution of glutathione to subcellular compartments. γ - EC(S), γ -glutamylcysteine (synthetase); BAG, Bcl2-associated anathogene; CYT, cytosol; PER, peroxisome; GSH-S, glutathione synthetase; GR, glutathione reductase; GSSG, glutathione disulfide, CLT, chloroquinone resistance transporter-like transporter; MIT, mitochondrion; MRP, multidrug resistance-associated protein; OPT, oligopeptide transporter. Adapted from Noctor et al., 2012.

62 This seems to be important during developmental process such as lateral roots, in which glutathione is localized in the nucleus of pericycle cells during lateral root development (Diaz-Vivancos et al., 2015). Other functions of glutathione can be observed with glutathione homologous compounds such as homoglutathione, which differs from glutathione for having a β-alanine instead of glycine in the amino acid sequence. This class of glutathione is found in leguminous plants, and is important for the growth of nodules in Medicago truncatula (Frendo et al., 1999; El Msehli et al., 2011). The different phenotypes presented in plants damage in the biosynthesis of glutathione and studies of the function of glutathione in plants, demonstrated the broad number of critical functions in which glutathione is involved (Figure I.21). This is mainly due for its capacity to control several pathways by different mechanisms and its distribution (Figure I.20).

I.6.2 Glutathione distribution

Glutathione is a peptide that is present in all subcellular compartments and it is transported in the cell by different types of carriers (Noctor et al., 2012). Immunocytochemical analysis on leaves and roots showed that glutathione is majorly accumulated in the mitochondrion, chloroplasts, cytosol, nucleus and peroxisomes (Figure I.20) (Zechmann et al., 2008; Delorme-Hinoux et al., 2016). However, it was hardly detected on vacuoles and apoplasts (Zechmann et al., 2008), which can be due to the fact that in those tissues there is a low level of glutathione, and it is more present as glutathione conjugates, which can represent difficulties for its detection (Vanacker et al., 1998; Ohkama-Ohtsu et al., 2007; Zechmann et al., 2008). The glutathione distribution is dependent on specialized proteins that are responsible for its transport. Chloroquine resistance transporter like (CLT), OLIGOPEPTIDE TRANSPORTERS (OPT), and MRP proteins from the ABC family of carriers have been shown to be involved in glutathione distribution in plants (Noctor et al., 2012; Bachhawat et al., 2013). CLT carriers are responsible for the subcellular distribution of glutathione. In Arabidopsis, there are three isoforms of CLT proteins named CLT1, CLT2, and CLT3, which are able to import glutathione from the cytosol and export glutathione intermediates from the chloroplast. Indeed, the mutation in the genes that encode CLT isoforms led to more oxidized glutathione in the cytosol rather than plastids. 63 Figure I.21. Glutathione functions. Glutathione is a tripeptide involved in a broad number of physiological functions. This is due to the importance of its endogenous level for signaling and detoxification process.

64 In addition, clt mutants presented less glutathione content and were more sensitive to heavy metal, and oomycete infections, which bold the important roles of CLT carriers in the control of glutathione synthesis (Maughan et al., 2010). In Arabidopsis, OPT6 transporter have been shown to be able to transport not only glutathione, but also GSH- conjugates to the apoplast, which seems to be important for glutathione distribution in lateral root initiation and vasculature tissues, in which there is an active cell division (Cagnac et al., 2004). Later studies confirmed that OPT6 indeed transport GSH but it is not able to transport the oxidized glutathione (Pike et al., 2009). The ATP-binding cassette transporter ATM3 located in the inner mitochondrial membrane is also suggested to export GSSG and glutathione trisulfide (GS-S-SG) from mitochondria to the cytosol (Schaedler et al., 2014). Finally, transport of glutathione conjugates and oxidized glutathione from the cytosol to the vacuole seems to be mediated by the ABC family of proteins of the class MRP. This has been suggested to be important for the redox state of glutathione in the cytosol (Noctor et al., 2012; Tommasini et al., 2018). Distribution of glutathione and glutathione-conjugates is critical for maintaining the redox state within the cell, which is important during the stress response and for redox signaling. In this regard, glutathione interaction with several specific targets during the physiological process by the so-called glutathionylation is an important aspect of the glutathione versatility in the control of plant development.

I.6.3 Glutathionylation

Glutathionylation is a posttranslational modification, in which glutathione forms a complex with cysteine residues, in a reversible manner, which is highly important for the activity control of target proteins (Michelet et al., 2005; Zaffagnini et al., 2007; Noctor et al., 2012; Diaz-Vivancos et al., 2015). Indeed, glutathionylation in a non-catalytic conserved cysteine of the chloroplastic thioredoxin f resulted in the negative control of its activity under oxidative environments (Michelet et al., 2005). In agreement to the important roles of glutathionylation in photosynthetic tissues, the activity of the chloroplastic A4-glyceraldehyde-3-phosphate dehydrogenase (A4-GAPDH), which is involved in the Calvin cycle, it is inhibited by glutathionylation and H2O2, which suggest a protective role of glutathione to A4-GAPDH under stress conditions (Zaffagnini et al., 2007). Cytoplasmic GAPDH known as GAPC involving in glycolysis is also a target of 65 glutathionylation in the presence of GSH and H2O2, suggesting a mechanism dependent of glutathione to control carbon metabolism (Zaffagnini et al., 2013). NADP-dependent isocitrate dehydrogenases (NADP-ICDH), are a family of enzymes that convert isocitrate to 2-oxoglutarate (2-OG) producing NADH or NADPH as a by-product. It was demonstrated that cytosolic cICDH isoform is under redox control in a conserved cysteine which is important for its enzymatic activity (Niazi et al., 2019). In addition, glutathionylation of conserved cysteine of RTL1 member of the RNase Three Like family is important for the control of its endonuclease activity endonucleases of double-stranded (ds)RNA (Charbonnel et al., 2017). Proteomics studies have identified many other glutathionylated proteins. However, for most of these proteins, the impact of glutathionylation has to be determined (Dixon et al., 2005; Aroca et al., 2018).

I.6.4 Glutathione-S-transferases

Several enzymes participate in the control of plant development by glutathione, such as glutathione transferases. Glutathione in its reduced form is able to conjugate with electrophilic molecules in a process mediated by glutathione transferase enzymes (GST) (Labrou et al., 2015). This process is important as a strategy to detoxify toxic compounds, and for modulating the activity of several targets proteins for the control of several signaling pathways. The GST family is characterized by harboring a N-terminal thioredoxin-like domain in which glutathione join to GST, and a α-helical C-terminal domain that is involved in the interaction with electrophilic substrates (Labrou et al., 2015; Nianiou-Obeidat et al., 2017). Based on the aminoacid sequences in photosynthetic organisms, GST are divided into 14 classes: Dehydroascorbate reductase (DHAR), Elongation factor 1Bγ (EF1Bγ), Glutathionyl hydroquinone reductase (GHR), Phi (GSTF), Hemerythrin (GSTH), Iota (GSTI), Lambda (GSTL), Theta (GSTT), Tau (GSTU), Zeta (GSTZ), Microsomal prostaglandin E synthase type 2 (mPGES-2), Tetrachloro- hydroquinone dehalogenase (TCHQD), and Ure2p (Lallement et al., 2014). In Arabidopsis, there are 55 sequences that code for GST in which 28 belong to the tau class, 13 are phi class and the rest are classified in theta, zeta, lambda, DHAR, TCHQD, and microsomal (Wagner et al., 2002; Labrou et al., 2015). They are localized in different sub- cellular compartments such as in peroxisome, mitochondria, nucleus, cytosol, plastid and 66 plasma membrane (Dixon et al., 2009; Lallement et al., 2014). Several of these GSTs are induced in the roots during biotic and abiotic stress conditions. Null mutation in the gene that encodes GSTU17 produced a decrease in the lateral root number, but at the same time an increase in the lateral root, in a mechanism dependent on glutathione and ABA, which is important during salt and drought stresses (Chen et al., 2012). In addition, GSTU5 is induced by hormone treatments in the roots like auxin, ABA, and cytokinin (Van Der Kop et al., 1996). Members of the phi family GSTF8 and GSTF2 are expressed in the root system after auxin treatment while GSTF10 is induced by salicylic acid. Family tau is expressed in roots where GSTU5 and GSTU19 have been shown to be induced by exogenous application of auxin (Moons, 2005). Moreover, it has been reported that IBA treatments induce the expression of several GST, particularly GSTU4 and GSTU8 that are also induced during phosphorous deprivation. This studies bold the importance of the control of the root system by glutathione mediated by glutathione-S-transferase enzymes (Thibaud et al., 2010; Xuan et al., 2015). Besides GSTs, glutathione is also able to interact with other class of enzymes known as glutaredoxins, in which GSH serves as a substrate for electron donors to GRXs.

I.6.5 Glutaredoxins system in the control of the root development

Besides the control of reactive oxygen species, glutathione is capable to modulate the activity of specific proteins with cysteine residues by a mechanism mediated by glutaredoxins (GRX). Glutaredoxins use glutathione as a source of reducing power to transfer electrons to target proteins that are involved in several metabolic processes. In the Arabidopsis genome, there are 50 genes that encode glutaredoxins and are divided into 5 classes. The class I of glutaredoxins include glutaredoxins GRXC1, GRXC2, GRXC3, GRXC4, GRXC5, and GRXS12. While GRXS14, GRXS15, GRXS16, and GRXS17 are members of the second class, the third class is the biggest group of glutaredoxins known as the ROXY family and the members are enumerated from ROXY1 to ROXY21. Glutaredoxins with the motif CXXC belong to the group IV and as the ROXY family, there are named from 4CXXC1 to 4CXXC13. Finally, the last class of glutaredoxins are composed by CPFC1, CPFC2, CPFC3, CPFA3, CPFA4, and CPFA5 (Figure I.22) (Meyer et al., 2012; Belin et al., 2015). 67 Figure I.22. Glutaredoxin family. In this scheme is shown the classification of the members that composed the family of glutaredoxin. In Arabidopsis, there are 50 genes that encode to glutaredoxins, which are divided into 5 classes according to their amino acid sequence. Figure adapted from Belin et al., 2015.

68 Few studies have addressed the importance of glutaredoxins in root development. However, it has been shown that several members of the glutaredoxin clade are highly expressed in different root tissues by which we could envisage that there are possible critical roles that GRXs play during the root development (Table I.1) (Belin et al., 2015). Genetic studies have shown that a null mutant of the glutaredoxin GRXS17 is critical during primary root growth and meristem activity. Biochemical and genetic analysis demonstrated that GRXS17 interact with Nuclear Factor Y Subunit C11/Negative Cofactor 2a (NF-YC11/NC2a) and BolA2 that are involved in abiotic stress and Fe-S cluster formation respectively (Cheng et al., 2011; Couturier et al., 2014; Knuesting et al., 2015; Yu et al., 2017). Glutaredoxins are also involved in the response to abiotic stress conditions (Patterson et al., 2016). Exogenous application of nitrogen sources induce the accumulation mRNA that encodes members of the ROXY family. Moreover, RNAi lines against ROXY16 were resistant to the effect of N-deprivation in the elongation of the primary root, a phenotype that seems to be dependent on the plant hormone cytokinin, suggesting a negative role of ROXY16 in the control of root responses to elevated accumulation of nitrogen in the rhizosphere (Patterson et al., 2016). Unravel the functions of Glutathione-Glutaredoxin system in plants have been a tough task due to the possible redundancy among the glutaredoxin members, and also the involvement of backup systems that can rescue the activity of glutathione as electron donor.

I.6.6 Transcriptomic control by glutathione

Studies regarding the control of glutathione in gene expression have be done in mutants of their synthesis, and in plants treated with the pharmacological inhibitor of GSH1 enzyme known as BSO (Ball et al., 2004; Koprivova et al., 2010; Schnaubelt et al., 2015; Fukushima et al., 2017). In the mutant rml1, there was a differential expression profile in both root and shoot organs in a broad number of biological pathways, especially in genes involved in stress responses, which is in agreement with the pivotal roles that glutathione play in ROS homeostasis. Moreover, there were several genes involved in hormone metabolism, such as auxins, abscisic acid, and ethylene, that were up and down- regulated in the mutant rml1.

69 Table I.1 Gene expression of GRX in root tissues. Information regarding the implications of glutaredoxins during root development is scarce in the research field. However, by in silico analysis, we showed that several genes that encode to glutaredoxins are differentially expressed in different tissues of the root system in Arabidopsis, suggesting putative roles of GRX in the development of the primary root. Future studies should be focused in the genetic characterization in the root system in members of class I and II as well as ROXY19 and CPFC3, due to the high level of expression in different root tissues. Adapted from Belin et al., 2015.

70 Intriguingly, the expression of the gene that encodes to the auxin transporter PIN5 was significantly decreased in the mutant rml1 compared to wild-type plants, which give more evidence of the roles of glutathione level in hormone pathways (Schnaubelt et al., 2015). In agreement with the transcriptomic results found in the mutant rml1, plants treated with the glutathione biosynthesis inhibitor buthionine sulfoximine (BSO), and in the mutants rax1-1 and cad2 showed differential expression profile in several pathways related with both stress response and developmental pathways (Ball et al., 2004; Koprivova et al., 2010). Moreover, comparing the transcriptomic data of the mutants gr1 and cat2 which accumulated more GSSG level showed an overlapping expression of genes involved in plant hormones such as jasmonic acid. Indeed, analysis of the double mutant cat2gr1 demonstrated the important roles of glutathione redox state in H2O2 signaling (Mhamdi et al., 2010a).

I.7 REACTIVE OXYGEN SPECIES AND AUXINS

It is well known that ROS affect auxin response, decreasing polar auxin transport and inducing the expression of auxin-related genes (Yang et al., 2014; Orman-Ligeza et al., 2016; Tognetti et al., 2017). For example, the expression of the glucosyltransferase

UGT74E2 gene that controls the auxin IBA metabolism is induced by H2O2, and the ectopic expression of UGT74E2 produces resistance to osmotic stress and other abiotic stress conditions (Tognetti et al., 2010). Moreover, increased ROS accumulation was monitored in the elongation zone of the root after few hours of incubation with IBA, which is dependent on the activity of the peroxisome-localized copper amine oxidase ζ (CuAOζ). This suggests a possible role of reactive oxygen species in the control of the IBA pathway (Qu et al., 2017). The link between ROS and auxin can be also seen in mutants that accumulated more

H2O2 such as in the case of cat2, which is not able to survive during photorespiratory conditions, a phenotype that is rescued by exogenous applications of auxin-related compounds (Kerchev et al., 2015). Moreover, inhibition of the CAT2 activity also decreases the level of IAA, by the oxidation of tryptophan during biotic stress conditions (Yuan et al., 2017). Enzymes involved in the tryptophan-dependent biosynthesis of IAA have been shown to be responsive to ROS (Cha et al., 2015). Overexpression of the gene that encodes the enzyme flavin monooxygenase YUC6 led to enhanced resistance to 71 drought stress. This was dependent on an increment in the antioxidant capacity of Arabidopsis plants, due to a moonlighting NADPH-dependent disulfide oxidoreductase activity of YUC6 (Cha et al., 2015). The control of ROS on the auxin transport is not fully elucidated yet. Nevertheless, it has been shown that phosphatidylinositol 3-kinase (PtdIns- 3-kinase) and Ca2+ that are required for the induction of RBOH activity is also able to control the PIN localization at the plasma membrane, suggesting that PtdIns-3-kinase and Ca2+ levels are key elements linking ROS and auxin transport (Zhang et al., 2011). In addition, it has been shown that ROS biosynthesis by RBOH activity is controlled by auxin, in a mechanism involving the auxin-inducible gene RSL4 which is also important for root hair elongation (Mangano et al., 2018). Exogenous application of ROS also induces lateral root formation, which is impaired in auxin mutants such as aux1 and lax3 in a process mediated by RBOH enzymes (Orman-Ligeza et al., 2016). Positive regulation of LR formation as well as in root hair development has been also seen with nitric oxide (NO), in a mechanism that is related to auxin transport and signaling (Sanz et al., 2015). Indeed, nitric oxide positivelycontrol the auxin receptor TIR1/AFB activity by S- nitrosylation (Terrile et al., 2012). However, elevated concentration of NO, as in the case of the mutant nox1, resulted in a decrease of auxin transport in a process involving the auxin carrier PIN, which resulted in a phenotype similar to pin mutants, suggesting that the effect of NO in the auxin pathway is highly dependent on the cellular level of NO (Fernandez-Marcos et al., 2011; Sanz et al., 2014).

72 OBJECTIVES OF THIS STUDY

As have been described so far in this chapter glutathione and auxin are important components in the control of the root system architecture. Indeed, it is well known that decreasing the level of endogenous glutathione by the GSH1 inhibitor buthionine sulfoximine (BSO) produced an impairment in the root system architecture, which have been attributed to the decrease of auxin signaling and transport (Bashandy et al., 2010; Koprivova et al., 2010; Tognetti et al., 2017). Moreover, decreasing glutathione results in a transcriptional alteration in a similar manner as when plants are altered in the auxin synthesis (Koprivova et al., 2010; Tognetti et al., 2017). However, the interplay between glutathione and auxin have been difficult to establish. For this reason, in the first chapter, I first focused in on the link between glutathione and the auxin pathway to control the root development. Secondly, I performed several genetic analysis to find players of the control of auxin by glutathione, and third I study the physiological meaning of glutathione controlling the auxin pathway. Finally, as some of the functions of glutathione are dependent of the glutaredoxin family and it was demonstrated that GRXS17 is critical for the development of the root system (Cheng et al., 2011; Knuesting et al., 2015), I studied the physiological pathways in which GRXS17 is involved to control the root architecture, specially in the primary root and lateral root development. For this reason, I performed genetic and bioinformatic analysis to find components GRXS17-dependent that are involved in the control of the root system architecture. Additionally, our group found that the mutant roxy19 presented an increase in the rate of lateral root elongation compared to wild-type plants (data not published). For this reason, I examined the roles of ROXY19 in the root system development. The next part of this manuscript is divided into three different chapters, each of them focuses on one of the objectives described above. Chapter I aim to describe the research concerning the roles of glutathione in control the auxin IBA, which is highly important for lateral roots and root hair growth. The second and third chapters are exploring the roles of some glutaredoxins in the control of the root development.

73 74 II RESULTS

75 76 II.1 CHAPTER 1 ROOT SYSTEM RESPONSES TO IBA ARE CONTROLLED BY GLUTATHIONE

The result of chapter I is presented in the form in which we are going to submit for publication. However, the enumerating of the figures are adapted to the thesis manuscript format.

IBA endogenous auxin regulates Arabidopsis root system development in a glutathione-dependent way and is important for adaptation to phosphate deprivation.

José A. Trujillo-Hernandez1,2, Laetitia Bariat2,1, Lucia C. Strader3, Jean-Philippe Reichheld2,1 and Christophe Belin1,2,*

Running title: IBA and glutathione interplay in root development Abstract: 207 words

1 Laboratoire Génome et Développement des Plantes, Université Perpignan Via Domitia, F- 66860 Perpignan, France.

2 Laboratoire Génome et Développement des Plantes, CNRS, F-66860 Perpignan, France. 3 NSF Science and Technology Center for Engineering Mechanobiology, Department of Biology, Washington University in St. Louis, St. Louis, MO 63130, USA

*Corresponding author E-mail: [email protected] tel. +33 4 68 66 17 54

77 Abstract

Root system architecture results from a highly plastic developmental process to perfectly adapt to environmental conditions. In particular, the development of lateral roots (LR) and root hair (RH) growth are constantly optimized to the rhizosphere properties, including biotic and abiotic constraints. Every step of root system development is tightly controlled by auxin, the major morphogenetic hormone in plants. Glutathione, a major thiol redox regulator, is also critical for root system development but its interplay with auxin is still scarcely understood. Indeed, previous works showed that glutathione deficiency does not alter root responses to exogenous indole acetic acid (IAA), the main active auxin in plants.

Because indole butyric acid (IBA), another endogenous auxinic compound, is an important source of IAA for the control of root development, we investigated the crosstalk between glutathione and IBA during root development. We show that glutathione deficiency alters

LR and RH responses to exogenous IBA but not IAA. Although many efforts have been deployed, we could not identify the precise mechanism responsible for this control.

However, we could show that both glutathione and IBA are required for the proper responses of RH to phosphate deprivation, suggesting an important role for this glutathione-dependent regulation of auxin pathway in plant developmental adaptation to its environment.

78 INTRODUCTION

The plasticity of root system development, combining root growth and root branching, is essential for plants to sustain, adapt and optimize their growth in changing environmental conditions, such as nutrient and water availability, rhizosphere microbiome or soil structure heterogeneity. Root growth relies on the activity of the root apical meristem that regulates histogenesis via the control of cell proliferation. Root branching is a complex organogenesis process that allows the development of new lateral roots (LR) from regularly spaced pericycle founder cells in Arabidopsis thaliana. These founder cells are originally specified by an oscillatory mechanism occurring in the root tip and involving cyclic programmed cell death of lateral root cap cells (Moreno-Risueno et al., 2010; Xuan et al., 2016). In addition to the root system architecture, the development of epidermal root hairs (RH) is particularly sensitive to changes in environmental conditions and contribute to the root system adaptation (Grierson et al., 2014).

Among nutrients, phosphorus has a key role in all living organisms and is one of the main limiting factors for plant growth and crop productivity (Leinweber et al., 2018). Its homeostasis is highly dependent on phosphate uptake by the root cap (Kanno et al., 2016).

Phosphate concentration strongly impacts root system development in many plant species including Arabidopsis (López-Bucio et al., 2002, 2003; Hodge, 2004). In particular, general low phosphate availability is well known to increase RH length (Bates and Lynch,

1996). Biotic factors also modulate root development. Among them, rhizospheric PGPR

(plant growth promoting rhizobacteria) are able to stimulate RH elongation (Poitout et al.,

2017).

Auxins, and particularly the most abundant endogenous one, indole acetic acid (IAA), play a pivotal and integrative role in all steps of root system development, both under optimal or

79 stress conditions (Saini et al., 2013; Korver et al., 2018). During LR development, auxin is responsible for the activation of founder cells, the development and organization of the LR primordia, and the emergence of the newly formed LR through the external layers of the primary root. Even earlier, the oscillatory positioning of pericycle founder cells depends on auxin maxima generated via auxin release by dying root cap cells (Moreno-Risueno et al.,

2010; Du and Scheres, 2018). Auxin also modulates RH elongation, which has been shown to mediate responses to stimuli such as abscisic acid, PGPR or phosphate deprivation

(Wang et al., 2017; Poitout et al., 2017; Nacry et al., 2005). Recently, a set of works on

Arabidopsis and rice evidenced that RH response to phosphate deprivation requires an increase in auxin de novo synthesis in the root cap, and its apico-basal transport to the epidermal cells via the auxin influx facilitator AUX1 (Giri et al., 2018; Bhosale et al.,

2018; Parry, 2018).

Indole-3-butyric Acid (IBA) is a structural derivative of IAA, differing by only two additional carbons in the side chain (Korasick et al., 2013). Although convincing evidence support a role for IAA in IBA biosynthesis, the mechanisms responsible and enzymes involved are still unknown (Ludwig-Müller et al., 1995; Frick and Strader, 2018). It is now broadly accepted that IBA solely acts as an IAA precursor through its -oxidative decarboxylation in peroxisomes (reviewed in Frick and Strader, 2018). This enzymatic process involves several enzymes, some shared with other -oxidation pathways such as

PED1, others apparently specific to the IBA-to-IAA conversion, such as IBR1, IBR3,

IBR10 and ECH2 (Strader and Bartel, 2011; Frick and Strader, 2018). IBA homeostasis is also regulated via its transport and conjugation, but only few regulators have been identified. Type G ABC transporters ABCG36 and ABCG37 can efflux IBA from the cells

(Strader and Bartel, 2009; Ruzicka et al., 2010), while the generic PXA1/COMATOSE

80 transporter is likely to import IBA into the peroxisome. Like other auxins, IBA is known to conjugate with sugars and amino acids. UGT74E2 and UGT75D1 can conjugate IBA to glucose (Tognetti et al., 2010; Zhang et al., 2016), whereas GH3.15 has recently been shown to be able to conjugate IBA to amino acids (Sherp et al., 2018). IBA-derived IAA plays important roles during root development, including root apical meristem maintenance and adventitious rooting. IBA conversion to IAA is also critical for RH elongation (Strader et al., 2010). Finally, recent works have reported the critical role of

IBA-to-IAA conversion in the root cap as a source of auxin for the oscillatory positioning of LR founder cells (De Rybel et al., 2012; Xuan et al., 2015). IBA-to-IAA conversion taking place in the LR primordium itself also likely participates in further LR development

(Strader and Bartel, 2011). Despite all these reported functions in root development, the importance of IBA-derived IAA relatively to other IAA sources is still poorly understood and scarcely documented, particularly in case of changing environmental constraints.

Changes in environmental conditions result in Reactive Oxygen Species (ROS) imbalance, thus affecting the general redox cellular homeostasis. For example, phosphate deficiency alters ROS production in roots (Tyburski et al., 2009). ROS modulate many aspects of plant development, including root system development (Singh et al., 2016; Tsukagoshi,

2016). Controlled ROS production by NADPH oxidases are particularly involved in RH elongation and LR development (Orman-ligeza et al., 2016; Mangano et al., 2016; Carol and Dolan, 2006; Mangano et al., 2018). Auxin and ROS pathways interact to control root development, since auxin can trigger ROS production and in turn ROS can affect IAA levels and transport (Orman-ligeza et al., 2016; Tognetti, 2017; Zwiewka et al., 2019).

Although ROS are important signalling molecules, they are also potent oxidants that can damage proteins, lipids or nucleic acids. Organisms have evolved many antioxidant

81 systems to prevent or repair such damages (Noctor et al., 2018). Among them are several detoxifying enzymes such as catalases or peroxidases, able to reduce the most stable ROS, hydrogen peroxide (H2O2), but also several antioxidant molecules such as ascorbate or glutathione. Glutathione is a small and stable thiol-containing tripeptide (Glu-Cys-Gly) essential for plant survival and present in large concentrations in cells, up to several millimolar (Noctor et al., 2012). It has many roles in almost all cellular compartments, including the detoxification of heavy metals and xenobiotics, sulfur homeostasis, ROS homeostasis and redox signalling. To ensure these functions, glutathione can act as a precursor (e.g. phytochelatins), be conjugated to other molecules via the Glutathione-S- transferase enzyme family, or serve as an electron donor for antioxidant systems (Noctor et al., 2012). Reduced glutathione (GSH) is therefore converted to oxidised glutathione

(GSSG) which is reduced back by glutathione reductases. In standard growth conditions

GSH is in large excess relative to GSSG. Glutathione biosynthesis is a 2-step reaction, with

GSH1 encoding the rate-limiting -glutamylcysteine synthetase and GSH2 encoding a glutathione synthetase (Noctor et al., 2012). Knock out mutations in any of these two genes are lethal. However, genetic screens identified several knock-down alleles of GSH1, allowing to get plants with reduced glutathione levels (Vernoux et al., 2000; Ball et al.,

2004; Jobe et al., 2012; Howden et al., 1995; Parisy et al., 2007; Shanmugam et al., 2012).

The importance of glutathione in root development is illustrated by the absence of root meristem maintenance in rml1 mutant, together with the impairment of primary root growth and LR development in less severe gsh1 alleles (Bashandy et al., 2010; Koprivova et al., 2010; Marquez-Garcia et al., 2014; Vernoux et al., 2000).

We previously reported that cad2 or pad2 mutants can respond almost normally to exogenous IAA for root development (Bashandy et al., 2010), although auxin transport

82 seems to be affected in glutathione-deficient plants (Koprivova et al., 2010; Bashandy et al., 2010). The role of glutathione in controlling root development is therefore still misunderstood. Given the importance of IBA-derived IAA in regulating root development, we chose to address the role of glutathione in root responses to IBA. We show that glutathione deficiency impairs LR and RH responses to IBA but not IAA. We then try to identify the glutathione-dependent mechanisms required for IBA responses but every IBA- related pathway examined does not reveal sensitivity to glutathione levels. We finally suggest a physiological role for this glutathione-dependent IBA response, showing that

IBA and glutathione pathways are critical for RH responses to phosphate deficiency.

RESULTS

Glutathione deficiency specifically alters LR and RH responses to IBA but not IAA

Since IBA plays important roles during root development, we investigated root responses to both IAA and IBA in several gsh1 weak alleles and in plants treated with Buthionine

Sulfoximine (BSO), a chemical specific inhibitor of GSH1 enzyme.

We first wanted to precisely know the glutathione levels in the different genotypes in our growth conditions, thus we quantified endogenous glutathione in whole 8-day old seedlings (Figure II.1.1A). We find the same amount of total glutathione (i.e. about 25% of wild-type content) in cad2, pad2 and zir1 mutants, which was expected for cad2 and pad2 whereas zir1 has previously been reported to have lower glutathione contents (about 15% relative to wild-type). We show that 0.5mM exogenous BSO also reduces endogenous glutathione levels of wild type plants to approximately the same levels. In addition, we report that exogenous IBA treatment does not impact endogenous glutathione levels.

83 84 Supplementary Figures II.1.1A and II.1.1B show that cad2 and pad2 mutants display the same primary root growth as the wild-type in our growth conditions, and that the primary root responds normally to both IAA and IBA. In the same way, BSO treatment (0.5mM) neither affects primary root growth nor its response to IAA or IBA. zir1 mutant shows a strongly reduced primary root growth under normal conditions but responds normally to both IAA and IBA. Since we measured the same total glutathione levels in zir1 than in cad2 or pad2 but observe a severely reduced growth of that allele, we decided to focus on pad2, cad2 and BSO treatment in the next experiments.

As expected, both IAA and IBA also induce LR density in the mature zone of the root in

10-day-old seedlings (Figure II.1.1B). LR density in both cad2 and pad2 mutants responds normally to IAA treatment but displays hyposensitivity to exogenous IBA. Interestingly, the addition of 0.5 mM BSO phenocopies cad2 and pad2 mutants, and this BSO phenotype is dose-sensitive (Supplementary Figure II.1.2A). Finally, we could revert cad2 hyposensitivity to IBA by adding exogenous GSH (Supplementary Figure II.1.2B).

Because emerged LR density depends both on LR initiation and subsequent LRP development, we also addressed the LRP density in the same conditions. Figure II.1.1C reveals that both LRP and emerged LR densities display IBA hyposensitivity in plants with low glutathione levels, suggesting that glutathione affects LR development upstream of

LRP development. Another way to investigate LRP development is to force and synchronize LRP initiation by gravistimulation (Péret et al., 2012). We therefore gravistimulated glutathione-deficient plants, in the absence or presence of exogenous IBA

(10 µM). We first observe that exogenous IBA treatment slightly slows down wild-type

LRP development upon gravistimulation in our growth conditions (Figure II.1.1D).

85 Figure II.1.1. Glutathione specifically regulates root responses to IBA. (A) Total glutathione content in 8-day-old wild-type (Col-0), cad2, pad2 or zir1 seedlings grown on standard ½ MS medium. Wild-type plants grown in the presence of 0.5 mM BSO were also assayed (BSO). Data represent the means of 3 biological repetitions. (B) Emerged lateral root density of 10-day-old wild-type (Col-0), cad2 or pad2 plants grown on standard ½ MS medium and wild-type plants grown in the presence of 0.5 mM BSO (n≥15). (C) Emerged LR and LR primordia density of 8-day-old wild- type plants grown on standard ½ MS medium (control), or in the presence of different combinations of BSO 0.5 mM, IBA 10 µM and/or IAA 50 nM, as mentioned (n≥15). (D) Developmental stages of LR primordia of 6-day-old wild-type (Col-0), cad2 or pad2 seedlings 48 hours after gravi-stimulation on standard ½ MS medium; wild-type plants grown in the presence of 0.5 mM BSO were also assayed (BSO). Data indicate the percentage of LRP in each developmental stage (n≥16) and are representative of 3 independent experiments. (E) Pictures of representative 8-day-old wild-type plants grown on standard ½ MS medium, or in the presence of different combinations of BSO 0.5 mM, IBA 10 µM and/or IAA 50 nM, as mentioned. (F) Quantification of root hair length of 8- day-old wild-type (Col-0), cad2 or pad2 mutant plants grown on standard ½ MS medium (Control), or in the presence of 50 nM IAA (IAA) or 10 µM IBA (IBA), . Wild-type plants in the presence of 0.5 mM BSO (BSO) were also assayed (n≥100). Histograms represent the mean and error bars represent the standard deviation. Asterisks indicate a significant difference, based on a two-tailed Student t-test (*=P≤0.01, **= P≤0.001 ).

86 We also observe that cad2, pad2 and BSO-treated plants behave exactly like wild-type in such assay. Taken together, these results tend to support a role for glutathione in IBA- dependent specification or activation of founder cells prior to the LR development process.

Finally, we monitored RH elongation responses to IBA and IAA in glutathione-deficient plants (Figure II.1.1E and II.1.1F). As for LR, we notice that cad2 and pad2 mutants are hyposensitive to the IBA-dependent induction of RH growth but respond normally to IAA.

Similarly, 0.5 mM BSO treatment also reduces the RH response to IBA but not IAA. We can therefore conclude that glutathione is also required for the IBA-dependent induction of

RH growth.

Glutathione levels affect auxin signalling in the basal part of the meristem.

We have shown that glutathione deficiency alters RH elongation and founder cells specification or activation in response to IBA. We know that root tip-derived IAA transits trough the lateral root cap to regulate both founder cell positioning and RH growth in the basal part of the meristem. We therefore investigated auxin response in the root tip, by using the DR5:GUS auxin signalling reporter (Ulmasov et al., 1997). In standard growth conditions, we only reveal GUS staining in the quiescent center and the columella, and

BSO has apparently no effect (Figure II.1.2A). As expected, exogenous IAA treatment for

24 hours induces auxin response in the whole root tip and root epidermis, and the presence of BSO has again no effect on this distribution. In response to IBA, GUS staining increases in the distal meristem and strongly appears specifically in the trichoblast epidermal cell files in the basal part of the meristem, up to the differentiation zones. In the presence of

BSO, this IBA-dependent strong signal in the basal meristem almost disappears while the signal in the distal meristem remains strong.

87 Figure II.1.2. Glutathione is critical for the IBA-derived IAA signalling in the basal meristem. (A) Representative Pictures of 8-day-old GUS-stained DR5:GUS plants grown on standard ½ MS medium (Control) or in the presence of 0.5 mM BSO (BSO), after a 24- hour long treatment with 10 µM IBA or 50 nM IAA as precised (n≥7). (B) Quantification of GFP and Tomato fluorescent signals from 8-day-old expressing the DR5:n3EGFP/DR5v2:ntdTomato double reporter, and grown in the same conditions as A (n≥7).

88 In order to confirm these results, we used another auxin signalling marker that allows quantification, the DR5:n3EGFP/DR5v2:ntdTomato double reporter line (Liao et al.,

2015). We quantified both GFP and Tomato fluorescence in the distal and basal parts of the meristem (Figure II.1.2B). In agreement with the DR5:GUS reporter line, both IBA and

IAA treatments increase auxin signalling in distal and basal parts of the meristem. The presence of BSO does not significantly alter auxin signalling, neither in standard conditions nor upon exogenous IAA treatment. However, we can confirm that BSO severely represses IAA signalling in the basal part of the meristem in response to exogenous IBA, while it has limited effect in the distal part.

Hence we show that glutathione is specifically required for IBA-derived IAA signalling in the basal part of the meristem, where LR founder cells are specified and RH growth is determined. Moreover, we show that glutathione does not affect IAA signalling components since auxin signalling reporters respond normally to exogenous IAA when glutathione content is depleted.

IBA-derived IAA responses are specifically affected by glutathione deficiency

Because glutathione is a critical regulator of general redox homeostasis, we addressed root responses to IBA in other redox-related mutants. We chose to analyse gr1 mutant, affected in cytosolic and peroxisomal glutathione reduction, cat2 mutant, affected in H2O2 detoxification, and ntra ntrb double mutant affected in the thioredoxin-dependent thiols reduction system. Quantification of both GSH and GSSG (Figure II.1.3A) shows that redox-related mutants have higher total glutathione concentrations than wild type plants, both in the absence or in the presence of exogenous IBA that has almost no effect. As expected, cat2 and gr1 mutants display high glutathione oxidation status.

89 90 In contrast with cad2, none of the other mutants display any LR density hyposensitivity to

IBA (Figure II.1.3B). In addition, the use of the roGFP2 probe allowed us to confirm that the presence of IBA does not generate any imbalance in glutathione redox status in root tissues (Figure II.1.3C and Supplementary Figure II.1.3). These results suggest that root responses to IBA specifically depend on glutathione overall levels rather than glutathione redox status or any general redox imbalance.

IAA transport from the distal to the basal meristem is not targeted by glutathione

We have just shown that IBA-derived IAA response in the basal meristem is dependent on glutathione levels. We also know that AUX1 is required for IAA to regulate RH growth in response to phosphate deprivation and that IBA-to-IAA conversion in the root cap is required for LR founder cells specification in the basal meristem. AUX1 auxin influx facilitator and PIN2 auxin efflux carrier are the IAA transporters that ensure the apico- basal IAA flux in the root cap and the root tip epidermis. Finally, we previously published that strong BSO treatment induces long-term decrease in the expression of several genes encoding IAA transporters, including AUX1 and PIN members.

In order to determine if AUX1 and/or PIN2 are the targets of glutathione to modulate root responses to IBA, we examined LR density in aux1 and pin2 mutants. As shown in Figure

II.1.4A, aux1 and pin2 mutants still display hyposensitivity to IBA when BSO is present.

In other words, the glutathione regulation still occurs in both aux1 and pin2 mutants, revealing that neither AUX1 nor PIN2 is regulated by glutathione to control root responses to IBA. Because AUX1 and PIN2 are members of multigene families, we wanted to ensure that other members of AUX/LAX or PIN families are not replacing AUX1 and PIN2 in respective mutants. We therefore addressed IBA responses with or without BSO treatment in the presence of specific inhibitors of both families (Figure II.1.4B).

91 Figure II.1.3. IBA hyposentitivity phenotype is specific to GSH-depleted plants. (A) Glutathione content in 8-day-old wild-type (Col-0), gr1, ntra ntrb or cat2 seedlings grown on standard ½ MS medium. Data represent the means of 3 biological repetitions. Reduced glutathione is represented in white (GSH) and oxidised glutathione in black (GSSG). (B) Emerged lateral root density of 10-day-old wild-type (Col-0), cad2, cat2, gr1 or ntra ntrb plants grown on standard ½ MS medium or the presence of 10 µM IBA (n≥16). (C) Representative pictures (n≥10) of roGFP2 reporter line grown on standard ½ MS medium (MS), submitted to a 30 min treatment with 1 mM H 2O2 then to an additional 30 min treatment with 10 mM DTT (top panel). The bottom panel represents roGFP2 reporter plants grown on standard ½ MS medium supplemented or not with 10 µM IBA. Pictures are made from the ratio between the oxidised roGFP2 signal (excitation at 405 nm) and the reduced roGFP2 signal (excitation at 488 nm). Ratio values are represented in the color scale. Histograms represent the mean and error bars represent the standard deviation. Asterisks indicate a significant difference, based on a two-tailed Student t-test (*P<0.01; **P<0,001).

92 We can observe that BSO still leads to LR hyposentitivity to IBA both in the presence of

NPA, that inhibits PIN-dependent auxin efflux, and NOA, a specific inhibitor of

AUX/LAX influx facilitators. All together these data suggest that the glutathione- dependent control of root responses to IBA does not affect IAA transport from the root apex to the basal meristem.

Looking for glutathione targets in IBA pathways.

Since IAA transport was not affected, we logically investigated IBA homeostasis in plants with low glutathione content. Interestingly, gsh1 mutants display pleiotropic phenotypes opposite to the pleiotropic phenotypes of mutant alleles in PDR8/PEN3/ABCG36 for IBA sensitivity, but also for sensitivity to Pseudomonas and to cadmium (Kim et al., 2007; Lu et al., 2015; Strader and Bartel, 2009). ABCG36 is responsible for IBA efflux from the cells, which prompted us to investigate the IBA import into plant cells in the cad2 mutant.

As shown in Figure II.1.5A, we cannot detect any impairment in 3H-IBA accumulation in cad2. This suggests that IBA import into the cell is not perturbed by glutathione deficiency.

In order to better investigate IBA transport, we analysed the glutathione-dependent LR and

RH response to IBA in abcg36/pdr8 and abcg37/pdr9 mutants. As expected, pdr8 mutant displays hypersensitivity to exogenous IBA (Figure II.1.5B and Supplementary Figure

II.1.4). However, both mutants are still clearly resistant to IBA in the presence of BSO.

Because of putative redundancy between these two proteins, we generated the pdr8 pdr9 double mutant. As for single mutants, LR density and RH length in the double mutant are induced by IBA but BSO is still able to decrease this response. This result means that IBA efflux transporters ABCG36 and ABCG37 are not the targets of glutathione-dependent regulation.

93 Figure II.1.4. IAA transport is not the target of glutathione. (A) Emerged lateral root density of 10-day-old wild-type (Col-0), aux1 or pin2 plants grown on standard ½ MS medium, or in the presence of different combinations of BSO 0.5 mM and IBA 10 µM (n≥10). (B) Emerged lateral root density of 10-day-old wild-type (Col-0) grown on standard ½ MS medium, supplemented or not with the transport inhibitors NPA 1 µM or NOA 1 µM (n≥15). Histograms represent the mean and error bars represent the standard deviation. Asterisks indicate a significant difference, based on a two-tailed Student t-test (**P<0,001).

94 We know that IBA homeostasis is also regulated via conjugation with glucose, and this depends on the activity of UGT74E2 and UGT75D1 glycosyltransferases. Again, we could observe that LR and RH responses to IBA are still BSO-sensitive in both mutants (Figure

II.1.5C and Supplementary Figure II.1.4). Here also, because of putative redundancy, we generated the double mutant. However, the response to IBA is still reduced in the presence of BSO in this double mutant. Therefore, glycosyltransferases are not the targets of glutathione-dependent regulation.

Finally, we also wanted to investigate the enzymatic pathway involved in the IBA-to-IAA conversion in the peroxisome. Figure II.1.6A shows that ibr1, ibr10 and ibr1 ibr3 ibr10 mutants are fully insensitive to exogenous IBA in our conditions, thus making impossible to genetically address the putative dependency of IBR1 and IBR10 to glutathione levels.

However, we can notice that in ech2 and ibr3 mutants, LR are still hyposensitive to IBA in the presence of BSO. Again, this suggests that ECH2 and IBR3 are not regulated in a glutathione-dependent manner. In order to detect an eventual defect in gene expression in glutathione-deficient plants, we analysed the expression level of genes involved in IBA-to-

IAA conversion (Figure II.1.6B). Surprisingly, ECH2, IBR3, IBR10, AIM1 and PED1 genes were all moderately (20 to 50 % increase) but consistently upregulated in cad2 mutant compared to the wild-type. Only IBR1 does not display any significant increase. In any case, this does not allow us to identify any target that could be transcriptionally down- regulated upon glutathione depletion. To complete this study and know if IBA conversion is affected, we tried to measure IBA-to-IAA conversion rate but unfortunately failed to obtain such data.

95 Figure II.1.5. IBA homeostasis does not display sensitivity to glutathione. (A) Uptake of [3H]IBA or [3H]IAA by wild type (Col), pen3-4, cad2-1, pcs1 or abcc10 excised root tips of 8-day-old plants. Data represent eight replicates with five root tips per genotype and per replicate. (B) Emerged lateral root density of 10-day-old wild- type (Col-0), pdr8, pdr9 or pdr8 pdr9 plants grown on standard ½ MS medium, or in the presence of different combinations of BSO 0.5 mM and IBA 10 µM (n≥15). (B) Emerged lateral root density of 10-day-old wild-type (Col-0), ugt74e2, ugt75d1 or ugt74e2 ugt75d1 plants grown on standard ½ MS medium, or in the presence of different combinations of BSO 0.5 mM and IBA 10 µM (n≥15). Histograms represent the mean and error bars represent the standard deviation. Asterisks indicate a significant difference, based on a two-tailed Student t-test (**P<0,001).

96 To conclude, we were not able to identify which IBA-dependent pathway is modulated by glutathione, but this is not surprising since the knowledge about IBA homeostasis is still very scarce.

IBA and glutathione control RH responses to phosphate deprivation

We have shown that LR and RH responses to exogenous IBA is dependent on glutathione levels. We then wanted to understand the physiological significance of such control. Since it had recently been shown that root-cap derived auxin regulates RH responses to phosphate deprivation, we decided to investigate RH responses to diverse environmental stimuli. Hence, we analysed RH elongation rate in response to phosphate starvation and to

Mesorhizobium loti (M. loti), a well-described PGPR. We can observe in Figures II.1.7A and 7B that both stimuli increase RH length in wild-type plants. Both cad2 and ibr1 ibr3 ibr10 mutants display wild-type like RH response to M. loti, suggesting that neither glutathione nor IBA participate in RH elongation response to PGPR (Figure II.1.7A).

In contrast we can observe that cad2 mutant is clearly hyposensitive to phosphate deprivation, and that RH elongation response to phosphate starvation is almost fully abolished in ibr1 ibr3 ibr10 mutant (Figure II.1.7B). We also confirm the reduction in glutathione levels in cad2 mutant while ibr1 ibr3 ibr10 mutant does not alter glutathione content and redox status (Supplementary Figure II.1.5). Interestingly, phosphate deprivation does not significantly affect total glutathione content in plants but increases the

GSH/GSSG ratio.

We wondered if cad2 and ibr1 ibr3 ibr10 mutants affect the general response to phosphate deficiency or specifically the response of root hairs to this abiotic stress. We therefore quantified the expression of marker genes known to be induced in response to phosphate deprivation.

97 Figure II.1.6. IBA conversion to IAA does not display sensitivity to glutathione. (A) Emerged lateral root density of 10-day-old wild-type (Col-0), ibr1, ibr3, ibr10, ibr1 ibr3 ibr10 or ech2 plants grown on standard ½ MS medium, or in the presence of different combinations of BSO 0.5 mM and IBA 10 µM (n≥15). (B) Expression levels (qRT-PCR) of IBR1, IBR3, IBR10, ECH2, AIM1 and PED1 genes in cad2 relative to wild-type 8-day-old plants grown on standard ½ MS medium. Data represent three independent replicates with 20 plants per replicate. Histograms represent the mean and error bars represent the standard deviation. Asterisks indicate a significant difference, based on a two-tailed Student t-test (*P<0,01;**P<0,001).

98 Figure II.1.7C shows that the induction of the expression of marker genes in response to phosphate starvation is not abolished in cad2 and ibr1 ibr3 ibr10 mutants. We can only notice a slight down-regulation of SPX1 in cad2 mutant. However, SPX1 encodes a negative regulator of phosphate starvation responses, which could explain why other responsive genes are slightly up-regulated in cad2.

We can conclude that both IBA conversion to IAA and glutathione are required for proper root hair response to phosphate starvation, without affecting general responses to this abiotic stress.

DISCUSSION

Glutathione regulation of IBA homeostasis or conversion to IAA in the root cap

In this work, we first report that root hair and lateral root responses to exogenous IBA, but not IAA, are impaired in case of glutathione deficiency. More precisely, the glutathione- dependent control of IBA response affects early steps of LR development, either founder cell specification or its activation to divide and form a LR primordium. We also show that low glutathione levels alter IBA-dependent auxin signalling in the basal meristem, where

RH elongation occurs and LR founder cells are specified. Previous works have reported the central role of the root cap for IBA-to-IAA conversion in the context of LR founder cell specification, and the role of IAA transport from the root cap to the basal meristem for controlling RH elongation (Bhosale et al., 2018; De Rybel et al., 2012; Giri et al., 2018).

Finally, we show in this manuscript that transporters known to be involved in the apico- basal flux of IAA to the basal meristem, namely AUX1 and PIN2, are not the targets of the glutathione-dependent control. All these data strongly suggest that glutathione modulates

IBA homeostasis or IBA-to-IAA conversion in the root cap, although it does not exclude that the same regulation can also occur in additional tissues.

99 Figure II.1.7. Glutathione and IBA are specifically required for RH growth response to low phosphate. (A) RH length in 8-day-old wild-type (Col-0), cad2 or ibr1 ibr3 ibr10 plants grown on standard ½ MS medium in the absence or the presence of the PGPR Mesorhizobium loti (see material and methods). (B) RH length in 8-day-old wild-type (Col-0), cad2 or ibr1 ibr3 ibr10 plants grown on 1/3 MS medium supplemented with 500 µM (NaCl, control) or 500 µM mM (NaH2PO4, minusP). (C) Expression levels (qRT-PCR) of PHT1, RNS1 and SPX1 low-phosphate responsive genes in wild-type (Col-0), cad2 or ibr1 ibr3 ibr10 8-day-old plants grown on 1/3 MS medium supplemented with 500 µM (NaCl, control) or 500 µM mM (NaH2PO4, minusP). Data represent three independent replicates with 20 plants per replicate and are normalised relatively to the wild-type value in control condition. Histograms represent the mean and error bars represent the standard deviation. Asterisks indicate a significant difference, based on a two-tailed Student t-test (*P<0,01;**P<0,001). 100 All the phenotypes reported in this manuscript concern root responses to exogenous IBA and one can wonder if the glutathione-dependent regulation normally occurs in physiological conditions, modulating pathways relying on endogenous IBA. Because of our scarce knowledge of IBA metabolism and the absence of genetic tools concerning IBA biosynthesis, it is rather difficult to address such question. However, we accumulated several clues that, taken together, strongly support the importance of glutathione for the control of endogenous IBA pathways. First, we observe in cad2 mutant seedlings grown under standard conditions (i.e. without exogenous IBA) a general increase in the expression of all genes involved in IBA-to-IAA conversion (Figure II.1.6B). This might reveal a feedback mechanism that would report an excess of IBA or a depletion in IBA- derived IAA in low glutathione conditions. Secondly, we observe that both ugt and pdr double mutants are hypersensitive to exogenous IBA (Figures II.1.5B and II.1.5C), which is consistent with their already reported overaccumulation of endogenous free IBA (Strader and Bartel, 2009; Ruzicka et al., 2010; Tognetti et al., 2010). Interestingly, it looks like

BSO has more effect in both mutant genotypes than for the wild-type plants, reducing IBA response by 20 to 25% in wild type plants, but by 28 to 33% in pdr and ugt double mutants, respectively (Figures II.1.5B and II.1.5C). This supports a role for glutathione in modulating responses to endogenous IBA. Finally, we show that both cad2 and ibr1 ibr3 ibr10 triple mutant display hyposensitivity in their RH response to phosphate, but not to

Mesorhizobium loti. In addition both are not impaired in the general plant response to phosphate deprivation, but only in the RH elongation response. This set of arguments not only proves the independent importance of endogenous IBA pathway on one hand and of glutathione in the other hand to mediate RH response to phosphate starvation, but also strongly suggests that the glutathione-dependent regulation of endogenous IBA pathway is

101 modulating the RH response.

Auxins fine-tuning of root architecture in response to phosphate availability

Although well-known to play important roles in plant development, the function of IBA as a source of IAA relatively to other IAA sources, such as de novo synthesis or conjugated forms, is still very mysterious and we might wonder “Why plants need more than one type of auxins?” (Frick and Strader, 2018; Simon and Petrášek, 2011). A previous work has reported that IBA regulates both plant development and resistance to water stress, suggesting an important function for IBA in adapting plant development to abiotic stress

(Tognetti et al., 2010). However, to our knowledge, our work is the first to clearly demonstrate that IBA-dependent pathway is necessary for IAA to adapt some developmental pathway, i.e. root hair elongation, to a specific abiotic stress (phosphate starvation).

In their recent work, Bhosale et al. (2018) already revealed the increase in IAA levels in the root cap in response to low external phosphate, necessary for RH elongation. However, because they observe a significant increase in TAA1 (TRYPTOPHAN

AMINOTRANSFERASE OF ARABIDOPSIS 1) gene expression in such conditions, they hypothesize that the increase in IAA levels in the root cap is due to the induction of Indole-

3-Pyruvic Acid (IPyA)-dependent pathway, the main one responsible for de novo IAA synthesis in Arabidopsis (Mashiguchi et al., 2011). Intriguingly, RH elongation response to external low phosphate is almost completely abolished in both ibr1 ibr3 ibr10 and taa1 mutants (Figure II.1.7B and Bhosale et al., 2018: Figures 2b and 2c). Could we imagine that two independent routes of IAA synthesis, i.e. the TAA1-dependent de novo synthesis and the IBA-to-IAA conversion, are both critical and induced in the root cap in response to phosphate deprivation? IBA can itself be derived from IAA and we can imagine that an

102 increase in IBA-to-IAA conversion would require an increase in IBA levels, and therefore more IAA to supply the IBA stock. However, thinking to a root cap increase in IAA to increase IBA in order to increase IAA gets almost no sense. Thus, we could suggest that external low phosphate induces TAA1 expression in the whole plant to increase global available auxin levels on the one hand, and in the other hand activates the local IBA-to-

IAA conversion in the root cap to ensure a local appropriate developmental response.

However, we must remind that we are working in vitro with homogenous media containing a given concentration of phosphate. A very elegant recent work has reported that, when subjected to an heterogenous environment, using a Dual-flow-root chip, a single root increases RH length in the medium having the highest phosphate concentration (Stanley et al., 2018). This observation contrasts with the induction of RH growth by low phosphate in homogenous in vitro media. RH adaptation to phosphate probably results from the crosstalk between a systemic information based on overall available phosphorus in the plant, and a local information based on phosphate availability in the environment, that we cannot differentiate in our in vitro homogenous conditions. Such dual control has been extensively documented for root responses to another important nutrient, nitrogen (Ruffel and Gojon, 2017). We might therefore envisage that both sources of IAA are differentially regulated by systemic and local signals. This might ensure a very acute dosage of free IAA to optimize root adaptation to the plant metabolism and environment. In such context, glutathione could be a central regulator of local IBA-to-IAA adaptation to external phosphate in the root cap.

Unravelling the glutathione-dependent regulation of IBA pathway

Although we explored most of known components of IBA pathways, we were not able to identify how glutathione regulates IBA homeostasis or conversion to IAA. This can first be

103 explained by the difficulties to work with IBA which is present in low amounts compared to IAA in plants, and is therefore difficult to detect and quantify (Frick and Strader, 2018).

Moreover, many components of the IBA pathways remain to be identified. For sure, the development of new tools, such as IBA probes or mutants in IBA biosynthesis would be very helpful to depict the important roles played by this auxin in plants. The other problem we faced is the complex multifaceted roles of glutathione in cells, that prevented us to try to decipher the mechanism from the glutathione starting point. The only clue we get is that

IBA hyposensitivity is specific to plants with reduced amounts of glutathione but does not occur in plants having more general redox problems (Figure II.1.3). This suggests that IBA regulation by glutathione may act through glutathionylation or via the activity of thiols reductases that specifically depend on glutathione, such as glutaredoxins. One hypothesis would be that IBA, or a precursor, could itself be glutathionylated, which would be responsible for its storage or transport. In Arabidopsis, the Tau class of Glutathione-S-

Transferases (GSTU) have been shown to glutathionylate some fatty acids, ranging from short to long acyl chains (Dixon and Edwards, 2009). Similarly, GSTU19 and GSTF8 have been proposed to glutathionylate OPDA, hence allowing its translocation from the chloroplast to the peroxisome where it is converted to Jasmonic Acid (Davoine, 2006;

Dixon and Edwards, 2009).

Because IBA-to-IAA conversion occurs in the peroxisome, we might also envisage that the glutathione-dependent regulation is not specific to IBA pathways but affects the peroxisomal machinery. One of the main peroxisomal functions is the -oxidation of fatty acids, which is essential to supply energy during seed germination. We never observed any problem for seed germination in cad2 or pad2 mutants, in contrast to mutants in peroxisomal functions which generally strongly affect seed germination. However, we

104 cannot exclude that the glutathione-dependent regulation only occurs in the root cap and therefore mainly affects IBA-ot-IAA conversion although regulating the peroxisome machinery. Among the PEROXINS proteins, many participate in the peroxisomal matrix protein import machinery (Cross et al., 2016). PEX5 is a central cargo protein that recognizes proteins targeted with the specific PTS1 (Peroxisomal Targeting Signal 1) signal peptide and import them into the peroxisome. PEX7 recognizes proteins harbouring another signal peptide (PTS2), and then binds to PEX5 for import into the peroxisome.

Interestingly, previous works on human and Pichia pastoris PEX5 revealed that its activity and oligomerization depend on a redox switch affecting a conserved N-terminal Cystein

(Ma et al., 2013; Apanasets et al., 2014). However, such regulation is not specific to glutathione but rather depends on the redox status and the content in ROS of peroxisomes.

Finally, the last hypothesis would be that glutathione regulates IBA homeostasis or conversion to IAA via a yet unidentified or not assayed component. New attempts in quantifying IBA-to-IAA conversion would be helpful in order to confirm this hypothesis.

Concerning the IBA-to-IAA -oxidation pathway in the peroxisome, PED1 has been reported to harbour a redox-sensitive switch affecting a conserved cysteine and that participates in activating the enzyme when reduced (Pye et al., 2010). In contrast to most of the other enzymes involved in the conversion pathway, PED1 is targeted to the peroxisome through a PTS2 signal peptide. It is interesting to notice that pex5-1, in contrast to pex5-10, is not altered in PTS1-dependent import but only in PTS2-dependent import of proteins into the peroxisome (Woodward and Bartel, 2004; Khan and Zolman,

2010). pex5-1, although not displaying any phenotype during germination, is highly affected in IBA responses. This might reveal that IBA to IAA conversion is highly dependent on a PTS2-targeted protein, and PED1 could therefore be a good candidate to

105 assay. However, PED1 is not specific to IBA-to-IAA pathway since it acts in almost all - oxidation pathways occurring in the peroxisome. We could imagine a local glutathione- dependent regulation of PED1 activity specifically in the root cap.

MATERIALS AND METHODS

Plant material and growth conditions

Wild-type plants and mutants used in this study were in the Arabidopsis thaliana ecotype

Col-0. Mutants of the biosynthesis of glutathione: cad2-1, pad2-1, and zir1 are described on Howden et al., 1995, Parisy et al., 2006, and Shanmugam et al., 2012, respectively, while gr1 and ntrantrb mutants are described on Mhamdi et al., 2010, and Bashandy et al.,

2010, respectively. Mutant cat2-1 impaired in catalase is described in Bueso et al., 2007.

The ibr3, ibr10, ech2, ibr1, and triple mutant ibr1ibr3ibr10 were donated by the group of

Lucia Strader (Strader et al., 2011). Mutant pin2 (eir1-1) is described in Roman et al.,

1995, Mutants aux1-21, pdr8 (Salk_000578), pdr9 (Salk_050885), ugt74e2

(Salk_091130), and ugt74d1 (Salk_011286) are Salk lines. Wild-type and mutant seeds were surface-sterilized by constant agitation with 0.05% SDS made on ethanol 70% (v/v) during 20 minutes. Thereafter, seeds were washed three times with ethanol 95% (v/v) and dried at room-temperature under sterile conditions. Seeds were placed on square Petri plates containing 50 ml of Murashige and Skoog medium diluted two fold (MS ½), and solidified with plant agar with a final concentration of 0.8% (w/v), without sucrose unless indicated. For the phosphorous deprivation experiments, we used MS 3/10 in which Pi- deficient plates contained 500µM NaCl, and control plates 500µM NaH2PO4 as indicated in

Arnaud et al., 2014. For the examination of the root hair responses in the experiments with

M. loti, seeds were placed vertically on standard MS ½ medium and after 5 days, seedlings were transferred to plates containing 0.1 OD of inoculum and root hair lengths were

106 quantified after 4 days of incubation. Plates with the inoculum were prepared according to

Poitout et al., 2017. All experiments were carried out on in vitro conditions with 160

µE.m-2.s-1 light intensity (16/8) at 20 °C.

Phenotypic analyses of the root system

Lateral root density was quantified 10 days after sowing by counting the number of visible lateral roots emerged divided by the length of the region in which the LR was counted.

Lateral root primordia density was calculated 6 days after sowing, in the same manner using a light microscope (Axioscop2, Zeiss). The study of the lateral root primordia development was performed by synchronization of the LRP initiation, that consisted of changing the orientation of the plates 90° and observing the stage of LRP after 48 hours.

Root hair lengths were quantified 8 days after sowing, excepted of M.loti experiments (see plant material and growth conditions). Primary root elongation rate was quantified by calculating the difference in the length of the root from day 8 and day 10. The analyses were made with ImageJ (https://imagej.nih.gov/ij/).

Glutathione measurements

Analysis of glutathione was performed using the enzymatic recycling methods, which is a modification of the classic Tietze recycling assay for total glutathione (Tietze, 1969;

Rahman et al., 2007). The method consists in the reduction of GSSG by glutathione reductase enzyme and NADPH to GSH. Total GSH is determined by the oxidation of GSH by 5,5′-dithio-bis(2-nitrobenzoic acid) (DTNB), that produce a yellow compound 5′-thio-2- nitrobenzoic acid (TNB) which is detectable at 412 nm (Rahman et al., 2007). Extraction of glutathione was performed on 8 days seedlings. In each experiment, we used three independent biological repetitions in which each biological repetition consisted of 20 plants.

107 Confocal analyses

Auxin response analyses using DR5v2 and the study of the redox state of glutathione with roGFP2 were performed by using a confocal laser scanning microscope Zeiss LSM 510

Meta (Carl Zeiss). Images were acquired in 16-bits with two average readings and using

10x lens. Confocal settings for DR5v2 were based on Liao et al., 2015. Settings and analyses for roGFP2 were based on the method published by Meyer et al., 2007, with small modifications. Excitation of roGFP2 was performed at 488 and 405 nm with a bandpass

(BP) of 490-555 nm. Correction of the 405 nm for its background noise was performed by removing the fluorescence of the 405 nm collecting with a BP 420-480 nm.

For the analysis with DR5v2, sterilized seeds were placed vertically in square plates containing standard MS ½ medium with either BSO 0.5mM (BSO) or without BSO

(Control). After 7 days of growth, seedlings were transferred to plates containing the following treatments: BSO = 0.5mM, IBA= 10µM, IAA=50nM, and combination of each auxin with BSO. After incubation with those treatments for 24 hours, pictures were taken in the confocal microscope.

Concerning the analyses with roGFP2, the reporter line was tested by performed a timeline experiment, that consisted in measuring the roGFP2 fluorescence during 10 minutes and then adding 1 mM of H2O2 for 30 minutes. Thereafter, reduction of the roGFP2 line with

10 mM DTT was performed in order to verify the functionality of the reporter line. For testing the effect of auxin IBA on the glutathione redox status, seeds were placed vertically for 8 days in mediums MS ½, containing IBA 10µM or without IBA (Control). Pictures were analyzed using ImageJ following the protocol published by Meyer et al., 2007 and

Morgan et al., 2011.

108 GUS staining

DR5-GUS activity was detected in root tissue from 8-days-old seedlings. The method is based on the work published by Jefferson et al., 1987, and Malamy and Benfey, 1997, with small modifications. Plants were fixed on acetone 80% (v/v) at 20 °C for 20 minutes, were washed with a buffer solution containing 25mM Na2HPO4, 25mM NaH2PO4, 2% Triton

(v/v), 1 mM K3Fe(CN)6, and 1 mM K4Fe(CN)6. Thereafter, seedlings were transferred to a new plate containing the buffer solution with 2mM of X-Gluc (5-bromo-4-chloro-3- indolyl-β-D-glucuronidase) and vacuum infiltrated for 1 minute and then incubate at 37 °C for 1 hour. The reaction was stopped by bathing seedlings in ethanol 70% (v/v). Pictures were taken with a light microscope (Axioscop2, Zeiss) on the next day. The experiment protocol was the same as in DR5v2 experiments.

Quantitative real-time PCR (qPCR)

For the quantification of the level of expression of gene of interest, total RNA was extracted using TriZol reagent (GE Healthcare, Littler Chalfont, Buckinghamshire, UK), and the RNA was purified with RNeasy Plant Mini Kit (Promega, USA). cDNA was synthesized by the GoScript™ Reverse Transcription System (Promega, USA).

Quantitative real-time PCR was done using Takyon™ No Rox SYBR® MasterMix dTTP

Blue (Eurogentec, Belgium).

Primers for phosphorous deprivation:

PHT1.4_LP 5’-CCTCGGTCGTATTTATTACCACG PHT1.4_RP 5’-

CCATCACAGCTTTTGGCTCATG, RNS1_LP 5’-TGATGCCTCTAAACCATTCGAT,

RNS1_RP 5’-TACCATGCTTCTCCCATTCG, SPX1_LP 5’-

CGGGTTTTGAAGGAGATCAG, SPX1_RP 5’-GCGGCAATGAAAACACACTA. For the IBA pathway: IBR1_LP 5’-ATCACCGGTAGCTCTGAAGTGAGG IBR1_RP 5’-

109 ATGTCTCCCGTTGTTCCCAACC IBR3_LP 5’-TTCTCGCAATGGCCAAGGTTGC

IBR3_RP 5’-GCTGCTCCATGAACTTGTATTGCC IBR10_LP 5’-

GCTTTCGATGCCGGATTATT IBR10_RP 5’-TTCCCACTCAACAACAGCTC

AIM1_LP 5’-GAAAGGCCCTCTCTTTGTTC AIM_RP 5’-

AATCACTGCCTCTATGACCA PED1_LP 5’-AGGAGCGGTTCTCCTAATGA

PED1_RP 5’-CCTGAATACACCAAGAACGG ECH2_LP 5’-

ATGGGCAGGATTTGATCCAGGTC ECH2_RP 5’-

ACTGCTGCCCATGCAACAAGAG.

Statistics analysis

All the experiments presented in this study were performed at least three independent biological repetitions. Each biological repetition from in vitro experiments consisted of two plates containing around 12 seeds for the mutant and the control in the same plate. Statistic analyses were performed with t-student test. Significant differences were considered when the p-value is ≤ to 0.01 (represented by one *) or ≤ 0.001 (represented with two **).

110 SUPPLEMENTARY INFORMATION

Supplementary Figure II.1.1. Glutathione does not affect primary root response to IBA. (A) Pictures of representative 8-day-old wild-type, cad2, pad2, and zir1 plants grown on standard ½ MS medium supplemented or not with 10 µM IBA or 50 nM IAA. Wild-type plants grown on 0.5 mM BSO are also presented. (B) Quantification of primary root elongation rate of 8-day-old wild-type (Col-0), cad2 or pad2 mutant plants grown on standard ½ MS medium (Control), or in the presence of 50 nM IAA (IAA) or 10 µM IBA (IBA). Wild-type plants in the presence of 0.5 mM BSO (BSO) were also assayed (n≥12). Histograms represent the mean and error bars represent the standard deviation. Asterisks indicate a significant difference, based on a two-tailed Student t-test (*P<0,01;**P<0,001).

111 Supplementary Figure II.1.2. Glutathione regulates LR responses to IBA. (A) Emerged LR density of 10-day-old wild-type plants grown on standard ½ MS medium (control), or in the presence of different concentrations of BSO, in the absence or the presence of IBA 10 µM (n≥15). (B) Emerged LR density of 10-day-old wild-type and cad2 plants grown on standard ½ MS medium (control), or in the presence of IBA 10 µM and 5 mM GSH (n≥15). Histograms represent the mean and error bars represent the standard deviation. Asterisks indicate a significant difference, based on a two-tailed Student t-test (*P<0,01).

112 Supplementary Figure II.1.3. IBA does not alter redox status of GSH in root tips.

(A) Cinetics of roGFP reporter response to 1 mM H2O2 then to an additional 10 min treatment with 10 mM DTT. The ratio between the oxidised roGFP signal (excitation at 405 nm) and the reduced roGFP signal (excitation at 488 nm) is used to monitor the glutathione redox status. (B) Quantification of roGFP ratio in root tips from wild-type plants grown on standard ½ MS medium or in the presence of 10 µM IBA for 8 days. Data represent the mean and error bars represent the standard deviation (n≥9).

113 Supplementary Figure II.1.4. RH response to IBA in mutants affected in IBA homeostasis.

BSO = 0.50mM, IBA = 10µM. Asterisks indicate a significant difference, based on a two-tailed Student t-test (*P<0,01;**P<0,001) n ≥ 50.

114 Supplementary Figure II.1.5. Glutathione content in response to low phosphate. Glutathione content in 8-day-old wild-type (Col-0), cad2, or ibr1 ibr3 ibr10 seedlings grown on 1/3 MS medium supplemented with 500 µM (NaCl, control) or 500

µM mM (NaH2PO4, minusP). Data represent the means of 3 biological repetitions. Reduced glutathione is represented in white (GSH) and oxidised glutathione in black (GSSG). Histograms represent the mean and error bars represent the standard deviation.

REFERENCES

Apanasets, O., Grou, C.P., Van Veldhoven, P.P., Brees, C., Wang, B., Nordgren, M., Dodt, G., Azevedo, J.E., and Fransen, M. (2014). PEX5, the Shuttling Import 115 Receptor for Peroxisomal Matrix Proteins, Is a Redox-Sensitive Protein. Traffic 15: 94–103. Ball, L., Accotto, G.-P., Bechtold, U., Creissen, G., Funck, D., Jimenez, A., Kular, B., Leyland, N., Mejia-carranza, J., Reynolds, H., Karpinski, S., and Mullineaux, P.M. (2004). Evidence for a Direct Link between Glutathione Biosynthesis and Stress Defense Gene Expression in Arabidopsis. Plant Cell 16: 2448–2462. Bashandy, T., Guilleminot, J., Vernoux, T., Caparros-Ruiz, D., Ljung, K., Meyer, Y., and Reichheld, J.-P. (2010). Interplay between the NADP-linked thioredoxin and glutathione systems in Arabidopsis auxin signaling. Plant Cell 22: 376–91. Bates, T.R. and Lynch, J.P. (1996). Stimulation of root hair elongation in Arabidopsis thaiiana by low phosphorus availability. Plant, Cell Environ. 19: 529–538. Bhosale, R. et al. (2018). A mechanistic framework for auxin dependent Arabidopsis root hair elongation to low external phosphate. Nat. Commun. 9: 1–9. Carol, R.J. and Dolan, L. (2006). The role of reactive oxygen species in cell growth : lessons from root hairs. J. Exp. Bot. 57: 1829–1834. Cross, L.L., Ebeed, H.T., and Baker, A. (2016). Peroxisome biogenesis, protein targeting mechanisms and PEX gene functions in plants. Biochim. Biophys. Acta - Mol. Cell Res. 1863: 850–862. Davoine, C. (2006). Adducts of Oxylipin Electrophiles to Glutathione Reflect a 13 Specificity of the Downstream Lipoxygenase Pathway in the Tobacco Hypersensitive Response. Plant Physiol. 140: 1484–1493. Dixon, D.P. and Edwards, R. (2009). Selective binding of glutathione conjugates of fatty acid derivatives by plant glutathione transferases. J. Biol. Chem. 284: 21249–21256. Du, Y. and Scheres, B. (2018). Lateral root formation and the multiple roles of auxin. J. Exp. Bot. 69: 155–167. Frick, E.M. and Strader, L.C. (2018). Roles for IBA-derived auxin in plant development. J. Exp. Bot. 69: 169–177. Giri, J. et al. (2018). Rice auxin influx carrier OsAUX1 facilitates root hair elongation in response to low external phosphate. Nat. Commun. 9: 1408. Grierson, C.S., Nielsen, E., Ketelaar, T., and Schiefelbein, J. (2014). Root Hairs. In The Arabidopsis Book, p. e0172. Hodge, A. (2004). The plastic plant: Root responses to heterogeneous supplies of nutrients. New Phytol. 162: 9–24. Howden, R., Andersen, C.R., Coldsbrough, B., and Cobbett, C.S. (1995). A Cadmium- Sensitive, Glutathione-Deficient Mutant of Arabidopsis thaliana. Plant Physiol. 107: 1067–1073. Jobe, T.O., Sung, D.-Y., Akmakjian, G., Pham, A., Komives, E.A., Mendoza-Cózatl, D.G., and Schroeder, J.I. (2012). Feedback inhibition by thiols outranks glutathione depletion : a luciferase-based screen reveals glutathione-deficient c -ECS and glutathione synthetase mutants impaired in cadmium- induced sulfate assimilation. Plant J. 70: 783–795. Kanno, S., Pe, B., Berthome, R., Bayle, V., Delannoy, E., Marin, E., and Nakanishi, T.M. (2016). A novel role for the root cap in phosphate uptake and homeostasis. Elife: 1–16. Khan, B.R. and Zolman, B.K. (2010). pex5 Mutants That Differentially Disrupt PTS1 and PTS2 Peroxisomal Matrix Protein Import in Arabidopsis. Plant Physiol. 154: 1602–1615. Kim, D.Y., Bovet, L., Maeshima, M., Martinoia, E., and Lee, Y. (2007). The ABC 116 transporter AtPDR8 is a cadmium extrusion pump conferring heavy metal resistance. Plant J. 50: 207–218. Koprivova, A., Mugford, S.T., and Kopriva, S. (2010). Arabidopsis root growth dependence on glutathione is linked to auxin transport. Plant Cell Rep. 29: 1157–67. Korasick, D.A., Enders, T.A., and Strader, L.C. (2013). Auxin biosynthesis and storage forms. J. Exp. Bot. 64: 2541–2555. Korver, R.A., Koevoets, I.T., and Testerink, C. (2018). Out of Shape During Stress : A Key Role for Auxin. Trends Plant Sci. 23: 783–793. Leinweber, P. et al. (2018). Handling the phosphorus paradox in agriculture and natural ecosystems: Scarcity, necessity, and burden of P. Ambio 47: 3–19. Liao, C., Smet, W., Brunoud, G., Yoshida, S., Vernoux, T., and Weijers, D. (2015). Reporters for sensitive and quantitative measurement of auxin response. Nat. Methods 12: 207–210. López-Bucio, J., Cruz-Ramírez, A., and Herrera-Estrella, L. (2003). The role of nutrient availability in regulating root architecture. Curr. Opin. Plant Biol. 6: 280– 287. López-Bucio, J., Hernandez-Abreu, E., Sanchez-Calderon, L., Nieto-jacobo, F., Simpson, J., and Herrera-estrella, L. (2002). Phosphate Availability Alters Architecture and Causes Changes in Hormone Sensitivity in the Arabidopsis. Plant Physiol. 129: 244–256. Lu, X., Dittgen, J., Piślewska-Bednarek, M., Molina, A., Schneider, B., Svatoš, A., Doubský, J., Schneeberger, K., Weigel, D., Bednarek, P., and Schulze-Lefert, P. (2015). Mutant Allele-Specific Uncoupling of PENETRATION3 Functions Reveals Engagement of the ATP-Binding Cassette Transporter in Distinct Tryptophan Metabolic Pathways. Plant Physiol. 168: 814–827. Ludwig-Müller, J., Hilgenberg, W., and Epstein, E. (1995). THE IN VITRO BIOSYNTHESIS OF INDOLE-3-BUTYRIC ACID IN MAIZE. Phytochemistry 40: 61–68. Ma, C., Hagstrom, D., Polley, S.G., and Subramani, S. (2013). Redox-regulated cargo binding and release by the peroxisomal targeting signal receptor, Pex5. J. Biol. Chem. 288: 27220–27231. Mangano, S., Denita-juarez, S.P., Marzol, E., Borassi, C., and Estevez, J.M. (2018). High Auxin and High Phosphate Impact on RSL2 Expression and ROS-Homeostasis Linked to Root Hair Growth in Arabidopsis thaliana. Front. Plant Sci. 9: 1164. Mangano, S., Paola, S., Juárez, D., and Estevez, J.M. (2016). Update on Polar Growth ROS Regulation of Polar Growth in Plant Cells. Plant Physiol. 171: 1593–1605. Marquez-Garcia, B., Njo, M., Beeckman, T., Goormachtig, S., and Foyer, C.H. (2014). A new role for glutathione in the regulation of root architecture linked to strigolactones. Plant. Cell Environ. 37: 488–98. Mashiguchi, K. et al. (2011). The main auxin biosynthesis pathway in Arabidopsis. Proc. Natl. Acad. Sci. 108: 18512–18517. Moreno-Risueno, M.A., Van Norman, J.M., Moreno, A., Zhang, J., Ahnert, S.E., and Benfey, P.N. (2010). Oscillating Gene Expression Determines Competence for Periodic Arabidopsis Root Branching. Science (80-. ). 329: 1306–1311. Nacry, P., Canivenc, G., Muller, B., Azmi, A., Onckelen, H. Van, Rossignol, M., and Doumas, P. (2005). A Role for Auxin Redistribution in the Responses of the Root System Architecture to Phosphate Starvation in Arabidopsis. Plant Physiol. 138: 2061–2074. 117 Noctor, G., Mhamdi, A., Chaouch, S., Han, Y.I., Neukermans, J., Marquez-garcia, B., Queval, G., and Foyer, C.H. (2012). Glutathione in plants : an integrated overview. Plant, Cell Environ. 35: 454–484. Noctor, G., Reichheld, J., and Foyer, C.H. (2018). ROS-related redox regulation and signaling in plants. Semin. Cell Dev. Biol. 80: 3–12. Orman-ligeza, B., Parizot, B., Rycke, R. De, Fernandez, A., Himschoot, E., Breusegem, F. Van, Bennett, M.J., Pe, C., Beeckman, T., and Draye, X. (2016). RBOH-mediated ROS production facilitates lateral root emergence in Arabidopsis. Development 143: 3328–3339. Parisy, V., Poinssot, B., Owsianowski, L., Buchala, A., Glazebrook, J., and Mauch, F. (2007). Identification of PAD2 as a gamma-glutamylcysteine synthetase highlights the importance of glutathione in disease resistance of Arabidopsis. Plant J. 49: 159– 72. Parry, G. (2018). Low Phosphate Puts Auxin in the Root Hairs. Trends Plant Sci. 23: 845– 847. Péret, B. et al. (2012). Auxin regulates aquaporin function to facilitate lateral root emergence. Nat. Cell Biol. 14: 991–998. Poitout, A., Martinière, A., Kucharczyk, B., Queruel, N., Mashkoor, S., Gamet, L., Varoquaux, F., Paris, N., Sentenac, H., Touraine, B., and Desbrosses, G. (2017). Local signalling pathways regulate the Arabidopsis root developmental response to Mesorhizobium loti inoculation. J. Exp. Bot. 68: 1199–1211. Pye, V.E., Christensen, C.E., Dyer, J.H., Arent, S., and Henriksen, A. (2010). Peroxisomal plant 3-ketoacyl-CoA thiolase structure and activity are regulated by a sensitive redox switch. J. Biol. Chem. 285: 24078–24088. Ruffel, S. and Gojon, A. (2017). Systemic nutrient signalling: On the road for nitrate. Nat. Plants 3: 17040. Ruzicka, K. et al. (2010). Arabidopsis PIS1 encodes the ABCG37 transporter of auxinic compounds including the auxin precursor indole-3-butyric acid. Proc. Natl. Acad. Sci. U. S. A. 107: 10749–53. De Rybel, B. et al. (2012). a role for the root cap in root branching revealed by the non- auxin probe naxillin. Nat. Chem. Biol. 8: 798–805. Saini, S., Sharma, I., Kaur, N., and Pati, P.K. (2013). Auxin: a master regulator in plant root development. Plant Cell Rep. 32: 741–57. Shanmugam, V., Tsednee, M., and Yeh, K.-C. (2012). ZINC TOLERANCE INDUCED BY IRON 1 reveals the importance of glutathione in the cross-homeostasis between zinc and iron in Arabidopsis thaliana. Plant J. 69: 1006–17. Sherp, A.M., Westfall, C.S., Alvarez, S., and Jez, J.M. (2018). Arabidopsis thaliana GH3.15 acyl acid amido synthetase has a highly specific substrate preference for the auxin precursor indole-3-butyric acid. J. Biol. Chem. 293: 4277–4288. Simon, S. and Petrášek, J. (2011). Why plants need more than one type of auxin. Plant Sci. 180: 454–60. Singh, R., Singh, S., Parihar, P., Mishra, R.K., Tripathi, D.K., Singh, V.P., Chauhan, D.K., and Prasad, S.M. (2016). Reactive Oxygen Species ( ROS ): Beneficial Companions of Plants ’ Developmental Processes. Front. Plant Sci. 7: 1299. Stanley, C.E., Shrivastava, J., Brugman, R., Heinzelmann, E., van Swaay, D., and Grossmann, G. (2018). Dual-flow-RootChip reveals local adaptations of roots towards environmental asymmetry at the physiological and genetic levels. New Phytol. 217: 1357–1369. 118 Strader, L.C. and Bartel, B. (2009). The Arabidopsis PLEIOTROPIC DRUG RESISTANCE8/ABCG36 ATP binding cassette transporter modulates sensitivity to the auxin precursor indole-3-butyric acid. Plant Cell 21: 1992–2007. Strader, L.C. and Bartel, B. (2011). Transport and metabolism of the endogenous auxin precursor indole-3-butyric Acid. Mol. Plant 4: 477–86. Strader, L.C., Culler, A.H., Cohen, J.D., and Bartel, B. (2010). Conversion of endogenous indole-3-butyric acid to indole-3-acetic acid drives cell expansion in Arabidopsis seedlings. Plant Physiol. 153: 1577–86. Tognetti, V.B. et al. (2010). Perturbation of Indole-3-Butyric Acid Homeostasis by the UDP-Glucosyltransferase UGT74E2 Modulates Arabidopsis Architecture and Water Stress Tolerance. Plant Cell 22: 2660–2679. Tognetti, V.B. (2017). Redox regulation at the site of primary growth : auxin , cytokinin and ROS crosstalk.: 2586–2605. Tsukagoshi, H. (2016). Control of root growth and development by reactive oxygen species. Curr. Opin. Plant Biol. 29: 57–63. Tyburski, J., Dunajska, K., and Tretyn, A. (2009). Reactive oxygen species localization in roots of Arabidopsis thaliana seedlings grown under phosphate deficiency. Plant Growth Regul. 59: 27–36. Ulmasov, T., Murfett, J., Hagen, G., and Guilfoyle, T.J. (1997). Aux/lAA Proteins Repress Expression of Reporter Genes Containing Natural and Highly Active Synthetic Auxin Response Elements. Plant Cell 9: 1963–1971. Vernoux, T., Wilson, R.C., Seeley, K.A., Reichheld, J., Muroy, S., Brown, S., Maughan, S.C., Cobbett, C.S., Montagu, M. Van, Inzé, D., May, M.J., and Sung, Z.R. (2000). The ROOT MERISTEMLESS1 / CADMIUM SENSITIVE2 Gene Defines a Glutathione-Dependent Pathway Involved in Initiation and Maintenance of Cell Division during Postembryonic Root Development. Plant Cell 12: 97–109. Wang, T., Li, C., Wu, Z., Jia, Y., Wang, H., Sun, S., Mao, C., and Wang, X. (2017). Abscisic Acid Regulates Auxin Homeostasis in Rice Root Tips to Promote Root Hair Elongation. Front. Plant Sci. 8: 1–18. Woodward, A.W. and Bartel, B. (2004). The Arabidopsis Peroxisomal Targeting Signal Type 2 Receptor PEX7 Is Necessary for Peroxisome Function and Dependent on PEX5 . Mol. Biol. Cell 16: 573–583. Xuan, W. et al. (2016). Cyclic programmed cell death stimulates hormone signaling and root development in Arabidopsis. Science (80-. ). 351: 384–387. Xuan, W., Audenaert, D., Parizot, B., M??ller, B.K., Njo, M.F., De Rybel, B., De Rop, G., Van Isterdael, G., M??h??nen, A.P., Vanneste, S., and Beeckman, T. (2015). Root cap-derived auxin pre-patterns the longitudinal axis of the arabidopsis root. Curr. Biol. 25: 1381–1388. Zhang, G.Z., Jin, S.H., Jiang, X.Y., Dong, R.R., Li, P., Li, Y.J., and Hou, B.K. (2016). Ectopic expression of UGT75D1, a glycosyltransferase preferring indole-3-butyric acid, modulates cotyledon development and stress tolerance in seed germination of Arabidopsis thaliana. Plant Mol. Biol. 90: 77–93. Zwiewka, M., Bielach, A., Tamizhselvan, P., Madhavan, S., Dobrev, P., Vankova, R., Ryad, E.E., and Tan, S. (2019). Root Adaptation to H2O2 -Induced Oxidative Stress by ARF-GEF BEN1- and Cytoskeleton-Mediated PIN2 Trafficking. Plant Cell Physiol. 60: 255–273.

119 PERSPECTIVES

We demonstrated the important roles that glutathione plays in the IBA pathway in the induction of lateral roots and in regulating root hair lengths. However, how glutathione controls the IBA pathway is a question that still elusive in this study. As it can be seen in chapter I, we investigated several known components of the IBA transport, conjugation, and conversion, in which glutathione might act to control the IBA functions, but not possible targets were found. Nevertheless, additional experiments need to be done in order to rule out such components. For example, regarding the IBA efflux, PDR8 and PDR9 belong to a big family of proteins with 150 members in Arabidopsis (Hwang et al., 2016). For this reason, a possible redundancy in their activity might occur in the absence of both proteins. The same remark is made for the IBA conjugation that dependent on UGT74E2 and UGT74D1. Thus, testing the auxin responses in combination with inhibitors of the whole family of ABCG and UGT in the presence of BSO, will confirm that both transport and conjugation are not involved in the control of IBA by glutathione. Genetic studies in known members of the IBA conversion to IAA were tested as well. However, those experiments that are based in the IBA responses in combination with BSO are hard to interprete due to the already IBA resistance of the IBA peroxisomal mutants. Another way to challenge the hypothesis that glutathione is controlling IBA through the peroxisomal conversion can be by testing that process with other tools, such as using the auxin reporter R2D2. R2D2 is a quantitative fluorescent reporter of the auxin signaling based in the DII-VENUS domain (see Introduction section 1.2.6) (Liao et al., 2015). This is examplified in figure II.1.8A, where incubation of plants with active auxin IAA induces the loss of the fluorescent in the DII-VENUS without affecting the internal control (mutated version mDII-ndTomato). A ratiometric quantification of DII-VENUS and mDII-ndTomato fluorescence allowed us to monitor the response to auxin incubation (figure II.1.8C). After IAA incubation, the mDII-ndTomato/DII-VENUS rapidly increases, which monitor the response to IAA uptake. The increase is much slower when plants are incubated with IBA, likely due to the time that takes IBA to be converted to IAA by the peroxisome. Then, we studied if the response to IAA and IBA are affected by BSO treatment. While BSO treatment prior to IAA incubation leads to a small increase of the ratio, a similar treatement has a slightly opposite effect on IBA incubation. This might indicate a negative effect of glutathione deficiency on IBA uptake or/and conversion. 120 Nevertheless, these data have to be repetead in order to confirm the results, and compared with other methods to measure the conversion, such as with mass spectrometric analyses. Besides the peroxisomal enzymes involved in the conversion, there are other components that could be involved. Peroxins are a group of proteins that are critical for the peroxisome activities, and the uptake of proteins with the peroxisomal label PTS1 and PTS2. Among them, PEX5 is one of the most important peroxins involved in the uptake of components with both tags PTS1 and PTS2. Intriguingly, it has been demonstrated that PEX5 activity in humans is depending on a cysteine in the position 11 that is subject to oxidation, forming a disulfide bond that prevents its monoubiquitination of the PEX5 protein, perturbing as a consequence the recycling of PEX5 to the cytosol (Apanasets et al., 2014). Moreover, it was also demonstrated that PEX5 in Pichia pastoris is under redox control on Cysteine 10 forming a disulfide bond. However, the mechanism seems to be different with respect to human. In yeast, oxidation of PEX5 in the cytosol allows its interaction with PTS1 proteins, and then, reduction of PEX5 in the peroxisome resultes in the release of PTS1 protein into the peroxisomal matrix (Ma et al., 2013). Interestingly, comparison of the amino acid sequences of PEX5 from human and yeast with plant PEX5 homologs (Arabidopsis and rice), showed conservation of this N-terminal cysteine (Cys13 in plants) (Figure II.1.9). The evidence that PEX5 is under redox control in yeast and in animals, and the identification of a close conserved cysteine in the N-terminal region of the amino acid sequence, place PEX5 as a good candidate for the control of IBA by glutathione (Ma et al., 2013; Apanasets et al., 2014). Indeed, when testing the response to IBA in the pex5-1 mutant allele, we observed an impaired inhibition of lateral root development by BSO, which was not observed in the pex7 mutant (Figure II.1.10). This might indicate a role of PEX5 in the glutathione-dependent IBA control of lateral root development. Nevertheless, this data has to be taken with caution due to the relative resistance of the mutant pex5-1 to IBA. In this regard, quantification of the uptake of PTS2 proteins with PTS2-GFP marker will give us more evidence of the possible roles that PEX5 might plays in the control of IBA by glutathione. It is important to mention that the IBA response seems not to be affected in mutants accumulating H2O2 like in the cat2 mutant, the major isoform of the three catalases in Arabidopsis (Bueso et al., 2007; Mhamdi et al., 2010b), although glutathione homeostasis is affected in this mutant. Other thiol reduction mutants like the ntra ntrb mutant seem also

121 unaffected. Therefore, it will be interesting further explore the specificity of the IBA in other mutant backgrounds or in plants treated with different oxidative agents. Besides the roles of glutathione in the IBA pathway, we also demonstrated that both glutathione and IBA are involved in the response to phosphorous deprivation. However, several aspects remain to be elucidated, for example, there is a lack of evidence of the mechanism in which both are participating in the response, it could even possible that glutathione and IBA are participating in the response independent of each other. For this reason, finding the target by which glutathione control the IBA pathway would be necessary to answer this question.

122 II.2 CHAPTER 2: CONTRIBUTION OF GLUTAREDOXINS IN LATERAL ROOT DEVELOPMENT

In chapter one, I demonstrated by genetic and molecular experiments that glutathione is critical for the auxin response to IBA, which illustrates the important roles of glutathione, not only as a ROS scavenging compound but also as a major regulator of developmental processes, interacting with hormonal pathways and modulating responses to nutrient deprivation. The capacity of glutathione to perform different roles in plants is in part due to its nature as an electron donor (Foyer and Noctor, 2011; Hasanuzzaman et al., 2017). Glutaredoxins are thiol enzymes that play a variety of functions in plants, and as their name suggests, they use glutathione as a substrate to reduce specific thiol proteins. This post-translational modification allows glutaredoxins to control the conformation, the subcellular localization or the activity of their targets depending on the cellular environment (Meyer et al., 2012; Belin et al., 2015). While several studies have described the roles of glutaredoxins in shoot development, little is known about the functions that GRX perform in the root system (Li et al., 2009a, 2011; Knuesting et al., 2015; Patterson et al 2015; Mhamdi and Van Breusegem, 2018). For this reason, this chapter aims to describe our findings related to the glutaredoxins functions during root development. This research was submitted to the journal Frontiers in Plant Science. The enumerating of the figures are adapted to the thesis manuscript format.

123 Contribution of Glutaredoxins in Lateral Root Development

José A. Trujillo-Hernandez1,2, Jana Cela-Udaondo1,2,3, Laetitia Bariat1,2, Véronique Sureda1,2, Jean-Philippe Reichheld1,2,*, Christophe Belin1,2,*

1Laboratoire Génome et Développement des Plantes, Université Perpignan Via Domitia, F- 66860 Perpignan, France. 2Laboratoire Génome et Développement des Plantes, CNRS, F-66860 Perpignan, France. 3Present address: Departament de Biologia Vegetal, Facultat de Biologia, Universitat de Barcelona, 08028 Barcelona, Spain

*To whom correspondence should be addressed: E-mail: [email protected]; [email protected]. Christophe Belin, Tel: +33 4 68661754; Fax: +33 4 68668499; E-mail: [email protected]. Laboratoire Génome et Développement des Plantes, Université Perpignan Via Domitia, F-66860 Perpignan, France. Jean-Philippe Reichheld, Tel: +33 4 68662225; Fax: +33 4 68668499; E-mail: jpr@univ- perp.fr. Laboratoire Génome et Développement des Plantes, Université Perpignan Via Domitia, F-66860 Perpignan, France.

Abstract

Lateral roots are essential structures of the root system. They are crucial for the uptake of nutrients and water from the soil. Lateral roots initiate from the primary root and their development is controlled by a broad number of molecular pathways. Glutaredoxins are thiol-disulfide proteins involved in the control of reactive oxygen species homeostasis, as well as the activity of several enzymes implicated in plant development. However, little is known about their functions in lateral root formation. By genetic an in silico analyses, we demonstrate the specific contribution of glutaredoxins ROXY19 and GRXS17 in lateral root initiation and growth. ROXY19 and GRXS17 genes are differentially and highly expressed in different stages of the lateral root primordia formation. Analysis of knock-out plants revealed a contrasted role of ROXY19 and GRXS17 in lateral root development. 124 Plants lacking ROXY19 increased both the density and the elongation rate of lateral roots, and modulates lateral roots responses to abscisic acid. However, in the grxs17 mutant, lateral root primordia development is impaired. Moreover, an EMS-based screening allowed us to isolate mutants that are able to suppress the primary root and/or lateral root phenotype of the grxs17 mutant. This study strengthens the importance of redox regulation during lateral root development and offers new perspectives to unravel these mechanisms.

Keywords: Lateral root, glutaredoxin, GRXS17, ROXY19, Abscisic Acid (ABA)

Introduction

The successful localization and uptake of nutrient sources by the terrestrial plants depend on their capacity of post-embryonic organogenesis and on the cellular plasticity of the primary root (PR). Lateral roots (LR) are new organs formed from xylem pools pericycle (XPP) founder cells through eight distinctive stages of development. Concisely, in the early stages (I to IV) XPP cells dedifferentiate and proliferate within the endodermis to form the lateral root primordia (LRP), which continually divide and grow through the cellular layers from the endodermis until emerging from the epidermis of the parent root in the latest stages of formation (Malamy and Benfey, 1997; Péret et al., 2009). Prebranch sites in which cells acquire competence to develop lateral root are marked by the auxin- dependent signal that takes place in the transition zone between proliferating and elongating cells of the PR meristem (Möller et al., 2017). Both initiation and development of LR are complex processes that are controlled by a broad number of molecular pathways. Reactive oxygen species (ROS) are signaling compounds produced as by-products of cellular metabolism as well as by biotic an abiotic stress that play a crucial role during the formation of lateral roots (Manzano et al., 2014). Indeed, overproduction of intracellular ROS by genetic manipulation increases the number of prebranch sites and facilitates the emergence of lateral root primordia (Manzano et al., 2014; Li et al., 2015; Orman-Ligeza et al., 2016). However, due to the high reactivity of ROS and their ability to damage many cellular components, their accumulation is tightly controlled by several mechanisms. Key components in the control of intracellular ROS levels and signaling in plants are thiol disulfide oxidoreductases that share the canonical CxxC motif at the active site, which

125 allows the transfer of electrons to cysteine residues of specific targets proteins. Members belonging to this family are thioredoxins (TRX) and glutaredoxins (GRX) that use as electron donors thioredoxin reductase (TR) and glutathione in its reduced state (GSH), respectively, forming the so-called NADPH-dependent thioredoxin reductase (NTS) and NADPH-dependent glutathione/glutaredoxin (NGS) systems. These systems mediate the redox control of proteins involved in many cellular functions, such as ATP synthesis, carbon metabolism and defense responses (Rouhier et al., 2008; Meyer et al., 2008; Geigenberger et al., 2017; Thormählen et al., 2018). Efforts to unravel physiological functions of the thiol reduction systems have been limited by the redundancy among both NGS and NTS pathways. However, the genetic manipulation of both genes that encode cytosolic thioredoxin reductases NTRA and NTRB (ntra ntrb double mutant), and the use of weak alleles (cad2 or pad2) of GSH1, encoding a rate-limiting enzyme in GSH synthesis, allowed to overcome this redundancy by altering the reduction capacity of cytosolic/nuclear TRX and GRX. Interestingly, both ntra ntrb double mutant and cad2 mutant are only mildly affected in their root system development, the ntra ntrb cad2 triple mutant synergistically displays severe defects in root system architecture, affecting both PR growth and LR development (Reichheld et al., 2007; Bashandy et al., 2010). This suggests that some cytosolic and/or nuclear TRX and GRX are key players controlling root system architecture in Arabidopsis. In this work, we focused on glutaredoxins, heat-stable oxidoreductases that catalyze the reduction of thiols residues in the presence of GSH. The number of GRX in land plants is high compared to the number of genes that encode them in other organisms. In Arabidopsis, there are thirty-one members compared to six GRX in yeast, and five in both E. coli and mammals (Meyer et al., 2012). They are divided in five classes according to the amino acid sequence of the active site and several of them are highly expressed across different root tissues (Rouhier et al., 2009; Meyer et al., 2009; Belin et al., 2015). Among those GRX, the iron-sulfur cluster-containing glutaredoxin GRXS17 is involved in plants adaptation against abiotic stresses such as heat, light, and iron deficiency (Cheng et al., 2011; Knuesting et al., 2015; Yu et al., 2017). It is localized in both nucleus and cytosol and is present in all organs in Arabidopsis but more highly expressed in stems, young leaves, flowers, and roots. Some of the targets of GRXS17 have been isolated and their interactions demonstrated in vivo like the Nuclear Factor Y Subunit C11/Negative Cofactor 2a (NF-YC11/NC2a) which is important during light stress (Cheng

126 et al., 2011; Knuesting et al., 2015; Yu et al., 2017), and the BolA2 isoform which interaction might be important in the Fe-S cluster formation in both cytosol and nucleus (Couturier et al., 2014; Dhalleine et al., 2014). The grxs17 knock-out mutant displays severe defects in root system development, in particular a clear reduction in PR growth and meristem size maintenance (Cheng et al., 2011; Knuesting et al., 2015). However, there is no report about its function during LR development. Interestingly, other nucleo cytoplasmic GRX have been shown to play major roles during plant development, particularly those from the ROXY class (subgroup III). For example, ROXY1 and ROXY2 are controlling Arabidopsis flower development by interacting with the TGA family of transcription factors (Li et al., 2009b; Murmu et al., 2010; Li et al., 2011). Another example in Arabidopsis root development has been reported by Patterson and coworkers in 2016, since a cluster of redundant ROXY genes (ROXY11, 12, 13 and 15) represses the PR growth in response to nitrate, in a cytokinin-dependent way. Several other studies on maize and rice support the involvement of GRX in controlling plant development (Wang et al., 2009; Hong et al., 2012; Yang et al., 2015; Mhamdi and Van Breusegem, 2018). Despite the increasing evidence of the important roles of glutaredoxins in modulating plant development, there was no example of any GRX or TRX involved in the control of lateral root development so far. Here, after an in silico overview of GRX gene expression during the development of Arabidopsis lateral root primordium, we concentrate on the functions of ROXY19 and GRXS17 in modulating the initiation and growth of lateral roots. We show that ROXY19 has a dual role in controlling both the initiation and the elongation of the lateral root, whereas GRXS17 specifically regulates the lateral root primordium development. In order to better understand how GRXS17 affects both PR meristem and LR primordium, we performed a genetic suppressor screening that revealed an uncoupling between both root system functions of GRXS17.

Plant material and growth conditions

The roxy19 (previously described in Huang et al., 2016) mutant carries a Ds element insertion 45bp before the Start codon and is derived from No-0 Arabidopsis thaliana ecotype. The transgenic line overexpressing ROXY19 in Col-0 ecotype (Huang et al., 2016) has been provided by Christiane Gatz. grxs17 SALK T-DNA insertion and

127 Pro35S:GRXS17-GFP in grxs17 lines were previously described in Knuesting et al., 2015. ROXY19 and GRXS17 correspond to the AGI numbers At1g28480 and At4g04950, respectively. Arabidopsis surface-sterilized seeds were placed in a 0.5X Murashige and Skoog medium without sucrose with the addition of 0.8% of plant agar, and placed vertically, without stratification, in a growth chamber (22°C with a 16-h photoperiod and a photon flux of 160 μmol.m -2 .s -1 ). ABA treatments were performed after 5 days of growth on normal MS1/2 medium, by transferring seedlings via a nylon mesh to fresh medium, containing or not the indicated concentrations of ABA.

Phenotypic analyses of root system

For the study of the root system, images of the plates were taken with a canon camera (PowerShot A620). Photographs were analyzed and quantified using the public domain image analysis program ImageJ version 1.52 (http://rsb.info.nih.gov/ij/) and the NeuronJ plugin (https://imagescience.org/meijering/software/neuronj/). Lateral root density is the result of the number of lateral roots visually identified with the help of a stereo-microscope (MZ12, Leica), divided by the length of the primary root part in which lateral roots are visible. In order to analyse LRP development, plates containing 6-day-old plants were turned 90° and plants were observed after 48 hours using a standard optical microscope (Axioscop2, Zeiss).

Isolation of suppressor mutants of the grxs17 phenotype

A population of 50,000 grxs17 seeds (M0) was chemically mutagenized using 0.3% ethyl methanesulfonate according to standard procedures (Lightner and Caspar, 1998). The progeny of 40,000 mutagenized seeds (M1) was harvested in 384 pools, and 400 seeds from each pool were used to screen for mutant M2 plants displaying reversion of PR and LR grxs17 phenotypes. 90 putative candidates were grown and propagated, and the M3 progeny was screened for the same phenotypes in order to validate the reverted phenotype. We selected the 12 best candidates that are presented in this manuscript.

Immunodetection of GRXS17

128 Total protein extraction was performed by grinding tissues in liquid nitrogen and resuspension in extraction buffer containing 25mM of Tris-HCl pH 7.6, 75 mM NaCl, 10mM DTT, 1mM phenylmethylsulfonyl fluoride (PMSF) and 0.15% NP-40. After measurements of the protein concentration by Bradford method (1976), 10 μg proteins were separated by SDS/PAGE and transferred to nitrocellulose membranes. GRXS17 were detected by western blot using rabbit polyclonal antibodies against GRXS17 diluted to 1:25 000. Goat anti-rabbit antibodies conjugated to horseradish peroxidase (1:10 000) were used as secondary antibodies and revealed with enhanced chemiluminescence reagents (Immobilon Western Chemiluminescent HRP Sbstrate, Millipore).

Results

Glutaredoxin gene expression during lateral root primordia development

In order to unravel the roles of glutaredoxins in lateral root primordia development, we conducted an in silico analysis of the expression profile of all members of the glutaredoxin family during the eight stages of the LRP formation. Members of the glutaredoxin groups I and II were particularly highly expressed during LRP development, while the majority of class III, the largest group of glutaredoxin composed by twenty-one members presented low accumulation of transcripts in each stage of the lateral root primordia formation (Figure II.2.1). However, ROXY18 and ROXY19 genes had a strong expression in early stages (I to VI) and then slightly decreased when the primordium reach the epidermis. ROXY1, ROXY5, ROXY7, and ROXY17 displayed differential expression, particularly ROXY5 and ROXY7 show a high level in the development that takes place within the endodermis and showed fluctuation in the last stages. Transcripts encoding members of the glutaredoxin class IV show contrasted level of expression during the LRP formation. Glutaredoxin 4CXXC2 was more expressed in stage VI and VII, while 4CXXC12 was high in the emergence stage. Finally, the four members of the group V were strongly expressed in all stages, particularly glutaredoxins CPFA4 and CPFC3. The differences in transcript accumulation profiles between genes of the glutaredoxin family suggest specific roles of GRX during the lateral root primordial development (Figure II.2.1).

129 Figure II.2.1 In silico glutaredoxins gene expression analysis during lateral root formation. The table shows the relative expression values from the BAR browser tool for each stage of lateral root primordial development (Winter et al., 2007). The data come from the work published by Voß et al., 2015. The scheme representation of each stage of the LRP is adapted from Peret et al., 2012.

130 ROXY19 regulates both LR density and LR elongation

To further investigate the function of glutaredoxins in lateral root formation, we decided to focus onthe importance of two specific GRX genes. First, GRXS17 is a serious candidate since it has previously been shown to be a critical component for normal root system development (Cheng et al., 2011; Knuesting et al., 2015). We also focused on ROXY19, the most expressed during LR development among the ROXY family, for which functions in regulating plant development have previously been reported. Since no T-DNA knock- out line is available, we analyzed a mutant line in Nossen (No-0) ecotype in which a Ds transposon is inserted 45bp before the Start codon and results in a strong decrease in ROXY19 mRNA levels (Sup. Figure II.2.1) (Huang et al., 2016). Similarly to the grxs17 mutant, down-regulation of ROXY19 resulted in a moderate negative effect on the PR growth compared to wild-type plants (Figure II.2.2A and II.2.2B). While grxs17 does not show any difference in lateral root density, roxy19 displays significant increase in LR density compared to wild-type plants 10 days after sowing (Figure II.2.2.C). At the same developmental stage, we also noticed longer lateral roots in roxy19 compared to wild type, which was not observed in grxs17 (Figure II.2.2A and Sup. Figure II.2.2). We further investigate this phenotype in the roxy19 mutant but also in plants overexpressing ROXY19 under the control of the strong 35S-CaMV promoter (Huang et al., 2016). We performed a detailed analysis of lateral root elongation between days 10 and 14. Interestingly, and in contrast to PR, LR elongate faster in roxy19 mutant than in the wild type. Consistently, the opposite phenotype was observed in Pro35S:ROXY19 plants in which LR elongation rate is slower than the respective wild-type Col-0 ecotype (Figure II.2.3A). This suggests a critical role for ROXY19 in inhibiting lateral root initiation and mature LR growth under normal growth conditions. Because abscisic acid (ABA) is known to modulate LR growth, we assayed LR elongation rate of roxy19 versus wild-type plants in the presence of ABA. After transferring 5 days-old platelets on media containing different concentrations of ABA (2-5 μM), we first noticed that the average length of LR was shorter in response to ABA at day 10, both in wild type and roxy19 mutant (Figure II.2.3B). This result is consistent with the well-documented negative effect of ABA on LR development (Duan et al., 2013; Zhao et al., 2014).

131 Figure II.2.2. Glutaredoxins GRXS17 and ROXY19 mediate lateral root development. (A) Root system phenotype of grxs17 and roxy19 mutants, their respective wild- type controls (Col-0, No-0), and transgenic plants Pro35S:GRXS17 in grxs17 background (35S:GRXS17). Scale bar = 0.5 mm. (B) Primary root elongation and (C) Lateral root density were calculated from at least 16 10-day-old plants. Error bars indicate the standard deviation; asterisks indicated a P-value ≤ 0.001 using Welch's t-test statistical analysis.

132 However, the quantification of LR elongation between day 10 and day 14 (Figure II.2.3C) shows a moderate but significant dose-dependent increase of wild-type lateral root elongation rate in the presence of ABA. Conversely, the same amounts of ABA significantly decrease the rate of elongation in roxy19 mutant. This result suggests that ABA can either repress or enhance LR growth, depending on the ROXY19 expression level.

GRXS17 is required for proper LR primordium development

Given the role of ROXY19 in inhibiting lateral root initiation, we further investigated LR primordium (LRP) development stages and measured the density of LRP and LR in earlier stage of lateral root development, in 8-day-old seedlings (Figure II.2.4A). We show that roxy19 mutant displays more emerged LR and less LRP than the wild-type, consistent with the repressor role of ROXY19 for LR growth described above. Conversely, grxs17 plants display an increase in LRP compared to wild-type plants while no emerged lateral roots are observed at that stage. Together with the previous results showing no difference in final LR density in older plants (Figure II.2.2C), it suggests that GRXS17 specifically regulates LRP development, but not LR initiation or elongation. The initiation of the lateral root primordium can be induced and synchronized by gravitropic stimulus (Ditengou et al., 2008; Péret et al., 2012). In order to further evaluate the role of GRXS17 during LRP development, we performed such a gravistimulation experiment. 48 hours after gravistimulation, we can notice a strong delay in LRP development in grxs17 mutant compared to wild-type plants. Indeed almost all LRP are in stages I to IV in the mutant whereas they are mostly emerged or in the latest stages in the wild-type (Figure II.2.4B). This delay was rescued in the lines overexpressing GRXS17 (CaMV35S promoter) in the grxs17 background (Figure II.2.4B). It took us almost six days after gravistimulation to see LR emergence in almost all grxs17 plants (data not shown). Altogether, these results show that GRXS17 specifically functions in controlling LRP development, in addition to regulate PR meristem activity.

133 Figure II.2.3. LR elongation rate and its response to abscisic acid depend on ROXY19. (A) LR elongation rate calculated between day 10 and day 14(n ≥ 9), (B) average length of emerged LR at 10 days (n ≥ 13), and (C) LR elongation rate in response to different concentrations of ABA (n ≥ 9). Error bars indicate the standard deviation; single asterisks indicated a P-value ≤ 0.01 while double asterisks indicated a P-value ≤ 0.001 using Welch's t-test statistical analysis.

134 Identification of GRXS17-dependent pathways during root development

In order to identify components of GRXS17-dependent pathways that regulate root system development, we decided to use a forward genetic strategy, looking for grxs17 suppressor mutations. We randomly mutagenized a grxs17 mutant population using ethyl methanesulfonate (EMS) and designed a genetic screen based on both PR and LR phenotypes of grxs17. M2 seeds from about 200 M1 pools were sown and grown vertically during 7 days. At this age, plants displaying longer roots compared to the whole population were identified, and gravistimulation was applied. 3 days after gravistimulation, we examined each root bend, looking for plants that displayed emerged LR (Figure II.2.5A). We selected and grew 274 M2 candidate plants, before performing a confirmation screening in the M3 offspring of each M2 candidate (Figure II.2.5B). We isolated 11 serious candidates suppressing either one or both PR and LR phenotypes of grxs17 mutant, and checked that GRXS17 protein is still absent in such mutants (Sup. Table II.2.1, Sup. Figure II.2.3). Hence, 41C3, 15D1, 20H1 38A1, 38F1, and 41D3 mutants are able to suppress both phenotypes, while 12F3, 22E1, 28A2, and 28H1 only overcome the PR growth inhibition. In particular, the strongest suppressor, 41C3, fully reverts PR elongation and LR emergence to the wild type levels (Figure II.2.5B and II.2.5C). Interestingly, the 33G2 candidate was selected only for suppression of the delay in LRP development. Fine analysis of the M3 population confirms that 33G2 only partially complement the LRP phenotype, but not the PR growth (Figure II.2.5B and II.2.5C). These results suggest that GRXS17-dependent pathways controlling both PR meristem and LRP involve both common and specific components, and that we are able to uncouple both phenotypes.

Discussion

There are thirty-one genes encoding glutaredoxins in the Arabidopsis genome and some of them have been shown to be important in plant development and in plant responses to biotic and abiotic stress conditions (Reichheld et al., 2007; Bashandy et al., 2010; Cheng et al., 2011; Patterson et al., 2016). However, despite several members of glutaredoxin family are highly and differentially expressed in all parts of the root system (Belin et al., 2015), most of their specific roles in root development remain unknown.

135 Figure II.2.4. GRXS17 functions during LRP development. (A) LR and LRP density in grxs17 and roxy19 8-day-old seedlings (n ≥ 20). Error bars indicate the standard deviation; double asterisks indicated a P-value ≤ 0.001 using Welch's t-test statistics analysis. (B) Percentage of LRP in each developmental stage, 48 hours after gravistimulation (n ≥ 20).

136 Discovery and characterization of localized new functions of thiol-disulfide proteins can help to understand the redox-dependent mechanisms that control root development plasticity. In this study we addressed the roles of GRX in LR development and unraveled the importance of GRXS17 and ROXY19 in different stages of the LR initiation and growth.

ROXY19 as a critical component of LR development

We show that ROXY19 has a dual function as a positive regulator of LR initiation and an inhibitor of LR elongation. ROXY19 is a CC-type glutaredoxin that is induced by salicylic acid (SA) and has been shown to be involved in the crosstalk of salicylic acid and jasmonic acid (JA) by interacting with basic/leucine zipper-type transcription factors TGA, leading to the suppression of the jasmonic/ethylene response pathway during detoxification (Ndamukong et al., 2007; Zander et al., 2012; Herrera-Vásquez et al., 2015; Huang et al., 2016). The role of SA on LR development is poorly studied. Some study report that the exogenous addition of SA inhibits LR density (Armengot et al., 2014) (Armengot et al., 2014), while other studies show that LR density is increased by Plant Growth-Promoting Rhizobacteria (PGPR) in a SA-dependent way (Shi et al., 2010). Because ROXY19 gene expression very much depends on SA and environmental cues affecting SA levels (UV, wounding, or bacteria), ROXY19 could be a key component in fine-tuning and adapting LR development to such biotic and abiotic constraints. Abscisic acid (ABA) is a major regulator of root system architecture since many studies showed effects of ABA on LR development, although contradictory effects have also been reported. The complexity of ABA-dependent LR control could be explained by some differences among species and by the influence of other environmental cues on ABA functions (Harris, 2015). In our standard in vitro growth conditions we show that ABA first inhibits LR development, but thereafter promote the LR elongation rate. This is perfectly consistent with the study from Zhao et al., (2014) that shows an inhibitory role of ABA by promoting LR quiescence, followed by a PYL8-dependent recovery phase during which ABA promotes LR elongation by boosting auxin pathways. We show that this positive effect of ABA on LR growth under standard growth conditions switches to a clear inhibitoryeffect of ABA when ROXY19 gene expression is severely decreased.

137 Figure II.2.5. Screening for grxs17 suppressor mutants. In (A) we show the double phenotype used to select EMS plants that suppress both primary root elongation (at 7 days) and LRP defects (at 10 days) in grxs17 background. Red arrows indicate good candidates; the right picture also displays a magnification of the bending zone (red circle) due to gravistimulation; (B) PR growth quantification in 33G2 and 41C3 suppressor mutants. Error bars indicate the standard deviation; double asterisks indicated a P-value ≤ 0.001 using Welch's t-test statistical analysis (n ≥ 35). (C) LR emergence, 3 days after gravistimulation; white bars represent percentage of plants with emerged lateral roots in the gravistimulation-induced bending area and red bars represent percentage of plants without emerged LR (n ≥ 35).

138 This suggests that ROXY19 is a major regulator of ABA effects on LR development. We propose that ROXY19 could therefore be a central component of the crosstalk allowing the LR system to adapt its development to the complex combination of biotic and abiotic environmental constraints. Further studies aiming to unravel the ROXY19-dependent mechanisms that control LR development and its responses to ABA would probably result in major advances in understanding the root system adaptation to its environment. We know that ROXY19 binds and regulates some TGA transcription factors, even though the molecular mechanism is not known yet. Such a regulation could also be important for the control of LR development. Interestingly, two of these TGA factors (TGA1 and TGA4) have been reported to regulate LR initiation in response to nitrate (Alvarez et al., 2014). Other TGA genes, such as TGA9 and TGA10, are also strongly expressed in LR (Winter et al., 2007). Genetic and biochemical interaction studies between ROXY19 and these candidate TGA would probably bring useful information about the thiol-dependent redox mechanisms controlling LR development.

Lateral root primordia development is depending on GRXS17 functions

The involvement of GRXS17 in the formation of new organs from the primary root is supported by the strong developmental delay affecting LRP in grxs17 mutant during the lateral root synchronization assay. This is consistent with the observed increase in LRP number and the decrease in emerged LR in young plantlets. Together with its already described function in the PR meristem, this suggests that GRXS17 plays a critical and general role in all meristems. Similar to primary meristems, LR primodium formation might rely from cell division impairment (Cheng et al., 2011, Knuesting et al., 2015, and personal unpublished data). Moreover, the identification of many suppressor mutants restoring both PR and LR wild-type phenotypes supports the hypothesis that GRXS17 plays similar functions during PR and LR development. Identification of the candidate suppressor mutations will help to decipher the way GRXS17 is acting to regulate LR development. This might occur through crosstalk between GRXS17 and auxin signaling which play a major role in LR development (Casimiro et al., 2003; Cheng et al., 2011). Another hypothesis could be related to iron metabolism in which GRXS17 has been shown to be involved (Yu et al., 2017).

139 Figure II.2.6. Proposed model of the involvement of ROXY19 and GRXS17 in the control of the lateral root development. The evidence reported in this work suggest a critical function of GRXS17 in the early stages of the LRP development, while ROXY19 seems to act negatively in the priming and elongation process after the emergence of the LRP.

140 Over-accumulation of iron decrease the number of mature lateral roots and increase the number of lateral root primordia, which phenocopy the grxs17 impairment in the lateral root primordia development (Manzano et al., 2014; Reyt et al., 2015). Future experiments have to address the link between iron metabolism and GRXS17 in lateral root development. The identification of the causal mutation in the suppressor mutants we identified here will be key for understanding the role of GRXS17 in LR development. Although most suppressor mutants (e.g. 41C3) are able to complement both grxs17 PR elongation and LR development phenotypes, the analysis of the 33G2 mutant allowed us to uncouple these two physiological pathways. In this particular case, the causal mutation might reveal a new key player in LRP development.

ROXY19 and GRXS17 are distinct regulators of the lateral root development

Based on the results showed in this study, we propose a model to explain the roles of both ROXY19 and GRXS17 in Arabidopsis during lateral root development. From the genetic studies we propose that glutaredoxin GRXS17 is required during lateral root primordia development in the stage in which the primordium is crossing the outer layers of the PR. ROXY19 is involved in the spacing between lateral roots, suggesting it functions during the induction process of LR (priming or activation of founder cells), and acts negatively in LR elongation after the emergence of the primordium (Figure II.2.6). The evidence that the control of the lateral root elongation by ABA is ROXY19-dependent let us envisage a hypothesis in which the glutaredoxin ROXY19 modulates the activity of the plant hormone abscisic acid to control the rate of elongation of the mature lateral roots.Intriguingly, we demonstrate that two redox proteins belonging to the same thiol reductase family are involved in LR development but in a very different way. GRXS17 might act as an intrinsic regulator of meristematic activity while ROXY19 might participate in the adaptation of LR development to its environment. Consistently, ROXY19 and GRXS17 are expected to act through very different mechanisms and pathways. Indeed, they differ in their already reported interacting proteins. Moreover, these two GRX have different structural characteristics. ROXY19 is a monodomain dicysteinic glutaredoxin having a CCMC motif, whereas GRXS17 contains one thioredoxin-like domain and three glutaredoxin domains harboring a CGFS active site able to coordinate an iron-sulfur cluster. This study is the first

141 report of the functions of thiols redox-regulating systems during LR development. Here we focused on GRXS17 and ROXY19 but we also report that other candidate GRX are highly or specifically expressed during LR development. The GRX-dependent regulation of LR is likely even more complex and critical than reported here. Future studies on these other candidates might unveil other GRX-dependent checkpoints during LR development. Other components of the redox homeostasis regulation, such as ROS, have already been reported to play important roles during LR development, supporting a key role for redox regulation in this process. ROS also relay ABA- dependent signaling during abiotic stress responses (Pei et al., 2000). However the connection between GRX pathways and ROS functions is not yet unveiled. Therefore, our study opens new avenues to study the involvement of such redox regulation in LR development and its adaptation to environmental cues, particularly in relation to hormonal signaling.

Acknowledgements:

We thank Christiane Gatz for the Pro35S:ROXY19 trangenic lines. This work was supported by the Centre National de la Recherche Scientifique, the Agence Nationale de la Recherche (ANR-Blanc Cynthiol 12-BSV6-0011), Agropolis Fondation through the ‘Investissements d’Avenir’ programme (ANR-10-LABX 0001-01) and a BQR grant from the University of Perpignan Via Domitia. This study is set within the framework of the "Laboratoires d’Excellences (LABEX)" TULIP (ANR-10-LABX-41). JA T-H is supported by a PhD grant from Consejo Nacional de Ciencia y Tecnología (CONACYT:254639/411761).

142 SUPPLEMENTARY INFORMATION

Supplementary figure II.2.1. Characterization of different insertion lines in ROXY19. The panels show RT-PCR for ROXY19 (top panel) and the control EF1 α (bottom panel) genes in the Ds insertion line (roxy19 in No-0) used in this study and its relative wild type ecotype (No-0). Other lanes show two T-DNA insertion lines from the SALK (in Col-0 ecotype) that do not alter ROXY19 expression levels and are not useful in our study. The last lane present a control for the experiment (without reverse transcriptase).

143 Supplementary figure II.2.2. Lateral root elongation rate in grxs17 and roxy19 mutants. Their respective wild-type controls (Col-0, No-0), and transgenic plants Pro35S:GRXS17 in grxs17 background (35S:GRXS17). Error bars indicate the standard deviation (n ≥ 16; double asterisks indicated a P- value ≤ 0.001 using Welch's t-test statistics analysis.

144 Supplementary figure II.2.3. Immunodetection of GRXS17 in suppressor mutants 41C3 and 33G2. A) Total protein extraction stained with Coomassie Brilliant Blue. B) Western-blot using GRXS17-specific antibodies.

145 Supplementary table II.2.1. Phenotype of the 11 suppressor mutant selected from the EMS screening. NO indicates that the candidate mutant does not revert either PR or LR grxs17 phenotype. +, ++, and +++ indicate increasing levels of phenotype reversion (+++ being equivalent to a wild-type phenotype).

146 References

Alvarez, J.M., Riveras, E., Vidal, E.A., Gras, D.E., Contreras-López, O., Tamayo, K.P. et al. (2014). Systems approach identifies TGA 1 and TGA 4 transcription factors as important regulatory components of the nitrate response of Arabidopsis thaliana roots. Plant J. 80, 1-13. doi: 10.1111/tpj.12618

Armengot, L., Marquès-Bueno, M.M., Soria-Garcia, A., Müller, M., Munné-Bosch, S. and Martínez, M.C. (2014). Functional interplay between protein kinase CK 2 and salicylic acid sustains PIN transcriptional expression and root development. Plant J. 78, 411-423. doi: 10.1111/tpj.12481

Bashandy, T., Guilleminot, J., Vernoux, T., Caparros-Ruiz, D., Ljung, K., Meyer, et al. (2010). Interplay between the NADP-linked thioredoxin and glutathione systems in Arabidopsis auxin signaling. The Plant Cell, 22, 376-391. doi: 10.1105/tpc.109.071225

Belin, C., Bashandy, T., Cela, J., Delorme‐Hinoux, V., Riondet, C. et al. (2015). A comprehensive study of thiol reduction gene expression under stress conditions in Arabidopsis thaliana. Plant Cell Environ. 38, 299-314. doi: 10.1111/pce.12276

Bradford, M.M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, doi: 248-254. 10.1016/0003-2697(76)90527-3

Casimiro, I., Beeckman, T., Graham, N., Bhalerao, R., Zhang, H., Casero, P. et al. (2003) Dissecting Arabidopsis lateral root development. Trends Plant Sci. 8, 165-171. doi: 10.1016/S1360-1385(03)00051-7

Cheng, N.H., Liu, J.Z., Liu, X., Wu, Q., Thompson, S.M., Lin, J., et al. (2011). Arabidopsis monothiol glutaredoxin, AtGRXS17, is critical for temperature-dependent postembryonic growth and development via modulating auxin response. J. Biol. Chem. 286, 20398-406. doi: 10.1074/jbc.M110.201707

Couturier, J., Wu, H.C., Dhalleine, T., Pégeot, H., Sudre, D., Gualberto, J.M., et al. (2014). Monothiol glutaredoxin–BolA interactions: redox control of Arabidopsis thaliana BolA2 and SufE1. Mol. Plant. 7, 187-205. doi: 10.1093/mp/sst156

Dhalleine, T., Rouhier, N. and Couturier, J. (2014). Putative roles of glutaredoxin-BolA holo-heterodimers in plants. Plant Signal Behav. 9, p.e28564

Ditengou, F.A., Teale, W.D., Kochersperger, P., Flittner, K.A., Kneuper, I., van der Graaff, E., et al. (2008). Mechanical induction of lateral root initiation in Arabidopsis thaliana. Proceedings of the Proc. Natl. Acad. Sci, 105, 18818-23. doi: 10.1073/pnas.0807814105

Duan, L., Dietrich, D., Ng, C.H., Chan, P.M.Y., Bhalerao, R., Bennett, M.J. et al. (2013). Endodermal ABA signaling promotes lateral root quiescence during salt stress in Arabidopsis seedlings. The Plant Cell, 25, 324–341. doi: 10.1105/tpc.112.107227 147 Geigenberger, P., Thormählen, I., Daloso, D.M. and Fernie, A.R. (2017). The Unprecedented Versatility of the Plant Thioredoxin System. Trends Plant Sci. 22, 249-262. doi: 10.1016/j.tplants.2016.12.008

Harris, J. (2015). Abscisic acid: hidden architect of root system structure. Plants. 4, 548- 572. doi:10.3390/plants4030548

Herrera-Vásquez, A., Carvallo, L., Blanco, F., Tobar, M., Villarroel-Candia, E., Vicente- Carbajosa, J. et al. (2015). Transcriptional control of glutaredoxin GRXC9 expression by a salicylic acid-dependent and NPR1-independent pathway in Arabidopsis. Plant Mol. Biol. Report. 33, 624-637, doi: 10.1007/s11105-014-0782-5

Hong, L., Tang, D., Zhu, K., Wang, K., Li, M. and Cheng, Z. (2012). Somatic and reproductive cell development in rice anther is regulated by a putative glutaredoxin. Plant Cell, 24, 577-88. doi: 10.1105/tpc.111.093740

Huang, L.J., Li, N., Thurow, C., Wirtz, M., Hell, R. and Gatz, C. (2016). Ectopically expressed glutaredoxin ROXY19 negatively regulates the detoxification pathway in Arabidopsis thaliana. BMC Plant Biol. 16, 200. doi: 10.1186/s12870-016-0886-1

Knuesting, J., Riondet, C., Maria, C., Kruse, I., Bécuwe, N., König, N., et al. (2015). Arabidopsis glutaredoxin S17 and its partner, the nuclear factor Y subunit C11/negative cofactor 2α, contribute to maintenance of the shoot apical meristem under long-day photoperiod. Plant Physiol. 167, 1643-1658. doi: 10.1104/pp.15.00049

Li, N., Sun, L., Zhang, L., Song, Y., Hu, P., Li, C. et al. (2015). AtrbohD and AtrbohF negatively regulate lateral root development by changing the localized accumulation of superoxide in primary roots of Arabidopsis. Planta, 241, 591-602. doi: 10.1007/s00425- 014-2204-1

Li, S., Gutsche, N., and Zachgo, S. (2011). The ROXY1 C-terminal L ∗∗ LL motif is essential for the interaction with TGA transcription factors. Plant Physiol. 157, 2056– 2068. doi: 10.1104/pp.111.185199

Li, S., Lauri, A., Ziemann, M., Busch, A., Bhave, M. and Zachgo, S. (2009). Nuclear activity of ROXY1, a glutaredoxin interacting with TGA factors, is required for petal development in Arabidopsis thaliana. Plant Cell, 21, 429-441. doi: 10.1105/tpc.108.064477

Lightner, J. and Caspar, T. (1998). 14Seed Mutagenesis of Arabidopsis. Methods Mol. Biol. 82, 91-102. Doi: 10.1385/0-89603-391-0:91

Malamy, J.E. and Benfey, P.N. (1997). Organization and cell differentiation in lateral roots of Arabidopsis thaliana. Development, 124, 33-44.

Manzano, C., Pallero, M., Casimiro, I., De Rybel, B., Orman-Ligeza, B., Van Isterdael, G. et al. (2014). The emerging role of ROS signalling during lateral root development. Plant 148 Physiol. 165, 1105-1119. doi: 10.1104/pp.114.238873

Meyer, Y., Belin, C., Delorme-Hinoux, V., Reichheld, J.P. and Riondet, C. (2012). Thioredoxin and glutaredoxin systems in plants: molecular mechanisms, crosstalks, and functional significance. Antioxid. Redox Signal. 17, 1124-1160. doi: 10.1089/ars.2011.4327

Meyer, Y., Buchanan, B.B., Vignols, F. and Reichheld, J.P. (2009). Thioredoxins and glutaredoxins: unifying elements in redox biology. Annu. Rev. Genet. 43, 335-367. doi: 10.1146/annurev-genet-102108-134201 Meyer, Y., Siala, W., Bashandy, T., Riondet, C., Vignols, F. and Reichheld, J.P. (2008). Glutaredoxins and thioredoxins in plants. Biochim. Biophys Acta. 1783, 589-600. doi: 10.1016/j.bbamcr.2007.10.017

Mhamdi, A. and Van Breusegem, F. (2018). Reactive oxygen species in plant development. Development. 145. doi: 10.1242/dev.164376.

Möller, B.K., Xuan, W. and Beeckman, T. (2017). Dynamic control of lateral root positioning. Curr Opin Plant Biol. 35, 1-7. doi: 10.1016/j.pbi.2016.09.001

Murmu, J., Bush, M.J., DeLong, C., Li, S., Xu, M., Khan, M. et al. (2010). Arabidopsis bZIP transcription factors TGA9 and TGA10 interact with floral glutaredoxins ROXY1 and ROXY2 and are redundantly required for anther development. Plant Physiol., 154, 1492-504. doi: 10.1104/pp.110.159111

Ndamukong, I., Abdallat, A.A., Thurow, C., Fode, B., Zander, M., Weigel, R. et al. (2007). SA-inducible Arabidopsis glutaredoxin interacts with TGA factors and suppresses JA- responsive PDF1.2 transcription. Plant J. 50, 128-139. doi: 10.1111/j.1365- 313X.2007.03039.x

Orman-Ligeza, B., Parizot, B., De Rycke, R., Fernandez, A., Himschoot, E., Van Breusegem, F., et al. (2016). RBOH-mediated ROS production facilitates lateral root emergence in Arabidopsis. Development, 143, 3328-39. doi: 10.1242/dev.136465

Patterson, K., Walters, L.A., Cooper, A.M., Olvera, J.G., Rosas, M.A., Rasmusson, A.G. et al. (2016). Nitrate-regulated glutaredoxins control Arabidopsis primary root growth. Plant physiol. 170, 989-999. doi: 10.1104/pp.15.01776

Pei, Z.M., Murata, Y., Benning, G., Thomine, S., Klüsener, B., Allen, G.J., et al. (2000). Calcium channels activated by hydrogen peroxide mediate abscisic acid signalling in guard cells. Nature, 406, 731-734.

Péret, B., De Rybel, B., Casimiro, I., Benková, E., Swarup, R., Laplaze, L., et al. (2009). Arabidopsis lateral root development: an emerging story. Trends Plant Sci. 14, 399-408. doi: 10.1016/j.tplants.2009.05.002

Péret, B., Li, G., Zhao, J., Band, L.R., Voß, U., Postaire, O., et al. (2012). Auxin regulates aquaporin function to facilitate lateral root emergence. Nat. Cell Biol. 14, 991-998. doi: 149 10.1038/ncb2573

Potters, G., Pasternak, T.P., Guisez, Y., Palme, K.J. and Jansen, M.A. (2007). Stress- induced morphogenic responses: growing out of trouble? Trends Plant Sci. 12, 98-105. doi: 10.1016/j.tplants.2007.01.004

Reichheld, J.P., Khafif, M., Riondet, C., Droux, M., Bonnard, G. and Meyer, Y. (2007). Inactivation of thioredoxin reductases reveals a complex interplay between thioredoxin and glutathione pathways in Arabidopsis development. Plant Cell, 19, 1851-1865. DOI: 10.1105/tpc.107.050849

Reyt, G., Boudouf, S., Boucherez, J., Gaymard, F. and Briat, J.F. (2015). Iron-and ferritin- dependent reactive oxygen species distribution: impact on Arabidopsis root system architecture. Mol. Plant, 8, 439-453. doi: 10.1016/j.molp.2014.11.014.

Rouhier, N., Couturier, J., Johnson, M.K. and Jacquot, J.P. (2010). Glutaredoxins: roles in iron homeostasis. Trends Biochem. Sci. 35, 43-52. doi:10.1016/j.tibs.2009.08.005 Rouhier, N., Lemaire, S.D. and Jacquot, J.P. (2008). The role of glutathione in photosynthetic organisms: emerging functions for glutaredoxins and glutathionylation. Annu. Rev. Plant Biol. 59, 143-166. doi: 10.1146/annurev.arplant.59.032607.092811

Shi, C.L., Park, H.B., Lee, J.S., Ryu, S. and Ryu, C.M. (2010). Inhibition of primary roots and stimulation of lateral root development in Arabidopsis thaliana by the rhizobacterium Serratia marcescens 90–166 is through both auxin-dependent and-independent signaling pathways. Mol. Cells. 29, 251-258. doi: 10.1007/s10059-010-0032-0

Thormählen, I., Naranjo, B., Trujillo-Hernandez, J.A., Reichheld, J.P., Cejudo, F.J. and Geigenberger, P. (2018). On the Elaborate Network of Thioredoxins in Higher Plants. In: Progress in Botany. Springer, Berlin, Heidelberg, 1-29. doi: 10.1007/124_2018_16

Voß, U., Wilson, M.H., Kenobi, K., Gould, P.D., Robertson, F.C., Peer, W.A., et al. (2015) The circadian clock rephases during lateral root organ initiation in Arabidopsis thaliana. Nat. Commun. 6, 7641. doi: 10.1038/ncomms8641

Wang, Z., Xing, S., Birkenbihl, R.P. and Zachgo, S. (2009). Conserved functions of Arabidopsis and rice CC-type glutaredoxins in flower development and pathogen response. Mol. Plant. 2, 323-335. doi: 10.1093/mp/ssn078

Winter, D., Vinegar, B., Nahal, H., Ammar, R., Wilson, G.V. and Provart, N.J. (2007). An “Electronic Fluorescent Pictograph” browser for exploring and analyzing large-scale biological data sets. PloS one. 2, 1-12. doi: 10.1371/journal.pone.0000718

Yang, F., Bui, H.T., Pautler, M., Llaca, V., Johnston, R., Lee, B.H. et al. (2015). A maize glutaredoxin gene, Abphyl2, regulates shoot meristem size and phyllotaxy. Plant Cell, 27, 121-31. doi: 10.1105/tpc.114.130393

Yu, H., Yang, J., Shi, Y., Donelson, J., Thompson, S.M., Sprague, S. et al. (2017). Arabidopsis glutaredoxin S17 contributes to vegetative growth, mineral accumulation, and 150 redox balance during iron deficiency. Front. Plant Sci. 8, 1045. doi: 10.3389/fpls.2017.01045

Zander, M., Chen, S., Imkampe, J., Thurow, C. and Gatz, C. (2012). Repression of the Arabidopsis thaliana jasmonic acid/ethylene-induced defense pathway by TGA-interacting glutaredoxins depends on their C-terminal ALWL motif. Mol. Plant. 5, 831-840. doi: 10.1093/mp/ssr113

Zhao, Y., Xing, L., Wang, X., Hou, Y.J., Gao, J., Wang, et al. (2014). The ABA receptor PYL8 promotes lateral root growth by enhancing MYB77-dependent transcription of auxin-responsive genes. Sci. signal. 7:ra53. doi:10.1126/scisignal.2005051

151 PERSPECTIVES

Based on the reviewers’s comments, this manuscript was not accepted for publication in the present form. I will summarize and analyze reviewers’s criticisms, propose perspectives to address these concerns which should improve the manuscript for future resubmission. After examining those observations from the reviewers, we agree that additional experimental data would be required to gain more information and to clarify the roles of glutaredoxins ROXY19 and GRXS17 in controlling root development. First, reviewers criticize the lack of information concerning ROS levels and glutathione redox status in the grx mutants. Indeed, it would be useful to get such information in order to better analyze GRX functions in the context of the general redox environment. This could be performed by using the fluorescent compound 2,7- Dichlorodihydrofluorescein (DCFH-DA), that measure ROS in a non-specific manner, but also by crossing the grx mutants with other redox reporters such as roGFP2 which reports glutathione redox status, and HyPer H2O2 specific genetically-encoded probe (Belousov et al., 2006). We therefore initiated such crosses. A second major criticism addresses the ABA phenotype reported in our manuscript. Indeed, we only present preliminary experiments assaying roxy19 LR responses to exogenous ABA. Further work is required in order to better understand the involvement of ROXY19 in regulating LR responses to both exogenous and endogenous ABA. First, endogenous ABA levels could be quantified in roxy19 versus wild-type plants. Global quantifications using LC-MS/MS or immunodetection assays could be performed and more precise tissue-specific detection of ABA using the FRET-based probe ABAleon, already available in the lab, would allow to monitor ABA during LR development (Waadt et al., 2014). In order to investigate ROXY19 functions in controlling endogenous ABA pathways, we should cross roxy19 with mutants affected in the ABA pathway abcg25, abi1-1, abi2-1, abi3-1, aba1-3, aba2-3 (Duan et al., 2013; Kuromori et al., 2018). Another possibility would be to address situations known to increase ABA levels such as wild-type plants submitted to water stress or plants overexpressing ZEP ABA-biosynthesis gene (Park et al., 2008). We should also analyze, as explained in the previous paragraph, ROS and glutathione redox status in response to ABA in both wild-type and roxy19 mutant all 152 along with LR development. Finally, ROXY19 expression should be analyzed in situations affecting ABA levels, either endogenously or exogenously. This set of experiments would help us in clarifying the function of ROXY19 in the context of ABA-dependent control of LR development.

The last major criticism of the reviewers concerns the genetic suppressor screening we set up to identify GRXS17-dependent pathways in regulating both primary root and lateral root development. In the proposed manuscript, we present the genetic screen and its direct output, the identification and characterization of several candidates that suppress either one or both root phenotypes of grxs17. However, we do not address the identification of loci affected in those candidates. Indeed, we wanted to present the different mutants identified in order to show that the GRXS17-dependent control of both primary root growth and LR development involves common components, although we were also able to uncouple both pathways since a few mutants specifically suppress grxs17 phenotypes either in primary root or lateral root. The identification of the causal mutation of the suppressor mutants was a clear objective of my PhD and will be presented in the next chapter.

153 154 II.3 CHAPTER 3: IDENTIFICATION OF THE CAUSAL MUTATION IN THE SUPPRESSOR VARIANTS OF THE grxs17 MUTANT

II.3.1 INTRODUCTION

The induction of point mutations in the genome by Ethyl methanesulfonate (EMS), is a powerful tool that has been used for a long time in forward genetics in order to identify genes that are involved in a particular physiological function (Greene et al., 2003; Kim et al., 2006; Lightner and Caspar, 2009). Overall, EMS generate random point mutations by the specific alkylation of the nucleotide guanine producing O6-ethylguanine, which pairs with thymine instead of cytosine, resulting in the transition from G-C to A-T pairs. It had been reported that 5% of the EMS-mutations consist of stop codons, while around 65% results in missense variants and the rest are silent mutations (Greene et al., 2003). Practically, the population of original M0 seeds is bathed in a solution containing EMS, and the resulting M1 seeds are sown and grown in order to get homozygous mutations in the M2 next generation. M2 seeds or plants are then phenotypically screened and candidates isolated. The historical procedure for the identification of the mutated gene that produces the phenotype of interest (causal mutation) involves several experimental steps, and is known as map-based cloning or positional cloning (Jander et al., 2002). Map-based cloning consists of crossing the M2 mutagenized plants with a divergent line (outcrossing), and use the F2 generation for mapping the homozygous locus by using molecular markers, thereafter, phenotypic analysis in the F3 generation allows the confirmation of the possible candidates, and the sequencing of the gene (Figure II.3.1) (Arondel et al., 1992; Giraudat et al., 1992; Jander et al., 2002). However, due to the advances in sequencing technologies that decreased the costs and time for genome sequencing, the development of many bioinformatic, and genomic tools have demonstrated to be useful in the identification of the causal mutation in forward genetics (Schneeberger, 2014; Candela et al., 2015).

155 Figure II.3.1. Map-based cloning. The identification of the causal mutation by the well known map-based cloning approach, consists in the localization of the homozygous locus in the F2 generation from the cross between the mutant, and a divergent line (around 7 months of experimentation). Thereafter, a new mapping in the F2 plants it is necessary for narrow the location of the causal mutation, and finally, the phenotypic analysis in the F3 progeny, and the sequencing of the DNA are necessary for the final identification of the gen responsible of the phenotype. The whole process without interruption or technical problems takes around 1 year. Scheme adapted from Jander et al., 2002.

156 SHOREmap, Genome Analysis Toolkit (GATK), Samtools, and Next Generation Mapping (NGM), are some of the pipelines based on Next-Generation Sequencing (NGS) technologies, that have been developed for the identification of the causal mutation, which is sometimes referred as mapping-by-sequencing (Schneeberger et al., 2009; Austin et al., 2011; James et al., 2013; Allen et al., 2013). SHOREmap, GATK, and Samtools are pipelines that are used to calculate the allele frequency of single nucleotide polymorphisms (SNPs) that are present in a pool of plants selected to harbor the phenotype of interest in the recombinant F2 generation, while NGS is able to identify the causal mutation based on the discordant chastity, which is an allelic frequency corrected for sequencing errors (Allen et al., 2013). Previous work has shown that all those pipelines are powerful tools to identify causal mutations in forward genetic screens, although minor differences exist (Figure II.3.2) (Allen et al., 2013). We have previously discussed the important roles of glutaredoxins during primary root growth, and lateral root formation. In order to identify GRXS17-dependent pathways involved in the control or root development, a genetic screen for suppressor mutations was done and several interesting candidates were isolated (see previous chapter). In addition, the mutagenized M2 population was used to look for grxs17 suppression of other developmental phenotypes. Because of the advantages that have been shown for the use of NGS in the identification of the causal mutation, we decided to use this strategy to identify the causal mutations in the best suppressor mutants we selected. All the candidates had been isolated and all mapping populations had been generated when I arrived for my PhD. I therefore prepared all DNA extractions of these populations and sent for sequencing, before analyzing the obtained sequences. This chapter is mainly dedicated to present you my bioinformatic analyses.

157 Figure II.3.2. Comparison of NGS pipelines in the detection of the causal mutation. This figure is adopted from the study published by Allen et al in 2013. They showed that NGM, SHOREmap, Samtools, and GATK are able to identify the causal mutation in the EMS-mutant BCF2 that produced leaf hyponasty in the mir159a plants (involved in the microRNA pathway). The causal mutation was located in chromosome 3, which produced a stop codon in the position 1405085, that corresponds to the gene HASTY involved in the microRNA biogenesis (black arrow).

158 II.3.2 RESULTS

II.3.2.1 Suppressor lines of the grxs17 mutant

Although candidates were selected before I joined the lab, I will recapitulate in this section the different mutants we are interested in. For the initial screening of suppressors of grxs17 root phenotypes, about 400 M2 seedlings were screened from each of 200 pools of M1 plants, each M1 pool containing about 100 M1 plants. 274 M2 candidates were picked-up, ranked from A (strong phenotype) to C (uncertain phenotype) and propagated to the M3 generation. We checked the phenotypes of M3 progeny for about 90 M2 candidates, after exclusion of sterile or almost sterile plants and all C-ranked M2 plants. We finally isolated twelve strong suppressor lines deriving from different M1 pools. As mentioned in chapter II, some of them are able to complement only one root phenotype of the grxs17 mutant (Table II.3.1). Indeed, 22E1, 28A2, and 28H1 only suppress the impairment in the primary root elongation, while 33G2 only complement the lateral root primordia phenotype (Table II.3.1). The other eight lines can suppress both root phenotypes of the grxs17 mutant. Among them, 41C3 line was the strongest suppressor mutant in both primary root elongation, and lateral root primordia phenotypes (Table II.3.1). In addition to the defects in the root system (Figure II.3.3), the grxs17 mutant also shows a delay in the development of the leaves (Figure II.3.4) (Knuesting et al., 2015). This delay is much more pronounced when plants are grown under long day (16h light/8h dark) and high light (500 µmol photons m-2 s-1) conditions. At a later developmental stage, leaves are elongated, show a thickened lamina and the development of the main floral spike is delayed (Figure II.3.5). Therefore, M2 plants from 24 pools of M1 mutagenized population were screened under long day/high light conditions. Potential suppressors were selected after 10 days of growth for the rescue of the development of the first two leaves (Figure II.3.6). The selected candidates were further grown under the same conditions. 15 suppressor mutants show an almost full rescue of the grxs17 mutant phenotype: normal leaves and main floral spike (Figure II.3.6). Moreover, 20 other plants had an intermediate phenotype between grxs17 and wild-type plants (not shown). The 15 suppressor lines were backcrossed with the parent grxs17 mutant in Col-0 (Table II.3.1). 159 Table II.3.1. Suppressor mutants of the grxs17. The headers: Primary root, lateral root, and shoot, indicate the phenotype of the grxs17 that the suppressor line is able to complement. Samples sent to sequencing are indicated in red. Mutant Primary root Lateral root Method 12F3 ++ NO Outcross 15D1 ++ ++ Outcross 20H1 ++ ++ Outcross 22E1 ++ NO Outcross 13B2 ++ + Outcross 28A2 ++ NO Outcross 28H1 ++ NO Outcross 33G2 NO ++ Outcross 38A1 ++ + Outcross 38F1 ++ + Outcross 41C3 +++ +++ Outcross 41D3 ++ + Outcross Mutant Shoot Backcross 23G2 +++ Backcross 1A1 +++ Backcross 7F1 ++ Backcross 21F1 ++ Backcross 29B1 +++ Backcross

160 In contrast to the root suppressor screen, we decided to use backcrosses instead of outcross (Table II.3.1), because of the milder shoot phenotype observed in the grxs17 mutant found in the Landsberg ecotype (Figure II.3.6A). This screen was performed in collaboration with the group of Pascal REY at the Biosciences and Biotechnologies Institute of Aix-Marseille (BIAM, CEA Cadarache).

II.3.2.2 NGM and SHOREmap failed to identify the causal mutation in the suppressor variants.

In order to prepare the mapping-by-sequencing population, we followed the experimental design shown in Figure II.3.7. Suppression variants of the root phenotype (in Col-0), were outcrossed with the grxs17 mutant within the Landsberg erecta (Ler) background to produce recombinant plants. The F1 generation was self-crossed, and the F2 progeny were screened to select a pool of homozygous plants, displaying the suppressor phenotype, which were used for genomic DNA extraction and sent for deep sequencing (Figure II.3.7). The same steps were performed for the suppression lines of the shoot phenotype of the mutant grxs17, but instead of outcrossing with the polymorphic variant Ler, those lines were backcrossed with grxs17 in Col-0 (Figure II.3.7). Row reads were mapped against the TAIR10 reference genome using bowtie2 pipeline, which produced a good coverage depth value for each of the samples (Table II.3.2). First, we analyzed both grxs17 alleles in Col-0 and Ler ecotypes. We could confirm that both mutants harbor a single T-DNA insertion at the At4g04950 locus, corresponding to GRXS17. In addition to the insertion in GRXS17, the grxs17 mutant in Col-0 displays 19605 SNPs and short insertions or deletions (INDELs), compared to the reference TAIR10 Col-0 genome, which means about 1 SNP or INDEL every 6 kb on average. As expected, the difference observed between both ecotypes is far bigger, since the grxs17 mutant in Ler background shows 860776 SNPs or short INDELs, compared to the TAIR10 reference Col-0 genome, which means about 1 SNP or INDEL every 138 bases on average, or 7.25 SNPs or INDELs per kb. This is almost twice higher frequency that what was previously reported by comparing Ler genome resequencing to the TAIR9 reference Col-0 genome sequence with a similar 19X coverage (Lu et al., 2012).

161 Figure II.3.3. Characterization of the grxs17 knock-out mutants. (A) Representation of the T-DNA insertions in the GRXS17 gene. SALK021301 in Col-0, CSHL GT22726 in Ler ecotypes. Lines and black bars represent transgenic sequences and exons respectively. (B) Phenotype of 12 day-old in vitro plants from wild- type and grxs17 in both Col-0 and Ler ecotypes. Plants were grown at 20°C under long day (16h/8h) conditions. (C) Primary root length from 12 day-old plants. Error bars represent the SD of the mean (n=20). * p<0.001 by Student 's T-test.

162 In order to find causal mutations, we first analyzed the data from genome sequencing with the online platform known as NGM (http://bar.utoronto.ca/ngm/). For this reason, vcf files produced by samtools mpileup were converted to *.emap extensions, and uploaded to the platform NGM. However, only one of the 17 samples analyzed showed a putative candidate on chromosome 2 (Figure II.3.8). The suppressor mutant named 12F3 showed a region with a relatively low SNPs frequency, supporting a non-recombinant area that might link to the causal mutation (Figure II.3.8A) (Austin et al., 2011). Subsequent analysis using the discordant chastity statistic (ChD), allowed to narrow the putative localization of the causal mutation, around the position 11818542, in which candidates were proposed by the NGM platform. Among the closest candidates, the gene At2g27050 that encodes ETHYLENE-INSENSITIVE3 protein (EIL1) shows a SNP at position 11547020, thus changing a proline to leucine (Figure II.3.8B). However, in order to validate the results from NGM, I decided to perform another analysis with SHOREmap pipeline. SHOREmap pipeline calculates the allele frequency of SNPs, which are used as markers to identify the causal mutation. Unfortunately, and in agreement with the NGM data, the results from SHOREmap did not show any positive result among the 15 samples. However, 12F3, 33G2, 41D3, and 22E1 samples showed a gradual increase of allele frequency in the chromosome 2, which could suggest that this region harbor the causal mutation (Figure II.3.9). Due to the failure of both NGM and SHOREmap to detect the causal mutation generated by EMS in most of the samples, I decided to perform a manual analysis based on the allele frequency concept.

163 Figure II.3.4. Phenotypes of the grxs17 mutant under long day and high light conditions. (A) Phenotype of wild-type and grxs17 mutant after 8 (B) and 20 days. Short days= 8 hours and long days = 16 hours under 200µmol m-2 s-1. Figure adapted from Knuesting et al., 2015.

164 II.3.2.3 Identification of the causal mutation using Samtools/SnpEff and R.

The methodology adopted to detect the causal mutation is illustrated in figure II.3.10. The identification of the SNPs was performed with Samtools, and the annotation of the variants was accessed by using the SnpEff command line. Thereafter, I calculated the allele frequency as the number of reads that support the SNP, divided by the total reads that were mapped at that position. The next step was the subtraction of the SNPs that belong to both parental grxs17 lines. In the case of the backcross strategy I only removed the SNPs from grxs17 in the Col-0 background. Finally, I only retained the EMS-SNPs that produced a transition from G-C to A-T pairs (Greene et al., 2003). The advantage of using SnpEff pipeline for the annotation of the vcf files is that it also adds the level of impact that the SNP might have on the gene product. This 4 levels classification (high, moderate, low, and modifier) is useful to identify possible candidates of the suppressor lines (Table II.3.3). It is likely that the causal mutation will produce a high or moderate impact in the gene product (Table II.3.3). For this reason, the identification of the causal mutation was addressed by considering the level of impact of the SNP, the allele frequency, and the distribution of the EMS-SNP across the chromosome. In order to validate my method, I run the analysis using a published causal mutation (Sun and Schneeberger, 2015). and I compared the data with SHOREmap, which also uses allele frequency (Figure II.3.11). Using the same data I was able to detect the causal mutation using my pipeline based in samtools and SnpEff. It is important to note that the area in which the causal mutation is located (Figure II.3.11, blue arrow) was devoid of SNPs in the Ler background. Conversely, there is an increase in the frequency of SNPs in this region (Figure II.3.11). Analysis of the allele distribution of the 15 samples shows in some cases similar profiles as those presented with the SHOREmap pipeline (data not shown). Indeed, 12F3 presents a particular increase in the allele frequency of several SNPs on the lower arm of chromosome 2, that was also indicated by the NGM pipeline (Figure II.3.12). However, there is no homozygous EMS-mutation located in chromosome 2.

165 Figure II.3.5. Screening of grxs17 suppressors. M2 seedlings growth at 10 days under long day/high light conditions (16 hours under 500 µmol photons m-2 s-1). The red circle represents a selected suppressor candidate.

166 The absence of homozygous SNPs in most of the samples, and the lack of areas with low Ler-SNPs frequency, led me envisage that there could be a contamination in the F2 population that was sequenced. This could explain why I am not able to find homozygous candidates in my analysis. For this reason, I decided to use another approach to find putative candidates for the causal mutation. The methodology consisted first in removing the SNPs that were common to the 15 suppressors. As EMS generated independent mutations in each M1, it is unlikely that the causal mutation will be in the same position even if it affects the same gene for several suppressor lines (Addo-Quaye et al., 2017). Moreover, since it is possible that the pools of F2 that were sent to sequencing contain contaminant non suppressor plants, I took in consideration SNPs that were between 0.80 to 1.0 in the AF scale, and SNPs with high and moderate impact. This strategy produced several possible candidates in the samples 12F3, 13B2, 22E1, 23G2, 28A2, 33G2, 38F1, 41C3, and 41D3 (Table II.3.4). The analysis of the allele distribution of the SNP frequency of those samples shows interesting results in the samples 12F3 and 23G2 in chromosome 2, 33G2 in chromosome 5, and 41C3 in chromosome 4 (Figures II.3.13-II.3.21). Interestingly, we observed an increase in the SNP frequency in the areas that are harboring those possible candidates.

167 Figure II.3.6. Phenotypes of suppressor lines of the grxs17 mutant. (A) Shoot phenotype of wild-type plants Col-0 and Ler, together with the T-DNA insertion lines in the GRXS17 gene in both ecotypes. (B-D) Suppressor lines isolated by EMS-screening. 1A1, 7F1, 23G2, 21F1, and 29B1 were selected due to the strong suppression of the shoot phenotype presented in the mutant grxs17.

168 II.3.3 DISCUSSION

As it was described in chapter II, disruption of the gene that encodes to the glutaredoxin GRXS17 reduces the growth of the primary root elongation, and impairs the formation of lateral root primordia. In order to study the pathways in which GRXS17 is involved, our group isolated several suppressor lines that can revert either the primary root or/and the lateral root primordia phenotype. Moreover, in collaboration with Dr. Pascal REY, we identified additional suppressor lines that are able to revert the shoot developmental phenotype of grxs17. Altogether, this set of suppressor mutants constitute a key toolbox to study the roles of the glutaredoxins GRXS17 during plant development (Figure II.3.22). The positive analysis of some of the suppressor lines described in this chapter was achieved by the implementation of homemade command lines, that are based principally in Samtools, SnpEff, and R. This is due to the failure of both NGM and SHOREmap to detect candidates in practically the whole set of suppressor lines. However, the identification of possible candidates with my homemade pipeline depends to a large extent on the subtraction of the EMS-derived SNPs that are common to the 15 samples, considering only SNPs that produced a high impact in the gene product, and by considering candidates within the range of 0.8 to 1.0 in the frequency scale. For this empirical measure, I considered the possibility of having contaminant or heterozygous plants in the pools of F2 that were sent for sequencing. This would decrease the homozygosity of the causal mutation, making difficult its identification by the NGM and SHOREmap platforms, that are based on the homozygosity of the candidates. This possible contamination of the mapping population by non-suppressor plants illustrates the difficulty to conduct a quantitative screen based on root length. However, we have to notice that we got exactly the same problems using a separate screen on shoot development, in another lab, and using another strategy to establish the mapping population (backcross instead of outcross). This might illustrate problems in the mutagenized population. However, we used a very standard protocol for EMS mutagenesis and the rate of usual marker phenotypes (such as white sectors), was apparently normal in the growing M1 population. Another hypothesis would be that grxs17 mutation itself, affecting a yet unidentified function, affects the feasibility of the screen.

169 Figure II.3.7. Mapping-by-sequencing methodology. This figure shows the two strategies to identify the causal mutation, from an EMS- screening that are based on NGS. Outcross and backcross are necessary in order to identify the causal mutation based on the allele frequency of linked SNPs, which are used as markers (Adapted from James et al 2013). In this study, we used as parent lines for outcross grxs17 mutant in Ler, which is crossed with the grxs17 mutant suppressor lines in Col-0 background. While in the backcross approach the suppressor lines were crossed only with grxs17 in Col-0 as a parent line. The resulting F2 generation that shows the suppression phenotype are collected and sent to sequencing.

170 Using the pipeline I developed, I was able to obtain several candidates in the suppressor lines (Table II.3.4). However, considering the allele frequency distribution (Figure II.3.13-II.3.21), samples 12F3, 41C3, 33G2, and 23G2 showed the most promising results. Those lines are characterized by their capacity to revert the root phenotypes of grxs17. 41C3 and 12F3 suppress the negative effect that grxs17 show in the primary root elongation growth, and the lateral root primordia formation, while 33G2 is only able to complement the lateral root primordia phenotype. 23G2 is another suppressor line that is capable to restore the normal shoot development that is compromised in the grxs17 mutant. The bioinformatic analysis of the genomic data for the suppressors described above, produced several candidates that are located in different chromosomes, in which there is an increase in the frequency of the EMS-derived SNPs. 41C3, the strongest suppressor line for the root phenotype, displays three candidate loci localized in chromosome 4. These consist in two missense and one nonsense mutations, in the genes encoding CRK8, a tripeptidyl peptidase enzyme, and an ATP-dependent DNA helicase respectively. Intriguingly, the only homozygous candidate identified among the suppressor lines with an allelic frequency equal to 1.0 is CRK8. The SNP identified in CRK8 produces a missense mutation, resulting in the change of the polar Threonine amino acid into the non-polar Isoleucine in the position 469. CRK8 belongs to the family of cysteine-rich RECEPTOR- like kinases, that are characterized by harboring four conserved cysteines in their amino acid structure, and to be involved in defense response against biotic stress. However, specific functions of CRK8 is still not known (Acharya et al., 2007; Wrzaczek et al., 2010; Idänheimo et al., 2014). In the case of 12F3, there were four missense candidates that are located in chromosome two in the genes At2g39980, At2g28850, At2g27050, and At2g28850. Interestingly, 12F3 was the only suppressor line showing a positive result in the NGM pipeline. At2g27050, that encodes ETHYLENE-INSENSITIVE3-like 1 (EIL-1), is a candidate gene identified via both pipelines. EIL-1 is a transcription factor that is involved in ethylene signaling, and we know that ethylene plays important functions during root development (Binder, 2007; Nziengui et al., 2018).

171 Table II.3.2. Alignment summary. Paired-end reads were alignment using bowtie2 pipeline to the TAIR10 reference genome. Coverage depth was calculated by multiplying the reads mapped by the length of the reads and divided by 134634692 bases (Arabidopsis genome size). Sample Total reads Reads mapped Coverage depth Mutant grxs17 in Col-0 32658340 32525218 24x Mutant grxs17 in Ler 26375896 25716761 19x 12F3 40681172 39935375 30x 20H1 41787580 40977145 31x 22E1 27703892 27251377 21x 13B2 33849592 33466053 25x 28A2 44187788 43342982 33x 33G2 36450154 35808297 27x 38A1 37490500 36831929 28x 38F1 30103614 29513284 22x 41C3 29690928 29244973 22x 41D3 35852870 35213851 27x 1A1 73460248 71382166 80x 21F1 78777228 74776000 84x 23G2 75044734 70954214 80x 29B1 67659386 64266186 72x 7F1 73521024 70136128 79x

172 33G2 suppressor line that only restores the normal lateral root primordia development, displays one candidate in a gene that encodes a hypothetical protein that is located in chromosome 5. The analysis of 23G2 that complements the shoot phenotype in the grxs17 mutant shows a nonsense mutation in the gene HAIRY MERISTEM 1 (HAM1), that encodes a member of the GRAS family transcription factors. Interestingly, it has been shown that HAM1 interacts with the homeodomain transcription factor WUS to activated the expression of CLV3 to control the shoot stem development. HAM1 is therefore a good candidate to harbor the causal mutation of the reversion phenotype observed in the 23G2 line (Zhou et al., 2015, 2018). Further studies have to be addressed to validate the candidates of the suppressor lines 41C3, 12F3, 33G2, and 23G2. In this regard, T-DNA SALK lines of those candidates were ordered. We will first examine the primary root elongation rate and lateral root primordium formation, or shoot development in high light, in such insertion lines. In parallel, RT-PCR will allow us to validate these insertion lines as knocked down or knocked out for the corresponding gene. Finally, another validation step of these candidates will consist in crossing the T-DNA insertion lines with grxs17 mutants to try to phenocopy the suppression observed in the corresponding EMS mutant line.

In conclusion, in this study, I showed the utility of my methodology based on Samtools, SnpEff, and R, in order to identify candidates for the causal mutation in our suppressor lines of the grxs17 mutant. Genetic and phenotypic analyses of mutants for each of the candidates have to be done in order to definitely validate the methodology and the causal mutations. This would offer the lab novel perspectives in understanding the important role of GRXS17 in regulating plant development.

173 Figure II.3.8. NGM results with suppressor line 12F3. (A) Allele frequency distribution in each chromosome of the sample 12F3. (B) Output of the analysis with the discordant chastity method for chromosome 2, that narrow the area in which the causal mutation might be located.

174 Figure II.3.9. SHOREmap results from chromosome 2. Samples 12F3, 22E1, 33G2, and 41D3 showed an increased in the allele frequency in chr2. However, the pipeline did not produce any possible target in each of those suppressor lines. Red dots represent the SNPs harboring by each sample, and the blue line show the window-based allele frequency (AF).

175 Figure II.3.10. Method to detect the causal mutation based on Samtools and SnpEff pipelines. The representation of the strategy to localize the causal mutation is based on the SNP detection by Samtools and the subsequent annotation of SnpEff pipeline. Thereafter, the allele frequency is calculated using the platform R. In order to narrow the possible candidates, SNPs that belong to the parent lines or that are not induced by EMS were filtered. The candidates are selected by the annotation class from the SnpEff, the allele frequency (homozygous = 1), and the distribution of the AF of SNP markers that are linked to the causal mutation.

176 Table II.3.3. SnpEff SNP classification. The annotation of the SNP in the vcf file contained also a distinction level, which depends on the impact that the SNP might produce in the gene product. Impact Meaning Example HIGH There is high probability that the Stop_gained SNP would affect the gene product. MODERATE It might alter the gene product. Missense variants LOW There is a low probability that Synonymous variants the SNP would alter the gene product. MODIFIER SNPs presented in non-coding Intergenic regions sequences.

177 Figure II.3.11. Comparison of SHOREmap and Samtools/snpEff pipeline. (A) Detection of the causal mutation in Chr2 using SHOREmap. The chart is adapted from (Sun and Schneeberger, 2015), in which an EMS mutation was found in the position 18,808,927 of Chr2 (yellow square), and which was responsible for the phenotype of the mutant named OCF2. (B) Dots represented the SNP frequency for EMS (red), non-EMS in Col-0 (black), and non-EMS in Ler (yellow). SNPs that belong to Ler background is plotted with negative values to facilitate the comparison.

178 Figure II.3.12. Allele frequency of SNPs in 12F3. Dots represented the SNP frequency for EMS (red), non-EMS in Col-0 (black), and non-EMS in Ler (yellow)

179 Table II.3.4. Possible candidates for the causal mutation. Highly samples in yellow indicated the strong candidates that I suggest due to they allele frequency distribution (see figures from 43-51). PR = Primary root, LR= Lateral root, and S = Shoot phenotypes.

Line Pos Chr AF Type Impact ID Description Phenotype 12F3 16688672 2 0.89 Missense Gly405Glu AT2G39980 HXXXD-type acyl- PR transferase family protein 12F3 12384307 2 0.84 Missense Ala219Thr AT2G28850 Cytochrome P450, family 710, subfamily A, PR polypeptide 3 12F3 11547020 2 0.8 Missense Pro236Leu AT2G27050 ETHYLENE- PR INSENSITIVE3-like 1 12F3 12383676 2 0.8 Missense Gly429Glu AT2G28850 Cytochrome P450, family PR 710, subfamily A, polypeptide 3 13B2 396165 5 0.84 Missense Gly138Arg AT5G02030 POX (plant homeobox) PR,LR family protein 13B2 25070824 1 0.83 Missense Ala5123Thr AT1G67120 Midasin-like protein PR,LR 13B2 1765918 3 0.83 Missense Leu165Phe AT3G05910 Pectinacetylesterase family PR,LR protein 13B2 8722818 4 0.82 Missense Ser446Asn AT4G15290 Cellulose synthase family PR,LR protein 22E1 11676305 2 0.8 Missense Ser249Asn AT2G27285 Nuclear speckle splicing PR regulatory-like protein (DUF2040) 22E1 13676315 4 0.8 Missense Pro155Ser AT4G27310 B-box type zinc finger PR family protein 22E1 18389785 4 0.8 Missense Cys252Tyr AT4G39600 Galactose oxidase/kelch PR repeat superfamily protein 23G2 18619957 2 0.81 nonsense Stop codon AT2G45160 GRAS family transcription S Gln26* factor 28A2 12549688 5 0.80 Missense Ala109Thr AT5G33280 Voltage-gated chloride PR channel family protein 33G2 9444194 1 0.90 Missense Ala1043Thr AT1G27180 Disease resistance protein LR (TIR-NBS-LRR class) 33G2 9108015 3 0.81 Missense Gly308Asp AT3G24982 Receptor like protein 40 LR

33G2 13336506 5 0.80 Missense Ala11Val AT5G35069 Hypothetical protein LR

38F1 7543402 5 0.81 nonsense Stop AT5G22690 Disease resistance protein PR,LR Trp627* (TIR-NBS-LRR class) family 41C3 6565314 2 1 Missense Arg54Lys AT2G15130 Plant basic secretory PR,LR protein (BSP) family protein 41C3 12130890 4 1 Missense Thr469Ile AT4G23160 cysteine-rich RECEPTOR- PR,LR like kinase 41C3 14991300 4 0.83 nonsense Trp174* AT4G30780 ATP-dependent DNA PR,LR 180 helicase 41C3 11168869 4 0.80 Missense Gly215Asp AT4G20850 tripeptidyl peptidase ii PR,LR

41D3 14667081 3 1 Missense Ala187Val AT3G42550 Eukaryotic aspartyl PR,LR protease family protein 41D3 17410231 5 1 Missense Pro195Ser AT5G43360 phosphate transporter 1;3 PR,LR

41D3 10952698 4 0.94 Missense Gly918Glu AT4G20270 Leucine-rich receptor-like PR,LR protein kinase family protein 41D3 22609997 1 0.90 Missense Arg60Lys AT1G61300 LRR and NB-ARC PR,LR domains-containing disease resistance protein

181 Figure II.3.13. EMS-SNP frequency of 12F3. Red dots represent the allele frequency of the EMS-SNPs, while the red line indicated the centromeric region of each chromosome.

182 Figure II.3.14. EMS-SNP frequency of 13B2. Red dots represent the allele frequency of the EMS-SNPs, while the red line indicated the centromeric region of each chromosome.

183 Figure II.3.15. EMS-SNP frequency of 22E1. Red dots represent the allele frequency of the EMS-SNPs, while the red line indicated the centromeric region of each chromosome.

184 Figure II.3.16. EMS-SNP frequency of 23G2. Red dots represent the allele frequency of the EMS-SNPs, while the red line indicated the centromeric region of each chromosome.

185 Figure II.3.17. EMS-SNP frequency of 28A2. Red dots represent the allele frequency of the EMS-SNPs, while the red line indicated the centromeric region of each chromosome.

186 Figure II.3.18. EMS-SNP frequency of 33G2. Red dots represent the allele frequency of the EMS-SNPs, while the red line indicated the centromeric region of each chromosome.

187 Figure II.3.19. EMS-SNP frequency of 38F1. Red dots represent the allele frequency of the EMS-SNPs, while the red line indicated the centromeric region of each chromosome.

188 Figure II.3.20. EMS-SNP frequency of 41C3. Red dots represent the allele frequency of the EMS-SNPs, while the red line indicated the centromeric region of each chromosome.

189 Figure II.3.21. EMS-SNP frequency of 41D3. Red dots represent the allele frequency of the EMS-SNPs, while the red line indicated the centromeric region of each chromosome.

190 Figure II.3.22. Suppressor mutants to study GRXS17 roles during plant development. Yellow boxes represent the lines that only complement the shoot phenotype, green box indicated the line that suppresses only the lateral root primordia impairment, blue boxes complement exclusive the primary root phenotype, and the black boxes indicated the lines that complement both primary root and lateral root phenotype of the mutant grxs17.

191 192 III GENERAL DISCUSSION

193 194 III.1 Glutathione as a mediator of root development

III.1.1 IBA control by glutathione

As it was demonstrated by genetic and molecular analyses presented in this work, glutathione is critical for root development in processes involving plant hormone signaling, as well as during abiotic stress conditions (Figure III.1). Indeed, by genetic and molecular analysis we demonstrated that glutathione is able to affect the IBA responses in lateral root and root hair. The situation looks different in the primary root. First, we showed that similarly to IAA, exogenous IBA application inhibits primary root growth, very likely due to IBA to IAA conversion and perception by the TIR1/AFB-auxin-Aux/IAA co-receptor complex. We also show that in the contrary to LR and RH, glutathione deficiency does not interfere with the IBA (and IAA)-dependent primary root growth inhibition. This might be due to the robustness of the auxin-dependent cell growth inhibition mechanism involving rapid and reversible root growth inhibition by the TIR auxin signaling pathway (Fendrych et al., 2018). Similarly, we have previously shown that the primary root growth response to IAA was not affected in glutathione deficient mutants (Bashandy et al., 2010). However, it is also likely that other thiol reduction pathways are able to compensate the glutathione deficiency on auxin response in primary root growth, as shown previously for the thioredoxin system (Bashandy et al., 2010). In another hand, the role of glutathione in primary root is well documented, particularly in the root apical meristem (RAM). Depletion of glutathione in the hypomorphic allele of GSH1 rml1 produced a strong defect in the primary root growth, due to a profoundly affected RAM. Still, the glutathione depletion in this mutant is much more pronounced than in our studies. To further explore the interaction between glutathione and IBA pathways in the primary root, it will be interesting to address the question in more affected glutathione deficiency conditions (e.g. zir1 mutant) or in mutants affected in complementary thiol reduction systems (e.g. ntra ntrb cad2 mutant, Bashandy et al., 2010).

195 Figure III.1. Glutathione as a key component to control the root system. This scheme aims to summarize the results presented in this manuscript. GSH = glutathione, IBA = Indole-3-butyric acid.

196 It is also possible that glutathione specifically interferes with IBA pathways responsible for the regulation of LR and root hairs. Control of lateral root development and root hairs by glutathione have been demonstrated in the past. A strong reduction of glutathione by high levels of BSO decreased the length of the root hairs and lateral roots numbers (Sanchez-Fernandez et al., 1997; Marquez-Garcia et al., 2014). This was also observed in my studies in which high levels of BSO treatments decreased the root hairs and lateral roots (data not shown). However, it is unknown if this particular phenotype is related to auxin signaling dependent on IBA. ROS accumulation has been demonstrated to take place in the tip of the root hairs as well as within the lateral roots, which is critical for the development of both lateral roots and root hairs (Carol and Dolan, 2006; Manzano et al., 2014; Orman-Ligeza et al., 2016; Mangano et al., 2018). Since glutathione is involved in ROS homeostasis through the Ascorbate/Glutathione cycle and the reduction of GRX/PRX pathways, glutathione deficiency might alter LR and RH growth in a ROS- dependent way (Manzano et al., 2014; Mangano et al., 2017; Noctor et al., 2018). Intriguingly, only plants with low reduce glutathione show impaired in the response to exogenous IBA, suggesting that GSH instead of GSSG is critical for the IBA response in the root architecture. A link between glutathione and auxin has been proposed before, in which depletion of glutathione decreased the polar auxin transport and signaling (Bashandy et al., 2010; Koprivova et al., 2010; Mhamdi and Van Breusegem, 2018). However, I showed that impairment in the biosynthesis of glutathione did not affect the responses to exogenous IAA, suggesting a specific role of glutathione in the IBA pathway. Nevertheless, as IAA can diffuse into cells, we cannot discard the possibility that glutathione controls downstream events of the IAA pathway, such as auxin transport mediated by polar carriers. Indeed, it has been described that low glutathione alters the auxin transport, and the expression of genes that encodes to PIN carriers. Still a redox regulation of PIN protein acyivity has not been clearly demonstrated (Bashandy et al., 2010; Koprivova et al., 2010). PIN2, through its importance for the acropetal distribution of auxin from the lateral root cap is of high interest (Bashandy et al., 2010; Koprivova et al., 2010). My experiments using inhibitors of the auxin transport (NPA and NOA as inhibitors of auxin efflux and influx, respectively) argue against a redox control of PIN and AUX/LAX proteins by glutathione (Parry et al., 2001; Kim et al., 2010; Kubeš et al., 2012). Moreover, depletion

197 of glutathione by genetic and pharmacological approaches did not produce impairment in the gravitropic phenotype as it is observed in pin2, pin1, aux1 mutants, and with auxin inhibitors (data not shown) (Swarup et al., 2001; Blilou et al., 2005). The control of IBA by glutathione seems to be particularly important in the basal meristem, in which the auxin signaling derived from the root cap is acting in the formation of new post-embryonic organs such as lateral roots (De Rybel et al., 2012; Xuan et al., 2015). Indeed, depletion of glutathione by BSO perturbes the auxin distribution, which I monitored with DR5-GUS after 24 hours treatment with IBA. Intriguingly, if the active IAA transport is not involved in the control of IBA, why is the DR5-GUS signal impaired in the basal meristem? This question is very important since no conversion in the basal meristem have been reported yet (Frick and Strader, 2018). For this reason, further experiments have to be performed in order to unravel the nature of the auxin signaling derived from IBA in the basal meristem. This could be done by pre-treating DR5-GUS lines with different concentration of NPA and visualizing the auxin signaling distribution after incubating them with IBA. Moreover, localization of IBA conversion enzymes in the basal meristem could give stronger evidence of the IBA conversion in this tissue. Indeed, visualization of the expression profile of genes that have been proposed to be specifically involved in the IBA conversion to IAA, such as IBR10 shown a strong expression in the basal meristem (Figure III.2). If IBA conversion would occur in the root cap, it will help to find the targets that we were not able to identify in this study. Our discovery that glutathione is controlling the IBA pathway in lateral root development and root hair growth is important in order to understand how IBA conversion to IAA is controlled at the molecular level. Such a yet almost unexplored topic could also open a new window to investigate which other auxin functions are redox regulated (e.g. by glutathione). In this regard, phenotypic characterization of other physiological processes in which IBA have been related such as in the hypocotyl growth, and cell expansion have to be done to address this question (Strader et al., 2011; Frick and Strader, 2018). Interestingly, before I arrived, the glutathione-dependent IBA response was first studied in hypocotyl elongation. IBA-increased hypocotyl length was almost abolished in cad2 mutant or in response to BSO. However, this phenotype was highly dependent on light intensity, suggesting even more complex regulation than the one shown in this work for LR and RH responses. This suggests that glutathione could regulate IBA pathways in the

198 whole plant, although this control could be integrated in more integrated regulation processes according to the tissue. LR and RH responses are very robust and reproducible and can therefore be used to unravel the mechanism used by glutathione to control IBA pathways.

III.1.2 IBA and glutathione involved in abiotic stress response

The impairment of root hair elongation of the mutants cad2 and ibr1ibr3ibr10 in response to phosphate deprivation give strong evidence for the role of IBA and glutathione on an important developmental adaptation to P carency (Figure III.1). The critical role that auxin plays in promoting root hair growth in response to low external P has been recently documented (Bhosale et al., 2018). Here, auxin synthesis, transport and response pathway components are involved. Therefore, my data suggesting that IBA conversion is implicated in root hair elongation upon P deprivation opens new perspective in this emerging field. Does IBA act in parallel to IAA in this pathway? Where is IBA conversion occurring? Is IBA-associated auxin signaling and transport involved under P deprivation? It is still not clear yet if the IBA functions in response to phosphorous deprivation is depending on glutathione. The full insensitivity of the ibr1ibr3ibr10 mutant to phosphorous deprivation make it difficult to challenge under glutathione deprivation (e.g. BSO treatment or cross between cad2 and ibr1ibr3ibr10 mutant). However, the P deprivation-dependent root hair phenotype could be explored in other IBA metabolism/conversion mutants in combination with glutathione deficiency. It would also be interesting to quantify the endogenous level of IBA and IAA in plants under phosphorous deprivation and in combination with BSO or mutants of the glutathione biosynthesis such as cad2 and pad2. As well, the discovery of the components involved in the control of IBA by glutathione such as direct targets will be critical to demonstrate a link between phosphorous deprivation and IBA modulation by glutathione. We can expect that an unknown target will be also impaired in the response to phosphorous deprivation. The discovery of the roles of glutathione for the response in root hairs under phosphorous deficiency, represent a big step to understand how the plant perceives the lack of phosphate, which is still not well understood (Ham et al., 2018).

199 Figure III.2. Tissue-specific expression of IBR10 in the root. The scheme represents the averages of absolute values from the BAR expression browser for the gene that encodes to IBR10. LRP = Lateral root primordia.

200 Moreover, this suggests that glutathione might be a part of a complex sensor mechanism to drive changes in the root system, depending on IBA and the nutrient status in the rhizosphere, and other abiotic stress conditions (Figure III.1). This would allow the plant to modify its root development in response to environmental stress based on glutathione activity (e.g. phosphorous deficiency). The glutathione measurements I performed in cad2 and ibr1ibr3ibr10 mutant plants did not reveal very significant modifications of the reduced and oxidized glutathione levels under P deprivation, although we notice a tendency to have lower oxidation. However, these measurements were done in whole plantlets which might not reveal local changes in the root cap for example. Therefore, glutathione redox homeostasis should be rather monitored in plants under P deprivation using redox-sensitive GFP reporter lines such as roGFP2. On the other hand, the fact that genes involved in the IBA pathway are up- regulated during abiotic stress conditions (Frick and Strader, 2018), makes interesting to see if other abiotic stresses are also dependent on IBA and glutathione, such as nitrogen deprivation and osmotic stress conditions, in which the genes that encode to IBR peroxisomal enzymes are up-regulated (Frick and Strader, 2018). Overall, the discovery of the control of the auxin signaling through the modulation of IBA by glutathione presented in this manuscript, give more evidence for the important roles that glutathione perform in developmental pathways, and give more clues about the way glutathione modulates the auxin signaling process and its roles during abiotic stress conditions.

III.2 Roles of glutaredoxin in the root system

Besides the roles of glutathione in the control of the root development by modulating the functions of plant hormones to abiotic stress conditions, glutathione is also highly important for the activity of a group of thiol enzymes known as glutaredoxins (GRX). This family of enzymes is composed in Arabidopsis by 50 members that are classified into five groups according to their amino acid sequence and the identity of their active site (Meyer et al., 2012; Belin et al., 2015). Indeed, the large diversity of the members of the glutaredoxin family in plants compared to other organisms is intriguing. Studies of their respective expression profiles in terms of tissue specific expression and expression under 201 various environmental stress conditions suggest that the members of these families are not redundant (Belin et al., 2015). Moreover, several functions of specific glutaredoxins have been documented in the literature, indicating that glutaredoxins regulate many developmental and stress response pathways (see introduction). However, their roles in the control of the root development are still poorly documented. In this study, I demonstrated that specific glutaredoxins performed uncharacterized roles in the control of the root development, especially in lateral root primordia formation. Indeed, in silico analysis showed that several members of the glutaredoxin family of enzymes are differentially expressed across the distinct stage of lateral root primordia formation (Chapter 2). However, phenotypic characterization of the root system in most of these grx mutants has not yet been performed. In this regard, we decided to study the functions of two glutaredoxins that are differentially expressed in lateral root primordia: ROXY19 and GRXS17. ROXY19 is a CC-type glutaredoxin that has been shown to interact with transcription factors TGA during detoxification process, involving salicylic acid and jasmonate (Ndamukong et al., 2007; Zander et al., 2012; Herrera-Vásquez et al., 2015; Huang et al., 2016). ROXY19 is one of the most expressed glutaredoxin during lateral root primordia formation. However, after studying the LRP development in the roxy19 mutant, we did not found any impairment in their LRP development formation. Nevertheless, the mutant showed an increase in both lateral root number and elongation, suggesting a negative function of ROXY19 in the control of the lateral root growth. Moreover, my genetic experiments showed that the induction of lateral roots by the plant hormone abscisic acid, a master regulator of plant responses during stress conditions (Finkelstein et al., 2002; Nemhauser et al., 2006; Xiong et al., 2006; Danquah et al., 2014; Kim et al., 2016), seems to be dependent on ROXY19. For this reason and as it was mentioned in chapter 2, genetic studies with ABA mutants such as abcg25, abi1-1, abi1-3, abi2-3, abi3-1, and aba2-1, crossed with roxy19 will bring new insides about, the link between glutaredoxin ROXY19 and ABA signaling, in the control of the lateral root growth. Besides the lateral root primordia phenotype of the roxy19 mutant, primary root elongation is also impaired in the mutant roxy19, which is similar to the phenotype found in the mutant grxs17. A cross between roxy19 and grxs17 mutant could also inform us if these two different GRX are acting independently in primary root growth. Indeed, beside the fact that both roxy19 and grxs17 seem to be involved in primary

202 root development, ROXY19 and GRXS17 are very different GRX in regard to their protein organization and activities. In contrast to the monodomain ROXY19, GRXS17 is a multidomain protein harbouring a Fe-S cluster, involved in different developmental processes like meristem activities, and in abiotic stress responses like tolerance to temperature stress (Cheng et al., 1995; Couturier et al., 2014; Knuesting et al., 2015; Yu et al., 2017). In my work, I also demonstrated that grxs17 presents a delay in the emergence of the LRP, suggesting a new role of GRXS17 in the root system development. The fact that the deglutathionylation activity of glutaredoxins is generally dependent on glutathione may suggest that the root development functions of ROXY19 and GRXS17 are glutathione-dependent. Accordingly, the primary root development of the grxs17 mutant is partially rescued by adding exogenous GSH in the growth medium (data not shown). However, ROXY19 and GRXS17 are atypical glutaredoxins for which a typical glutathione-dependent deglutathionylation activities has not been fully established (our unpublished data). Therefore, further studies have to be performed to establish the link between glutathione, ROXY19 and GRXS17 in root development phenotypes (Figure III.1).

III.3 Identification of the causal mutation in grxs17 suppressor lines

Identification of the possible pathways in which GRXS17 is involved to control plant development is an important aspect to understand the different roles that this glutaredoxin plays in plants. For this reason, the isolation of the suppressor mutants of the grxs17 phenotype represents an important tool to accomplish this goal. In this study, I performed several bioinformatic analyses in order to identify the causal mutation of the 15 suppressor lines of the grxs17 phenotype in both shoot and root. We were able to identify candidate SNPs in the lines 12F3 (primary root phenotype), 41C3 (both primary root and lateral root primordia phenotype), 33G2 (only lateral root phenotype), and 23G2 (shoot phenotype). The most promising candidate SNPs are located in genes encoding ETHYLENE- INSENSITIVE3-like 1(EIL-1) for suppressor 12F3, CRK8 identified in 41C3 suppressor, the hypothetical protein of 33G2 line, and HAIRY MERISTEM 1 in the suppressor 23G2.

203 Validation of the causal mutations are required using phenotypic characterization and complementation assays. Mutant lines in the different genes of interest are now under characterization. It is important to bold that EIL-1 is involved in the root growth (Nziengui et al., 2018), and HAM1 is critical for the shoot development (Zhou et al., 2015, 2018), which perfectly correlated with the phenotypes of the respective suppressor lines. If the complementation of grxs17 with the mutant lines is corroborated, further studies have to be done in order to elucidate the molecular mechanism by which GRXS17 interact with those genes. CRK8 is characterized by harboring four conserved cysteines in its sequences, which make him a good direct target of GRXS17 because of the possibility of interacting with CRK8 through the formation of thiol bonds. Since it is known that members of the CRK proteins have been involved in stress conditions, this will give evidence for the physiological meaning of the control of GRXS17 and CRK8 in the control of root system under biotic stress conditions (Acharya et al., 2007; Wrzaczek et al., 2010; Idänheimo et al., 2014). In silico analysis with the amino acid structure of EIL-1 and HAM1 need to be done in order to see if there are exposed cysteine residues that could be targets of GRXS17. However, it could be possible as well that they are downstream components in the control of GRXS17 in the root system.

III.4 Conclusions

In conclusion, the results showed in this Ph.D. work highlight the roles that glutathione and glutaredoxin play in the control of the root development, and during abiotic stress conditions. Moreover, the genetic and molecular experiments presented in this work revealed new functions for the three components studied in this thesis. Glutathione as a mediator of the IBA pathway, roles of both glutathione and IBA in the response to phosphorous deprivation, and glutaredoxins functions in the control of the root development. Those three major discoveries made in this Ph.D. work will be important to unravel the functions of glutathione, IBA and glutaredoxins in the control of the root development.

204 IV MATERIAL AND METHODS

205 206 IV.1 CHAPTER 1

Plant material and growth conditions

Wild-type plants and mutants used in this study were in the Arabidopsis thaliana ecotype Col-0. Mutants of the biosynthesis of glutathione: cad2-1, pad2-1, and zir1 are described on Howden et al., 1995, Parisy et al., 2006, and Shanmugam et al., 2012, respectively, while gr1 and ntrantrb mutants are described on Mhamdi et al., 2010, and Bashandy et al., 2010, respectively. Mutant cat2-1 impaired in catalase is described in Bueso et al., 2007. The ibr3, ibr10, ech2, ibr1, and triple mutant ibr1ibr3ibr10 were donated by the group of Lucia Strader (Strader et al., 2011). Mutant pin2 (eir1-1) is described in Roman et al., 1995, Mutants aux1-21, pdr8 (Salk_000578), pdr9 (Salk_050885), ugt74e2 (Salk_091130), and ugt74d1 (Salk_011286) are Salk lines. Wild- type and mutant seeds were surface-sterilized by constant agitation with 0.05% SDS made on ethanol 70% (v/v) during 20 minutes. Thereafter, seeds were washed three times with ethanol 95% (v/v) and dried at room-temperature under sterile conditions. Seeds were placed on square Petri plates containing 50 ml of Murashige and Skoog medium diluted two fold (MS ½), and solidified with plant agar with a final concentration of 0.8% (w/v), without sucrose unless indicated. For the phosphorous deprivation experiments, we used

3 MS /10 in which Pi-deficient plates contained 500µM NaCl, and control plates 500µM

NaH2PO4 as indicated in Arnaud et al., 2014. For the examination of the root hair responses in the experiments with M. loti, seeds were placed vertically on standard MS ½ medium and after 5 days, seedlings were transferred to plates containing 0.1 OD of inoculum and root hair lengths were quantified after 4 days of incubation. Plates with the inoculum were prepared according to Poitout et al., 2017. All experiments were carried out on in vitro conditions with 160 µE.m-2.s-1 light intensity (16/8) at 20 °C.

Phenotypic analyses of the root system

Lateral root density was quantified 10 days after sowing by counting the number of visible lateral roots emerged divided by the length of the region in which the LR was counted. Lateral root primordia density was calculated 6 days after sowing, in the same 207 manner using a light microscope (Axioscop2, Zeiss). The study of the lateral root primordia development was performed by synchronization of the LRP initiation, that consisted of changing the orientation of the plates 90° and observing the stage of LRP after 48 hours. Root hair lengths were quantified 8 days after sowing, excepted of M.loti experiments (see plant material and growth conditions). Primary root elongation rate was quantified by calculating the difference in the length of the root from day 8 and day 10. The analyses were made with ImageJ (https://imagej.nih.gov/ij/).

Glutathione measurements

Analysis of glutathione was performed using the enzymatic recycling methods, which is a modification of the classic Tietze recycling assay for total glutathione (Tietze, 1969; Rahman et al., 2007). The method consists in the reduction of GSSG by glutathione reductase enzyme and NADPH to GSH. Total GSH is determined by the oxidation of GSH by 5,5′-dithio-bis(2-nitrobenzoic acid) (DTNB), that produce a yellow compound 5′-thio-2- nitrobenzoic acid (TNB) which is detectable at 412 nm (Rahman et al., 2007). Extraction of glutathione was performed on 8 days seedlings. In each experiment, we used three independent biological repetitions in which each biological repetition consisted of 20 plants.

Confocal analyses

Auxin response analyses using DR5v2 and the study of the redox state of glutathione with roGFP2 were performed by using a confocal laser scanning microscope Zeiss LSM 510 Meta (Carl Zeiss). Images were acquired in 16-bits with two average readings and using 10x lens. Confocal settings for DR5v2 were based on Liao et al., 2015. Settings and analyses for roGFP2 were based on the method published by Meyer et al., 2007, with small modifications. Excitation of roGFP2 was performed at 488 and 405 nm with a bandpass (BP) of 490-555 nm. Correction of the 405 nm for its background noise was performed by removing the fluorescence of the 405 nm collecting with a BP 420-480 nm. For the analysis with DR5v2, sterilized seeds were placed vertically in square plates containing standard MS ½ medium with either BSO 0.5mM (BSO) or without BSO (Control). After 7 days of growth, seedlings were transferred to plates containing the

208 following treatments: BSO = 0.5mM, IBA= 10µM, IAA=50nM, and combination of each auxin with BSO. After incubation with those treatments for 24 hours, pictures were taken in the confocal microscope. Concerning the analyses with roGFP2, the reporter line was tested by performed a timeline experiment, that consisted in measuring the roGFP2 fluorescence during 10 minutes and then adding 1 mM of H2O2 for 30 minutes. Thereafter, reduction of the roGFP2 line with 10 mM DTT was performed in order to verify the functionality of the reporter line. For testing the effect of auxin IBA on the glutathione redox status, seeds were placed vertically for 8 days in mediums MS ½, containing IBA 10µM or without IBA (Control). Pictures were analyzed using ImageJ following the protocol published by Meyer et al., 2007 and Morgan et al., 2011.

GUS staining

DR5-GUS activity was detected in root tissue from 8-days-old seedlings. The method is based on the work published by Jefferson et al., 1987, and Malamy and Benfey, 1997, with small modifications. Plants were fixed on acetone 80% (v/v) at 20 °C for 20 minutes, were washed with a buffer solution containing 25mM Na2HPO4, 25mM

NaH2PO4, 2% Triton (v/v), 1 mM K3Fe(CN)6, and 1 mM K4Fe(CN)6. Thereafter, seedlings were transferred to a new plate containing the buffer solution with 2mM of X-Gluc (5- bromo-4-chloro-3-indolyl-β-D-glucuronidase) and vacuum infiltrated for 1 minute and then incubate at 37 °C for 1 hour. The reaction was stopped by bathing seedlings in ethanol 70% (v/v). Pictures were taken with a light microscope (Axioscop2, Zeiss) on the next day. The experiment protocol was the same as in DR5v2 experiments.

Quantitative real-time PCR (qPCR)

For the quantification of the level of expression of gene of interest, total RNA was extracted using TriZol reagent (GE Healthcare, Littler Chalfont, Buckinghamshire, UK), and the RNA was purified with RNeasy Plant Mini Kit (Promega, USA). cDNA was synthesized by the GoScript™ Reverse Transcription System (Promega, USA). Quantitative real-time PCR was done using Takyon™ No Rox SYBR® MasterMix dTTP Blue (Eurogentec, Belgium).

209 Primers for phosphorous deprivation: PHT1.4_LP 5’- CCTCGGTCGTATTTATTACCACG, PHT1.4_RP 5’- CCATCACAGCTTTTGGCTCATG, RNS1_LP 5’-TGATGCCTCTAAACCATTCGAT, RNS1_RP 5’-TACCATGCTTCTCCCATTCG, SPX1_LP 5’- CGGGTTTTGAAGGAGATCAG, SPX1_RP 5’-GCGGCAATGAAAACACACTA. For the IBA pathway: IBR1_LP 5’-ATCACCGGTAGCTCTGAAGTGAGG IBR1_RP 5’- ATGTCTCCCGTTGTTCCCAACC IBR3_LP 5’-TTCTCGCAATGGCCAAGGTTGC IBR3_RP 5’-GCTGCTCCATGAACTTGTATTGCC IBR10_LP 5’- GCTTTCGATGCCGGATTATT IBR10_RP 5’-TTCCCACTCAACAACAGCTC AIM1_LP 5’-GAAAGGCCCTCTCTTTGTTC AIM_RP 5’- AATCACTGCCTCTATGACCA PED1_LP 5’-AGGAGCGGTTCTCCTAATGA PED1_RP 5’-CCTGAATACACCAAGAACGG ECH2_LP 5’- ATGGGCAGGATTTGATCCAGGTC ECH2_RP 5’- ACTGCTGCCCATGCAACAAGAG.

Statistics analysis

All the experiments presented in this study were performed at least three independent biological repetitions. Each biological repetition from in vitro experiments consisted of two plates containing around 12 seeds for the mutant and the control in the same plate. Statistic analyses were performed with t-student test. Significant differences were considered when the p-value is ≤ to 0.01 (represented by one *) or ≤ 0.001 (represented with two **).

IV.2 CHAPTER 2

Plant material and growth conditions

The roxy19 (previously described in Huang et al., 2016) mutant carries a Ds element insertion 45bp before the Start codon and is derived from No-0 Arabidopsis thaliana ecotype. The transgenic line overexpressing ROXY19 in Col-0 ecotype (Huang et al., 2016) has been provided by Christiane Gatz. grxs17 SALK T-DNA insertion and 210 Pro35S:GRXS17-GFP in grxs17 lines were previously described in Knuesting et al., 2015. ROXY19 and GRXS17 correspond to the AGI numbers At1g28480 and At4g04950, respectively. Arabidopsis surface-sterilized seeds were placed in a 0.5X Murashige and Skoog medium without sucrose with the addition of 0.8% of plant agar, and placed vertically, without stratification, in a growth chamber (22°C with a 16-h photoperiod and a photon flux of 160 μmol.m -2 .s -1 ). ABA treatments were performed after 5 days of growth on normal MS1/2 medium, by transferring seedlings via a nylon mesh to fresh medium, containing or not the indicated concentrations of ABA.

Phenotypic analyses of root system

For the study of the root system, images of the plates were taken with a canon camera (PowerShot A620). Photographs were analyzed and quantified using the public domain image analysis program ImageJ version 1.52 (http://rsb.info.nih.gov/ij/) and the NeuronJ plugin (https://imagescience.org/meijering/software/neuronj/). Lateral root density is the result of the number of lateral roots visually identified with the help of a stereo-microscope (MZ12, Leica), divided by the length of the primary root part in which lateral roots are visible. In order to analyse LRP development, plates containing 6-day-old plants were turned 90° and plants were observed after 48 hours using a standard optical microscope (Axioscop2, Zeiss).

Isolation of suppressor mutants of the grxs17 phenotype

A population of 50,000 grxs17 seeds (M0) was chemically mutagenized using 0.3% ethyl methanesulfonate according to standard procedures (Lightner and Caspar, 1998). The progeny of 40,000 mutagenized seeds (M1) was harvested in 384 pools, and 400 seeds from each pool were used to screen for mutant M2 plants displaying reversion of PR and LR grxs17 phenotypes. 90 putative candidates were grown and propagated, and the M3 progeny was screened for the same phenotypes in order to validate the reverted phenotype. We selected the 12 best candidates that are presented in this manuscript.

Immunodetection of GRXS17

Total protein extraction was performed by grinding tissues in liquid nitrogen and 211 resuspension in extraction buffer containing 25mM of Tris-HCl pH 7.6, 75 mM NaCl, 10mM DTT, 1mM phenylmethylsulfonyl fluoride (PMSF) and 0.15% NP-40. After measurements of the protein concentration by Bradford method (1976), 10 μg proteins were separated by SDS/PAGE and transferred to nitrocellulose membranes. GRXS17 were detected by western blot using rabbit polyclonal antibodies against GRXS17 diluted to 1:25 000. Goat anti-rabbit antibodies conjugated to horseradish peroxidase (1:10 000) were used as secondary antibodies and revealed with enhanced chemiluminescence reagents (Immobilon Western Chemiluminescent HRP Sbstrate, Millipore).

IV.3 CHAPTER 3

Whole-genome sequencing

DNA extraction from grxs17 suppressor mutants in both Col-0 and Ler background were performed using Qiagen DNeasy® Plant Mini kit. Samples were sequenced by Illumina platform (HudsonAlpha, USA) by the paired-ends method which produced reads with a length of 102 bases for outcross samples, while for backcross the sequencing results in reads with a length of 152 bases. The outputs of both classes of samples were produced on the Fastq format.

Bioinformatic analysis

Reads from Illumina sequencing were revised for quality with fastqc command line, and then, they were assembling using Bowtie 2, with the sensitive-local configuration option, and the reference genome TAIR10 (Table 4.1) (Langmead and Salzberg, 2012). Samtools mpileup was performed in order to identify SNPs with the input options - A, and -B, which include all reads in the sorted .bam file for the identification of the SNPs (Li et al., 2009). Bcftools was used to convert the output file of samtools mpileup to .vcf format, and snpEff pipeline was used for the annotation and classification of the SNPs detected (Cingolani et al., 2012). Further calculation data such as allele frequency, comparison with parent lines, filtration of the EMS mutation, and charts were performed 212 with R version 3.5.1, and vcftools (https://www.r-project.org/). For the analysis with the NGM pipeline, the results from the samtools mpileup were converted to .emap files, and analyzed following the instructions of the web site (http://bar.utoronto.ca/ngm/). Identification of the causal mutation with the SHOREmap package, was performed following the instructions of the authors (Schneeberger et al., 2009b; Sun and Schneeberger, 2015).

213 214 V REFERENCES

215 This is a complementary bibliography list

Acharya BR, Raina S, Maqbool SB, Jagadeeswaran G, Mosher SL, Appel HM, Schultz JC, Klessig DF, Raina R. 2007. Overexpression of CRK13, an Arabidopsis cysteine-rich receptor-like kinase, results in enhanced resistance to Pseudomonas syringae. Plant Journal 50, 488–499. Addo-Quaye C, Buescher E, Best N, Chaikam V, Baxter I, Dilkes BP. 2017. Forward Genetics by Sequencing EMS Variation-Induced Inbred Lines. Genes, Genomes, Genetics 7, 413–425. Ahemad M, Kibret M. 2014. Mechanisms and applications of plant growth promoting rhizobacteria: Current perspective. Journal of King Saud University - Science 26, 1–20. Allen RS, Nakasugi K, Doran RL, Millar AA, Waterhouse PM. 2013. Facile mutant identification via a single parental backcross method and application of whole genome sequencing based mapping pipelines. Frontiers in Plant Science 4. Anderson CT, Carroll A, Akhmetova L, Somerville C. 2010. Real-Time Imaging of Cellulose Reorientation during Cell Wall Expansion in Arabidopsis Roots. PLANT PHYSIOLOGY 152, 787–796. Apanasets O, Grou CP, Van Veldhoven PP, Brees C, Wang B, Nordgren M, Dodt G, Azevedo JE, Fransen M. 2014. PEX5, the Shuttling Import Receptor for Peroxisomal Matrix Proteins, Is a Redox-Sensitive Protein. Traffic 15, 94–103. Aroca A, Gotor C, Romero LC. 2018. Hydrogen Sulfide Signaling in Plants: Emerging Roles of Protein Persulfidation. Frontiers in Plant Science 9. Arondel V, Lemieux B, Hwang I, Gibson S, Goodman HM, Somerville CR. 1992. Map-based cloning of a gene controlling omega-3 fatty acid desaturation in Arabidopsis. Science 258, 1353–1355. Asada K. 1999. THE WATER-WATER CYCLE IN CHLOROPLASTS: Scavenging of Active Oxygens and Dissipation of Excess Photons. Annual Review of Plant Physiology and Plant Molecular Biology 50, 601–639. Austin RS, Vidaurre D, Stamatiou G, et al. 2011. Next-generation mapping of Arabidopsis genes. Plant Journal 67, 715–725. Baars C. 2017. Review of plant evolution and its effect on climate during the time of the Old Red Sandstone. Proceedings of the Geologists’ Association 128, 431–437. Bachhawat AK, Thakur A, Kaur J, Zulkifli M. 2013. Glutathione transporters. Biochimica et Biophysica Acta 1830, 3154–3164. 216 Balcerowicz D, Schoenaers S, Vissenberg K. 2015. Cell Fate Determination and the Switch from Diffuse Growth to Planar Polarity in Arabidopsis Root Epidermal Cells. Frontiers in Plant Science 6, 1–13. Ball L, Accotto G-P, Bechtold U, et al. 2004. Evidence for a Direct Link between Glutathione Biosynthesis and Stress Defense Gene Expression in Arabidopsis. THE PLANT CELL ONLINE 16, 2448–2462. Barbez E, Kubeš M, Rolčík J, et al. 2012. A novel putative auxin carrier family regulates intracellular auxin homeostasis in plants. Nature 485, 119–122. Bargmann BOR, Vanneste S, Krouk G, et al. 2013. A map of cell type-specific auxin responses. Molecular Systems Biology 9, 1–13. Barkawi LS, Tam Y-Y, Tillman JA, Pederson B, Calio J, Al-Amier H, Emerick M, Normanly J, Cohen JD. 2007. A high-throughput method for the quantitative analysis of indole-3-acetic acid and other auxins from plant tissue. Analytical Biochemistry 372, 177–188. Barrada A, Montané MH, Robaglia C, Menand B. 2015. Spatial regulation of root growth: Placing the plant TOR pathway in a developmental perspective. International Journal of Molecular Sciences 16, 19671–19697. Bartoli CG, Buet A, Grozeff GG, Galatro A, Simontacchi M. 2018. Ascorbate- Glutathione Cycle and Abiotic Stress Tolerance in Plants Carlos. Ascorbic Acid in Plant Growth, Development and Stress Tolerance. Springer International Publishing, 177–200. Bashandy T, Guilleminot J, Vernoux T, Caparros-Ruiz D, Ljung K, Meyer Y, Reichheld J-P. 2010. Interplay between the NADP-Linked Thioredoxin and Glutathione Systems in Arabidopsis Auxin Signaling. The Plant Cell 22, 376–391. Bates TR, Lynch JP. 1996. Stimulation of root hair elongation in Arabidopsis thaliana by low phosphorus availability. Plant, Cell and Environment 19, 529–538. Batt S, Wilkins MB. 1976. Auxin Binding to Corn Coleoptile Membranes: Kinetics and Specificity. Planta 130, 7–13. Bela K, Horváth E, Gallé Á, Szabados L, Tari I, Csiszár J. 2015. Plant glutathione peroxidases: Emerging role of the antioxidant enzymes in plant development and stress responses. Journal of Plant Physiology 176, 192–201. Belin C, Bashandy T, Cela J, Delorme-Hinoux V, Riondet C, Reichheld JP. 2015. A comprehensive study of thiol reduction gene expression under stress conditions in Arabidopsis thaliana. Plant, Cell and Environment 38, 299–314. Bellini C, Pacurar DI, Perrone I. 2014. Adventitious Roots and Lateral Roots: 217 Similarities and Differences. Annual Review of Plant Biology 65, 639–666. Belousov V V., Fradkov AF, Lukyanov KA, Staroverov DB, Shakhbazov KS, Terskikh A V., Lukyanov S. 2006. Genetically encoded fluorescent indicator for intracellular hydrogen peroxide. Nature Methods 3, 281–286. Bender J, Celenza JL. 2009. Indolic glucosinolates at the crossroads of tryptophan metabolism. Phytochemistry Reviews 8, 25–37. Benková E, Bielach A. 2010. Lateral root organogenesis — from cell to organ. Nervenheilkunde 13, 677–683. Bennett MJ, Marchant A, Green HG, May ST, Ward SP, Millner PA, Walker AR, Schulz B, Feldmann KA. 1996. Arabidopsis AUX1 Gene: A Permease-Like Regulator of Root Gravitropism. Berendsen RL, Pieterse CMJ, Bakker PAHM. 2012. The rhizosphere microbiome and plant health. Trends in Plant Science 17, 478–486. Van Den Berg C, Willemsen V, Hendriks G, Weisbeek P, Scheres B. 1997. Short- range control of cell differentiation in the Arabidopsis root meristem. Nature 390, 287–289. Bernhardt C, Lee MM, Gonzalez A, Zhang F, Lloyd A, J. S. 2003. The bHLH genes GLABRA3 (GL3) and ENHANCER OF GLABRA3 (EGL3) specify epidermal cell fate in the Arabidopsis root. Development 130, 6431–6439. Bhosale R, Giri J, Pandey BK, et al. 2018. A mechanistic framework for auxin dependent Arabidopsis root hair elongation to low external phosphate. Nature Communications 9. Bibikova TN, Jacob T, Dahse I, Gilroy S. 1998. Localized changes in apoplastic and cytoplasmic pH are associated with root hair development in Arabidopsis thaliana. 125, 2925–2934. Binder BM. 2007. Rapid kinetic analysis of ethylene growth responses in seedlings: New insights into ethylene signal transduction. Journal of Plant Growth Regulation 26, 131–142. Blancaflor EB, Fasano JM, Gilroy S. 1998. Mapping the Functional Roles of Cap Cells in the Response of Arabidopsis Primary Roots to Gravity 1. Plant physiology 116, 213–222. Blilou I, Xu J, Wildwater M, Willemsen V, Paponov I, Friml J, Heidstra R, Aida M, Palme K, Scheres B. 2005. The PIN auxin efflux facilitator network controls growth and patterning in Arabidopsis roots. Nature 433, 39–44. Boer DR, Freire-Rios A, Van Den Berg WAM, et al. 2014. Structural basis for DNA 218 binding specificity by the auxin-dependent ARF transcription factors. Cell 156, 577–589. Bowman JL, Briginshaw LN, Florent SN. 2019. Evolution and co-option of developmental regulatory networks in early land plants. Current Topics in Developmental Biology. Elsevier Inc., 35–53. Braun N, Wyrzykowska J, Muller P, David K, Couch D, Perrot-Rechenmann C, Fleming AJ. 2008. Conditional Repression of AUXIN BINDING PROTEIN1 Reveals That It Coordinates Cell Division and Cell Expansion during Postembryonic Shoot Development in Arabidopsis and Tobacco. THE PLANT CELL ONLINE 20, 2746–2762. Bruex A, Kainkaryam RM, Wieckowski Y, et al. 2012. A gene regulatory network for root epidermis cell differentiation in Arabidopsis. PLoS Genetics 8, 1–20. Brumos J, Robles LM, Yun J, Vu TC, Jackson S, Alonso JM, Stepanova AN. 2018. Local Auxin Biosynthesis Is a Key Regulator of Plant Development. Developmental Cell 47, 306–318. Brunoud G, Wells DM, Oliva M, et al. 2012. A novel sensor to map auxin response and distribution at high spatio-temporal resolution. Nature 482, 103–108. Bueso E, Alejandro S, Carbonell P, Perez-Amador MA, Fayos J, Bellés JM, Rodriguez PL, Serrano R. 2007. The lithium tolerance of the Arabidopsis cat2 mutant reveals a cross-talk between oxidative stress and ethylene. Plant Journal 52, 1052–1065. Cagnac O, Bourbouloux A, Chakrabarty D, Zhang M-Y, Delrot S. 2004. AtOPT6 Transports Glutathione Derivatives and Is Induced by Primisulfuron. PLANT PHYSIOLOGY 135, 1378–1387. Cairns NG, Pasternak M, Wachter A, Cobbett CS, Meyer AJ. 2006. Maturation of Arabidopsis Seeds Is Dependent on Glutathione Biosynthesis within the Embryo. PLANT PHYSIOLOGY 141, 446–455. Calderón Villalobos LIA, Lee S, De Oliveira C, et al. 2012. A combinatorial TIR1/AFB–Aux/IAA co-receptor system for differential sensing of auxin. Nature Chemical Biology 8, 477–485. Campanella JJ, Smith SM, Leibu D, Wexler S, Ludwig-Müller J. 2007. The Auxin Conjugate Hydrolase Family of Medicago truncatula and Their Expression During the Interaction with Two Symbionts. Journal of Plant Growth Regulation 27, 26– 38. Candela H, Casanova-Sáez R, Micol JL. 2015. Getting started in mapping-by- sequencing. Journal of Integrative Plant Biology 57, 606–612. 219 Carol RJ, Dolan L. 2002. Building a hair: Tip growth in Arabidopsis thaliana root hairs. Philosophical Transactions of the Royal Society B: Biological Sciences 357, 815–821. Carol RJ, Dolan L. 2006. The role of reactive oxygen species in cell growth: Lessons from root hairs. Journal of Experimental Botany.1829–1834. Casimiro I, Beeckman T, Graham N, Bhalerao R, Zhang H, Casero P, Sandberg G, Bennett MJ. 2003. Dissecting Arabidopsis lateral root development. Trends in Plant Science 8, 165–171. Cha JY, Kim WY, Kang S Bin, et al. 2015. A novel thiol-reductase activity of Arabidopsis YUC6 confers drought tolerance independently of auxin biosynthesis. Nature Communications 6. Chandrika NNP, Sundaravelpandian K, Yu SM, Schmidt W. 2013. ALFIN-LIKE 6 is involved in root hair elongation during phosphate deficiency in Arabidopsis. New Phytologist 198, 709–720. Charbonnel C, Niazi AK, Elvira-Matelot E, et al. 2017. The siRNA suppressor RTL1 is redox-regulated through glutathionylation of a conserved cysteine in the double-stranded-RNA-binding domain. Nucleic Acids Research 45, 11891– 11907. Chen Z, Gallie DR. 2004. The Ascorbic Acid Redox State Controls Guard Cell Signaling and Stomatal Movement. The Plant Cell 16, 1143–1162. Chen J-H, Jiang H-W, Hsieh E-J, Chen H-Y, Chien C-T, Hsieh H-L, Lin T-P. 2012. Drought and Salt Stress Tolerance of an Arabidopsis Glutathione S- Transferase U17 Knockout Mutant Are Attributed to the Combined Effect of Glutathione and Abscisic Acid. PLANT PHYSIOLOGY 158, 340–351. Cheng JC, Seeley KA, Sung ZR. 1995. RML1 and RML2, Arabidopsis Genes Required for Cell Proliferation at the Root Tip. Plant Physiology 107, 365–376. Chuberre C, Plancot B, Driouich A, Moore JP, Bardor M, Gügi B, Vicré M. 2018. Plant Immunity Is Compartmentalized and Specialized in Roots. Frontiers in Plant Science 9, 1–13. Cobbett CS, May MJ, Howden R, Rolls B. 1998. The glutathione-deficient, cadmium- sensitive mutant, cad2-1, of Arabidopsis thaliana is deficient in γ- glutamylcysteine synthetase. Cosio C, Dunand C. 2009. Specific functions of individual class III peroxidase genes. Journal of Experimental Botany 60, 391–408. Couturier J, Wu HC, Dhalleine T, Pégeot H, Sudre D, Gualberto JM, Jacquot JP, 220 Gaymard F, Vignols F, Rouhier N. 2014. Monothiol Glutaredoxin-BolA interactions: Redox control of Arabidopsis thaliana BolA2 and SufE1. Molecular Plant 7, 187–205. Czarnocka W, Karpiński S. 2018. Friend or foe? Reactive oxygen species production, scavenging and signaling in plant response to environmental stresses. Free Radical Biology and Medicine 122, 4–20. Danquah A, de Zelicourt A, Colcombet J, Hirt H. 2014. The role of ABA and MAPK signaling pathways in plant abiotic stress responses. Biotechnology Advances 32, 40–52. Darwin C, Darwin F. 1881. THE POWER OF MOVEMENT IN PLANTS. New York: Appleton and Company. Das K, Roychoudhury A. 2014. Reactive oxygen species (ROS) and response of antioxidants as ROS-scavengers during environmental stress in plants. Frontiers in Environmental Science 2. Datta S, Prescott H, Dolan L. 2015. Intensity of a pulse of RSL4 transcription factor synthesis determines Arabidopsis root hair cell size. Nature Plants 1, 1–6. Delorme-Hinoux V, Bangash SAK, Meyer AJ, Reichheld JP. 2016. Nuclear thiol redox systems in plants. Plant Science 243, 84–95. Diaz-Vivancos P, De Simone A, Kiddle G, Foyer CH. 2015. Glutathione - Linking cell proliferation to oxidative stress. Free Radical Biology and Medicine 89, 1154–1164. Dixon DP, Hawkins T, Hussey PJ, Edwards R. 2009. Enzyme activities and subcellular localization of members of the Arabidopsis glutathione transferase superfamily. Journal of Experimental Botany 60, 1207–1218. Dixon DP, Skypsey M, Grundy NM, Edwards R. 2005. Stress-Induced Protein S- Glutathionylation in Arabidopsis. Plant physiology 138, 2233–2244. Dolan L, Janmaat K, Willemsen V, Linstead P, Poethig S, Roberts K, Scheres B. 1993. Cellular organisation of the Arabidopsis thaliana root. Development 119, 71–84. Doornbos RF, Van Loon LC, Bakker PAHM. 2012. Impact of root exudates and plant defense signaling on bacterial communities in the rhizosphere. A review. Agronomy for Sustainable Development 32, 227–243. Duan L, Dietrich D, Ng CH, Chan PMY, Bhalerao R, Bennett MJ, Dinneny JR. 2013. Endodermal ABA Signaling Promotes Lateral Root Quiescence during Salt Stress in Arabidopsis Seedlings. The Plant Cell 25, 324–341. 221 Dubrovsky JG, Doerner PW, Colón-Carmona A, Rost TL. 2000. Pericycle Cell Proliferation and Lateral Root Initiation in Arabidopsis. Plant Physiology 124, 1648–1657. Dunand C, Crèvecoeur M, Penel C. 2007. Distribution of superoxide and hydrogen peroxide in Arabidopsis root and their influence on root development: Possible interaction with peroxidases. New Phytologist 174, 332–341. Edwards D. 2004. Embryophytic sporophytes in the Rhynie and Windyfield cherts. Transactions of the Royal Society of Edinburgh: Earth Sciences 94, 397–410. Enders TA, Strader LC. 2015. Auxin activity: Past, present, and future. American Journal of Botany 102, 180–196. Epstein E, Ludwig-Müller J. 1993. Indole‐3‐butyric acid in plants: occurrence, synthesis, metabolism and transport. Physiologia Plantarum 88, 382–389. Esch JJ, Chen M, Sanders M, Hillestad M, Ndkium S, Idelkope B, Neizer J, Marks MD. 2003. A contradictory GLABRA3 allele helps define gene interactions controlling trichome development in Arabidopsis. Development 130, 5885–5894. Fendrych M, Akhmanova M, Merrin J, Glanc M, Hagihara S, Takahashi K, Uchida N, Torii KU, Friml J. 2018. Rapid and reversible root growth inhibition by TIR1 auxin signalling. Nature Plants 4, 453–459. Feng M, Kim J-Y. 2015. Revisiting Apoplastic Auxin Signaling Mediated by AUXIN BINDING PROTEIN 1. Molecules and Cells 38, 829–835. Fernandez-Marcos M, Sanz L, Lewis DR, Muday GK, Lorenzo O. 2011. Nitric oxide causes root apical meristem defects and growth inhibition while reducing PIN-FORMED 1 (PIN1)-dependent acropetal auxin transport. Proceedings of the National Academy of Sciences 108, 18506–18511. Finkelstein RR, Gampala SSL, Rock CD. 2002. Abscisic Acid Signaling in Seeds and Seedlings. The Plant Cell S15–S45. Foreman J, Demidchik V, Bothwell JHF, et al. 2003. Reactive oxygen species produced by NADPH oxidase regulate plant cell growth. Nature 422, 422–426. Foyer CH, Halliwell B. 1976. The Presence of Glutathione and Glutathione Reductase in Chloroplasts: A Proposed Role in Ascorbic Acid Metabolism. Planta 133, 21– 25. Foyer CH, Noctor G. 2011. Ascorbate and Glutathione: The Heart of the Redox Hub. PLANT PHYSIOLOGY 155, 2–18. Frendo P, Gallesi D, Turnbull R, Van De Sype G, Hérouart D, Puppo A. 1999. Localisation of glutathione and homoglutathione in Medicago truncatula is 222 correlated to a differential expression of genes involved in their synthesis. Plant Journal 17, 215–219. Frick EM, Strader LC. 2018. Roles for IBA-derived auxin in plant development. Journal of Experimental Botany 69, 169–177. Friml J. 2003. Auxin transport - shaping the plant. Current Opinion in Plant Biology 6, 7–12. Friml J, Wisniewska J, Benková E, Mendgen K, Palme K. 2002. Lateral relocation of auxin efflux regulator PIN3 mediates tropism in Arabidopsis. Nature 415, 806– 809. Fukaki H, Tameda S, Masuda H, Tasaka M. 2002. Lateral root formation is blocked by a gain-of-function mutation in the SOLITARY-ROOT/IAA14 gene of Arabidopsis. Plant Journal 29, 153–168. Fukushima A, Iwasa M, Nakabayashi R, Kobayashi M, Nishizawa T, Okazaki Y, Saito K, Kusano M. 2017. Effects of Combined Low Glutathione with Mild Oxidative and Low Phosphorus Stress on the Metabolism of Arabidopsis thaliana. Frontiers in Plant Science 8. Galway ME, Masucci JD, Lloyd AM, Walbot V, Davis RW, Shiefelbein JW. 1994. The TTG Gene Is Required to Specify Epidermal Cell Fate and Cell Patterning in the Arabidopsis Root. DEVELOPMENTAL BIOLOGY 166, 740–754. Gälweiler L, Changhui G, Andreas M, Ellen W, Kurt M, Yephremov A, Palme K. 1998. Regulation of Polar Auxin Transport by AtPIN1 in Arabidopsis Vascular Tissue. Science 282, 2226–2230. Gao Y, Zhang Y, Zhang D, Dai X, Estelle M, Zhao Y. 2015. Auxin binding protein 1 (ABP1) is not required for either auxin signaling or Arabidopsis development. Proceedings of the National Academy of Sciences 112, 2275–2280. Gelhaye E, Rouhier N, Jacquot JP. 2003. Evidence for a subgroup of thioredoxin h that requires GSH/Grx for its reduction. FEBS Letters 555, 443–448. Geng Y, Wu R, Wee CW, Xie F, Wei X, Chan PMY, Tham C, Duan L, Dinneny JR. 2013. A Spatio-Temporal Understanding of Growth Regulation during the Salt Stress Response in Arabidopsis. The Plant Cell 25, 2132–2154. Gill SS, Tuteja N. 2010. Reactive oxygen species and antioxidant machinery in abiotic stress tolerance in crop plants. Plant Physiology and Biochemistry 48, 909–930. Giraudat J, Hauge BM, Valon C, Smalle J, Parcy F, Goodman HM. 1992. Isolation of the Arabidopsis ABI3 Gene by Positional Cloning. The Plant Cell 4, 1251– 1261. 223 Glazebrook J, Ausubel FM. 1994. Isolation of phytoalexin-deficient mutants of Arabidopsis thaliana and characterization of their interactions with bacterial pathogens. Proceedings of the National Academy of Sciences 91, 8955–8959. Glick BR. 2012. Plant Growth-Promoting Bacteria: Mechanisms and Applications. Scientifica 2012, 1–15. Greene EA, Codomo CA, Taylor NE, et al. 2003. Spectrum of Chemically Induced Mutations From a Large-Scale Reverse-Genetic Screen in Arabidopsis. Genetics 164, 731–740. Grieneisen VA, Xu J, Marée AFM, Hogeweg P, Scheres B. 2007. Auxin transport is sufficient to generate a maximum and gradient guiding root growth. Nature 449, 1008–1013. Guilfoyle TJ, Hagen G. 2007. Auxin response factors. Plant Cell and Environment 10, 453–460. Gupta AS, Webb RP, Holaday AS, Allen RD. 1993. Overexpression of Superoxide Dismutase Protects Plants from Oxidative Stress. Plant Physiology 103, 1067– 1073. Haagen-Smit AJ, Leech WC, Bergen WR. 1941. ESTIMATION, ISOLATION AND IDENTIFICATION OF AUXINS IN PLANT MATERIAL. Science 93, 624–625. Haas D, Défago G. 2005. Biological Control of Soil-Borne Pathogens by Fluorescent Pseudomonads. Nature Reviews Microbiology 3, 307–319. Ham BK, Chen J, Yan Y, Lucas WJ. 2018. Insights into plant phosphate sensing and signaling. Current Opinion in Biotechnology 49, 1–9. Hao S, Xue J, Guo D, Wang D. 2010. Earliest rooting system and root : SShoot ratio from a new Zosterophyllum plant. New Phytologist 185, 217–225. Hasanuzzaman M, Nahar K, Anee TI, Fujita M. 2017. Glutathione in plants: biosynthesis and physiological role in environmental stress tolerance. Physiology and Molecular Biology of Plants 23, 249–268. Hayashi K -i., Nakamura S, Fukunaga S, Nishimura T, Jenness MK, Murphy AS, Motose H, Nozaki H, Furutani M, Aoyama T. 2014. Auxin transport sites are visualized in planta using fluorescent auxin analogs. Proceedings of the National Academy of Sciences 111, 11557–11562. Herrera-Vásquez A, Carvallo L, Blanco F, Tobar M, Villarroel-Candia E, Vicente- Carbajosa J, Salinas P, Holuigue L. 2015. Transcriptional Control of Glutaredoxin GRXC9 Expression by a Salicylic Acid-Dependent and NPR1- Independent Pathway in Arabidopsis. Plant Molecular Biology Reporter 33, 624– 224 637. Hetherington AJ, Dolan L. 2017. Bilaterally symmetric axes with rhizoids composed the rooting structure of the common ancestor of vascular plants. Philosophical Transactions of the Royal Society B: Biological Sciences 373. Himanen K, Boucheron E, Vanneste S, Engler J de A, Inzé D, Beeckman T. 2002. Auxin-Mediated Cell Cycle Activation during Early Lateral Root Initiation. THE PLANT CELL ONLINE 14, 2339–2351. Honkanen S, Jones VAS, Morieri G, Champion C, Hetherington AJ, Kelly S, Proust H, Saint-Marcoux D, Prescott H, Dolan L. 2017. The Mechanism Forming the Cell Surface of Tip-Growing Rooting Cells Is Conserved among Land Plants. Current Biology 27, 3238–3244. Howden R, Andersen CR, Coldsbrough PB, Cobbett CS. 1995a. A Cadmium- Sensitive, Glutathione-Deficient Mutant of Arabidopsis thaliana. Plant Physiol 1, 1067–1073. Howden R, Coldsbrough PB, Andersen CR, Cobbett CS. 1995b. Cadmium- Sensitive, cad1 Mutants of Arabidopsis thaliana Are Phytochelatin Deficient. PLANT PHYSIOLOGY 107, 1059–1060. Huang LJ, Li N, Thurow C, Wirtz M, Hell R, Gatz C. 2016. Ectopically expressed glutaredoxin ROXY19 negatively regulates the detoxification pathway in Arabidopsis thaliana. BMC Plant Biology 16. Hwang JU, Song WY, Hong D, et al. 2016. Plant ABC Transporters Enable Many Unique Aspects of a Terrestrial Plant’s Lifestyle. Molecular Plant 9, 338–355. Idänheimo N, Gauthier A, Salojärvi J, Siligato R, Brosché M, Kollist H, Mähönen AP, Kangasjärvi J, Wrzaczek M. 2014. The Arabidopsis thaliana cysteine-rich receptor-like kinases CRK6 and CRK7 protect against apoplastic oxidative stress. Biochemical and Biophysical Research Communications 445, 457–462. Iqbal A, Yabuta Y, Takeda T, Nakano Y, Shigeoka S. 2006. Hydroperoxide reduction by thioredoxin-specific glutathione peroxidase isoenzymes of Arabidopsis thaliana. FEBS Journal 273, 5589–5597. Jackson RG, Kowalczyk M, Li Y, Higgins G, Ross J, Sandberg G, Bowles DJ. 2002a. Over-expression of an Arabidopsis gene encoding a glucosyltransferase of indole-3-acetic acid: phenotypic characterisation of transgenic lines. The Plant Journal 32, 573–583. Jackson RG, Lim E-K, Li Y, Kowalczyk M, Sandberg G, Hoggett J, Ashford DA, Bowles DJ. 2002b. Identification and Biochemical Characterization of an Arabidopsis Indole-3-acetic Acid Glucosyltransferase. Journal of Biological 225 Chemistry 276, 4350–4356. James GV, Patel V, Nordström KJV, Klasen JR, Salomé PA, Weigel D, Schneeberger K. 2013. User guide for mapping-by-sequencing in Arabidopsis. Genome Biology 14, 1–13. Jander G, Norris SR, Rounsley SD, Bush DF, Levin IM, Last RL. 2002. Arabidopsis Map-Based Cloning in the Post-Genome Era. PLANT PHYSIOLOGY 129, 440–450. Jiang K, Meng YL, Feldman LJ. 2003. Quiescent center formation in maize roots is associated with an auxin-regulated oxidizing environment. Development 130, 1429–1438. Jing H, Yang X, Zhang J, et al. 2015. Peptidyl-prolyl isomerization targets rice Aux/IAAs for proteasomal degradation during auxin signalling. Nature Communications 6. Jones AR, Kramer EM, Knox K, Swarup R, Bennett MJ, Lazarus CM, Leyser HMO, Grierson CS. 2009. Auxin transport through non-hair cells sustains root- hair development. Nature Cell Biology 11, 78–84. Jones MA, Shen J-J, Fu Y, Li H, Yang Z, Grierson CS. 2002. The Arabidopsis Rop2 GTPase Is a Positive Regulator of Both Root Hair Initiation and Tip Growth. THE PLANT CELL ONLINE 14, 763–776. Jung JKH, McCouch S. 2013. Getting to the roots of it: Genetic and hormonal control of root architecture. Frontiers in Plant Science 4, 1–32. Jurado S, Abraham Z, Manzano C, López-Torrejón G, Pacios LF, Del Pozo JC. 2010. The Arabidopsis Cell Cycle F-Box Protein SKP2A Binds to Auxin. The Plant Cell 22, 3891–3904. Kamimoto Y, Terasaka K, Hamamoto M, et al. 2012. Arabidopsis ABCB21 is a Facultative Auxin Importer/Exporter Regulated by Cytoplasmic Auxin Concentration. Plant and Cell Physiology 53, 2090–2100. Kant S, Bi YM, Rothstein SJ. 2011. Understanding plant response to nitrogen limitation for the improvement of crop nitrogen use efficiency. Journal of Experimental Botany 62, 1499–1509. Karol KG, McCourt RM, Cimino MT, Delwiche CF. 2001. The closest living relatives of land plants. Science 294, 2351–2353. Katano M, Takahashi K, Hirano T, Kazama Y, Abe T, Tsukaya H, Ferjani A. 2016. Suppressor Screen and Phenotype Analyses Revealed an Emerging Role of the Monofunctional Peroxisomal Enoyl-CoA Hydratase 2 in Compensated Cell 226 Enlargement. Frontiers in Plant Science 7, 1–11. Keeling PJ. 2010. The endosymbiotic origin, diversification and fate of plastids. Philosophical Transactions of the Royal Society B: Biological Sciences 365, 729– 748. Kenrick P, Strullu-Derrien C. 2014. The Origin and Early Evolution of Roots. PLANT PHYSIOLOGY 166, 570–580. Kerchev P, Mühlenbock P, Denecker J, Morreel K, Hoeberichts FA, Van Der Kelen K, Vandorpe M, Nguyen L, Audenaert D, Van Breusegem F. 2015. Activation of auxin signalling counteracts photorespiratory H2O2-dependent cell death. Plant, Cell and Environment 38, 253–265. Kerchev PI, Pellny TK, Vivancos PD, Kiddle G, Hedden P, Driscoll S, Vanacker H, Verrier P, Hancock RD, Foyer CH. 2011. The Transcription Factor ABI4 Is Required for the Ascorbic Acid–Dependent Regulation of Growth and Regulation of Jasmonate-Dependent Defense Signaling Pathways in Arabidopsis . The Plant Cell 23, 3319–3334. Kim T-H, Bohmer M, Hu H, Nishimura N, Schroeder JI. 2016. Guard Cell Signal Transduction Network: Advances in Understanding Abscisic Acid, CO2 , and Ca2+ Signaling. Annual Review of Plant Biology 61, 561–591. Kim J-Y, Henrichs S, Bailly A, Vincenzetti V, Sovero V, Mancuso S, Pollmann S, Kim D, Geisler M, Nam H-G. 2010. Identification of an ABCB/P-glycoprotein- specific Inhibitor of Auxin Transport by Chemical Genomics. Journal of Biological Chemistry 285, 23309–23317. Kim Y, Schumaker KS, Zhu J-K. 2006. EMS Mutagenesis of Arabidopsis Seed. Arabidopsis Protocols. Humana Press, 101–103. Kirik V, Simon M, Huelskamp M, Schiefelbein J. 2004. The ENHANCER of TRY and CPC1 gene acts redundantly with TRIPTYCHON and CAPRICE in trichome and root hair cell patterning in Arabidopsis. Developmental Biology 268, 506– 513. Kleine-Vehn J, Dhonukshe P, Sauer M, Brewer PB, Wiśniewska J, Paciorek T, Benková E, Friml J. 2008. ARF GEF-Dependent Transcytosis and Polar Delivery of PIN Auxin Carriers in Arabidopsis. Current Biology 18, 526–531. Knuesting J, Riondet C, Maria C, et al. 2015. Arabidopsis Glutaredoxin S17 and Its Partner, the Nuclear Factor Y Subunit C11/Negative Cofactor 2α, Contribute to Maintenance of the Shoot Apical Meristem under Long-Day Photoperiod. Plant Physiology 167, 1643–1658. Kögl F, Kostermaas DGFR. 1934. Hetero-auxin als Stoffwechselprodukt niederer 227 pflanzlicher Organismen. Isolierung aus Hefe. Hoppe-Seyler’s Zeitschrift fur Physiologische Chemie. 228, 113–121. Kong X, Liu G, Liu J, Ding Z. 2018. The Root Transition Zone: A Hot Spot for Signal Crosstalk. Trends in Plant Science 23, 403–409. Van Der Kop DAM, Schuyer M, Scheres B, Van Der Zaal BJ, Hooykaas PJJ. 1996. Isolation and characterization of an auxin-inducible glutathione S-transferase gene of Arabidopsis thaliana. Plant Molecular Biology 30, 839–844. Koprivova A, Mugford ST, Kopriva S. 2010. Arabidopsis root growth dependence on glutathione is linked to auxin transport. Plant Cell Reports 29, 1157–1167. Krouk G, Lacombe B, Bielach A, et al. 2010. Nitrate-Regulated Auxin Transport by NRT1.1 Defines a Mechanism for Nutrient Sensing in Plants. Developmental Cell 18, 927–937. Kubeš M, Yang H, Richter GL, et al. 2012. The Arabidopsis concentration-dependent influx/efflux transporter ABCB4 regulates cellular auxin levels in the root epidermis. Plant Journal 69, 640–654. Kuromori T, Seo M, Shinozaki K. 2018. ABA Transport and Plant Water Stress Responses. Trends in Plant Science 23, 513–522. Kuźniak E, Kopczewski T, Chojak-Koźniewska J. 2018. Ascorbate-Glutathione Cycle and Biotic Stress Tolerance in Plants. Ascorbic Acid in Plant Growth, Development and Stress Tolerance. Springer International Publishing, 201–231. Kwak SH, Schiefelbein J. 2007. The role of the SCRAMBLED receptor-like kinase in patterning the Arabidopsis root epidermis. Developmental Biology 302, 118–131. Kwak SH, Schiefelbein J. 2008. A Feedback Mechanism Controlling SCRAMBLED Receptor Accumulation and Cell-Type Pattern in Arabidopsis. Current Biology 18, 1949–1954. Kwak S-H, Woo S, Lee MM, Schiefelbein J. 2014. Distinct Signaling Mechanisms in Multiple Developmental Pathways by the SCRAMBLED Receptor of Arabidopsis. PLANT PHYSIOLOGY 166, 976–987. Labrou NE, Papageorgiou AC, Pavli O, Flemetakis E. 2015. Plant GSTome: Structure and functional role in xenome network and plant stress response. Current Opinion in Biotechnology 32, 186–194. Lallement PA, Brouwer B, Keech O, Hecker A, Rouhier N. 2014. The still mysterious roles of cysteine-containing glutathione transferases in plants. Frontiers in Pharmacology 5, 1–22. Laloi C, Havaux M. 2015. Key players of singlet oxygen-induced cell death in plants. 228 Frontiers in Plant Science 6, 1–9. Lan P, Li W, Schmidt W. 2015. ‘Omics’ Approaches Towards Understanding Plant Phosphorus Acquisition and Use. Phosphorus Metabolism in Plants. John Wiley & Sons, Ltd, 65–97. LeClere S, Tellez R, Rampey RA, Matsuda SPT, Bartel B. 2002. Characterization of a Family of IAA-Amino Acid Conjugate Hydrolases from Arabidopsis. Journal of Biological Chemistry 227, 20446–20452. Lee MM, Schiefelbein J. 1999. WEREWOLF, a MYB-related protein in Arabidopsis, is a position-dependent regulator of epidermal cell patterning. Cell 99, 473–483. Leliaert F, Verbruggen H, Zechman FW. 2011. Into the deep: New discoveries at the base of the green plant phylogeny. BioEssays 33, 683–692. Lewis DR, Miller ND, Splitt BL, Wu G, Spalding EP. 2007. Separating the Roles of Acropetal and Basipetal Auxin Transport on Gravitropism with Mutations in Two Arabidopsis Multidrug Resistance-Like ABC Transporter Genes . The Plant Cell 19, 1838–1850. Li WF, Perry PJ, Prafulla NN, Schmidt W. 2010. Ubiquitin-Specific Protease 14 (UBP14) is Involved in Root Responses to Phosphate Deficiency in Arabidopsis. Molecular Plant 3, 212–223. Li S-B, Xie Z-Z, Hu C-G, Zhang J-Z. 2016. A Review of Auxin Response Factors (ARFs) in Plants. Frontiers in Plant Science 7, 1–7. Liao CY, Smet W, Brunoud G, Yoshida S, Vernoux T, Weijers D. 2015. Reporters for sensitive and quantitative measurement of auxin response. Nature Methods 12, 207–210. Lightner J, Caspar T. 2009. Seed Mutagenesis of Arabidopsis. Arabidopsis Protocols. Humana Press, 91–102. Linkohr BI, Williamson LC, Fitter AH, Leyser HMO. 2002. Nitrate and phosphate availability and distribution have different effects on root system architecture of Arabidopsis. Plant Journal 29, 751–760. Liu X, Hegeman AD, Gardner G, Cohen JD. 2012. Protocol: High-throughput and quantitative assays of auxin and auxin precursors from minute tissue samples. Plant Methods 8, 1–17. Ljung K. 2013. Auxin metabolism and homeostasis during plant development. Development 140, 943–950. López-Arredondo DL, Leyva-González MA, González-Morales SI, López-Bucio J, Herrera-Estrella L. 2014. Phosphate Nutrition: Improving Low-Phosphate 229 Tolerance in Crops. Annual Review of Plant Biology 65, 95–123. Lu P, Han X, Qi J, Yang J, Wijeratne AJ, Li T, Ma H. 2012. Analysis of Arabidopsis genome-wide variations before and after meiosis and meiotic recombination by resequencing Landsberg erecta and all four products of a single meiosis. Genome Research 22, 508–518. Ludwig-Müller J. 2007. Indole-3-butyric acid synthesis in ecotypes and mutants of Arabidopsis thaliana under different growth conditions. Journal of Plant Physiology 164, 47–59. Ludwig-Müller J. 2011. Auxin conjugates: Their role for plant development and in the evolution of land plants. Journal of Experimental Botany 62, 1757–1773. Ludwig-Muller J, Epstein E. 1991. Occurrence and in Vivo Biosynthesis of Indole-3- Butyric Acid in Corn (Zea mays L.). Plant Physiology 97, 765–770. Ma C, Hagstrom D, Polley SG, Subramani S. 2013. Redox-regulated cargo binding and release by the peroxisomal targeting signal receptor, Pex5. Journal of Biological Chemistry 288, 27220–27231. Ma W, Penrose DM, Glick BR. 2002. Strategies used by rhizobia to lower plant ethylene levels and increase nodulation. Canadian Journal of Microbiology 48, 947–954. Malamy JE, Benfey PN. 1997. Organization and cell differentiation in lateral roots of Arabidopsis thaliana. Development 124, 33–34. Mangano S, Denita-Juarez SP, Choi H-S, et al. 2017. Molecular link between auxin and ROS-mediated polar growth. Proceedings of the National Academy of Sciences 114, 5289–5294. Mangano S, Denita-Juarez SP, Marzol E, Borassi C, Estevez JM. 2018. High Auxin and High Phosphate Impact on RSL2 Expression and ROS-Homeostasis Linked to Root Hair Growth in Arabidopsis thaliana. Frontiers in Plant Science 9. Mano Y, Nemoto K. 2012. The pathway of auxin biosynthesis in plants. Journal of Experimental Botany 63, 2853–2872. Manzano C, Pallero-Baena M, Casimiro I, De Rybel B, Orman-Ligeza B, Van Isterdael G, Beeckman T, Draye X, Casero P, del Pozo JC. 2014. The Emerging Role of Reactive Oxygen Species Signaling during Lateral Root Development. PLANT PHYSIOLOGY 165, 1105–1119. Marquez-Garcia B, Njo M, Beeckman T, Goormachtig S, Foyer CH. 2014. A new role for glutathione in the regulation of root architecture linked to strigolactones. Plant, Cell and Environment 37, 488–498. 230 Marty L, Siala W, Schwarzlander M, Fricker MD, Wirtz M, Sweetlove LJ, Meyer Y, Meyer AJ, Reichheld J-P, Hell R. 2009. The NADPH-dependent thioredoxin system constitutes a functional backup for cytosolic glutathione reductase in Arabidopsis. Proceedings of the National Academy of Sciences 106, 9109–9114. Maughan SC, Pasternak M, Cairns N., et al. 2010. Plant homologs of the Plasmodium falciparum chloroquine-resistance transporter, PfCRT, are required for glutathione homeostasis and stress responses. Proceedings of the National Academy of Sciences 107, 2331–2336. McCourt RM, Delwiche CF, Karol KG. 2004. Charophyte algae and land plant origins. Trends in Ecology & Evolution 19, 661–666. Meister A. 1983. Selective Modification of Glutathione Metabolism. Menand B, Yi K, Jouannic S, Hoffmann L, Ryan E, Linstead P, Schaefer DG, Dolan L. 2007. An Ancient Mechanism Controls the Development of Cells with a Rooting Function in Land Plants. Science 316, 1477–1480. Menon SG, Goswami PC. 2007. A redox cycle within the cell cycle: Ring in the old with the new. Oncogene 26, 1101–1109. Meyer Y, Belin C, Delorme-Hinoux V, Reichheld J-P, Riondet C. 2012. Thioredoxin and Glutaredoxin Systems in Plants: Molecular Mechanisms, Crosstalks, and Functional Significance. Antioxidants & Redox Signaling 17, 1–37. Mhamdi A, Van Breusegem F. 2018. Reactive oxygen species in plant development. Development 145, 1–12. Mhamdi A, Hager J, Chaouch S, et al. 2010a. Arabidopsis GLUTATHIONE REDUCTASE1 Plays a Crucial Role in Leaf Responses to Intracellular Hydrogen Peroxide and in Ensuring Appropriate Gene Expression through Both Salicylic Acid and Jasmonic Acid Signaling Pathways. Plant physiology 153, 1144–1160. Mhamdi A, Queval G, Chaouch S, Vanderauwera S, Van Breusegem F, Noctor G. 2010b. Catalase function in plants: A focus on Arabidopsis mutants as stress- mimic models. Journal of Experimental Botany 61, 4197–4220. Michalko J, Glanc M, Perrot-Rechenmann C, Friml J. 2016. Strong morphological defects in conditional Arabidopsis abp1 knock-down mutants generated in absence of functional ABP1 protein. F1000Research 5, 86. Michelet L, Zaffagnini M, Marchand C, et al. 2005. Glutathionylation of chloroplast thioredoxin f is a redox signaling mechanism in plants. Proceedings of the National Academy of Sciences 102, 16478–16483. Millet YA, Danna CH, Clay NK, Songnuan W, Simon MD, Werck-Reichhart D, 231 Ausubel FM. 2010. Innate Immune Responses Activated in Arabidopsis Roots by Microbe-Associated Molecular Patterns. The Plant Cell 22, 973–990. Mittler R. 2017. ROS Are Good. Trends in Plant Science 22, 1–19. Molendijk AJ, Bischoff F, Rajendrakumar CSV, Friml J, Braun M, Gilroy S, Palme K. 2001. Arabidopsis thaliana Rop GTPases are localized to tips of root hairs and control polar growth. EMBO Journal 20, 2779–2788. Moons A. 2005. Regulatory and Functional Interactions of Plant Growth Regulators and Plant Glutathione S-Transferases (GSTs). Vitamins and Hormones.155–202. Moreno-Risueno MA, Van Norman JM, Moreno A, Zhang J, Ahnert SE, Benfey PN. 2010. Oscillating Gene Expression Determines Competence for Periodic Arabidopsis Root Branching. Science 329, 1306–1311. Morffy NJ, Strader LC. 2018. Locally Sourced: Auxin Biosynthesis and Transport in the Root Meristem. Developmental Cell 47, 262–264. Mravec J, Kubes M, Bielach A, Gaykova V, Petrasek J, Skupa P, Chand S, Benkova E, Zazimalova E, Friml J. 2008. Interaction of PIN and PGP transport mechanisms in auxin distribution-dependent development. Development 135, 3345–3354. El Msehli S, Lambert A, Baldacci-Cresp F, Hopkins J, Boncompagni E, Smiti SA, Hérouart D, Frendo P. 2011. Crucial role of (homo)glutathione in nitrogen fixation in Medicago truncatula nodules. New Phytologist 192, 496–506. Müller A, Guan C, Gälweiler L, Tänzler P, Huijser P, Marchant A, Parry G, Bennett M, Wisman E, Palme K. 1998. AtPIN2 defines a locus of Arabidopsis for root gravitropism control. Napier RM. 1995. Towards an understanding of ABP1. Journal of Experimental Botany 46, 1787–1795. Ndamukong I, Abdallat A Al, Thurow C, Fode B, Zander M, Weigel R, Gatz C. 2007. SA-inducible Arabidopsis glutaredoxin interacts with TGA factors and suppresses JA-responsive PDF1.2 transcription. Plant Journal 50, 128–139. Nemhauser JL, Hong F, Chory J. 2006. Different Plant Hormones Regulate Similar Processes through Largely Nonoverlapping Transcriptional Responses. Cell 126, 467–475. Nemoto K, Hara M, Suzuki M, Seki H, Muranaka T, Mano Y. 2009. The NtAMI1 gene functions in cell division of tobacco BY-2 cells in the presence of indole-3- acetamide. FEBS Letters 583, 487–492. Nianiou-Obeidat I, Madesis P, Kissoudis C, Voulgari G, Chronopoulou E, 232 Tsaftaris A, Labrou NE. 2017. Plant glutathione transferase-mediated stress tolerance: functions and biotechnological applications. Plant Cell Reports 36, 791–805. Niu YF, Chai RS, Jin GL, Wang H, Tang CX, Zhang YS. 2013. Responses of root architecture development to low phosphorus availability: A review. Annals of Botany 112, 391–408. Noctor G, Mhamdi A, Chaouch S, Han Y, Neukermans J, Marquez-Garcia B, Queval G, Foyer CH. 2012. Glutathione in plants: An integrated overview. Plant, Cell and Environment 35, 454–484. Noctor G, Reichheld JP, Foyer CH. 2018. ROS-related redox regulation and signaling in plants. Seminars in Cell and Developmental Biology 80, 3–12. Noh B, Bandyopadhyay A, Peer WA, Spalding EP, Murphy AS. 2003. Enhanced gravi-and phototropism in plant mdr mutants mislocalizing the auxin efflux protein PIN1. Nature 423, 999–1002. Van Norman JM, Xuan W, Beeckman T, Benfey PN. 2013. To branch or not to branch: the role of pre-patterning in lateral root formation. Development 140, 4301–4310. Normanly J. 2010. Approaching Cellular and Molecular Resolution of Auxin Biosynthesis and Metabolism. Cold Spring Harbor perspectives in biology. Novák O, Hényková E, Sairanen I, Kowalczyk M, Pospíšil T, Ljung K. 2012. Tissue-specific profiling of the Arabidopsis thaliana auxin metabolome. Plant Journal 72, 523–536. Nziengui H, Lasok H, Kochersperger P, Ruperti B, Rébeillé F, Palme K, Ditengou FA. 2018. Root gravitropism is regulated by a crosstalk between para- aminobenzoic acid, ethylene, and auxin. Plant Physiology 178, 1370–1389. O’Brien JA, Daudi A, Finch P, Butt VS, Whitelegge JP, Souda P, Ausubel FM, Bolwell GP. 2012. A Peroxidase-Dependent Apoplastic Oxidative Burst in Cultured Arabidopsis Cells Functions in MAMP-Elicited Defense. Plant physiology 158, 2013–2027. Ohkama-Ohtsu N, Zhao P, Xiang C, Oliver DJ. 2007. Glutathione conjugates in the vacuole are degraded by γ-glutamyl transpeptidase GGT3 in Arabidopsis. Plant Journal 49, 878–888. Okada K, Ueda J, Komaki MK, Bell CJ, Shimura Y. 1991. Requirement of the Auxin Polar Transport System in Early Stages of Arabidopsis Floral Bud Formation. THE PLANT CELL ONLINE 3, 677–684.

233 Okushima Y, Overvoorde P, Arima K, et al. 2005. Functional Genomic Analysis of the AUXIN RESPONSE FACTOR Gene Family Members in Arabidopsis thaliana: Unique and Overlapping Functions of ARF7 and ARF19. The Plant Cell 17, 444–463. Orman-Ligeza B, Parizot B, de Rycke R, Fernandez A, Himschoot E, Van Breusegem F, Bennett MJ, Périlleux C, Beeckman T, Draye X. 2016. RBOH- mediated ROS production facilitates lateral root emergence in Arabidopsis. Development 143, 3328–339. Overvoorde P, Fukaki H, Beeckman T. 2010. Auxin Control of Root Development. Cold Spring Harbor perspectives in biology 2. Parisy V, Poinssot B, Owsianowski L, Buchala A, Glazebrook J, Mauch F. 2006. Identification of PAD2 as a γ-glutamylcysteine synthetase highlights the importance of glutathione in disease resistance of Arabidopsis. Plant Journal 49, 159–172. Park HY, Seok HY, Park BK, Kim SH, Goh CH, Lee B ha, Lee CH, Moon YH. 2008. Overexpression of Arabidopsis ZEP enhances tolerance to osmotic stress. Biochemical and Biophysical Research Communications 375, 80–85. Parry G. 2018. Low Phosphate Puts Auxin in the Root Hairs. Trends in Plant Science 23, 845–847. Parry G, Delbarre A, Marchant A, Swarup R, Napier R, Perrot-Rechenmann C, Bennett MJ. 2001. Novel auxin transport inhibitors phenocopy the auxin influx carrier mutation aux1. The Plant Journal 25, 399–406. Passaia G, Queval G, Bai J, Margis-Pinheiro M, Foyer CH. 2014. The effects of redox controls mediated by glutathione peroxidases on root architecture in Arabidopsis thaliana. Journal of Experimental Botany 65, 1403–1413. Pasternak M, Lim B, Wirtz M, Hell R, Cobbett CS, Meyer AJ. 2008. Restricting glutathione biosynthesis to the cytosol is sufficient for normal plant development. Plant Journal 53, 999–1012. Payne CT, Zhang F, Lloyd AM. 2000. GL3 Encodes a bHLH Protein That Regulates Trichome Development in Arabidopsis Through Interaction With GL1 and TTG1. Genetics 156, 1349–1362. Pečenková T, Janda M, Ortmannová J, Hajná V, Stehlíková Z, Žárský V. 2017. Early Arabidopsis root hair growth stimulation by pathogenic strains of Pseudomonas syringae. Annals of Botany 120, 437–446. Petrášek J, Friml J. 2009. Auxin transport routes in plant development. Development 136, 2675–2688. 234 Pike S, Patel A, Stacey G, Gassmann W. 2009. Arabidopsis OPT6 is an oligopeptide transporter with exceptionally broad substrate specificity. Plant and Cell Physiology 11, 1923–1932. Poitout A, Martinière A, Kucharczyk B, et al. 2017. Local signalling pathways regulate the Arabidopsis root developmental response to Mesorhizobium loti inoculation. Journal of Experimental Botany 68, 1199–1211. Powers SK, Strader LC. 2016. Up in the air: Untethered Factors of Auxin Response. F1000Research 133, 1–8. Del Pozo JC, Boniotti MB, Gutierrez C. 2002. Arabidopsis E2Fc Functions in Cell Division and Is Degraded by the Ubiquitin-SCFAtSKP2 Pathway in Response to Light. THE PLANT CELL ONLINE 14, 3057–3071. Pracharoenwattana I, Smith SM. 2008. When is a peroxisome not a peroxisome? Trends in Plant Science 13, 522–525. Puga MI, Mateos I, Charukesi R, et al. 2014. SPX1 is a phosphate-dependent inhibitor of PHOSPHATE STARVATION RESPONSE 1 in Arabidopsis. Proceedings of the National Academy of Sciences 111, 14947–14952. Puga MI, Rojas-Triana M, de Lorenzo L, Leyva A, Rubio V, Paz-Ares J. 2017. Novel signals in the regulation of Pi starvation responses in plants: facts and promises. Current Opinion in Plant Biology 39, 40–49. Purdue PE, Lazarow PB. 2001. PEROXISOME BIOGENESIS. Annu. Rev. Cell Dev. Biol. 17, 701–752. Puttick MN, Morris JL, Williams TA, et al. 2018a. The Interrelationships of Land Plants and the Nature of the Ancestral Embryophyte. Current Biology 28, 733– 745. Puttick MN, Morris JL, Yang Z, Donoghue PCJ, Edwards D, Clark JW, Pressel S, Schneider H, Kenrick P, Wellman CH. 2018b. The timescale of early land plant evolution. Proceedings of the National Academy of Sciences 115. Qu Y, Wang Q, Guo J, Wang P, Song P, Jia Q, Zhang X, Kudla J, Zhang W, Zhang Q. 2017. Peroxisomal CuAOζ and its product H2O2 regulate the distribution of auxin and IBA-dependent lateral root development in Arabidopsis. Journal of Experimental Botany 68, 4851–4867. Quittenden LJ, Davies NW, Smith JA, Molesworth PP, Tivendale ND, Ross JJ. 2009. Auxin Biosynthesis in Pea: Characterization of the Tryptamine Pathway. PLANT PHYSIOLOGY 151, 1130–1138. Rampey RA, LeClere S, Kowalczyk M, Ljung K, Sandberg G, Bartel B. 2004. A 235 Family of Auxin-Conjugate Hydrolases That Contributes to Free Indole-3-Acetic Acid Levels during Arabidopsis Germination. PLANT PHYSIOLOGY 135, 978– 988. Raven JA, Edwards D. 2017. Roots: evolutionary origins and biogeochemical significance. Journal of Experimental Botany 52, 381–401. Ray PM. 1977. Specificity of Auxin-binding Sites on Maize Coleoptile Membranes as Possible Receptor Sites for Auxin Action. Plant Physiology 60, 585–591. Razem FA, El-Kereamy A, Abrams SR, Hill RD. 2006. The RNA-binding protein FCA is an abscisic acid receptor. Nature 439, 290–294. Reichheld J-P, Khafif M, Riondet C, Droux M, Bonnard G, Meyer Y. 2007. Inactivation of Thioredoxin Reductases Reveals a Complex Interplay between Thioredoxin and Glutathione Pathways in Arabidopsis Development. The Plant Cell 19, 1851–1865. Remans T, Nacry P, Pervent M, Filleur S, Diatloff E, Mounier E, Tillard P, Forde BG, Gojon A. 2006. The Arabidopsis NRT1.1 transporter participates in the signaling pathway triggering root colonization of nitrate-rich patches. Proceedings of the National Academy of Sciences 103, 19206–19211. Rerie WG, Feldmann KA, David Marks M. 1994. The GLABRA2 gene encodes a homeo domain protein required for normal tnchome development in Arabidopsis. Genes and Development 8, 1388–1399. Rubinstein C V., Gerrienne P, de la Puente GS, Astini RA, Steemans P. 2010. Early Middle Ordovician evidence for land plants in Argentina (eastern Gondwana). New Phytologist 188, 365–369. Ruzicka K, Strader LC, Bailly A, et al. 2010. Arabidopsis PIS1 encodes the ABCG37 transporter of auxinic compounds including the auxin precursor indole-3-butyric acid. Proceedings of the National Academy of Sciences 107, 10749–10753. De Rybel B, Audenaert D, Xuan W, et al. 2012. A role for the root cap in root branching revealed by the non-auxin probe naxillin. Nature Chemical Biology 8, 798–805. Rymen B, Kawamura A, Schäfer S, et al. 2017. ABA Suppresses Root Hair Growth via the OBP4 Transcriptional Regulator. Plant Physiology 173, 1750–1762. Sagan L. 1967. On the Origin of Mitosing Cdls. Salamon MA, Gerrienne P, Steemans P, et al. 2018. Putative Late Ordovician land plants. New Phytologist. Salazar-Henao JE, Vélez-Bermúdez IC, Schmidt W. 2016. The regulation and 236 plasticity of root hair patterning and morphogenesis. Development 143, 1848– 1858. Sanchez-Fernandez R, Fricker M, Corben LB, White NS, Sheard N, Leaver CJ, Van Montagu M, Inze D, May MJ. 1997. Cell proliferation and hair tip growth in the Arabidopsis root are under mechanistically different forms of redox control. Proceedings of the National Academy of Sciences 94, 2745–2750. Santelia D, Vincenzetti V, Azzarello E, Bovet L, Fukao Y, Düchtig P, Mancuso S, Martinoia E, Geisler M. 2005. MDR-like ABC transporter AtPGP4 is involved in auxin-mediated lateral root and root hair development. FEBS Letters 579, 5399–5406. Sanz L, Albertos P, Mateos I, Sánchez-Vicente I, Lechón T, Fernández-Marcos M, Lorenzo O. 2015. Nitric oxide (NO) and phytohormones crosstalk during early plant development. Journal of Experimental Botany 66, 2857–2868. Sanz L, Fernandez-Marcos M, Modrego A, Lewis DR, Muday GK, Pollmann S, Duenas M, Santos-Buelga C, Lorenzo O. 2014. Nitric Oxide Plays a Role in Stem Cell Niche Homeostasis through Its Interaction with Auxin. Plant physiology 166, 1972–1984. Schaedler TA, Thornton JD, Kruse I, Schwarzländer M, Meyer AJ, van Veen HW, Balk J. 2014. A Conserved Mitochondrial ATP-binding Cassette Transporter Exports Glutathione Polysulfide for Cytosolic Metal Cofactor Assembly. Journal of Biological Chemistry 289, 23264–23274. Schellmann S, Schnittger A, Kirik V, Wada T, Okada K, Beermann A, Thumfahrt J, Jurgens G, Hulskamp M. 2002. TRIPTYCHON and CAPRICE mediate lateral inhibition during trichome and root hair patterning in Arabidopsis. The EMBO Journal 21, 5036–5046. Scheres B, Benfey P, Dolan L. 2002. Root Development. The Arabidopsis Book 1, 1– 18. Schiefelbein JW, Shipley A, Rowse P. 1992. Calcium influx at the tip of growing root- hair cells of Arabidopsis thaliana. Planta 187, 455–459. Schlatter D, Kinkel L, Thomashow L, Weller D, Paulitz T. 2017. Disease Suppressive Soils: New Insights from the Soil Microbiome. Phytopathology 107, 1284–1297. Schnall JA, Quatrano RS. 1992. Abscisic Acid Elicits the Water-Stress Response in Root Hairs of Arabidopsis thaliana’. Schnaubelt D, Queval G, Dong Y, Diaz-Vivancos P, Makgopa ME, Howell G, De Simone A, Bai J, Hannah MA, Foyer CH. 2015. Low glutathione regulates gene 237 expression and the redox potentials of the nucleus and cytosol in Arabidopsis thaliana. Plant, Cell and Environment 38, 266–279. Schneeberger K. 2014. Using next-generation sequencing to isolate mutant genes from forward genetic screens. Nature Reviews Genetics. Schneeberger K, Ossowski S, Lanz C, Juul T, Petersen AH, Nielsen KL, Jørgensen J-E, Weigel D, Andersen SU. 2009. SHOREmap: simultaneous mapping and mutation identification by deep sequencing. Nature Methods 6, 550–551. Shanmugam V, Tsednee M, Yeh KC. 2012. ZINC TOLERANCE INDUCED BY IRON 1 reveals the importance of glutathione in the cross-homeostasis between zinc and iron in Arabidopsis thaliana. Plant Journal 69, 1006–1017. Shekhar V, Stӧ ckle D, Thellmann M, Vermeer JEM. 2018. The role of plant root systems in evolutionary adaptation. Current Topics in Developmental Biology. Elsevier Inc., 55–80. Shimizu-Mitao Y, Kakimoto T. 2014. Auxin Sensitivities of All Arabidopsis Aux/IAAs for Degradation in the Presence of Every TIR1/AFB. Plant and Cell Physiology 55, 1450–1459. Shimotohno A, Heidstra R, Blilou I, Scheres B. 2018. Root stem cell niche organizer specification by molecular convergence of PLETHORA and SCARECROW transcription factor modules. Genes and Development 32, 1085–1100. Shin R, Berg RH, Schachtman DP. 2005. Reactive Oxygen Species and Root Hairs in Arabidopsis Root Response to Nitrogen, Phosphorus and Potassium Deficiency. Plant and Cell Physiology 46, 1350–1357. Simon S, Petrášek J. 2011. Why plants need more than one type of auxin. Plant Science 180, 454–460. De Smet I, Lau S, Voß U, et al. 2010. Bimodular auxin response controls organogenesis in Arabidopsis. Proceedings of the National Academy of Sciences 107, 2705–2710. De Smet I, Signora L, Beeckman T, Inzé D, Foyer CH, Zhang H. 2003. An abscisic acid-sensitive checkpoint in lateral root development of Arabidopsis. Plant Journal 33, 543–555. De Smet I, Zhang H, Inzé D, Beeckman T. 2006. A novel role for abscisic acid emerges from underground. Trends in Plant Science 11, 434–439. Smith AP, Jain A, Deal RB, Nagarajan VK, Poling MD, Raghothama KG, Meagher RB. 2010. Histone H2A.Z Regulates the Expression of Several Classes of Phosphate Starvation Response Genes But Not as a Transcriptional Activator. 238 PLANT PHYSIOLOGY 152, 217–225. Smith S, de Smet I. 2012. Root system architecture: Insights from Arabidopsis and cereal crops. Philosophical Transactions of the Royal Society B: Biological Sciences 367, 1441–1452. Spartz AK, Ren H, Park MY, Grandt KN, Lee SH, Murphy AS, Sussman MR, Overvoorde PJ, Gray WM. 2014. SAUR Inhibition of PP2C-D Phosphatases Activates Plasma Membrane H+-ATPases to Promote Cell Expansion in Arabidopsis. The Plant Cell 26, 2129–2142. Staswick PE. 2009. The Tryptophan Conjugates of Jasmonic and Indole-3-Acetic Acids Are Endogenous Auxin Inhibitors. PLANT PHYSIOLOGY 150, 1310– 1321. Staswick PE, Serban B, Rowe M, Tiryaki I, Maldonado MT, Maldonado MC, Suza W. 2005. Characterization of an Arabidopsis Enzyme Family That Conjugates Amino Acids to Indole-3-Acetic Acid. The Plant Cell 17, 616–627. Strader LC, Bartel B. 2009. The Arabidopsis PLEIOTROPIC DRUG RESISTANCE8/ABCG36 ATP Binding Cassette Transporter Modulates Sensitivity to the Auxin Precursor Indole-3-Butyric Acid. THE PLANT CELL ONLINE 21, 1992–2007. Strader LC, Bartel B. 2011. Transport and Metabolism of the Endogenous Auxin Precursor Indole-3-Butyric Acid. Molecular Plant 4, 477–486. Strader LC, Culler AH, Cohen JD, Bartel B. 2010. Conversion of Endogenous Indole-3-Butyric Acid to Indole-3-Acetic Acid Drives Cell Expansion in Arabidopsis Seedlings. PLANT PHYSIOLOGY 153, 1577–1586. Strader LC, Wheeler DL, Christensen SE, Berens JC, Cohen JD, Rampey RA, Bartel B. 2011. Multiple Facets of Arabidopsis Seedling Development Require * Indole-3-Butyric Acid-Derived Auxin. The Plant Cell 23, 984–999. Strader LC, Zhao Y. 2016. Auxin perception and downstream events. Current Opinion in Plant Biology 33, 8–14. Suen D-F, Tsai Y-H, Cheng Y-T, Radjacommare R, Ahirwar RN, Fu H, Schmidt W. 2018. The Deubiquitinase OTU5 Regulates Root Responses to Phosphate Starvation. Plant Physiology 176, 2441–2455. Sun H, Schneeberger K. 2015. SHOREmap v3.0: Fast and accurate identification of causal mutations from forward genetic screens. Plant Functional Genomics: Methods and Protocols: Second Edition. Sutter EG, Cohen JD. 1992. Measurement of Indolebutyric Acid in Plant Tissues by 239 Isotope Dilution Gas Chromatography-Mass Spectrometry Analysis. Plant Physiology 99, 1719–1722. Suzuki N, Miller G, Morales J, Shulaev V, Torres MA, Mittler R. 2011. Respiratory burst oxidases: The engines of ROS signaling. Current Opinion in Plant Biology 14, 691–699. Swarup K, Benková E, Swarup R, et al. 2008. The auxin influx carrier LAX3 promotes lateral root emergence. Nature Cell Biology 10, 946–954. Swarup R, Friml J, Marchant A, Ljung K, Sandberg G, Palme K, Bennett M. 2001. Localization of the auxin permease AUX1 suggests two functionally distinct hormone transport pathways operate in the Arabidopsis root apex. Genes and Development 15, 2648–2653. Szövényi P, Waller M, Kirbis A. 2019. Evolution of the plant body plan. Current Topics in Developmental Biology. Elsevier Inc., 1–34. Tao Y, Ferrer JL, Ljung K, et al. 2008. Rapid Synthesis of Auxin via a New Tryptophan-Dependent Pathway Is Required for Shade Avoidance in Plants. Cell 133, 164–176. Teale WD, Paponov IA, Palme K. 2006. Auxin in action: Signalling, transport and the control of plant growth and development. Nature Reviews Molecular Cell Biology 7, 847–859. Terrile MC, París R, Calderõn-Villalobos LIA, Iglesias MJ, Lamattina L, Estelle M, Casalongué CA. 2012. Nitric oxide influences auxin signaling through S- nitrosylation of the Arabidopsis TRANSPORT INHIBITOR RESPONSE 1 auxin receptor. Plant Journal 70, 492–500. Thibaud MC, Arrighi JF, Bayle V, Chiarenza S, Creff A, Bustos R, Paz-Ares J, Poirier Y, Nussaume L. 2010. Dissection of local and systemic transcriptional responses to phosphate starvation in Arabidopsis. Plant Journal 64, 775–789. Thormählen I, Naranjo B, Trujillo-Hernandez JA, Reichheld P, Cejudo FJ, Geigenberger P. 2018. On the Elaborate Network of Thioredoxins in Higher Plants. Progress in Botany 80, 223–251. Tognetti VB, Van Aken O, Morreel K, et al. 2010. Perturbation of Indole-3-Butyric Acid Homeostasis by the UDP-Glucosyltransferase UGT74E2 Modulates Arabidopsis Architecture and Water Stress Tolerance. The Plant Cell 22, 1–20. Tognetti VB, Bielach A, Hrtyan M. 2017. Redox regulation at the site of primary growth: auxin, cytokinin and ROS crosstalk. Plant Cell and Environment 40, 2586–2605.

240 Tommasini R, Martinoia E, Grill E, Dietz K-J, Amrhein N. 2018. Transport of Oxidized Glutathione into Barley Vacuoles: Evidence for the Involvement of the Glutathione-S-Conjugate ATPase. Zeitschrift für Naturforschung C 48c, 867–871. Tzin V, Galili G. 2010. New Insights into the shikimate and aromatic amino acids biosynthesis pathways in plants. Molecular Plant 3, 956–972. Ulmasov T, Murfett J, Hagen G, Guilfoyle TJ. 1997. Aux/IAA Proteins Repress Expression of Reporter Genes Containing Natural and Highly Active Synthetic Auxin Response Elements. The Plant Cell 9, 1963–1971. Uzunova V V., Quareshy M, Del Genio CI, Napier RM. 2016. Tomographic docking suggests the mechanism of auxin receptor TIR1 selectivity. Open biology 6. Vanacker, Carver, Foyer. 1998. Pathogen-Induced Changes in the Antioxidant Status of the Apoplast in Barley leaves. Plant physiology 117, 1103–1114. Vernoux T, Wilson RC, Seeley KA, et al. 2000. The ROOT MERISTEMLESS1/CADMIUM SENSITIVE2 Gene Defines a Glutathione- Dependent Pathway Involved in Initiation and Maintenance of Cell Division during Postembryonic Root Development. The Plant Cell 12, 97–109. Vilches-Barro A, Maizel A. 2015. Talking through walls: Mechanisms of lateral root emergence in Arabidopsis thaliana. Current Opinion in Plant Biology 23, 31–38. Vivancos PD, Dong Y, Ziegler K, Markovic J, Pallardó F V., Pellny TK, Verrier PJ, Foyer CH. 2010. Recruitment of glutathione into the nucleus during cell proliferation adjusts whole-cell redox homeostasis in Arabidopsis thaliana and lowers the oxidative defence shield. Plant Journal 64, 825–838. Waadt R, Hitomi K, Nishimura N, Hitomi C, Adams SR, Getzoff ED, Schroeder JI. 2014. FRET-based reporters for the direct visualization of abscisic acid concentration changes and distribution in Arabidopsis. eLife 3. Wachter A, Wolf S, Steininger H, Bogs J, Rausch T. 2005. Differential targeting of GSH1 and GSH2 is achieved by multiple transcription initiation: Implications for the compartmentation of glutathione biosynthesis in the Brassicaceae. Plant Journal 41, 15–30. Wada T, Kurata T, Tominaga R, Koshino-Kimura, Yoshihiro Tachibana T, Koji G, David M, Shimura Y, Okada K. 2002. Role of a positive regulator of root hair development, CAPRICE, in Arabidopsis root epidermal cell differentiation. Development 129, 5409–5419. Wagner U, Edwards R, Dixon DP, Mauch F. 2002. Probing the Diversity of the Arabidopsis glutathione S-Transferase Gene Family. Plant Molecular Biology 49, 515–532. 241 Wagner D, Przybyla D, Op Den Camp R, et al. 2004. The genetic basis of singlet oxygen-induced stress response of Arabidopsis thaliana. Science 306, 1183–1185. Walker AR, Davison P, Bolognesi-Winfield AC, James CM, Srinivasan N, Blundell TL, Esch JJ, Marks MD, Gray JC. 1999. The TRANSPARENT TESTA GLABRA1 Locus, Which Regulates Trichome Differentiation and Anthocyanin Biosynthesis in Arabidopsis, Encodes a WD40 Repeat Protein. THE PLANT CELL ONLINE 11, 1337–1349. Went , F. W. 1928 . Wuchsstoff und Wachstum. Extrait du Recueil des Travaux Botaniques Neerlandais 25, 1–116 . Wild R, Gerasimaite R, Jung J-Y, et al. 2016. Control of eukaryotic phosphate homeostasis by inositol polyphosphate sensor domains. Science 352, 986–990. Wrzaczek M, Brosché M, Salojärvi J, Kangasjärvi S, Idänheimo N, Mersmann S, Robatzek S, Karpiński S, Karpińska B, Kangasjärvi J. 2010. Transcriptional regulation of the CRK/DUF26 group of Receptor-like protein kinases by ozone and plant hormones in Arabidopsis. Xiong L, Wang R-G, Mao G, Koczan JM. 2006. Identification of Drought Tolerance Determinants by Genetic Analysis of Root Response to Drought Stress and Abscisic Acid. PLANT PHYSIOLOGY 142, 1065–1074. Xuan W, Audenaert D, Parizot B, et al. 2015. Root Cap-Derived Auxin Pre-patterns the Longitudinal Axis of the Arabidopsis Root. Current Biology 25, 1381–1388. Yamada M, Greenham K, Prigge MJ, Jensen PJ, Estelle M. 2009. The TRANSPORT INHIBITOR RESPONSE2 Gene Is Required for Auxin Synthesis and Diverse Aspects of Plant Development. PLANT PHYSIOLOGY 151, 168– 179. Yang L, Zhang J, He J, Qin Y, Hua D, Duan Y, Chen Z, Gong Z. 2014. ABA- Mediated ROS in Mitochondria Regulate Root Meristem Activity by Controlling PLETHORA Expression in Arabidopsis. PLoS Genetics 10. Yanni YG, Rizk RY, Abd El-Fattah FK, et al. 2001. Natural Beneficial Association of Rhizobium leguminosarum bv. trifolii and Rice Roots and its Relevance to Sustainable Agriculture. Journal of plant physiology 28, 845–870. Yi K, Menand B, Bell E, Dolan L. 2010. A basic helix-loop-helix transcription factor controls cell growth and size in root hairs. Nature Genetics 42, 264–269. Yu H, Yang J, Shi Y, et al. 2017. Arabidopsis Glutaredoxin S17 Contributes to Vegetative Growth, Mineral Accumulation, and Redox Balance during Iron Deficiency. Frontiers in Plant Science 8, 1–11.

242 Yu H, Zhang Y, Moss BL, Bargmann BOR, Wang R, Prigge M, Nemhauser JL, Estelle M. 2015. Untethering the TIR1 auxin receptor from the SCF complex increases its stability and inhibits auxin response. Nature Plants 1, 1–8. Yuan HM, Liu WC, Lu YT. 2017. CATALASE2 Coordinates SA-Mediated Repression of Both Auxin Accumulation and JA Biosynthesis in Plant Defenses. Cell Host and Microbe 21, 143–155. Yuhashi KI, Ichikawa N, Ezura H, Akao S, Minakawa Y, Nukui N, Yasuta T, Minamisawa K. 2000. Rhizobitoxine production by Bradyrhizobium elkanii Enhances Nodulation and Competitiveness on Macroptilium atropurpureum. Applied and Environmental Microbiology 66, 2658–2663. Zaffagnini M, Fermani S, Costa A, Lemaire SD, Trost P. 2013. Plant cytoplasmic GAPDH: redox post-translational modifications and moonlighting properties. Frontiers in Plant Science 4. Zaffagnini M, Michelet L, Marchand C, Sparla F, Decottignies P, Le Maréchal P, Miginiac-Maslow M, Noctor G, Trost P, Lemaire SD. 2007. The thioredoxin- independent isoform of chloroplastic glyceraldehyde-3- phosphate dehydrogenase is selectively regulated by glutathionylation. FEBS Journal 274, 212–226. Zandalinas SI, Mittler R. 2018. ROS-induced ROS release in plant and animal cells. Free Radical Biology and Medicine 122, 21–27. Zander M, Chen S, Imkampe J, Thurow C, Gatz C. 2012. Repression of the Arabidopsis thaliana Jasmonic Acid/Ethylene-Induced Defense Pathway by TGA- Interacting Glutaredoxins Depends on Their C-Terminal ALWL Motif. Molecular Plant 5, 831–840. Zechmann B, Mauch F, Sticher L, Müller M. 2008. Subcellular immunocytochemical analysis detects the highest concentrations of glutathione in mitochondria and not in plastids. Journal of Experimental Botany 59, 4017–4027. Zhang F, Gonzalez A, Zhao M, Payne T, Lloyd A. 2003. A network of redundant bHLH proteins functions in all TTG1-dependent pathways of Arabidopsis. Development 130, 4859–4869. Zhang J, Vanneste S, Brewer PB, et al. 2011. Inositol Trisphosphate-Induced Ca2+ Signaling Modulates Auxin Transport and PIN Polarity. Developmental Cell 20, 855–866. Zhao Y, Hull AK, Gupta NR, Goss KA, Alonso J, Ecker JR, Normanly J, Chory J, Celenza JL. 2002. Trp-dependent auxin biosynthesis in Arabidopsis: Involvement of cytochrome P450s CYP79B2 and CYP79B3. Genes and Development 16, 3100–3112. 243 Zhao Y, Xing L, Wang X, Hou Y-J, Gao J, Wang P, Duan C-G, Zhu X, Zhu J-K. 2014. The ABA Receptor PYL8 Promotes Lateral Root Growth by Enhancing MYB77-Dependent Transcription of Auxin-Responsive Genes. Plant Biology 7, 1–11. Zhou Y, Liu X, Engstrom EM, Nimchuk ZL, Pruneda-Paz JL, Tarr PT, Yan A, Kay SA, Meyerowitz EM. 2015. Control of plant stem cell function by conserved interacting transcriptional regulators. Nature 517, 377–380. Zhou Y, Yan A, Han H, Li T, Geng Y, Liu X, Meyerowitz EM. 2018. HAIRY MERISTEM with WUSCHEL confines CLAVATA3 expression to the outer apical meristem layers. Science 361, 502–506. Zhou ZY, Zhang CG, Wu L, et al. 2011. Functional characterization of the CKRC1/TAA1 gene and dissection of hormonal actions in the Arabidopsis root. Plant Journal 66, 516–527. Zimmerman PW, Wilcoxon F. 1993. SEVERAL CHEMICAL GROWTH SUBSTANC ES WHI CH CAUSE INITIATION OF ROOTS AND OTHER RESPONSES IN PLANTS’. CONTRIBUTIONS FROM BOYCE THOMPSON I NSTITUTE 7, 209–229. Zolman BK, Martinez N, Millius A, Adham AR, Bartel B. 2008. Identification and characterization of Arabidopsis indole-3-butyric acid response mutants defective in novel peroxisomal enzymes. Genetics 180, 237–251. Zolman BK, Nyberg M, Bartel B. 2007. IBR3, a novel peroxisomal acyl-CoA dehydrogenase-like protein required for indole-3-butyric acid response. Plant Molecular Biology 64, 59–72. Zolman BK, Yoder A, Bartel B. 2000. Genetic Analysis of Indole-3-butyric Acid Responses in Arabidopsis thaliana Reveals Four Mutant Classes. Genetics 156, 1323–1337.

244 VI ANNEXES

245 VI.1 Annex I: Thiol Based Redox Signaling in Plant Nucleus

Laura Martins*, José Abraham Trujillo-Hernandez*, and Jean-Philippe Reichheld

246 MINI REVIEW published: 28 May 2018 doi: 10.3389/fpls.2018.00705

Thiol Based Redox Signaling in Plant Nucleus

Laura Martins1,2†, José Abraham Trujillo-Hernandez1,2† and Jean-Philippe Reichheld1,2*

1 Laboratoire Génome et Développement des Plantes, Université Perpignan Via Domitia, Perpignan, France, 2 Laboratoire Génome et Développement des Plantes, Centre National de la Recherche Scientifique, Perpignan, France

Reactive oxygen species (ROS) are well-described by-products of cellular metabolic activities, acting as signaling molecules and regulating the redox state of proteins. Solvent exposed thiol residues like cysteines are particularly sensitive to oxidation and their redox state affects structural and biochemical capacities of many proteins. While thiol redox regulation has been largely studied in several cell compartments like in the plant chloroplast, little is known about redox sensitive proteins in the nucleus. Recent works have revealed that proteins with oxidizable thiols are important for the regulation of many nuclear functions, including gene expression, transcription, epigenetics, and chromatin remodeling. Moreover, thiol reducing molecules like glutathione and specific isoforms of thiols reductases, thioredoxins and glutaredoxins were found in different nuclear subcompartments, further supporting that thiol-dependent systems are active Edited by: in the nucleus. This mini-review aims to discuss recent progress in plant thiol redox field, Christian Lindermayr, Helmholtz Zentrum München – taking examples of redox regulated nuclear proteins and focusing on major thiol redox Deutsches Forschungszentrum für systems acting in the nucleus. Gesundheit und Umwelt, Germany Keywords: nucleus, thiol, glutathione, thioredoxin, glutaredoxin, ROS Reviewed by: Renu Deswal, University of Delhi, India Sixue Chen, INTRODUCTION University of Florida, United States *Correspondence: Oxygen is one of the most important molecules for aerobic organisms. It is necessary Jean-Philippe Reichheld for cell metabolism, but it also generates reactive oxygen species (ROS) as by-products of •− [email protected] oxidoreduction pathways. ROS include free radical species like superoxides (O2 ), hydroxyl • • † radicals (OH ), or nitric oxide (NO ), and non-radical species like hydrogen peroxide These authors should be considered − as co-first authors. (H2O2) and peroxynitrite (ONOO )(Sies et al., 2017). In plants, major sources of ROS are photosynthetic and respiratory chains in chloroplasts and mitochondria. ROS are also Specialty section: generated by plasma membrane NADPH oxidases and peroxisomal xanthine oxidases. Oxidative This article was submitted to eustress is playing important signaling functions by inducing post-translational modifications Plant Physiology, (PTM) and by regulating protein redox state. ROS can also trigger oxidative distress a section of the journal which can damage the cell (Foyer and Noctor, 2016; Choudhury et al., 2017). Plant cells Frontiers in Plant Science display a large panel of ROS scavenging enzymes like catalases, peroxidases, and superoxide Received: 09 February 2018 dismutases. They also generate compounds that reverse ROS-induced oxidations. Among Accepted: 09 May 2018 these compounds are antioxidant molecules like glutathione and ascorbate, which both play Published: 28 May 2018 important roles as cofactors for thiol reduction enzymes like peroxidases and reductases Citation: (Noctor, 2017; Rahantaniaina et al., 2017). Glutathione and ascorbate are themselves reduced by Martins L, Trujillo-Hernandez JA and glutathione reductases (GRs) and dehydroascorbate reductases (DHARs). Thioredoxins (TRXs) Reichheld J-P (2018) Thiol Based Redox Signaling in Plant Nucleus. and glutaredoxins (GRXs) are key thiol reduction enzymes. They act as reducing power Front. Plant Sci. 9:705. of metabolic enzymes and ROS scavenging systems but they also regulate thiol-based post- doi: 10.3389/fpls.2018.00705 transcriptional redox modifications in proteins (Meyer et al., 2012). Oxidized TRXs are generally

Frontiers in Plant Science | www.frontiersin.org 1 May 2018 | Volume 9 | Article 705 Martins et al. Nuclear Redox Signaling in Plants reduced by NADPH-dependent thioredoxin reductases (NTRs), thiol-containing tripeptide (γ-glutamyl-cysteinyl-glycine). whereas the reduction of GRXs is dependent on glutathione. Glutathione biosynthesis is performed in chloroplasts and in the Due to their multifunctional thiol reduction capacities, TRXs cytosol but is found in almost all cell compartments, including and GRXs have been involved in many metabolic functions, the nucleus. Nuclear pores are generally assumed to allow controlling plant developmental programs and acting as key unrestricted bidirectional diffusion of glutathione across the signaling molecules in response to abiotic and biotic stresses nuclear envelope. Therefore, nuclear glutathione translocation (Meyer et al., 2009; Rouhier et al., 2015). In this mini-review, to the nucleus can be passive. Glutathione quantification we aim to give an updated overview of nuclear thiol-based ROS at the subcellular level is technically challenging due to its signaling in plants. highly dynamic compartmentation (Delorme-Hinoux et al., 2016). Thiol-specific dyes and genetically encoded probes such as reduction-oxidation-sensitive green fluorescent proteins ROS AND Cys Ox-PTMs IN THE (roGFPs) have consistently detected glutathione in the nucleus. NUCLEUS Immunocytochemistry (ICC) coupled to electronic microscopy were also used to address its location at a subcompartment level Some data suggest that ROS are actively generated in the nucleus (Zechmann et al., 2008; Zechmann and Müller, 2010). This (Ashtamker et al., 2007), but they principally accumulate in study found glutathione uniformly spread in the nucleoplasm, the nucleus through transfer from other cell compartments. without distinction between euchromatin and heterochromatin. Genetically encoded fluorescent H2O2 sensors (e.g., HyPer) have In Arabidopsis thaliana, a glutathione reductase (GR1) is found consistently shown that cytosolic H2O2 freely diffuses in the in the nucleus, suggesting that oxidized glutathione (GSSG) is nucleus through nuclear pores (Møller et al., 2007; Rodrigues actively reduced in the nucleus (Delorme-Hinoux et al., 2016). et al., 2017). It is also transferred from the chloroplasts to the The functions of glutathione in the nucleus are still nucleus under pathogen and high light (HL) conditions (Caplan poorly understood. Thiol-labeling experiments using the 5- et al., 2015; Exposito-Rodriguez et al., 2017). chloromethylfluorescein diacetate (CMFDA) dye and glutathione The chemical characteristics of the sulfur atom make Cys redox state measured by roGFP, suggest that a redox cycle is and Met residues major sites of oxidation within proteins occurring during the cell cycle progression and is critical for (Davies, 2005). Depending on their pKa and on the pH of cell cycle progression (Diaz Vivancos et al., 2010; Vivancos the medium, thiol residues are deprotonated into a thiolate et al., 2010; de Simone et al., 2017). Consistently, sustained mild − residue (R-S ) which is prone to oxidation. This is leading to oxidation observed in ascorbate mutants also restricts nuclear successive oxidations to sulfenic (R-SOH), sulfinic (R-SO2H), functions and impairs progression through the cell cycle (de and sulfonic (R-SO3H) acids (Davies, 2005). Thiol groups Simone et al., 2017). More than acting as a general redox buffer, can also form a disulfide bridge (S-S) or react with reactive glutathione could also provide reducing moiety for anti-oxidant nitrogen species (RNS) or oxidized glutathione (GSSG) resulting enzymes like GRXs (i.e., GRXC1, GRXC2, GRXS17, GRXS13, in S-nitrosylation (R-SNO) or S-glutathionylation (R-S-SG). and ROXY1/2/4) and Glutathione S-Transferases (i.e., GSTF5- Depending on their nature, most of these thiol modifications 10, GSTU19/20, and GSTT19/20), several of them having been can be reversed by dedicated thiol reduction systems (TRX, identified in the nucleus (Rouhier et al., 2015; Palm et al., 2016). GRX, and GSNO Reductase) which exhibit disulfide bond, Consistent levels of ascorbate have also been found in the deglutathionylation or denitrosylation activities (Figure 1). Thiol nucleus but little evidence for its nuclear functions has been modifications can alter the structure and/or the activity of many described (Zechmann et al., 2011; Considine and Foyer, 2014; de proteins like transcription factors, MAP kinases, and chromatin Simone et al., 2017; Zechmann, 2017). modification proteins (see below). Proteomic approaches aiming to identify oxidized thiol targets have been developed in Thiols Reductases in the Nucleus plants. Hundreds of nuclear candidate proteins were found Thioredoxins (TRXs) and Glutaredoxins (GRXs) are major sulfenylated, nitrosylated, or glutathionylated (Supplementary classes of thiols reductases. Plants harbor a complex TRX and Table 1; Zaffagnini et al., 2012, 2016; Morisse et al., 2014; GRX network (Meyer et al., 2012). Among the 40 TRXs and Waszczak et al., 2014; Chaki et al., 2015; Pérez-Pérez et al., 50 GRXs isoforms were found in Arabidopsis, at least 4–8 2017). These approaches also revealed the complexity of the thiol of them have been assigned to the nucleus, although often redox modification networks in plants (Pérez-Pérez et al., 2017). in association with a cytosolic localization (Delorme-Hinoux However, among all these candidates, redox regulation has been et al., 2016). Moreover, NTR isoforms were also found in the validated only in a few cases (see below). nucleus, where they reduce TRXh, TRXo1, and Nucleoredoxin 1 (NRX1) (Serrato et al., 2001; Serrato and Cejudo, 2003; Marty et al., 2009; Marchal et al., 2014). Some thiol reductases are THIOL REDOX SYSTEMS IN THE constitutively located in the nucleus, but others were found to NUCLEUS shuttle between the cytosol and nucleus. In tomato subjected to heat stress, the predominant cytosolic GRXS17 was found Glutathione to relocate in the nucleus (Wu et al., 2012). In wheat and Plants exhibit a large panel of thiol reduction systems (Meyer Chlamydomonas, TRXh isoforms accumulate in the nucleus et al., 2012). Among them is glutathione, a low molecular weight upon oxidative or genotoxic stress (Serrato and Cejudo, 2003;

Frontiers in Plant Science | www.frontiersin.org 2 May 2018 | Volume 9 | Article 705 Martins et al. Nuclear Redox Signaling in Plants

FIGURE 1 | H2O2-induced thiol modifications and scavenging activities in the nucleus. H2O2 accumulating in the nucleus can be detoxified by a NTR/TRX/PRX system. While many other H2O2 detoxification enzymes are active in plants (e.g., catalases and ascorbate peroxidases), their presence in the nucleus is not demonstrated yet. They are not represented here. H2O2 can oxidize thiol residues in protein. Sulfenic acid reacts with GSH, NO or with adjacent thiol residues. Putative nuclear proteins prone to S-glutathionylated, S-nitrosylated, or disulfide bonds formation have been identified by proteomic and biochemical approaches (see Supplementary Table 1; Delorme-Hinoux et al., 2016; Pérez-Pérez et al., 2017). S-glutathionylated, S-nitrosylated, or intra/intermolecular disulfide bonds can be reduced by NTR/TRX, GR/GSH/GRX, or GSNOR. NTR, NADPH-dependent thioredoxin reductase; TRX, thioredoxin; GR, glutathione reductase; GRX, glutaredoxin; PRX, peroxiredoxin; GSNOR, S-nitrosoglutathione reductase; GSH, glutathione; NO, nitric oxide; H2O2, hydrogen peroxide.

Sarkar et al., 2005). Little is known about the subnuclear thiol-containing target proteins are shown as well. Most of these localization of these respective proteins. This is due to the components have been recently reviewed by Delorme-Hinoux low resolution of localization techniques like ICC and GFP- et al. (2016) and will not be further discussed here. fusion coupled with confocal microscopy analyses. In most The nucleolus, a nuclear subcompartment responsible for cases, these proteins are detected in the nucleoplasm, without rRNA biosynthesis, might also be subjected to redox regulation distinction between heterochromatin and euchromatin, and are (Saez-Vasquez and Medina, 2009). Significant accumulation apparently excluded from the nucleolus. Recently, ICC analyses of H2O2 has been detected in the nucleolus in tobacco cell have detected NRX1 and NTRA in the nucleolar cavity, but suspension subjected to elicitor treatments (Ashtamker et al., the functional significance of this localization is still unknown 2007). Intriguingly, the nucleolus also accumulates high amounts (Marchal et al., 2014). Major thiol redox components found in the of iron, which might provide substrates for ROS generation nucleus are presented in Figure 2. ROS scavenging enzymes and by Fenton reactions (Roschzttardtz et al., 2011). In addition,

Frontiers in Plant Science | www.frontiersin.org 3 May 2018 | Volume 9 | Article 705 Martins et al. Nuclear Redox Signaling in Plants

FIGURE 2 | Major thiol redox components found in the nucleus. Thiol reduction systems, thiol-containing target proteins were represented as well as ROS scavenging enzymes. Proteins which were only found in proteomic data (according to Montrichard et al., 2009; Waszczak et al., 2014; Delorme-Hinoux et al., 2016; Palm et al., 2016; Montacié et al., 2017; Pérez-Pérez et al., 2017) but not further validated in the nucleus were represented in italic. NTR, NADPH-dependent thioredoxin reductase; TRX, thioredoxin; GR, glutathione reductase; GRX, glutaredoxin; PRX, peroxiredoxin; PDI, protein disulfide isomerase; GST, glutathione S-transferase; APX, ascorbate peroxidase; DHAR, dehydroascorbate reductase; Ran-GTPase, Ran-guanosine triphosphatase; PCNA, proliferating cell nuclear antigen; GAPDH-C, glyceraldehyde 3-phosphate dehydrogenase-C; RTL1, RNAse III-like 1; TGA, TGA-transcription factor; NF-YC11, nuclear factor-YC11; GSH, glutathione; NO, nitric oxide; H2O2, hydrogen peroxide. glutathione and several isoforms of PRXs, DHARs, APXs, TRXs, Among the molecules involved in the retrograde signaling and GRX-like proteins were enriched in this compartment (Suzuki et al., 2012; Vogel et al., 2014; Dietz et al., 2016), singlet 1 1 (Zechmann et al., 2008; Zechmann and Müller, 2010; Palm oxygen ( O2) induces expression of subsets of O2-responsive et al., 2016; Montacié et al., 2017). Whether redox activities are genes and enhances tolerance to HL and to other abiotic and occurring in the nucleolus will need further investigations. biotic stress (Wagner et al., 2004; Carmody et al., 2016). H2O2 originating by dismutation of superoxide in the chloroplast has also been recently shown to be involved in retrograde REDOX-REGULATED NUCLEAR signaling upon HL exposure (Exposito-Rodriguez et al., 2017). FUNCTIONS Another retrograde signaling was suggested to involve a redox regulation of the chloroplastic cyclophilin Cyp20.3, leading to Transcriptomic Control by ROS stimulation of Cys synthesis, accumulation of non-protein thiols Reactive oxygen species causes drastic changes in nuclear gene and activation of defense gene expression (Dominguez-Solis expression (Gadjev et al., 2006; Willems et al., 2016; Shaikhali et al., 2008; Park et al., 2013). Presumably, ROS generated in other and Wingsle, 2017). Oxidative stress affects many pathways cell compartments (mitochondria, peroxisomes, and apoplast) involved in RNA processing, including splicing, polyadenylation, can also exert similar retrograde signaling (Noctor and Foyer, exporting, and editing. It is also involved in RNA degradation 2016; Rodríguez-Serrano et al., 2016). and protein translation (Van Ruyskensvelde et al., 2018). Under Photorespiration produces H2O2 in peroxisomes. In this High-light (HL) conditions, ROS originated in chloroplasts are compartment, catalases play an important role in removing associated with chloroplast-to-nucleus (retrograde) signaling. H2O2. The cat2 mutant inactivated in the major peroxisomal

Frontiers in Plant Science | www.frontiersin.org 4 May 2018 | Volume 9 | Article 705 Martins et al. Nuclear Redox Signaling in Plants

catalase accumulates a high level of peroxisomal H2O2 and are occurring in the nucleus and affect TGA-regulated gene impacts nuclear gene expression extensively, rapidly inducing expression (Xing et al., 2005; Xing and Zachgo, 2008; Li et al., subsets of stress and hormonal response genes (Queval et al., 2009, 2011; Murmu et al., 2010; reviewed by Dietz, 2014 and 2007). In this case, regulation of gene expression involves Delorme-Hinoux et al., 2016). glutathione signaling also, as the transcriptomic response is R2R3-type MYB transcription factors from maize require partly abolished in a glutathione-defective (cad2) cat2 cad2 reducing conditions for DNA binding. Under non-reducing double mutant (Han et al., 2013a,b). A signaling role of conditions, Cys49 and Cys53 form a disulfide bond that prevents glutathione on nuclear gene expression was also suggested the R2R3 MYB domain from binding DNA (Williams and by transcriptomic data in genetically or pharmacologically Grotewold, 1997; Heine et al., 2004). More recently, the structure manipulated glutathione backgrounds (Xiang and Oliver, 1998; and the DNA binding activity of a AtMYB30 transcription Ball et al., 2004; Schnaubelt et al., 2015). Other transcriptomic factor were shown to be influenced by S-nitrosylation analyses also showed involvement of thiol reduction systems (Tavares et al., 2014). in modulating nuclear gene expression (Bashandy et al., 2009; AP2/ethylene response factor (ERF) is another class of Martins et al., unpublished data). Whether these actors are transcription factors which undergoes redox regulation (Welsch directly involved in gene expression needs further investigations. et al., 2007; Shaikhali et al., 2008; Vogel et al., 2014). One of the most striking examples was described for the Rap2.12-dependent regulation of hypoxia response genes. Under aerobic conditions, Redox Regulation of Transcription Rap2.12 is bound to the plasma membrane within an acyl-CoA Factors binding protein 1 or 2 (ACBP1/2) complex. In low oxygen, A likely impact of ROS on nuclear gene expression relies on Rap2.12 is released from the plasma membrane by a mechanism the regulation of redox-sensitive transcription factors (Considine involving a N-terminal Cys2 residue, and is translocated to the and Foyer, 2014; Dietz, 2014; Rouhier et al., 2015; Waszczak et al., nucleus where it activates hypoxia response genes (Gibbs et al., 2015). In most cases, redox regulation induces conformation 2011; Licausi et al., 2011; Licausi, 2013). changes in transcription factors or associated proteins. Such Comelli and Gonzalez (2007) also reported a redox regulation modifications can occur in the cytosol and trigger nuclear of conserved Cys in the homeodomain (HD) DNA of plant class translocation, e.g., by uncovering of a nuclear localization III HD-Zip proteins. Here, DNA binding capacities are only sequence (NLS). A well-documented example is the thiol redox- maintained when an intramolecular disulfide bond is reduced by dependent nuclear translocation of the glycolytic enzyme Glucose a thioredoxin (Comelli and Gonzalez, 2007). A Cys-dependent 6-Phosphate Dehydrogenase C (GAPDH-C) which impacts redox regulation of the DNA binding activity of basic region both its metabolic activity and its moonlighting function as leucine zipper (bZIP) transcription factors has also been reported a transcriptional activator of glycolytic genes (Holtgrefe et al., (Shaikhali et al., 2012). 2008; Vescovi et al., 2013; Zaffagnini et al., 2013; Testard et al., Finally, a subunit of the Nuclear Factor-Y (NF-Y) 2016; Zhang et al., 2017). The HL- and H2O2-dependent nuclear transcription factor complex (NY-YC11) physically interacts translocation of Heat-Shock Factors (HSFA1D and HSFA8) is with the iron-sulfur cluster glutaredoxin GRXS17 in the nucleus. also dependent on specific Cys residues (Miller and Mittler, It is not known yet if this interaction is redox-dependent 2006; Jung et al., 2013; Giesguth et al., 2015; Dickinson et al., (Knuesting et al., 2015). In the cytosol and the nucleus, GRXS17 2018). The pathogen-induced Salicylic Acid (SA)-dependent also interacts with and reduces BolA2, a factor involved in iron transcriptional response is mediated by redox-dependent nuclear metabolism (Couturier et al., 2014; Qin et al., 2015). translocation of NON-EXPRESSOR OF PR GENES1 (NPR1). In this particular case, NPR1 is kept in the cytosol in a disulfide- Epigenetic Regulation bound oligomeric homocomplex. Upon pathogen attack, SA Redox regulation of epigenetic processes has mostly been induces TRXh5 expression which counteracts NPR1 oligomer addressed in mammals (García-Giménez et al., 2017), but this formation by reducing NPR1 disulfides. Moreover, through its field is poorly explored in plants (Delorme-Hinoux et al., 2016; denitrosylase activity, TRXh5 also suppresses the stimulatory Shen et al., 2016). Nevertheless, due to the ubiquity of these effect of Cys156 S-nitrosylation on formation of disulfide-linked basic mechanisms in living organisms, it is likely that such NPR1 oligomer (Tada et al., 2008; Kneeshaw et al., 2014). NPR1 is regulation occurs in plants as well. Different enzymes involved in translocated in the nucleus where it promotes PR gene expression histone methylation are prone to redox regulation, affecting both through interaction with TGA transcription factors such as positive and negative histone marks (e.g., H3K4me2, H3K4me3, TGA1 (Després et al., 2003; Mou et al., 2003; Tada et al., 2008; H3K79me3, H3K27me2, and H3K9me2) (Chen et al., 2006; Zhou Kneeshaw et al., 2014). et al., 2008, 2010; Niu et al., 2015). In mammals, nuclear histone Indeed, other members of the TGA transcription factors are acetylation activities are redox sensitive, affecting chromatin likely redox regulated in the nucleus. Among the 10 TGA factors conformation and transcription (Ito et al., 2004; Ago et al., found in Arabidopsis, several of them (i.e., TGA1, TGA2, TGA3, 2008; Nott et al., 2008; Doyle and Fitzpatrick, 2010). During TGA7, and Perianthia) interact with type III GRXs (ROXY1 brain development, neurotrophic factors induce S-nitrosylation and 2) and are involved in the development of petals, anthers at conserved Cys of HDAC2 in neurons, resulting in changes and microspores. Although the redox dependent control of of histone modification and gene expression (Nott et al., 2008). these TGA is not fully established, ROXY/TGA interactions Upon cardiac hypertrophy, a ROS/TRX-dependent redox switch

Frontiers in Plant Science | www.frontiersin.org 5 May 2018 | Volume 9 | Article 705 Martins et al. Nuclear Redox Signaling in Plants of key Cys residues affects nuclear trafficking of a class II HDAC CONCLUSION AND PERSPECTIVES and subsequent gene expression (Ago et al., 2008). Within the large family of HDAC identified in plants (Pandey et al., 2002), Data supporting the role of redox regulation in nuclear functions members of the class I RPD-3 like HDAC (HDAC9, 19) have been are rapidly increasing. ROS are key actors of this regulation, shown to be sensitive to oxidation (Liu et al., 2015; Mengel et al., influencing gene expression at multiple levels (transcription 2017), but the physiological significance of those modifications and post-transcription), notably by modulating activities of is still poorly understood. NO-induced HDAC inhibition is transcription regulators. While H2O2 has been detected in the proposed to operate in plant stress response by facilitating the nucleus, little is known about ROS metabolism and dynamics stress-induced transcription of genes (Mengel et al., 2017). in this cell compartment. The recent discovery of a H2O2 flux In addition to methylation and acetylation, mammalian from the chloroplast to the nucleus opens new perspectives histone H3 has been shown to be glutathionylated on a to decipher the role of ROS in gene expression. In this way, conserved and unique Cys residue (García-Giménez et al., 2014). the identification of a redox regulation of key transcriptional Histone H3 glutathionylation increases during cell proliferation regulators (e.g., transcription factors, HDAC) will shed light on and decreases during aging. This produces structural changes the ways ROS are acting on gene expression. Key for these affecting nucleosome stability and leading to a more open questions are the proteomic approaches aiming to identify chromatin structure (García-Giménez et al., 2013, 2014, 2017; Xu nuclear PTM occurring on redox-sensitive residues, and the et al., 2014). structure biology techniques designed to visualize redox-based Small RNAs (siRNA and miRNA) are key regulators of gene modifications on protein structure (Waszczak et al., 2014; Parker expression, involved in most developmental and stress response et al., 2015; Zaffagnini et al., 2016; Pérez-Pérez et al., 2017). While processes in eukaryotic cells (Leisegang et al., 2017). Biogenesis those redox proteome approaches have identified hundreds of of small RNAs is orchestrated by DICER-LIKE (DCL) and nuclear proteins which could be prone to redox modifications, RNASE THREE-LIKE (RTL) endonucleases that process almost biochemical and functional evidence is missing to support the every class of double-stranded RNA precursors. Charbonnel biological significance of these redox switches. This will be a et al. (2017) have recently demonstrated that members of DCL major challenge for future research in redox biology. and RTL families in Arabidopsis are glutathionylated on a conserved Cys which affects their RNase III activity. R-S-SG of RTL1 is reversed by type I GRXs, suggesting that small RNA biogenesis and subsequent gene expression responses are under AUTHOR CONTRIBUTIONS the control of the cell redox environment (Charbonnel et al., 2017). Indeed, the RNase activity of another member of the family LM, JT-H and J-PR wrote the paper. LM and JT-H performed the fi (RTL2) was previously shown to be regulated by its dimerization gures. state through an intermolecular disulfide bond (Comella et al., 2008), showing that a redox switch might regulate small RNA biogenesis. FUNDING Epigenetic regulation of gene expression is performed by DNA methylation. Some key metabolic enzymes involved in DNA This work was supported by the Centre National de la Recherche methylation are suspected to be redox regulated. Among them Scientifique, the Agence Nationale de la Recherche (ANR-Blanc are enzymes of the S-Adenosyl Methionine (SAM) cycle which Cynthiol 12-BSV6-0011). LM was supported by a Ph.D. grant provide precursors for DNA and histone methylation (Shen from the Université de Perpignan Via Domitia (Ecole Doctorale et al., 2016). Other nuclear candidates are the DNA demethylases Energie et Environnement ED305). JT-H was supported by a Repressor of Silencing1 (ROS1) and Demeter-like (DME, DML2, Ph.D. grant from Consejo Nacional de Ciencia y Tecnología and DML3) enzymes which remove methylated bases from the (CONACYT: 254639/411761). DNA backbone (Zhu, 2009). All these enzymes contain an iron- sulfur (Fe-S) cluster which might be susceptible to oxidation by ROS. Moreover, different members of the cytosolic Fe-S cluster assembly machinery (i.e., MET18 and AE7) are involved in SUPPLEMENTARY MATERIAL DNA methylation, likely because they affect the nuclear DNA demethylases Fe-S cluster metabolism (Luo et al., 2012; Duan The Supplementary Material for this article can be found online et al., 2015). Therefore, all these examples show an emerging link at: https://www.frontiersin.org/articles/10.3389/fpls.2018.00705/ between redox regulation and epigenetic regulation. full#supplementary-material

REFERENCES Ashtamker, C., Kiss, V., Sagi, M., Davydov, O., and Fluhr, R. (2007). Diverse subcellular locations of cryptogein-induced reactive oxygen species production Ago, T., Liu, T., Zhai, P., Chen, W., Li, H., Molkentin, J. D., et al. in tobacco Bright Yellow-2 cells. Plant Physiol. 143, 1817–1826. doi: 10.1104/ (2008). A redox-dependent pathway for regulating class II HDACs pp.106.090902 and cardiac hypertrophy. Cell 133, 978–993. doi: 10.1016/j.cell.2008. Ball, L., Accotto, G.-P., Bechtold, U., Creissen, G., Funck, D., Jimenez, A., et al. 04.041 (2004). Evidence for a direct link between glutathione biosynthesis and stress

Frontiers in Plant Science | www.frontiersin.org 6 May 2018 | Volume 9 | Article 705 Martins et al. Nuclear Redox Signaling in Plants

defense gene expression in Arabidopsis. Plant Cell 16, 2448–2462. doi: 10.1105/ stress responses in chloroplasts. Proc. Natl. Acad. Sci. U.S.A. 105, 16386–16391. tpc.104.022608 doi: 10.1073/pnas.0808204105 Bashandy, T., Taconnat, L., Renou, J.-P., Meyer, Y., and Reichheld, J.-P. (2009). Doyle, K., and Fitzpatrick, F. A. (2010). Redox signaling, alkylation (carbonylation) Accumulation of flavonoids in an ntra ntrb mutant leads to tolerance to UV-C. of conserved cysteines inactivates class I histone deacetylases 1, 2, and 3 Mol. Plant 2, 249–258. doi: 10.1093/mp/ssn065 and antagonizes their transcriptional repressor function. J. Biol. Chem. 285, Caplan, J. L., Kumar, A. S., Park, E., Padmanabhan, M. S., Hoban, K., Modla, S., 17417–17424. doi: 10.1074/jbc.M109.089250 et al. (2015). Chloroplast stromules function during innate immunity. Dev. Cell Duan, C.-G., Wang, X., Tang, K., Zhang, H., Mangrauthia, S. K., Lei, M., 34, 45–57. doi: 10.1016/j.devcel.2015.05.011 et al. (2015). MET18 connects the cytosolic iron-sulfur cluster assembly Carmody, M., Crisp, P. A., d’Alessandro, S., Ganguly, D., Gordon, M., Havaux, M., pathway to active DNA demethylation in Arabidopsis. PLoS Genet. 11:e1005559. et al. (2016). Uncoupling high light responses from singlet oxygen retrograde doi: 10.1371/journal.pgen.1005559 signaling and spatial-temporal systemic acquired acclimation. Plant Physiol. Exposito-Rodriguez, M., Laissue, P. P., Yvon-Durocher, G., Smirnoff, N., and 171, 1734–1749. doi: 10.1104/pp.16.00404 Mullineaux, P. M. (2017). Photosynthesis-dependent H2O2 transfer from Chaki, M., Shekariesfahlan, A., Ageeva, A., Mengel, A., von Toerne, C., Durner, J., chloroplasts to nuclei provides a high-light signalling mechanism. Nat. et al. (2015). Identification of nuclear target proteins for S-nitrosylation in Commun. 8:49. doi: 10.1038/s41467-017-00074-w pathogen-treated Arabidopsis thaliana cell cultures. Plant Sci. 238, 115–126. Foyer, C. H., and Noctor, G. (2016). Stress-triggered redox signalling: what’s in doi: 10.1016/j.plantsci.2015.06.011 pROSpect? Plant Cell Environ. 39, 951–964. doi: 10.1111/pce.12621 Charbonnel, C., Niazi, A. K., Elvira-Matelot, E., Nowak, E., Zytnicki, M., de Gadjev, I., Vanderauwera, S., Gechev, T. S., Laloi, C., Minkov, I. N., Shulaev, V., Bures, A., et al. (2017). The siRNA suppressor RTL1 is redox-regulated through et al. (2006). Transcriptomic footprints disclose specificity of reactive oxygen glutathionylation of a conserved cysteine in the double-stranded-RNA-binding species signaling in Arabidopsis. Plant Physiol. 141, 436–445. doi: 10.1104/pp. domain. Nucleic Acids Res. 45, 11891–11907. doi: 10.1093/nar/gkx820 106.078717 Chen, H., Ke, Q., Kluz, T., Yan, Y., and Costa, M. (2006). Nickel ions increase García-Giménez, J. L., Ibañez-Cabellos, J. S., Seco-Cervera, M., and Pallardó, F. V. histone H3 lysine 9 dimethylation and induce transgene silencing. Mol. Cell. (2014). Glutathione and cellular redox control in epigenetic regulation. Free Biol. 26, 3728–3737. doi: 10.1128/MCB.26.10.3728-3737.2006 Radic. Biol. Med. 75(Suppl. 1):S3. doi: 10.1016/j.freeradbiomed.2014.10.828 Choudhury, F. K., Rivero, R. M., Blumwald, E., and Mittler, R. (2017). Reactive García-Giménez, J. L., Òlaso, G., Hake, S. B., Bönisch, C., Wiedemann, S. M., oxygen species, abiotic stress and stress combination. Plant J. 90, 856–867. Markovic, J., et al. (2013). Histone h3 glutathionylation in proliferating doi: 10.1111/tpj.13299 mammalian cells destabilizes nucleosomal structure. Antioxid. Redox Signal. 19, Comella, P., Pontvianne, F., Lahmy, S., Vignols, F., Barbezier, N., Debures, A., et al. 1305–1320. doi: 10.1089/ars.2012.5021 (2008). Characterization of a ribonuclease III-like protein required for cleavage García-Giménez, J. L., Romá-Mateo, C., Pérez-Machado, G., Peiró-Chova, L., and of the pre-rRNA in the 3’ETS in Arabidopsis. Nucleic Acids Res. 36, 1163–1175. Pallardó, F. V. (2017). Role of glutathione in the regulation of epigenetic doi: 10.1093/nar/gkm1130 mechanisms in disease. Free Radic. Biol. Med. 112, 36–48. doi: 10.1016/j. Comelli, R. N., and Gonzalez, D. H. (2007). Conserved homeodomain cysteines freeradbiomed.2017.07.008 confer redox sensitivity and influence the DNA binding properties of plant class Gibbs, D. J., Lee, S. C., Isa, N. M., Gramuglia, S., Fukao, T., Bassel, G. W., III HD-Zip proteins. Arch. Biochem. Biophys. 467, 41–47. doi: 10.1016/j.abb. et al. (2011). Homeostatic response to hypoxia is regulated by the N-end rule 2007.08.003 pathway in plants. Nature 479, 415–418. doi: 10.1038/nature10534 Considine, M. J., and Foyer, C. H. (2014). Redox regulation of plant development. Giesguth, M., Sahm, A., Simon, S., and Dietz, K.-J. (2015). Redox-dependent Antioxid. Redox Signal. 21, 1305–1326. doi: 10.1089/ars.2013.5665 translocation of the heat shock transcription factor AtHSFA8 from the cytosol Couturier, J., Wu, H.-C., Dhalleine, T., Pégeot, H., Sudre, D., Gualberto, J. M., et al. to the nucleus in Arabidopsis thaliana. FEBS Lett. 589, 718–725. doi: 10.1016/j. (2014). Monothiol glutaredoxin-BolA interactions: redox control of Arabidopsis febslet.2015.01.039 thaliana BolA2 and SufE1. Mol. Plant 7, 187–205. doi: 10.1093/mp/ss Han, Y., Chaouch, S., Mhamdi, A., Queval, G., Zechmann, B., and Noctor, G. t156 (2013a). Functional analysis of Arabidopsis mutants points to novel roles for Davies, M. J. (2005). The oxidative environment and protein damage. Biochim. glutathione in coupling H2O2 to activation of salicylic acid accumulation and Biophys. Acta 1703, 93–109. doi: 10.1016/j.bbapap.2004.08.007 signaling. Antioxid. Redox Signal. 18, 2106–2121. doi: 10.1089/ars.2012.5052 de Simone, A., Hubbard, R., de la Torre, N. V., Velappan, Y., Wilson, M., Han, Y., Mhamdi, A., Chaouch, S., and Noctor, G. (2013b). Regulation of basal and Considine, M. J., et al. (2017). Redox changes during the cell cycle in the oxidative stress-triggered jasmonic acid-related gene expression by glutathione. embryonic root meristem of Arabidopsis thaliana. Antioxid. Redox Signal. 27, Plant Cell Environ. 36, 1135–1146. doi: 10.1111/pce.12048 1505–1519. doi: 10.1089/ars.2016.6959 Heine, G. F., Hernandez, J. M., and Grotewold, E. (2004). Two cysteines in plant Delorme-Hinoux, V., Bangash, S. A. K., Meyer, A. J., and Reichheld, J.-P. (2016). R2R3 MYB domains participate in REDOX-dependent DNA binding. J. Biol. Nuclear thiol redox systems in plants. Plant Sci. 243, 84–95. doi: 10.1016/j. Chem. 279, 37878–37885. doi: 10.1074/jbc.M405166200 plantsci.2015.12.002 Holtgrefe, S., Gohlke, J., Starmann, J., Druce, S., Klocke, S., Altmann, B., Després, C., Chubak, C., Rochon, A., Clark, R., Bethune, T., Desveaux, et al. (2008). Regulation of plant cytosolic glyceraldehyde 3-phosphate D., et al. (2003). The Arabidopsis NPR1 disease resistance protein is a dehydrogenase isoforms by thiol modifications. Physiol. Plant. 133, 211–228. novel cofactor that confers redox regulation of DNA binding activity to doi: 10.1111/j.1399-3054.2008.01066.x the basic domain/leucine zipper transcription factor TGA1. Plant Cell 15, Ito, K., Hanazawa, T., Tomita, K., Barnes, P. J., and Adcock, I. M. (2004). Oxidative 2181–2191. doi: 10.1105/tpc.012849 stress reduces histone deacetylase 2 activity and enhances IL-8 gene expression: Diaz Vivancos, P., Wolff, T., Markovic, J., Pallardó, F. V., and Foyer, C. H. (2010). role of tyrosine nitration. Biochem. Biophys. Res. Commun. 315, 240–245. A nuclear glutathione cycle within the cell cycle. Biochem. J. 431, 169–178. doi: 10.1016/j.bbrc.2004.01.046 doi: 10.1042/BJ20100409 Jung, H.-S., Crisp, P. A., Estavillo, G. M., Cole, B., Hong, F., Mockler, T. C., Dickinson, P. J., Kumar, M., Martinho, C., Yoo, S. J., Lan, H., Artavanis, G., et al. et al. (2013). Subset of heat-shock transcription factors required for the early (2018). Chloroplast signaling gates thermotolerance in Arabidopsis. Cell Rep. 22, response of Arabidopsis to excess light. Proc. Natl. Acad. Sci. U.S.A. 110, 1657–1665. doi: 10.1016/j.celrep.2018.01.054 14474–14479. doi: 10.1073/pnas.1311632110 Dietz, K.-J. (2014). Redox regulation of transcription factors in plant stress Kneeshaw, S., Gelineau, S., Tada, Y., Loake, G. J., and Spoel, S. H. (2014). Selective acclimation and development. Antioxid. Redox Signal. 21, 1356–1372. doi: 10. protein denitrosylation activity of Thioredoxin-h5 modulates plant Immunity. 1089/ars.2013.5672 Mol. Cell 56, 153–162. doi: 10.1016/j.molcel.2014.08.003 Dietz, K.-J., Mittler, R., and Noctor, G. (2016). Recent progress in understanding Knuesting, J., Riondet, C., Maria, C., Kruse, I., Bécuwe, N., König, N., et al. the role of reactive oxygen species in plant cell signaling. Plant Physiol. 171, (2015). Arabidopsis glutaredoxin S17 and its partner, the nuclear factor Y 1535–1539. doi: 10.1104/pp.16.00938 subunit C11/negative cofactor 2α, contribute to maintenance of the shoot Dominguez-Solis, J. R., He, Z., Lima, A., Ting, J., Buchanan, B. B., and Luan, S. apical meristem under long-day photoperiod. Plant Physiol. 167, 1643–1658. (2008). A cyclophilin links redox and light signals to cysteine biosynthesis and doi: 10.1104/pp.15.00049

Frontiers in Plant Science | www.frontiersin.org 7 May 2018 | Volume 9 | Article 705 Martins et al. Nuclear Redox Signaling in Plants

Leisegang, M. S., Schröder, K., and Brandes, R. P. (2017). Redox regulation and Noctor, G., and Foyer, C. H. (2016). Intracellular redox compartmentation and noncoding RNAs. Antioxid. Redox Signal. doi: 10.1089/ars.2017.7276 [Epub ROS-related communication in regulation and signaling. Plant Physiol. 171, ahead of print]. 1581–1592. doi: 10.1104/pp.16.00346 ∗∗ Li, S., Gutsche, N., and Zachgo, S. (2011). The ROXY1 C-terminal L LL motif is Nott, A., Watson, P. M., Robinson, J. D., Crepaldi, L., and Riccio, A. (2008). essential for the interaction with TGA transcription factors. Plant Physiol. 157, S-Nitrosylation of histone deacetylase 2 induces chromatin remodelling in 2056–2068. doi: 10.1104/pp.111.185199 neurons. Nature 455, 411–415. doi: 10.1038/nature07238 Li, S., Lauri, A., Ziemann, M., Busch, A., Bhave, M., and Zachgo, S. (2009). Palm, D., Simm, S., Darm, K., Weis, B. L., Ruprecht, M., Schleiff, E., et al. Nuclear activity of ROXY1, a glutaredoxin interacting with TGA factors, is (2016). Proteome distribution between nucleoplasm and nucleolus and its required for petal development in Arabidopsis thaliana. Plant Cell 21, 429–441. relation to ribosome biogenesis in Arabidopsis thaliana. RNA Biol. 13, 441–454. doi: 10.1105/tpc.108.064477 doi: 10.1080/15476286.2016.1154252 Licausi, F. (2013). Molecular elements of low-oxygen signaling in plants. Physiol. Pandey, R., Müller, A., Napoli, C. A., Selinger, D. A., Pikaard, C. S., Richards, E. Plant. 148, 1–8. doi: 10.1111/ppl.12011 J., et al. (2002). Analysis of histone acetyltransferase and histone deacetylase Licausi, F., Kosmacz, M., Weits, D. A., Giuntoli, B., Giorgi, F. M., Voesenek, L. A., families of Arabidopsis thaliana suggests functional diversification of chromatin et al. (2011). Oxygen sensing in plants is mediated by an N-end rule pathway modification among multicellular eukaryotes. Nucleic Acids Res. 30, 5036–5055. for protein destabilization. Nature 479, 419–422. doi: 10.1038/nature10536 Park, S.-W., Li, W., Viehhauser, A., He, B., Kim, S., Nilsson, A. K., et al. Liu, P., Zhang, H., Yu, B., Xiong, L., and Xia, Y. (2015). Proteomic identification (2013). Cyclophilin 20-3 relays a 12-oxo-phytodienoic acid signal during stress of early salicylate- and flg22-responsive redox-sensitive proteins in Arabidopsis. responsive regulation of cellular redox homeostasis. Proc. Natl. Acad. Sci. U.S.A. Sci. Rep. 5:8625. doi: 10.1038/srep08625 110, 9559–9564. doi: 10.1073/pnas.1218872110 Luo, D., Bernard, D. G., Balk, J., Hai, H., and Cui, X. (2012). The DUF59 family Parker, J., Balmant, K., Zhu, F., Zhu, N., and Chen, S. (2015). cysTMTRAQ- gene AE7 acts in the cytosolic iron-sulfur cluster assembly pathway to maintain An integrative method for unbiased thiol-based redox proteomics. Mol. Cell. nuclear genome integrity in Arabidopsis. Plant Cell 24, 4135–4148. doi: 10.1105/ Proteomics 14, 237–242. doi: 10.1074/mcp.O114.041772 tpc.112.102608 Pérez-Pérez, M. E., Mauriès, A., Maes, A., Tourasse, N. J., Hamon, M., Lemaire, Marchal, C., Delorme-Hinoux, V., Bariat, L., Siala, W., Belin, C., Saez-Vasquez, J., S. D., et al. (2017). The deep thioredoxome in Chlamydomonas reinhardtii: new et al. (2014). NTR/NRX define a new thioredoxin system in the nucleus of insights into redox regulation. Mol. Plant 10, 1107–1125. doi: 10.1016/j.molp. Arabidopsis thaliana cells. Mol. Plant 7, 30–44. doi: 10.1093/mp/sst162 2017.07.009 Marty, L., Siala, W., Schwarzländer, M., Fricker, M. D., Wirtz, M., Sweetlove, Qin, L., Wang, M., Zuo, J., Feng, X., Liang, X., Wu, Z., et al. (2015). Cytosolic BolA L. J., et al. (2009). The NADPH-dependent thioredoxin system constitutes a plays a repressive role in the tolerance against excess iron and MV-induced functional backup for cytosolic glutathione reductase in Arabidopsis. Proc. Natl. oxidative stress in plants. PLoS One 10:e0124887. doi: 10.1371/journal.pone. Acad. Sci. U.S.A. 106, 9109–9114. doi: 10.1073/pnas.0900206106 0124887 Mengel, A., Ageeva, A., Georgii, E., Bernhardt, J., Wu, K., Durner, J., et al. (2017). Queval, G., Issakidis-Bourguet, E., Hoeberichts, F. A., Vandorpe, M., Gakière, B., Nitric oxide modulates histone acetylation at stress genes by inhibition of Vanacker, H., et al. (2007). Conditional oxidative stress responses in the histone deacetylases. Plant Physiol. 173, 1434–1452. doi: 10.1104/pp.16.01734 Arabidopsis photorespiratory mutant cat2 demonstrate that redox state is a key Meyer, Y., Belin, C., Delorme-Hinoux, V., Reichheld, J.-P., and Riondet, C. (2012). modulator of daylength-dependent gene expression, and define photoperiod Thioredoxin and glutaredoxin systems in plants: molecular mechanisms, as a crucial factor in the regulation of H2O2-induced cell death. Plant J. 52, crosstalks, and functional significance. Antioxid. Redox Signal. 17, 1124–1160. 640–657. doi: 10.1111/j.1365-313X.2007.03263.x doi: 10.1089/ars.2011.4327 Rahantaniaina, M.-S., Li, S., Chatel-Innocenti, G., Tuzet, A., Issakidis-Bourguet, E., Meyer, Y., Buchanan, B. B., Vignols, F., and Reichheld, J.-P. (2009). Thioredoxins Mhamdi, A., et al. (2017). Cytosolic and chloroplastic DHARs cooperate in and glutaredoxins: unifying elements in redox biology. Annu. Rev. Genet. 43, oxidative stress-driven activation of the salicylic acid pathway. Plant Physiol. 335–367. doi: 10.1146/annurev-genet-102108-134201 174, 956–971. doi: 10.1104/pp.17.00317 Miller, G., and Mittler, R. (2006). Could heat shock transcription factors function Rodrigues, O., Reshetnyak, G., Grondin, A., Saijo, Y., Leonhardt, N., Maurel, C., as hydrogen peroxide sensors in plants? Ann. Bot. 98, 279–288. doi: 10.1093/ et al. (2017). Aquaporins facilitate hydrogen peroxide entry into guard cells to aob/mcl107 mediate ABA- and pathogen-triggered stomatal closure. Proc. Natl. Acad. Sci. Møller, I. M., Jensen, P. E., and Hansson, A. (2007). Oxidative modifications to U.S.A. 114, 9200–9205. doi: 10.1073/pnas.1704754114 cellular components in plants. Annu. Rev. Plant Biol. 58, 459–481. doi: 10.1146/ Rodríguez-Serrano, M., Romero-Puertas, M. C., Sanz-Fernández, M., Hu, J., and annurev.arplant.58.032806.103946 Sandalio, L. M. (2016). Peroxisomes extend peroxules in a fast response to stress Montacié, C., Durut, N., Opsomer, A., Palm, D., Comella, P., Picart, C., et al. via a reactive oxygen species-mediated induction of the peroxin PEX11a. Plant (2017). Nucleolar proteome analysis and proteasomal activity assays reveal a Physiol. 171, 1665–1674. doi: 10.1104/pp.16.00648 link between nucleolus and 26S proteasome in A. thaliana. Front. Plant Sci. Roschzttardtz, H., Grillet, L., Isaure, M.-P., Conéjéro, G., Ortega, R., Curie, C., 8:1815. doi: 10.3389/fpls.2017.01815 et al. (2011). Plant cell nucleolus as a hot spot for iron. J. Biol. Chem. 286, Montrichard, F., Alkhalfioui, F., Yano, H., Vensel, W. H., Hurkman, W. J., and 27863–27866. doi: 10.1074/jbc.C111.269720 Buchanan, B. B. (2009). Thioredoxin targets in plants: the first 30 years. J. Rouhier, N., Cerveau, D., Couturier, J., Reichheld, J.-P., and Rey, P. (2015). Proteomics 72, 452–474. doi: 10.1016/j.jprot.2008.12.002 Involvement of thiol-based mechanisms in plant development. Biochim. Morisse, S., Zaffagnini, M., Gao, X.-H., Lemaire, S. D., and Marchand, C. H. (2014). Biophys. Acta 1850, 1479–1496. doi: 10.1016/j.bbagen.2015.01.023 Insight into protein S-nitrosylation in Chlamydomonas reinhardtii. Antioxid. Saez-Vasquez, J., and Medina, F. J. (2009). The plant nucleolus. Adv. Bot. Res. 47, Redox Signal. 21, 1271–1284. doi: 10.1089/ars.2013.5632 1–46. Mou, Z., Fan, W., and Dong, X. (2003). Inducers of plant systemic acquired Sarkar, N., Lemaire, S., Wu-Scharf, D., Issakidis-Bourguet, E., and Cerutti, H. resistance regulate NPR1 function through redox changes. Cell 113, 935–944. (2005). Functional specialization of Chlamydomonas reinhardtii cytosolic doi: 10.1016/S0092-8674(03)00429-X thioredoxin h1 in the response to alkylation-induced DNA damage. Eukaryot. Murmu, J., Bush, M. J., DeLong, C., Li, S., Xu, M., Khan, M., et al. (2010). Cell 4, 262–273. doi: 10.1128/EC.4.2.262-273.2005 Arabidopsis basic leucine-zipper transcription factors TGA9 and TGA10 Schnaubelt, D., Queval, G., Dong, Y., Diaz-Vivancos, P., Makgopa, M. E., interact with floral glutaredoxins ROXY1 and ROXY2 and are redundantly Howell, G., et al. (2015). Low glutathione regulates gene expression and the required for anther development. Plant Physiol. 154, 1492–1504. doi: 10.1104/ redox potentials of the nucleus and cytosol in Arabidopsis thaliana. Plant Cell pp.110.159111 Environ. 38, 266–279. doi: 10.1111/pce.12252 Niu, Y., DesMarais, T. L., Tong, Z., Yao, Y., and Costa, M. (2015). Oxidative stress Serrato, A. J., and Cejudo, F. J. (2003). Type-h thioredoxins accumulate in the alters global histone modification and DNA methylation. Free Radic. Biol. Med. nucleus of developing wheat seed tissues suffering oxidative stress. Planta 217, 82, 22–28. doi: 10.1016/j.freeradbiomed.2015.01.028 392–399. doi: 10.1007/s00425-003-1009-4 Noctor, G. (2017). ROS-related redox regulation and signaling in plants. Semin. Serrato, A. J., Crespo, J. L., Florencio, F. J., and Cejudo, F. J. (2001). Cell Dev. Biol. doi: 10.1016/j.semcdb.2017.07.013 [Epub ahead of print]. Characterization of two thioredoxins h with predominant localization in the

Frontiers in Plant Science | www.frontiersin.org 8 May 2018 | Volume 9 | Article 705 Martins et al. Nuclear Redox Signaling in Plants

nucleus of aleurone and scutellum cells of germinating wheat seeds. Plant Mol. Williams, C. E., and Grotewold, E. (1997). Differences between plant and animal Biol. 46, 361–371. doi: 10.1023/A:1010697331184 Myb domains are fundamental for DNA binding activity, and chimeric Myb Shaikhali, J., Heiber, I., Seidel, T., Ströher, E., Hiltscher, H., Birkmann, S., et al. domains have novel DNA binding specificities. J. Biol. Chem. 272, 563–571. (2008). The redox-sensitive transcription factor Rap2.4a controls nuclear doi: 10.1074/jbc.272.1.563 expression of 2-Cys peroxiredoxin A and other chloroplast antioxidant Wu, Q., Lin, J., Liu, J.-Z., Wang, X., Lim, W., Oh, M., et al. (2012). Ectopic enzymes. BMC Plant Biol. 8:48. doi: 10.1186/1471-2229-8-48 expression of Arabidopsis glutaredoxin AtGRXS17 enhances thermotolerance Shaikhali, J., Norén, L., de Dios Barajas-López, J., Srivastava, V., König, J., Sauer, in tomato. Plant Biotechnol. J. 10, 945–955. doi: 10.1111/j.1467-7652.2012. U. H., et al. (2012). Redox-mediated mechanisms regulate DNA binding activity 00723.x of the G-group of basic region leucine zipper (bZIP) transcription factors in Xiang, C., and Oliver, D. J. (1998). Glutathione metabolic genes coordinately Arabidopsis. J. Biol. Chem. 287, 27510–27525. doi: 10.1074/jbc.M112.361394 respond to heavy metals and jasmonic acid in Arabidopsis. Plant Cell 10, Shaikhali, J., and Wingsle, G. (2017). Redox-regulated transcription in plants: 1539–1550. doi: 10.1105/tpc.10.9.1539 emerging concepts. AIMS Mol. Sci. 4, 301–338. doi: 10.3934/molsci.2017. Xing, S., Rosso, M. G., and Zachgo, S. (2005). ROXY1, a member of the plant 3.301 glutaredoxin family, is required for petal development in Arabidopsis thaliana. Shen, Y., Issakidis-Bourguet, E., and Zhou, D.-X. (2016). Perspectives on the Development 132, 1555–1565. doi: 10.1242/dev.01725 interactions between metabolism, redox, and epigenetics in plants. J. Exp. Bot. Xing, S., and Zachgo, S. (2008). ROXY1 and ROXY2, two Arabidopsis glutaredoxin 67, 5291–5300. doi: 10.1093/jxb/erw310 genes, are required for anther development. Plant J. 53, 790–801. doi: 10.1111/ Sies, H., Berndt, C., and Jones, D. P. (2017). Oxidative stress. Annu. Rev. Biochem. j.1365-313X.2007.03375.x 86, 715–748. doi: 10.1146/annurev-biochem-061516-045037 Xu, Y.-M., Du, J.-Y., and Lau, A. T. Y. (2014). Posttranslational modifications of Suzuki, N., Koussevitzky, S., Mittler, R., and Miller, G. (2012). ROS and redox human histone H3: an update. Proteomics 14, 2047–2060. doi: 10.1002/pmic. signalling in the response of plants to abiotic stress. Plant Cell Environ. 35, 201300435 259–270. doi: 10.1111/j.1365-3040.2011.02336.x Zaffagnini, M., Bedhomme, M., Marchand, C. H., Morisse, S., Trost, P., and Tada, Y., Spoel, S. H., Pajerowska-Mukhtar, K., Mou, Z., Song, J., Wang, C., Lemaire, S. D. (2012). Redox regulation in photosynthetic organisms: focus on et al. (2008). Plant immunity requires conformational changes of NPR1 via glutathionylation. Antioxid. Redox Signal. 16, 567–586. doi: 10.1089/ars.2011. S-nitrosylation and thioredoxins. Science 321, 952–956. doi: 10.1126/science. 4255 1156970 Zaffagnini, M., De Mia, M., Morisse, S., Di Giacinto, N., Marchand, C. H., Tavares, C. P., Vernal, J., Delena, R. A., Lamattina, L., Cassia, R., and Terenzi, H. Maes, A., et al. (2016). Protein S-nitrosylation in photosynthetic organisms: a (2014). S-nitrosylation influences the structure and DNA binding activity of comprehensive overview with future perspectives. Biochim. Biophys. Acta 1864, AtMYB30 transcription factor from Arabidopsis thaliana. Biochim. Biophys. 952–966. doi: 10.1016/j.bbapap.2016.02.006 Acta 1844, 810–817. doi: 10.1016/j.bbapap.2014.02.015 Zaffagnini, M., Fermani, S., Costa, A., Lemaire, S. D., and Trost, P. (2013). Plant Testard, A., Da Silva, D., Ormancey, M., Pichereaux, C., Pouzet, C., Jauneau, A., cytoplasmic GAPDH: redox post-translational modifications and moonlighting et al. (2016). Calcium- and nitric oxide-dependent nuclear accumulation of properties. Front. Plant Sci. 4:450. doi: 10.3389/fpls.2013.00450 cytosolic glyceraldehyde-3-phosphate dehydrogenase in response to long chain Zechmann, B. (2017). Compartment-specific importance of ascorbate during bases in tobacco BY-2 cells. Plant Cell Physiol. 57, 2221–2231. doi: 10.1093/pcp/ environmental stress in plants. Antioxid. Redox Signal. doi: 10.1089/ars.2017. pcw137 7232 [Epub ahead of print]. Van Ruyskensvelde, V., Van Breusegem, F., and Van Der Kelen, K. (2018). Zechmann, B., Mauch, F., Sticher, L., and Müller, M. (2008). Subcellular Post-transcriptional regulation of the oxidative stress response in plants. Free immunocytochemical analysis detects the highest concentrations of glutathione Radic. Biol. Med. doi: 10.1016/j.freeradbiomed.2018.02.032 [Epub ahead of in mitochondria and not in plastids. J. Exp. Bot. 59, 4017–4027. doi: 10.1093/ print]. jxb/ern243 Vescovi, M., Zaffagnini, M., Festa, M., Trost, P., Lo Schiavo, F., and Zechmann, B., and Müller, M. (2010). Subcellular compartmentation of Costa, A. (2013). Nuclear accumulation of cytosolic glyceraldehyde-3- glutathione in dicotyledonous plants. Protoplasma 246, 15–24. doi: 10.1007/ phosphate dehydrogenase in cadmium-stressed Arabidopsis roots. Plant s00709-010-0111-2 Physiol. 162, 333–346. doi: 10.1104/pp.113.215194 Zechmann, B., Stumpe, M., and Mauch, F. (2011). Immunocytochemical Vivancos, P. D., Dong, Y., Ziegler, K., Markovic, J., Pallardó, F. V., Pellny, determination of the subcellular distribution of ascorbate in plants. Planta 233, T. K., et al. (2010). Recruitment of glutathione into the nucleus during cell 1–12. doi: 10.1007/s00425-010-1275-x proliferation adjusts whole-cell redox homeostasis in Arabidopsis thaliana and Zhang, H., Zhao, Y., and Zhou, D.-X. (2017). Rice NAD+-dependent histone lowers the oxidative defence shield. Plant J. 64, 825–838. doi: 10.1111/j.1365- deacetylase OsSRT1 represses glycolysis and regulates the moonlighting 313X.2010.04371.x function of GAPDH as a transcriptional activator of glycolytic genes. Nucleic Vogel, M. O., Moore, M., König, K., Pecher, P., Alsharafa, K., Lee, J., et al. (2014). Acids Res. 45, 12241–12255. doi: 10.1093/nar/gkx825 Fast retrograde signaling in response to high light involves metabolite export, Zhou, X., Sun, H., Chen, H., Zavadil, J., Kluz, T., Arita, A., et al. (2010). Hypoxia MITOGEN-ACTIVATED PROTEIN KINASE6, and AP2/ERF transcription induces trimethylated H3 lysine 4 by inhibition of JARID1A demethylase. factors in Arabidopsis. Plant Cell 26, 1151–1165. doi: 10.1105/tpc.113.12 Cancer Res. 70, 4214–4221. doi: 10.1158/0008-5472.CAN-09-2942 1061 Zhou, X., Sun, H., Ellen, T. P., Chen, H., and Costa, M. (2008). Arsenite alters Wagner, D., Przybyla, D., Op den Camp, R., Kim, C., Landgraf, F., Lee, K. P., global histone H3 methylation. Carcinogenesis 29, 1831–1836. doi: 10.1093/ et al. (2004). The genetic basis of singlet oxygen-induced stress responses of carcin/bgn063 Arabidopsis thaliana. Science 306, 1183–1185. doi: 10.1126/science.1103178 Zhu, J.-K. (2009). Active DNA demethylation mediated by DNA glycosylases. Waszczak, C., Akter, S., Eeckhout, D., Persiau, G., Wahni, K., Bodra, N., et al. Annu. Rev. Genet. 43, 143–166. doi: 10.1146/annurev-genet-102108-134205 (2014). Sulfenome mining in Arabidopsis thaliana. Proc. Natl. Acad. Sci. U.S.A. 111, 11545–11550. doi: 10.1073/pnas.1411607111 Conflict of Interest Statement: The authors declare that the research was Waszczak, C., Akter, S., Jacques, S., Huang, J., Messens, J., and Van Breusegem, F. conducted in the absence of any commercial or financial relationships that could (2015). Oxidative post-translational modifications of cysteine residues in plant be construed as a potential conflict of interest. signal transduction. J. Exp. Bot. 66, 2923–2934. doi: 10.1093/jxb/erv084 Welsch, R., Maass, D., Voegel, T., Dellapenna, D., and Beyer, P. (2007). Copyright © 2018 Martins, Trujillo-Hernandez and Reichheld. This is an open-access Transcription factor RAP2.2 and its interacting partner SINAT2: stable article distributed under the terms of the Creative Commons Attribution License elements in the carotenogenesis of Arabidopsis leaves. Plant Physiol. 145, (CC BY). The use, distribution or reproduction in other forums is permitted, provided 1073–1085. doi: 10.1104/pp.107.104828 the original author(s) and the copyright owner are credited and that the original Willems, P., Mhamdi, A., Stael, S., Storme, V., Kerchev, P., Noctor, G., et al. (2016). publication in this journal is cited, in accordance with accepted academic practice. The ROS wheel: refining ROS transcriptional footprints. Plant Physiol. 171, No use, distribution or reproduction is permitted which does not comply with these 1720–1733. doi: 10.1104/pp.16.00420 terms.

Frontiers in Plant Science | www.frontiersin.org 9 May 2018 | Volume 9 | Article 705 VI.2 Annex II: On the Elaborate Network of Thioredoxins in Higher Plants

Ina Thormählen, Belén Naranjo, José Abraham Trujillo-Hernandez, Jean- Philippe Reichheld, Francisco Javier Cejudo, Peter Geigenberger

256 On the Elaborate Network of Thioredoxins in Higher Plants

Ina Thormählen, Belén Naranjo, José Abraham Trujillo-Hernandez, Jean-Philippe Reichheld, Francisco Javier Cejudo, and Peter Geigenberger

Contents 1 Introduction 2 The Plastidial Thioredoxin Systems 2.1 Ferredoxin Thioredoxin Reductase-Dependent Thioredoxin System 2.2 NADPH-Dependent Thioredoxin Reductase C System 2.3 NADPH-Dependent Thioredoxin Reductase C Acts Concertedly with Other Chloroplast Thioredoxins 3 The Extraplastidial Thioredoxin Systems 3.1 Cytosolic Thioredoxin System 3.2 Nuclear Thioredoxin System 3.3 Mitochondrial Thioredoxin System 3.4 Endomembrane Thioredoxin System 4 Conclusions and Perspectives References

Communicated by Ulrich Lüttge

I. Thormählen (*) and P. Geigenberger Fakultät für Biologie – Pflanzenmetabolismus, Ludwig-Maximilians-Universität München, Munich, Germany e-mail: [email protected] B. Naranjo Fakultät für Biologie – Molekularbiologie der Pflanzen, Ludwig-Maximilians-Universität München, Munich, Germany J. A. Trujillo-Hernandez and J.-P. Reichheld Laboratoire Génome et Développement des Plantes, Université de Perpignan Via Domitia, Perpignan, France Laboratoire Génome et Développement des Plantes, Centre National de la Recherche Scientifique, Paris, France F. J. Cejudo Instituto de Bioquímica Vegetal y Fotosíntesis, Universidad de Sevilla and CSIC, Sevilla, Spain

© Springer International Publishing AG 2018 Progress in Botany, DOI 10.1007/124_2018_16 I. Thormählen et al.

Abstract Thioredoxins represent ubiquitous small proteins acting as redox regula- tors of diverse metabolic and developmental processes in almost all organisms. These proteins contain highly conserved cysteines in their redox-active sites, which enable the modification of target enzyme conformation and activity by reversible thiol- disulfide exchanges. Since their discovery in plants around 40 years ago, the number of thioredoxin family members as well as the knowledge about their distinct functions are still increasing and under investigation. Originally, the first plant thioredoxin was found in chloroplasts, while further analyses demonstrated additional cytosolic, nuclear, mitochondrial, endomembrane, and non-photosynthetic plastid locations. This chapter provides an overview on the complexity of the thioredoxin family in higher plants and discusses its role in integrating metabolism, stress responses, development, and gene expression. This will help to understand why plants harbor the most versatile thioredoxin system among all organisms.

Abbreviations

ACHT Atypical cysteine/histidine-rich thioredoxin ADP Adenosine diphosphate AGPase ADP-glucose pyrophosphorylase AMP Adenosine monophosphate AOX Alternative oxidase APS AGPase small subunit ATP Adenosine triphosphate ATPase ATP synthase CBC Calvin-Benson cycle cDNA Complementary DNA CDSP Chloroplastic drought-induced stress protein CHLM Mg-protoporphyrin methyl transferase CxxS Atypical thioredoxin h with cysteine-x-x-serine active site Cys Cysteine Cyt b6f Cytochrome b6f complex DNA Deoxyribonucleic acid ER Endoplasmic reticulum FAD Flavin adenine dinucleotide FBPase Fructose 1,6-bisphosphatase FDX Ferredoxin FNR Ferredoxin NADP+ reductase FTR Ferredoxin thioredoxin reductase FUM Fumarase Gb Gossypium barbadense GFP Green fluorescent protein GGLC Glycine-glycine-leucine-cysteine motif Gly Glycine GRX Glutaredoxin On the Elaborate Network of Thioredoxins in Higher Plants

GSH Glutathione HCF High-chlorophyll-fluorescence-mutant protein HCGPC Histidine-cysteine-glycine-proline-cysteine motif His Histidine LOV Locus orchestrating victorin effects protein Met Methionine MSR Methionine sulfoxide reductase NADP Nicotinamide adenine dinucleotide phosphate NADP+-MDH NADP+-dependent malate dehydrogenase NPQ Non-photochemical quenching NPR Non-pathogenesis-related protein expressor NRX Nucleoredoxin NTR NADPH-dependent thioredoxin reductase OAA Oxaloacetate ox Oxidized P Phosphate PC Plastocyanin PCNA Proliferating cell nuclear antigen PGR Proton gradient regulation complex Pi Inorganic phosphate PPi Inorganic pyrophosphate PQ Plastoquinon PRK Phosphoribulokinase protoMME Protomonomethylester PRX Peroxiredoxin PS Photosystem Ps Pisum sativum PsbS Photosystem II subunit S qE Energy- or ΔpH-dependent quenching RbcS Gene of the Rubisco small subunit red Reduced RNA Ribonucleic acid ROS Reactive oxygen species Rubisco Ribulose 1,5-bisphosphate carboxylase/oxygenase SAR Systemic acquired resistance SBPase Sedoheptulose 1,7-bisphosphatase SDH Succinate dehydrogenase SP Peroxidatic cysteine residue SR Resolving cysteine residue SRX Sulfiredoxin Ssb Single-stranded DNA binding protein TCA Tricarboxylic acid TDX Tetratricoredoxin THL h-type thioredoxins in Brassica napus TRX Thioredoxin I. Thormählen et al.

WCEVC Tryptophan-cysteine-glutamic acid-valine-cysteine motif WCGPC Tryptophan-cysteine-glycine-proline-cysteine motif WCRKC Atypical thioredoxin with tryptophan-cysteine-arginine-lysine- cysteine active site YF Tyrosine-phenylalanine motif

1 Introduction

Redox regulation plays a crucial role in a large number of plant cellular processes (Geigenberger and Fernie 2014). Thioredoxins (TRX), small polypeptides of 12–13 kDa with protein disulfide reductase activity, catalyze Cys-based posttrans- lational modifications, which affect the conformation and function of a large number of proteins. Plants harbor a multiplicity of TRX isoforms and reduction pathways (Lemaire et al. 2007; Meyer et al. 2012; Geigenberger et al. 2017). In Arabidopsis thaliana, the gene family of typical TRX encodes for up to 20 different isoforms located in different subcellular compartments. While the plastid uses ferredoxin (FDX) to reduce TRX via the FDX TRX reductase (FTR), extraplastidial TRX are reduced by NADPH-dependent TRX reductases (NTR). Plastids also contain an unusual NADPH-dependent TRX system termed NTRC, which is characterized by a tethered TRX domain, providing a complete NTR-TRX system in a single polypep- tide. In the last years, progress has been made to improve our understanding of the organization and biological roles of this complex thiol-based redox network. The aim of this review is to provide a synthesis of these recent evolvements focusing on the emerging roles of TRX in regulating metabolism, stress responses, development, and gene expression of higher plants.

2 The Plastidial Thioredoxin Systems

2.1 Ferredoxin Thioredoxin Reductase-Dependent Thioredoxin System

2.1.1 Plastids Contain the Most Diverse TRX System of Higher Plant Cells

About half of the total TRX isoforms in Arabidopsis cells are localized in the stroma of plastids (Meyer et al. 2012; Geigenberger et al. 2017). Typical TRX consist of a small single TRX domain carrying a highly conserved motif (WCGPC) in the redox- active site, which modulates the redox state of target enzymes by the reversible exchange of an oxidized disulfide bridge to two reduced thiols. The ten typical TRX isoforms localized in plastids are divided into five types, namely f (1–2), m (1–4), x, y (1–2), and z. Interestingly, the two TRX isoforms f1 and f2 contain an additional On the Elaborate Network of Thioredoxins in Higher Plants

PSI FDXox TRXox Targetox S S S FTR S S S

SH SH SH SH SH SH

FDXred TRXred Targetred

Fig. 1 The light-dependent reduction pathway of TRX and their targets in chloroplasts of higher plants. Absorbed light enables the photosynthetic electron transport chain to provide reducing power from the PSI to FDX. FDX transmits electrons to the FTR, which reduces oxidized TRX (blue). Reduced TRX modify the redox activation states of distinct target enzymes (red) cleaving the disulfide bridge between Cys residues to two thiols (see above for Abbreviations) conserved Cys and are suggested to be of eukaryotic origin based on their high homology with cytosolic TRX, while all other plastidial TRX are considered to have prokaryotic ancestors (Sahrawy et al. 1996; Schürmann and Buchanan 2008). Tissue-specific gene expression analyses revealed different patterns of expression for each of the single TRX isoforms in Arabidopsis (Bohrer et al. 2012; Belin et al. 2015). TRXm3, y1, and z were almost not detectable in green tissues, while TRXf1, m1, m2, m4, and x were expressed at high levels. In comparison to leaves, roots contained overall low plastidial TRX transcript abundance; however, TRXm2 displayed the highest expression values. The source of reducing power for TRX reduction differs between plastids and other compartments (Meyer et al. 2012; Geigenberger et al. 2017). While in plastids the delivery of reducing power to TRX depends on reduced FDX, TRX in other compartments rely on NADPH and are reduced via NTR of which Arabidopsis contains two isoforms termed NTRA and NTRB (see below). In chloroplasts, photosynthetically reduced FDX provides electrons to the stromal FTR system, which in turn reduces TRX and is thus a strictly light-dependent process (see Fig. 1). In non-photosynthetic plastids, FDX is reduced by the FDX NADP+ reductase (FNR) with NADPH as electron donor. However, the reduction concept of the plastidial TRXz isoform is not clear so far. Chibani et al. (2011) observed a reduction of TRXz only by the FTR system, while Bohrer et al. (2012) stated that it is exclusively reduced by its plastidial TRX family members. More recently, Yoshida and Hisabori (2016b) demonstrated an additional reduction pathway of TRXz via the alternative NADPH-dependent TRX system NTRC (see below) present in plastids.

2.1.2 The Redox Regulation of Target Enzymes Involves Multiple Functions of TRX in Plastidial Metabolism

The first discovery of TRX in plants led to a deeper understanding of the light- dependent regulation of the photosynthetic carbon assimilation reactions, namely the I. Thormählen et al.

Calvin-Benson cycle (CBC) enzymes (Buchanan 1980). It was observed that chloroplast TRX act as electron carrier able to modulate enzyme activity by chang- ing the redox state of the target proteins dependent on light. Thus, the previously suggested idea that the increased CO2 fixation under light is due to the provision of the energy and redox carriers ATP and NADPH during photosynthesis, which might facilitate a promoted protein synthesis of the CBC enzymes, could be ruled out. Following the initial finding about TRX around 40 years ago, extensive in vitro studies with isolated or recombinant TRX proteins were performed (reviewed by Meyer et al. 2012; Geigenberger et al. 2017). These studies extended the list of potential TRX targets and demonstrated distinct functions for the single isoforms. TRX types f and m were found to be predominantly important for the activation of anabolic pathways, like the CBC (Wolosiuk and Buchanan 1977; Breazale et al. 1978; Collin et al. 2003; Marri et al. 2009; Yoshida and Hisabori 2016b), the biosynthesis of starch (Ballicora et al. 2000; Skryhan et al. 2015; Thormählen et al. 2013) and lipids (Sasaki et al. 1997; Yamaryo et al. 2006), and for balancing the energy and redox homeostasis in chloroplasts by participating in ATP synthesis (McKinney et al. 1978) and the malate valve (Wolosiuk et al. 1977; Collin et al. 2003; Yoshida and Hisabori 2016b). Additionally, many other enzymes involved in diverse chloroplast processes were observed to be redox regulated by TRXf and m in vitro, like the oxidative pentose phosphate (Scheibe and Anderson 1981; Née et al. 2009) and shikimate pathway (Entus et al. 2002), the nitrogen assimilation (Lichter and Häberlein 1998) or protein import processes (Bartsch et al. 2008). In vitro studies investigating the regulation of the cyclic electron transport in the photosyn- thetic light reaction showed a preference for TRXm isoforms, but described so far contrasting effects leading to activation (Hertle et al. 2013) or inhibition (Courteille et al. 2013) of this process. TRXx, y, and z showed higher affinities for enzymes involved in oxidative stress responses (Collin et al. 2003, 2004; Navrot et al. 2006; Vieira Dos Santos et al. 2007; Chibani et al. 2011; Bohrer et al. 2012). The TRX-dependent redox activation of starch degradation (Mikkelsen et al. 2005; Sparla et al. 2006; Valerio et al. 2011; Seung et al. 2013; Silver et al. 2013) and chlorophyll synthesis (Ikegami et al. 2007; Luo et al. 2012; Yoshida and Hisabori 2016b) presented no clear distinction in the preference to different TRX isoforms until now. Additionally, it remains unclear how the starch degradation process in leaves could be activated by TRX reduction, since mobilization of transient starch occurs mainly during the night, when the photosynthetic reduction pathway of TRX is inactive.

2.1.3 The Plastidial TRX System Shows Surprisingly Redundant Activities In Vivo

Within the last decade, reverse genetic approaches were used to investigate Arabidopsis mutant lines lacking or overexpressing single or multiple TRX, so that in vivo evidences for the functional roles of the individual TRX in planta could be addressed. Surprisingly, under normal light conditions most single mutant On the Elaborate Network of Thioredoxins in Higher Plants lines deficient in plastidial TRX showed no or only minor visible phenotypes as compared to the wild type (Pulido et al. 2010; Thormählen et al. 2013, 2017; Laugier et al. 2013), despite the diverse roles previously proposed for these TRX based on analyses in vitro. For example, TRXf was known to be the only TRX type, which activates the redox regulated CBC enzymes fructose 1,6-bisphosphatase (FBPase) or glyceraldehyde 3-phosphate dehydrogenase (Collin et al. 2003; Marri et al. 2009). However, growth under moderate light conditions and CO2 fixation rates were not impaired in the Arabidopsis trxf1 single mutant, showing a strong deficiency in TRXf protein level in leaves (Thormählen et al. 2013; Naranjo et al. 2016a). Even the double mutant trxf1 trxf2 with complete lack of TRXf presented only slightly decreased rosette fresh weights under short day conditions (Naranjo et al. 2016a). The low impact of single TRX deficiencies on growth is most probably due to underestimated redundant functions of the other TRX isoforms or of additional enzymes in the plastidial redox regulatory network, like NTRC (see below). There is, however, one exception. The recently discovered Arabidopsis trxz single mutant presented an albino phenotype with strongly impaired growth compared to wild type plants (Arsova et al. 2010). Here, obviously, other TRX isoforms were not able to compensate for the deficiency in TRXz. Some of the observed metabolic phenotypes in Arabidopsis mutant lines devoid of TRXf or m isoforms were confirmatory of the in vitro results mentioned above. The redox activation state of ADP-glucose pyrophosphorylase (AGPase), which catalyzes the first committed step in starch synthesis, was attenuated in leaves of trxf1 mutants during the day, accompanied by a reduced accumulation of starch (Thormählen et al. 2013) in agreement with previous in vitro results (Ballicora et al. 2000; Geigenberger et al. 2005). However, leaves of Nicotiana tabacum overexpression lines with elevated levels of TRXf accumulated higher levels of starch, while no change in AGPase redox activation was observed (Sanz-Barrio et al. 2013). Plants lacking TRXf isoforms additionally showed an impaired light- dependent activation of the CBC enzyme FBPase (Thormählen et al. 2015; Yoshida et al. 2015; Naranjo et al. 2016a), and the ribulose 1,5-bisphosphate carboxylase/ oxigenase (Rubisco) activase (Naranjo et al. 2016a). Okegawa and Motohashi (2015) stated that the TRXm isoforms might be the predominant redox regulators of the CBC enzymes, since Arabidopsis triple mutant lines simultaneously deficient in TRXm1, m2, and m4 showed attenuated CO2 fixation rates, smaller rosettes, and impaired redox activation states of FBPase and sedoheptulose 1,7-bisphosphatase (SBPase) as well as the malate valve enzyme NADP+-dependent malate dehydro- genase (NADP+-MDH). Since the plastidial FBPase is not activated by TRXm isoforms in vitro (Collin et al. 2003; Yoshida and Hisabori 2016b), the in vivo observation on the impaired FBPase redox activation (Okegawa and Motohashi 2015) might not be due to direct TRX effects, but most probably due to secondary reasons, like an inhibited electron flow to downstream metabolic processes. The latter suggestion would be in agreement with additional in vivo studies, which found TRXm impacts on processes involved in the photosynthetic light reaction, like an impaired biogenesis of photosystem (PS) II in Arabidopsis mutant lines simulta- neously lacking TRXm1, m2, and m4 (Wang et al. 2013) or the inhibition of cyclic I. Thormählen et al. electron transport by TRXm4 orthologues in N. tabacum (Courteille et al. 2013). Very recently Arabidopsis trxm1 and trxm2 single and double mutants were com- pared to wild type plants with respect to the transient activation of the NADP+-MDH after the onset of light (Thormählen et al. 2017). It was demonstrated that the light- dependent redox activation of the NADP+-MDH decreased upon combined defi- ciencies of TRXm1 and m2, which corresponded to an impaired acclimation to fluctuating light intensities during the day. In vitro experiments revealed the ability of TRXf and at least partially TRXm to reductively activate the NADP+-MDH (Collin et al. 2003; Yoshida and Hisabori 2016b). However, this contrasts with illuminated leaves of trxf Arabidopsis mutant lines exhibiting even higher NADP+- MDH activation states (Thormählen et al. 2015). Further studies are necessary to investigate the distinct involvements of TRXf and m isoforms in redox regulated processes playing a role in the fluctuating light response. More recently, the findings about TRX-dependent redox regulation of the chlorophyll biosynthesis pathway were extended in vivo. Transgenic pea plants with simultaneously silenced TRXf and m genes were impaired in redox activation of Mg-chelatase subunits (Luo et al. 2012). Da et al. (2017) demonstrated in Arabidopsis that the combined silencing of TRXm1, m2, and m4 impairs the chlorophyll biosynthesis pathway by attenuating the redox control of the Mg-protoporphyrin methyl transferase (CHLM). Furthermore, the in vivo insights about TRXm3 revealed functions in meristem development by modulating symplastic protein transport (Benitez-Alfonso et al. 2009). In response to oxidative stress, mutant lines lacking TRXy2 showed attenuated protein repair mechanisms in leaves due to impaired redox activation of methionine sulfoxide reductases (MSR) (Laugier et al. 2013). TRXz was demonstrated to be essential for autotrophic growth by influencing plastidial gene expression as structural component of the plastid- encoded RNA polymerase (Arsova et al. 2010), although this effect seems to be independent of TRXz redox activity (Wimmelbacher and Börnke 2014). Taken together, the diverse functions of typical TRX in plastidial processes investigated within the last decades illustrate a complex redox regulatory system, which is still not fully elucidated.

2.1.4 Plastids Contain Additional Atypical TRX with Mainly Unresolved Functions

Contrary to the TRX described above, the members of the atypical TRX are not yet well analyzed in their functional roles of plastidial metabolism (Meyer et al. 2012; Belin et al. 2015). These unusual TRX have different protein properties such as atypical active site motifs, multiple TRX domains, or combinations of TRX and other domains. The proteins of the atypical Cys/His-rich TRX (ACHT) family consist of six isoforms carrying diverse non-canonical active site motifs (Dangoor et al. 2009; Meyer et al. 2012; Belin et al. 2015). The chloroplastic drought-induced stress protein CDSP32 contains two TRX domains and several additional conserved Cys (Rey et al. 1998). One of the TRX domains is functional with an unusual redox- On the Elaborate Network of Thioredoxins in Higher Plants active motif (HCGPC), while the second domain is degenerated. The transmembrane high-chlorophyll-fluorescence-mutant protein HCF164 is localized in the thylakoid membranes with atypical TRX motifs (WCEVC) at the stromal as well as at the lumen side (Motohashi and Hisabori 2006; Meyer et al. 2012). About the two isoforms WCRKC1 and WCRKC2 not much is known; however, these proteins contain non-canonical redox sites as their names suggest (Meyer et al. 2012). ACHT2, ACHT4, ACHT6, CDSP32, HCF164, and WCRKC1 are mostly expressed in green tissues (Belin et al. 2015). The genes encoding ACHT1, ACHT3, and ACHT5 have an overall low expression in tested tissues, except for mature and germinating pollen, in which the ACHT3 gene showed very high transcript level. How the reduction pathways for the atypical TRX are organized is not yet suffi- ciently investigated. One tested member was HCF164, which was reduced by TRXm (Motohashi and Hisabori 2006). TRXm enables a transmembrane electron transfer to the luminal HCF164 part, so that HCF164 acts as a transducer of reducing equiva- lents to the inner parts of the thylakoids. Whether the FTR-dependent reduction pathway in chloroplasts is directly interacting with atypical TRX needs to be tested. The functional insights about atypical TRX are weak compared to the other plastidial TRX. Members of the ACHT family have a less reducing redox midpoint potential than typical TRX and prefer 2-Cys peroxiredoxins (PRX; see below) in comparison to NADP+-MDH as redox targets in vitro (Dangoor et al. 2009). Eliyahu et al. (2015) observed in Arabidopsis mutant studies that ACHT4 is deactivating the key enzyme of starch synthesis AGPase through oxidation of the regulatory Cys in the AGPase small subunit APS1 under low light conditions. CDSP32 was shown to be involved in photooxidative stress responses by interacting with plastidial PRX (PRXQ and 2-Cys PRX) and MSRB1 (Rey et al. 2005; Tarrago et al. 2010). HCF164 acts as reductant for luminal target proteins of the photosynthetic electron transport chain and is essential for assembling the chloroplast cytochrome b6f complex (Lennartz et al. 2001; Motohashi and Hisabori 2006). Future studies are needed to identify additional biological roles of atypical TRX in plastidial processes and the connection to their typical relatives.

2.2 NADPH-Dependent Thioredoxin Reductase C System

2.2.1 The NTRC Gene Is Exclusive of Aerobic Photosynthetic Organisms

In heterotrophic organisms, and non-photosynthetic plant cells and compartments, the reduction of TRX relies on NADPH with the participation of an NTR. NTR are flavoenzymes able to transfer reducing equivalents from NADPH via FAD to a double Cys at their active site, which reduce the disulfide at the active site of TRX. Like TRX, NTR are widely distributed in prokaryotic and eukaryotic organisms; however, two forms of the enzyme have evolved with different molecular and kinetic properties (Williams et al. 2000). While the enzyme from prokaryotes is, in its native I. Thormählen et al. form, a homodimer formed by subunits of approximately 35 kDa with NADPH, FAD, and central domains, the enzyme from mammals has a larger molecular mass (approximately 55 kDa) and contains selenocysteine, which is active in catalysis (Arscott et al. 1997). In plants, the first NTR cloned and characterized, from Arabidopsis (Jacquot et al. 1994) and wheat (Serrato et al. 2002), revealed that the enzyme is similar to the one in prokaryotes. When the genome sequences from Arabidopsis and rice were available, a search of the genes encoding NTR in these plants revealed the presence of three genes termed NTRA, NTRB, and NTRC (Serrato et al. 2002). The NTRA and NTRB genes encode polypeptides corresponding to those previously reported from Arabidopsis and wheat. Further analyses in Arabidopsis revealed that NTRB is the predominant isoform in mitochondria whereas NTRA is predominant in the cytosol (Reichheld et al. 2005). The polypeptide deduced of the NTRC gene contained the expected active site double Cys as well as the NADPH and FAD domains, hence being very similar to NTRA and NTRB. However, in contrast to these proteins, the deduced NTRC polypeptide contained a putative transit peptide at the N-terminus and, most intriguingly, a putative TRX domain at the C-terminus; thus, it contained a complete NTR-TRX system in a single polypeptide (Serrato et al. 2004). Both the NTR and TRX domains were expressed in the bacterium Escherichia coli as N-terminal His-tagged proteins, and the biochemical analysis confirmed that these truncated polypeptides displayed NTR and TRX activity, respectively, in vitro, leading to the proposal that NTRC is a bifunctional enzyme (Serrato et al. 2004). Moreover, the search in genomes available at the moment, and recently confirmed, showed that the NTRC gene is exclusively found in aerobic photosynthetic organisms including some, but not all, cyanobacteria, algae, and plants (Serrato et al. 2004; Najera et al. 2017). Furthermore, it was shown that NTRC is localized in the chloroplast stroma both in Arabidopsis and rice (Serrato et al. 2004; Moon et al. 2006), hence implying that the putative transit peptide deduced of the NTRC gene sequence actually serves to target the enzyme to this organelle. More recently, the localization of NTRC was analyzed by expressing NTRC::GFP fusion proteins in Arabidopsis. This study confirmed the localization of NTRC in chloroplasts and showed the presence of the enzyme in plastids from non-photosynthetic tissues such as root amyloplasts, though NTRC is much less abundant in these tissues as compared with green tissues (Kirchsteiger et al. 2012). The phylogenetic analysis suggests that the NTRC gene originated in cyanobacteria most probably by mutational joining of adjacent but independent NTR and TRX genes. The combination of NTR and TRX activities in a single enzyme might provide selective advantage, which explains the presence of the NTRC gene in all eukaryotic photosynthetic organisms (Najera et al. 2017).

2.2.2 NTRC Is an Efficient Reductant of 2-Cys PRX

An important breakthrough for the biochemical characterization of NTRC was the identification of 2-Cys PRX as a target of the enzyme. 2-Cys PRX are thiol- dependent PRX able to reduce hydrogen or organic peroxides to water or the On the Elaborate Network of Thioredoxins in Higher Plants corresponding alcohol (Perkins et al. 2015). There are different types of 2-Cys PRX, the one found to be a target of NTRC belongs to the typical form, which is a homodimer each of the monomers containing two catalytically active Cys residues, peroxidatic (SP) and resolving (SR). A catalytic cycle is initiated by the reaction of hydrogen peroxide with the thiolate form of SP, which is oxidized to sulfenic (–SOH). This intermediate is then attacked by SR thus forming a disulfide with the liberation of water (see Fig. 2). For a new catalytic cycle, the SP-SR disulfide has to be reduced. This reaction is in chloroplasts efficiently performed by NTRC. Indeed, it is possible to reconstitute in vitro a system formed by NTRC and 2-Cys PRX which is able to use NADPH to reduce hydrogen peroxide (Perez-Ruiz et al. 2006; Moon et al. 2006). The finding that NTRC is an efficient reductant of 2-Cys PRX was important because it allowed an in-depth characterization of the biochemical properties of this novel enzyme. It was first shown that NTRC conjugates its NTR and TRX activities to efficiently support the hydrogen peroxide scavenging activity of 2-Cys PRX. However, equal amount of the truncated polypeptides containing the NTR and TRX domains of NTRC showed a very poor activity, indicating that the conformation of the enzyme is important to display its biochemical activity. Moreover, it was established that NTRC interacts with 2-Cys PRX by the TRX domain and that the

NADPH+H+ NADP+

S SPHSRH S H2O2 NTRC SS SH H O SH 2 SRX NTRC 2-Cys PRX

catalytic + + SP SR SPO +H SPO2+H cycle SRH SRH

SS SS SS

H2O2 H2O

Fig. 2 NTRC controls the redox state of 2-Cys PRX. The catalytically active form of NTRC (blue) is a dimer arranged in a head-to-tail conformation. NTRC interacts with 2-Cys PRX (red) through its TRX domain and efficiently reduces the disulfide bond linking the peroxidatic (SP) and resolving (SR) Cys residues of oxidized 2-Cys PRX using NADPH as source of reducing equivalents. The thiolic form of SP reacts with H2O2 releasing H2O and rendering the peroxidatic Cys oxidized as a sulfenic intermediate (SPO�), which is then attacked by the resolving Cys (SRH) to form the disulfide form. Under oxidizing conditions, the sulfenic intermediate (SPO�) can be overoxidized to sulfinic (SPO2�), hence yielding an inactive form of the enzyme. This state is reversed by the reductive action of SRX (see above for Abbreviations) I. Thormählen et al. redox-sensitive Cys residues at the active sites of the NTR and the TRX domains are essential for activity (Perez-Ruiz and Cejudo 2009). Based on these data it was proposed that the catalytically active form of NTRC is a homodimer arranged in a head-to-tail conformation. The enzyme shows much higher affinity for NADPH than for NADH. Thus, NADPH is the source of reducing power, which is transferred to FAD and then to the double Cys at the active site of the NTR domain of one of the subunits. There is then an inter-subunit transfer of electrons to the active site double Cys of the TRX domain of the other subunit of the enzyme, which interacts with the disulfide of the 2-Cys PRX (Perez-Ruiz and Cejudo 2009). According to this model, stating that NTRC interacts with its targets by the TRX domain, it could be considered that NTRC is a TRX with its own reductase, which might be the reason of the high catalytic efficiency of this enzyme (Cejudo et al. 2012). The high affinity of NTRC for NADPH and the proposed electron transfer pathway was further confirmed by stopped-flow spectroscopy (Bernal-Bayard et al. 2012), and the interaction between NTRC and 2-Cys PRX was also demonstrated in vivo by bimolecular fluorescence complementation analyses (Bernal-Bayard et al. 2014). In line with this notion it was found that NTRC is unable to interact with other plastidial TRX, hence, having no activity as NTR (Bohrer et al. 2012). However, the Arabidopsis ntrc mutant, which is devoid of NTRC, was partially complemented by the expression of a mutated version of the NTRC gene encoding an enzyme with intact NTR and inactive TRX domains, indicating that the NTR domain of NTRC might interact with other plastidial TRX among which TRXf was proposed as the most likely partner (Toivola et al. 2013). Therefore, whether NTRC acts exclusively as a TRX with its reductase incorporated or is able to interact with other plastidial TRX, hence showing NTR activity, still requires further analysis. Finally, it should be mentioned that NTRC shows ability to oligomerize in vitro. While for the enzyme from rice this ability is redox sensitive, the enzyme being detected as dimer in the presence of NADPH and as oligomer in its absence (Perez- Ruiz et al. 2006), the oligomerization capacity of the enzyme from barley is redox insensitive (Wulff et al. 2011). The oligomeric form of NTRC shows chaperone and holdase activity, which might be the reason of the tolerance to elevated temperature of plants overexpressing NTRC (Chae et al. 2013).

2.2.3 NTRC Controls the Redox State of 2-Cys PRX

Based on the fact that NTRC is an efficient reductant of 2-Cys PRX, hence supporting the hydrogen peroxide scavenging activity of these enzymes, it was proposed that NTRC has antioxidant activity (Perez-Ruiz et al. 2006). However, it was also shown that different chloroplast TRX are able to reduce 2-Cys PRX, the most efficient one being TRXx (Collin et al. 2003). Thus, a relevant issue was to establish the hierarchy of the different chloroplast redox systems for the reduction of 2-Cys PRX. The redox state of the 2-Cys PRX in vivo, determined as the ratio of dimeric (oxidized) to monomeric (reduced) forms of the enzyme, was severely impaired in an Arabidopsis mutant devoid of NTRC, which contained almost no On the Elaborate Network of Thioredoxins in Higher Plants monomeric form of 2-Cys PRX (Kirchsteiger et al. 2009). In contrast, mutant plants devoid of TRXx showed levels of oxidized and reduced 2-Cys PRX indistinguish- able of the wild type, indicating that NTRC is more relevant reductant of 2-Cys PRX than TRXx (Pulido et al. 2010). In the presence of excess hydrogen peroxide, 2-Cys PRX may be inactivated by over-oxidation of the sulfenic acid intermediate of SP (–SOH) to sulfinic acid (–SO2H), which can be reverted by sulfiredoxin (SRX) (Biteau et al. 2003), or even to sulfonic acid (–SO3H), which is irreversible. Interestingly, 2-Cys PRX from eukaryotes are more sensitive to over-oxidation than those from prokaryotes due to structural constraints imposed by the presence of two motifs, GGLC and YF, in the enzyme from eukaryotes (Wood et al. 2003). Since over-oxidation causes the inactivation of 2-Cys PRX, it was proposed that this feature is a gain-of-function of the enzymes from eukaryotes that allows the accumulation of hydrogen peroxide and, thus, its signaling function, the so-called floodgate hypothesis (Wood et al. 2003). Chloroplast 2-Cys PRX are sensitive to over-oxidation, hence behaving as eukaryotic type enzymes (Kirchsteiger et al. 2009). However, since plant chloro- plasts have prokaryotic origin similar to modern cyanobacteria (Gould et al. 2008) the question arising was whether or not 2-Cys PRX from cyanobacteria are sensitive to over-oxidation. The finding that the enzyme from Anabaena is sensitive to over- oxidation while the one from Synechocystis is not (Pascual et al. 2010) showed the existence of sensitive 2-Cys PRX in prokaryotes and revealed that cyanobacteria have developed two strategies to cope with hydrogen peroxide, the strategy of Anabaena being remarkably similar to the one in chloroplasts. The level of over-oxidation of chloroplast 2-Cys PRX and, thus, the balance of antioxidant and signaling activities of these enzymes depends on the reductants (NTRC and TRX of this organelle), but also on SRX, which catalyzes the reversion of the over-oxidized to the reduced form of the enzyme (Puerto-Galan et al. 2013). However, a double mutant of Arabidopsis devoid of NTRC and SRX showed the phenotype of the ntrc mutant (Puerto-Galan et al. 2015). Altogether, the results of the biochemical analyses of NTRC in conjunction with the genetic interaction of NTRC with TRXx and SRX indicate that a central function of NTRC is the control of the redox state of the 2-Cys PRX, which is important for the balance of the harmful and signaling activities of hydrogen peroxide.

2.2.4 NTRC Participates in the Redox Regulation of Light Energy Utilization and Several Chloroplast Biosynthetic Pathways

To address the issue of the function of NTRC the use of Arabidopsis mutants has been a useful tool. The Arabidopsis knock-out mutant ntrc shows pale green leaves with a lower content of chlorophylls and retarded growth, as compared to wild type plants. Moreover, the retarded growth phenotype of the ntrc mutant is dependent on the photoperiod, short-day conditions aggravating the phenotype (Serrato et al. 2004; Lepistö et al. 2009). More in-depth analyses revealed that the ntrc mutant displays lower CO2 fixation rates, especially at low light intensities, and abnormal I. Thormählen et al. chloroplast structure (Perez-Ruiz et al. 2006; Lepistö et al. 2009). In addition, plants devoid of NTRC have been reported to be hypersensitive to different abiotic stresses, such as salinity and drought (Serrato et al. 2004), prolonged darkness (Perez-Ruiz et al. 2006), and heat (Chae et al. 2013), and also to biotic stress (Ishiga et al. 2012, 2016). These phenotypic characteristics of the ntrc mutant might be explained by the antioxidant function of NTRC as an efficient reductant of 2-Cys PRX. In this regard, it is worth mentioning that the NTRC-2-Cys PRX system has also been reported to exist in other photosynthetic organisms including green algae, such as Chlorella (Machida et al. 2012), and cyanobacteria, such as Anabaena (Pascual et al. 2011). Indeed, an Anabaena mutant strain devoid of NTRC (ΔntrC) is more sensitive to oxidative stress, especially to treatments with hydrogen peroxide, and the redox balance of 2-Cys PRX is impaired, more oxidized, in the mutant (Sanchez-Riego et al. 2016). Beside the phenotypic features related with the response to stress, which might be due to the antioxidant activity of the NTRC-2-Cys PRX system, it should be noted that the Arabidopsis ntrc mutant has a rather pleiotropic phenotype, suggesting additional functions for the enzyme. This notion was further supported by the phenotype of an Arabidopsis mutant (Δ2cp) with severely decreased levels of 2-Cys PRX (Pulido et al. 2010), which is almost like the wild type. It is well known that AGPase, a key regulatory enzyme of the starch biosynthesis pathway, is subject to allosteric regulation but also to sugar and light-dependent redox regulation (Geigenberger et al. 2005). On the other hand, as stated above, NTRC uses as source of reducing power NADPH, which can be generated from sugars during the night by the oxidative pentose phosphate pathway. These results suggested the possibility that NTRC participates in the redox regulation of starch biosynthesis, a notion further supported by the lower content of starch in roots and leaves of the ntrc mutant. In vitro and in vivo analyses confirmed the participation of NTRC in the redox regulation of AGPase (Michalska et al. 2009; Lepistö et al. 2013) and, thus, in starch biosynthesis. Interestingly, the redox state of the AGPase was found to be altered not only in leaf chloroplasts but also in roots, suggesting a role of NTRC in non-photosynthetic tissues (Michalska et al. 2009), in line with the localization of the enzyme in any type of plastids. However, Arabidopsis transgenic plants expressing NTRC exclusively in green tissues, by expression of the NTRC cDNA under the control of the RbcS gene promoter in the ntrc background, was sufficient to rescue the wild type phenotype with respect to growth and pigmenta- tion, while expression exclusively in roots had no effect (Kirchsteiger et al. 2012). These results show the essential function of chloroplasts for the growth of photo- synthetic and non-photosynthetic organs. The characteristic pale green leaf phenotype of the ntrc mutant, which is due to lower chlorophyll content than in the leaves of the wild type (Serrato et al. 2004; Lepistö et al. 2009), suggested the participation of NTRC in chlorophyll biosynthe- sis. Several reports have shown the NTRC-dependent redox regulation of different enzymes involved in the tetrapyrrole biosynthesis pathway including the glutamyl- transfer RNA reductase GluTR1, CHLM (Richter et al. 2013) and the Mg-chelatase subunit CHLI (Perez-Ruiz et al. 2014). On the Elaborate Network of Thioredoxins in Higher Plants

More recently, it has been reported that NTRC is required for efficient light energy utilization. Under growth light conditions, the ntrc mutant displays high levels of energy dissipation by non-photochemical quenching (NPQ) with the concomitant decrease of the photosynthetic performance (Thormählen et al. 2015; Naranjo et al. 2016b; Carrillo et al. 2016). Plants lacking NTRC accumulate more protons in the thylakoid lumen at low and moderate light intensities leading to the synthesis of zeaxanthin and the activation of the rapid inducible and reversible component of NPQ, namely the energy- or ΔpH-dependent quenching (qE) (Naranjo et al. 2016b; Carrillo et al. 2016). The higher accumulation of protons in the absence of NTRC has been attributed to a lower reduction of the ATP synthase (ATPase) γ-subunit (Naranjo et al. 2016b; Carrillo et al. 2016), though other mechanisms may be also involved. The lower photosynthetic performance of the ntrc mutant leads to a state of permanent starvation for light energy, which may explain its retarded growth phenotype, particularly under conditions of light limitation like short-day photope- riod. Blocking NPQ in the ntrc mutant background by the lack of the PS II subunit S (PsbS), in the double mutant ntrc psbs, causes the partial recovery of the photo- synthetic performance and growth (Naranjo et al. 2016b). Furthermore, since the induction and recovery of qE are essential to cope with rapid changes in light intensities, the growth of the ntrc mutant is even more compromised under fluctu- ating light conditions (Thormählen et al. 2017).

2.3 NADPH-Dependent Thioredoxin Reductase C Acts Concertedly with Other Chloroplast Thioredoxins

The finding of the participation of NTRC in the redox regulation of biosynthetic pathways previously known to be regulated by TRX raised the issue of the relation- ship of NTRC with the different TRX in chloroplasts (see Fig. 3). This issue has been addressed in recent years by the analysis of Arabidopsis mutants combining the deficiency of NTRC and TRX. The different mutants analyzed lacking NTRC and f-type TRX (Thormählen et al. 2015; Ojeda et al. 2017) or NTRC and TRXx (Ojeda et al. 2017) have in common a very severe growth inhibition phenotype, in agree- ment with the severe impairment of the efficiency of use of light energy and redox regulation of CBC enzymes in these mutants. In line with these results, Arabidopsis mutants combining the lack of NTRC and the deficiency of m-type TRX show as well a severe growth retard and impairment of the redox regulation of enzymes of the tetrapyrrole biosynthesis pathway (Da et al. 2017). Altogether, these results support the notion of a concerted action of NTRC with other plastidial TRX, such as those of types f, x, and m in chloroplast redox regulation. The finding that the double mutant combining the deficiency of NTRC and the catalytic subunit of FTR is not viable (Yoshida and Hisabori 2016b) lends further support to this notion. An initial possibility to explain the concerted action of NTRC and TRX in chloroplast redox regulation is that both systems share common targets, so that the I. Thormählen et al.

TRX-dependent regulation Starch, ATP, chlorophyll biosynthesis NO 2 Oxidative stress response ATP Cyclic electron transport ADP Calvin-Benson cycle OAA Malate Malate valve NADP+-MDH Glutamate P TRXf i NADPH NADPH NADP+ TRXm NTRC Triose-P TRXy + NADP TRXx Protoporphyrin IX ATP FBP TRXz + NADP NADPH ase Mg-chelatase H O + ½ O 2 2 Calvin- ADP Benson 2-Cys PRX cycle Mg-protoporphyrin IX FTR FNR PRK SBP H2O2 CO2 CHLM ase - O2 Methionine AMP + PPi Mg-protoMME Glucose 1-P NADPH FDX H+ AGPase Met-SO MSR ADP + P ATP PGR i NADP+ H+ ADP-glucose Chlorophyll Starch Cyt PQ PSI b6f ATP PSII PC ase Stroma O - + 2 H O2 Lumen H2O + + ½ O2 + 2H H

Fig. 3 TRX-dependent redox regulation of metabolic pathways in chloroplasts of higher plants. Light enables the energy-driven electron flow (red arrows) through the thylakoid membranes to FDX, which distributes the reducing power to different systems. The FNR system produces NADPH consumed by the Calvin-Benson cycle, but also by the malate valve or NTRC (bright orange). Secondly, the FTR system is providing electrons to TRX (deep orange). Additionally, the nitrogen assimilation pathway and the cyclic electron transport rely on the reducing ability of FDX. Reduced TRX and NTRC proteins facilitate the reversible redox activation of their distinct target enzymes (deep purple). Exclusively, in vitro as well as in vivo confirmed targets are shown. These include enzymes involved in different biosynthesis pathways (AGPase, ATPase, CHLM, Mg-chelatase), the oxidative stress system (2-Cys PRX, MSR), the cyclic electron transport (PGR), the Calvin-Benson cycle (FBPase, PRK, Rubisco activase, SBPase), and the malate valve (NADP+- MDH). The function of the proposed interaction between the TRX and NTRC systems, as well as the reduction pathway of TRXz and its role as redox regulator in general are not yet fully resolved (dotted red arrows). Whether TRX have an activating or inhibiting effect on the PGR target (bright purple) is also not clear so far (see above for Abbreviations) lack of one of them might be complemented by the other, while the combined deficiency of both systems would cause the severe impairment of the redox regula- tion of these targets, hence with strong consequences on plant phenotype. In support of this notion, Nikkanen et al. (2016) showed the interaction of NTRC in planta with well-established TRX targets such as FBPase. However, in vitro assay analyses showed that NTRC is unable to reduce FBPase (Ojeda et al. 2017), suggesting that the effect of NTRC on the redox regulation of this enzyme might be indirect. In a recent report it was shown that the content of 2-Cys PRX affects the phenotype of Arabidopsis mutants devoid of NTRC, so that decreased contents of 2-Cys PRX alleviated the phenotype while increased contents had the opposite effect (Perez- On the Elaborate Network of Thioredoxins in Higher Plants

Ruiz et al. 2017). Based on these results it was proposed that the NTRC-dependent redox balance of 2-Cys PRX modulates the redox state of the pool of plastid TRX and, consequently, the redox regulation of the TRX targets. These results provide an explanation for the indirect participation of NTRC in the redox regulation of the TRX-dependent processes of the chloroplast.

3 The Extraplastidial Thioredoxin Systems

While functions of chloroplastic TRX have been extensively studied, extraplastidial TRX systems are much less characterized in plants. Among the around 40 TRX found in Arabidopsis, at least 20 of them are extraplastidial isoforms, based on the absence of a putative plastidial targeting sequence (Meyer et al. 2005, 2012). However, for some of them, the respective localization is still unexplored. It is now clearly established that functional TRX systems are present in most cellular compartments, including mitochondria, cytosol, nucleus, and endomembrane sys- tems (see Fig. 4). In contrast to the large number of extraplastidial TRX, only two NTR genes (NTRA and NTRB in Arabidopsis) are found in the genome of most higher plants, suggesting multiple subcellular localizations of these isoforms. Con- sistently, NTR isoforms were found dual or even triple targeted in cytosol, mito- chondria, and nucleus in Arabidopsis and pea (Laloi et al. 2001; Serrato and Cejudo 2003; Reichheld et al. 2005; Marchal et al. 2014).

3.1 Cytosolic Thioredoxin System

Cytosolic TRX are composed of h-type (h for heterotrophic) and other type TRX. TRXh isoforms have been mostly characterized in Arabidopsis but orthologues are found in other plants including rice and poplar (Meyer et al. 2006). All TRXh isoforms consist of a single TRX domain harboring a WCG(P)C active site and, according to their respective clustering on phylogenetic trees, have been divided into three different subtypes (I, II, III). TRX from type I (TRXh1, h3, h4, h5) are mainly cytosolic and typically reduced by NTR (Rivera-Madrid et al. 1995). Type II (h2, h7, h8) and type III (h9, h10, atypical CxxS1 and 2) TRXh are less characterized isoforms. While predominantly located in the cytosol, TRXh2, h8, and h9 are also located in the endomembrane system, presumably through the fact that they harbor myristoylated/palmitoylated residues in their N-terminal extensions (see below) (Traverso et al. 2013). Remarkably, while generally reduced by NTR, some TRXh isoforms are also reduced by the alternative thiol reduction systems glutathione (GSH)/glutaredoxin (GRX) (Gelhaye et al. 2003; Reichheld et al. 2007; Koh et al. 2008; Meng et al. 2010). For example, in Arabidopsis and poplar, TRXh9 is reduced through a disulfide cascade mechanism involving the transient glutathionylation of a Cys located in the N-terminal extension of TRXh9 (Koh et al. 2008). I. Thormählen et al.

Fig. 4 Extraplastidial TRX systems in higher plants. Different plant TRX isoforms (yellow) are represented in their respective subcellular localization. When established, their respective reducers are represented (blue). Some key TRX target proteins (green) as well as the respective regulated functions (red spots) are also shown (see above for Abbreviations)

Only a few functions have been assigned to TRXh isoforms based on gene expression and genetic studies. Tissue-specific gene expression analyses of TRXh isoforms have revealed overlapping expression patterns in Arabidopsis (Reichheld et al. 2002; Belin et al. 2015). Overlapping gene expression between TRXh isoforms is likely explaining the lack of phenotype of most of the trxh knock-out mutants analyzed so far (our personal data). However, the characterization of a ntra ntrb double mutant deficient in the reduction of TRXh isoforms has revealed several plant development functions like pollen fitness, seed development, and plant growth (Reichheld et al. 2007). Moreover, it also shed light on cross talks acting between TRX and GRX systems in flower development and hormonal signaling (Reichheld et al. 2007; Marty et al. 2009; Bashandy et al. 2009, 2010). On the Elaborate Network of Thioredoxins in Higher Plants

Interestingly, the TRXh5 isoform is specifically involved in pathogen response pathways. TRXh5 is a major actor of plant immunity, regulating the oligomeric state of the non-pathogenesis-related protein expressor NPR1, a master regulator of the systemic acquired resistance (SAR). NPR1 normally exists in the cytosol as a thiol- bound oligomer. Upon pathogen challenge or treatment with salicylic acid, TRXh5 is upregulated (Laloi et al. 2004; Tada et al. 2008). TRXh5 then reduces NPR1 oligomers resulting in monomer release, translocation to the nucleus, and regulation of defense gene expression associated with both local and systemic resistance. Consequently, a trxh5 knock-out mutant is impaired in establishing a SAR and is therefore more susceptible to pathogen infection (Tada et al. 2008). More recently, Kneeshaw et al. (2014) have shown that TRXh5 acts in the SAR signaling by selective NPR1 protein S-denitrosylation. TRXh acting as denitrosylase has been additionally shown in ntra mutants in which higher levels of S-nitrosylated proteins were found (Tada et al. 2008). TRXh5 is also involved in the plant tolerance to the necrotrophic fungal pathogen Cochliobolus victoriae. The redox mechanism involves binding of victorin, the virulent effector of C. victoriae, to the first Cys residue in the TRXh5 active site (Sweat and Wolpert 2007; Lorang et al. 2012). Victorin binding to TRXh5 activates the locus orchestrating victorin effects protein LOV1, an Arabidopsis susceptibility protein, and elicits a resistance-like response that confers disease susceptibility. Lorang et al. (2012) proposed that victorin mimics a conventional pathogen virulence effector and confers virulence to C. victoriae solely because it incites plant defense mechanisms. Specifically, in self-incompatible species, TRXh isoforms (called THL in Brassica napus) have been involved in recognition and rejection of self-incompatible pollen through redox regulation of a key actor (S-locus receptor kinase) of the self- incompatible signaling pathway (Bower et al. 1996; Cabrillac et al. 2001; Ivanov and Gaude 2009). While this specific regulatory mechanism has been questioned (Yamamoto and Nasrallah 2009), works in other self-incompatible species (i.e., Solanaceae and Poaceae) have also established a role of TRXh isoforms in self-incompatible pollen recognition, while acting by other ways (Li et al. 1996; Juárez-Díaz et al. 2006). A function of cytosolic TRX isoforms in thermotolerance has been suggested. When overexpressed in plants, TRXh3 and the multidomain TRX tetratricoredoxin (TDX) enhance heat stress tolerance due to their chaperone activities (Park et al. 2009; Lee et al. 2009). Interestingly, TDX was shown to interact with the single- stranded DNA binding protein Ssb2, a yeast heat-shock protein 70 chaperone, suggesting that the TDX-dependent heat stress tolerance might be related to a co-chaperone activity (Vignols et al. 2003). Of note, none of the trxh3 and tdx knock-out mutants does exhibit a heat stress sensitive phenotype, suggesting that this function is compensated by other proteins sharing chaperone activities.

3.2 Nuclear Thioredoxin System

Although several TRX isoforms have been identified in the nucleus, evidence of their functions in the nucleus is still scarce. Nucleoredoxins (NRX) are multidomain I. Thormählen et al. proteins consisting of two or three TRX-like domains fused to a Cys-rich C-terminal domain. In Zea mays and Arabidopsis, characterized NRX isoforms share a dual cytosolic/nuclear localization and harbor putative nuclear localization signals (Laughner et al. 1998; Marchal et al. 2014). On the other hand, a cotton NRX orthologue (GbNRX1) was preferentially found in the secretory pathway and the apoplast (see below) (Li et al. 2016). Both NRX isoforms (NRX1 and 2) found in Arabidopsis exhibit disulfide reduction activities. While NRX1 can be reduced by NTR in vitro, constituting a functional TRX reduction pathway in the nucleus (Marchal et al. 2014), the NRX2 physiological reducer is still unknown. Genetic studies have assigned a function of NRX1 in plant fertility. It is required for pollen tube navigation through the pistil, guiding it to the ovule (Qin et al. 2009; Marchal et al. 2014). Still it is not yet established, if this function is dependent on the nuclear or the cytosolic localization of NRX1 and what is the pathway involved. Recent works have involved NRX1 in plant immunity responses (Kneeshaw et al. 2017). NRX1 gene expression is responsive to salicylic acid in Arabidopsis and poplar and is compromised in the SAR induced by pathogenic bacterial strains of Pseudomonas syringae (Kneeshaw et al. 2017). Upon pathogen trigger, NRX1 is induced and binds target enzymes of major hydrogen peroxide scavenging path- ways, including catalases. Mutant nrx1 plants displayed reduced catalase activity and are hypersensitive to oxidative stress. Remarkably, catalase is maintained in a reduced state by interaction with NRX1, a process necessary for its hydrogen peroxide scavenging activity. Thus, the works by Kneeshaw et al. (2017) conclude that hydrogen peroxide scavenging enzymes, like catalase, experience oxidative distress in reactive oxygen species (ROS)-induced environments and require reduc- tive protection from NRX1 for optimal activity. A pea orthologue of mitochondrial o-type TRX (PsTRXo1) shows a dual mito- chondrial and nuclear localization (Martí et al. 2009). Recent works by Calderón et al. (2017) show that PsTRXo1 interacts with proliferating cell nuclear antigen (PCNA) and that its overexpression in cell suspension affects cell growth and cell cycle progression. While coinciding with an upregulation of PCNA protein, the effect of PsTRXo1 expression on cell growth is not yet fully established (Calderón et al. 2017). TRXh isoforms accumulate in the nuclear compartment in tissues subjected to oxidative stress or genotoxic conditions (Serrato et al. 2001; Serrato and Cejudo 2003; Sarkar et al. 2005; Pulido et al. 2009) or during seed germination in which the NTR-TRXh system serves as a source of reducing power to regenerate 1-Cys (Pulido et al. 2009).

3.3 Mitochondrial Thioredoxin System

Although TRX and TRX reductase activities were reported in plant mitochondrial extracts two decades ago (Konrad et al. 1996; Banze and Follmann 2000), the first mitochondrial TRX system has been identified in Arabidopsis by Laloi et al. (2001). On the Elaborate Network of Thioredoxins in Higher Plants

In Arabidopsis, two NTR isoforms are localized in the mitochondria (NTRA and NTRB). Both of them reduce TRXo1, one of the two o-type TRX isoforms in Arabidopsis, which is targeted to the mitochondrial matrix through a cleavable N-terminal signal (Laloi et al. 2001; Reichheld et al. 2005). Mitochondrial localiza- tion of TRXo1 was later confirmed in other plants including poplar and pea (Gelhaye et al. 2004; Martí et al. 2009). Moreover, in poplar and Arabidopsis, the TRXh2 isoform has been identified in mitochondria, although its submitochondrial locali- zations have not been addressed yet (Gelhaye et al. 2004; Meng et al. 2010). Proteomic affinity approaches have been used to identify potential mitochondrial TRX using purified mitochondrial protein extracts (Balmer et al. 2004; Martí et al. 2009; Marchand et al. 2010; Yoshida et al. 2013). These approaches have been highly valuable to identify putative TRX mitochondrial targets and open the field for further validation. Nevertheless, a few enzymatic and genetic evidences have further validated those interactions yet. In pea, TRXo1 has been shown to be an efficient reducer of the PRXII-F, suggesting that TRXo is involved in mitochondrial ROS metabolism (Barranco-Medina et al. 2008). Pea TRXh2 and Arabidopsis TRXo1 were also shown to affect in vitro the dimerization state of the alternative oxidase (AOX) by reducing a heterodimeric disulfide (Cys78–Cys78), and further restoring the AOX activity (Gelhaye et al. 2004; Yoshida et al. 2013). Nevertheless, evidence is lacking to clearly favor the interpretation of a regulatory function over that of a maintenance function, i.e., to avoid Cys over-oxidation upon ROS exposure. More- over, genetic evidences are lacking to support this redox regulation in vivo. Both proteomic and in vitro works have pinpointed a potent TRX-dependent regulation of tricarboxylic acid (TCA) cycle enzymes (Balmer et al. 2004; Martí et al. 2009; Yoshida et al. 2013; Schmidtmann et al. 2014). Thiol switching of mitochondrial citrate synthase and isocitrate dehydrogenase activity were reported in vitro (Schmidtmann et al. 2014; Yoshida and Hisabori 2014), while thiol switching of malate dehydrogenase activity was found unchanged (Yoshida and Hisabori 2016a; Huang et al. 2017). In order to further address the function of TRX on the TCA cycle in vivo, Daloso et al. (2015) have characterized Arabidopsis knock-out mutants for mitochondrial TRXo1 (trxo1) and mitochondrial NTR (dou- ble mutant ntra ntrb). Both mutants were affected in mitochondrial metabolic fingerprints and flux patterns, suggesting a role of the matrix TRX system in regulating metabolism in vivo. In a complementary approach, Daloso et al. (2015) evaluated in vitro activities of TCA cycle and associated enzymes in mutants deficient in mitochondrial TRX. Among TCA cycle enzyme activities affected in the mutants, both succinate dehydrogenase (SDH) and fumarase (FUM) were pro- posed to be deactivated by reduced TRXo1, suggesting that TRXo is acting as a direct regulator of carbon flow through the TCA cycle (Daloso et al. 2015).

3.4 Endomembrane Thioredoxin System

While TRXh were previously described as cytosolic isoforms, more recent works have established alternative cellular localizations for some members of this family. I. Thormählen et al.

In Arabidopsis, Traverso et al. (2013) have reexamined the subcellular localiza- tions of members of the subtype II and III. By using a combination of predictive software and in vitro assays, they showed that members of the subtype II (TRXh2, h7, h8) and III (TRXh9, h10, atypical CxxS2) are subjected to N-terminal (Gly2 and Cys4) fatty acid acylations (N-myristoylation and S-palmitoylation). The role of these posttranslational modifications is to target proteins to membranes. Accord- ingly, TRX::GFP fusions revealed that myristoylation targeted proteins to the endomembrane system including the endoplasmic reticulum (ER) and Golgi, and that partitioning between this membrane compartment and the cytosol correlated with the catalytic efficiency of the N-myristoyltransferase acting at the N-terminus of the TRX. Moreover, a palmitoylatable Cys residue (only present in subtype III) flanking the N-myristoylation site was crucial to target proteins to the plasma membrane (Traverso et al. 2013). Independently, Meng et al. (2010) showed that N-myristoylation and S-palmitoylation of TRXh9 at Gly2/Cys4 residues is associ- ated with the capacity of the protein to move from cell-to-cell, suggesting a role of this TRXh isoform in intercellular communication and cell growth. In spite of the presence of distinct TRXh isoforms in the endomembrane system, their role is still unknown. While they might sustain other thiol oxidoreductases (i.e., protein disulfide isomerases, ER oxidases, etc.) in disulfide reduction/isomerization in the ER, no putative target proteins have been identified yet in these compartments. Concerning their reduction pathway, neither putative endomembrane nor plasma membrane NTR have been identified so far. Alternatively, GSH of the ER could constitute a physiological reducer of endomembrane TRX, as shown for subtype III isoform TRXh9 in poplar (Gelhaye et al. 2003; Koh et al. 2008). Of note, an atypical TRXh member of subtype III (CxxS2) is acting as a protein disulfide isomerase in vitro, which might be physiologically relevant (Serrato et al. 2008). Finally, an NRX1 isoform (GbNRX1) has recently been involved in the apoplast immune response to the pathogenic fungus Verticillium dahliae. Here, GbNRX1 functions in apoplastic ROS scavenging after the ROS burst that occurs upon recognition of the pathogen (Li et al. 2016).

4 Conclusions and Perspectives

In this review, we have discussed emerging topics of higher plant TRX. It is becoming clear that TRX form an elaborate network in plants being important to integrate metabolism, stress responses, development, and gene expression in a fluctuating environment. The chloroplast contains light- and NADPH-dependent TRX systems, which are combined via the redox balance of 2-Cys PRX to allow a cooperative control of chloroplast functions and photoautotrophic growth in higher plants. Extraplastidial TRX have been found to play critical and integrative roles in controlling pathogen defense and abiotic stress tolerance of plants interfacing with other signals such as ROS and phytohormones. Additional emerging fields in TRX functions are the regulation of plant development and the interaction of TRX with On the Elaborate Network of Thioredoxins in Higher Plants transcription factors and genetic control elements in the nucleus. Future studies are required to examine the interorganellar cross talk between TRX at the cellular level and to explore further in vivo functions in planta.

References

Arscott LD, Gromer S, Schirmer RH et al (1997) The mechanism of thioredoxin reductase from human placenta is similar to the mechanisms of lipoamide dehydrogenase and glutathione reductase and is distinct from the mechanism of thioredoxin reductase from Escherichia coli. Proc Natl Acad Sci U S A 94:3621–3626 Arsova B, Hoja U, Wimmelbacher M et al (2010) Plastidial thioredoxin z interacts with two fructokinase-like proteins in a thiol-dependent manner: evidence for an essential role in chlo- roplast development in Arabidopsis and Nicotiana benthamiana. Plant Cell 22:1498–1515 Ballicora MA, Frueauf JB, Fu Y et al (2000) Activation of the potato tuber ADP-glucose pyrophosphorylase by thioredoxin. J Biol Chem 275:1315–1320 Balmer Y, Vensel WH, Tanaka CK et al (2004) Thioredoxin links redox to the regulation of fundamental processes of plant mitochondria. Proc Natl Acad Sci U S A 101:2642–2647 Banze M, Follmann H (2000) Organelle-specific NADPH thioredoxin reductase in plant mitochon- dria. J Plant Physiol 156:126–129 Barranco-Medina S, Krell T, Bernier-Villamor L et al (2008) Hexameric oligomerization of mitochondrial peroxiredoxin PrxIIF and formation of an ultrahigh affinity complex with its electron donor thioredoxin Trx-o. J Exp Bot 59:3259–3269 Bartsch S, Monnet J, Selbach K, Quigley F, Gray J, von Wettstein D, Reinbothe S, Reinbothe C (2008) Three thioredoxin targets in the inner envelope membrane of chloroplasts function in protein import and chlorophyll metabolism. Proc Natl Acad Sci U S A 105:4933–4938 Bashandy T, Taconnat L, Renou J-P et al (2009) Accumulation of flavonoids in an ntra ntrb mutant leads to tolerance to UV-C. Mol Plant 2:249–258 Bashandy T, Guilleminot J, Vernoux T et al (2010) Interplay between the NADP-linked thioredoxin and glutathione systems in Arabidopsis auxin signaling. Plant Cell 22:376–391 Belin C, Bashandy T, Cela J et al (2015) A comprehensive study of thiol reduction gene expression under stress conditions in Arabidopsis thaliana. Plant Cell Environ 38:299–314 Benitez-Alfonso Y, Cilia M, San Roman A et al (2009) Control of Arabidopsis meristem develop- ment by thioredoxin-dependent regulation of intercellular transport. Proc Natl Acad Sci U S A 106:3615–3620 Bernal-Bayard P, Hervas M, Cejudo FJ et al (2012) Electron transfer pathways and dynamics of chloroplast NADPH-dependent thioredoxin reductase C (NTRC). J Biol Chem 287:33865–33872 Bernal-Bayard P, Ojeda V, Hervas M et al (2014) Molecular recognition in the interaction of chloroplast 2-Cys peroxiredoxin with NADPH-thioredoxin reductase C (NTRC) and thioredoxin x. FEBS Lett 588:4342–4347 Biteau B, Labarre J, Toledano MB (2003) ATP-dependent reduction of cysteine-sulphinic acid by S. cerevisiae sulphiredoxin. Nature 425:980–984 Bohrer A-S, Massot V, Innocenti G et al (2012) New insights into the reduction systems of plastidial thioredoxins point out the unique properties of thioredoxin z from Arabidopsis. J Exp Bot 63:6315–6323 Bower MS, Matias DD, Fernandes-Carvalho E (1996) Two members of the thioredoxin-h family interact with the kinase domain of a Brassica S locus receptor kinase. Plant Cell 8:1641–1650 Breazale VD, Buchanan BB, Wolosiuk RA (1978) Chloroplast sedoheptulose 1,7-bisphosphatase: evidence for regulation by the ferredoxin/thioredoxin system. Z Naturforsch 33c:521–528 Buchanan BB (1980) Role of light in the regulation of chloroplast enzymes. Annu Rev Plant Physiol 31:341–374 I. Thormählen et al.

Cabrillac D, Cock JM, Dumas C et al (2001) The S-locus receptor kinase is inhibited by thioredoxins and activated by pollen coat proteins. Nature 410:220–223 Calderón A, Ortiz-Espín A, Iglesias-Fernández R et al (2017) Thioredoxin (Trxo1) interacts with proliferating cell nuclear antigen (PCNA) and its overexpression affects the growth of tobacco cell culture. Redox Biol 11:688–700 Carrillo LR, Froehlich JE, Cruz JA et al (2016) Multi-level regulation of the chloroplast ATP synthase: the chloroplast NADPH thioredoxin reductase C (NTRC) is required for redox modulation specifically under low irradiance. Plant J 87:654–663 Cejudo FJ, Ferrandez J, Cano B et al (2012) The function of the NADPH thioredoxin reductase C-2- Cys peroxiredoxin system in plastid redox regulation and signalling. FEBS Lett 586:2974–2980 Chae HB, Moon JC, Shin MR et al (2013) Thioredoxin reductase type C (NTRC) orchestrates enhanced thermotolerance to Arabidopsis by its redox-dependent holdase chaperone function. Mol Plant 6:323–336 Chibani K, Tarrago L, Schürmann P et al (2011) Biochemical properties of poplar thioredoxin z. FEBS Lett 585:1077–1081 Collin V, Issakidis-Bourguet E, Marchand C et al (2003) The Arabidopsis plastidial thioredoxins: new functions and new insights into specificity. J Biol Chem 278:23747–23752 Collin V, Lamkemeyer P, Miginiac-Maslow M et al (2004) Characterization of plastidial thioredoxins from Arabidopsis belonging to the new y-type. Plant Physiol 136:4088–4095 Courteille A, Vesa S, Sanz-Barrio R et al (2013) Thioredoxin m4 controls photosynthetic alterna- tive electron pathways in Arabidopsis. Plant Physiol 161:508–520 Da Q, Wang P, Wang M et al (2017) Thioredoxin and NADPH-dependent thioredoxin reductase C regulation of tetrapyrrole biosynthesis. Plant Physiol 175:652–666 Daloso DM, Müller K, Obata T et al (2015) Thioredoxin, a master regulator of the tricarboxylic acid cycle in plant mitochondria. Proc Natl Acad Sci U S A 112:E1392–E1400 Dangoor I, Peled-Zehavi H, Levitan A et al (2009) A small family of chloroplast atypical thioredoxins. Plant Physiol 149:1240–1250 Eliyahu E, Rog I, Inbal D et al (2015) ACHT4-driven oxidation of APS1 attenuates starch synthesis under low light intensity in Arabidopsis plants. Proc Natl Acad Sci U S A 112:12876–12881 Entus R, Poling M, Herrmann KM (2002) Redox regulation of Arabidopsis 3-deoxy-D-arabino- heptulosonate 7-phosphate synthase. Plant Physiol 129:1866–1871 Geigenberger P, Fernie A (2014) Metabolic control of redox and redox control of metabolism in plants. Antioxid Redox Signal 21:1389–1421 Geigenberger P, Kolbe A, Tiessen A (2005) Redox regulation of carbon storage and partitioning in response to light and sugars. J Exp Bot 56:1469–1479 Geigenberger P, Thormählen I, Daloso DM et al (2017) The unprecedented versatility of the plant thioredoxin system. Trends Plant Sci 22:249–262 Gelhaye E, Rouhier N, Jacquot JP (2003) Evidence for a subgroup of thioredoxin h that requires GSH/Grx for its reduction. FEBS Lett 555:443–448 Gelhaye E, Rouhier N, Gérard J et al (2004) A specific form of thioredoxin h occurs in plant mitochondria and regulates the alternative oxidase. Proc Natl Acad Sci U S A 101:14545–14550 Gould SB, Waller RF, McFadden GI (2008) Plastid evolution. Annu Rev Plant Biol 59:491–517 Hertle AP, Blunder T, Wunder T et al (2013) PGRL1 is the elusive ferredoxin-plastoquinone reductase in photosynthetic cyclic electron flow. Mol Cell 49:511–523 Huang J, Niazi AK, Young D et al (2017) Self-protection of Arabidopsis cytosolic malate dehydrogenase against oxidative stress. J Exp Bot. https://doi.org/10.1093/jxb/erx396 Ikegami A, Yoshimura N, Motohashi K et al (2007) The CHLI1 subunit of Arabidopsis thaliana magnesium chelatase is a target protein of the chloroplast thioredoxin. J Biol Chem 282:19282–19291 Ishiga Y, Ishiga T, Wangdi T et al (2012) NTRC and chloroplast-generated reactive oxygen species regulate Pseudomonas syringae pv. tomato disease development in tomato and Arabidopsis. Mol Plant-Microbe Interact 25:294–306 On the Elaborate Network of Thioredoxins in Higher Plants

Ishiga Y, Ishiga T, Ikeda Y et al (2016) NADPH-dependent thioredoxin reductase C plays a role in nonhost disease resistance against Pseudomonas syringae pathogens by regulating chloroplast- generated reactive oxygen species. Peer J 4:e1938 Ivanov R, Gaude T (2009) Endocytosis and endosomal regulation of the S-receptor kinase during the self-incompatibility response in Brassica oleracea. Plant Cell 21:2107–2117 Jacquot JP, Rivera-Madrid R, Marinho P et al (1994) Arabidopsis thaliana NAPHP thioredoxin reductase. cDNA characterization and expression of the recombinant protein in Escherichia coli. J Mol Biol 235:1357–1363 Juárez-Díaz JA, McClure B, Vázquez-Santana S et al (2006) A novel thioredoxin h is secreted in Nicotiana alata and reduces S-RNase in vitro. J Biol Chem 281:3418–3424 Kirchsteiger K, Pulido P, Gonzalez M et al (2009) NADPH thioredoxin reductase C controls the redox status of chloroplast 2-Cys peroxiredoxins in Arabidopsis thaliana. Mol Plant 2:298–307 Kirchsteiger K, Ferrandez J, Pascual MB et al (2012) NADPH thioredoxin reductase C is localized in plastids of photosynthetic and nonphotosynthetic tissues and is involved in lateral root formation in Arabidopsis. Plant Cell 24:1534–1548 Kneeshaw S, Gelineau S, Tada Y et al (2014) Selective protein denitrosylation activity of thioredoxin-h5 modulates plant immunity. Mol Cell 56:153–162 Kneeshaw S, Keyani R, Delorme-Hinoux V et al (2017) Nucleoredoxin guards against oxidative stress by protecting antioxidant enzymes. Proc Natl Acad Sci U S A 114:8414–8419 Koh CS, Navrot N, Didierjean C et al (2008) An atypical catalytic mechanism involving three cysteines of thioredoxin. J Biol Chem 283:23062–23072 Konrad A, Banze M, Follmann H (1996) Mitochondria of plant leaves contain two thioredoxins. Completion of the thioredoxin profile of higher plants. J Plant Physiol 149:317–321 Laloi C, Rayapuram N, Chartier Y et al (2001) Identification and characterization of a mitochon- drial thioredoxin system in plants. Proc Natl Acad Sci U S A 98:14144–14149 Laloi C, Mestres-ortega D, Marco Y et al (2004) The Arabidopsis cytosolic thioredoxin h5 gene induction by oxidative stress and its W-box-mediated response to pathogen elicitor. Plant Physiol 134:1006–1016 Laughner BJ, Sehnke PC, Ferl RJ (1998) A novel nuclear member of the thioredoxin superfamily. Plant Physiol 118:987–996 Laugier E, Tarrago L, Courteille A et al (2013) Involvement of thioredoxin y2 in the preservation of leaf methionine sulfoxide reductase capacity and growth under high light. Plant Cell Environ 36:670–682 Lee JR, Lee SS, Jang HH et al (2009) Heat-shock dependent oligomeric status alters the function of a plant-specific thioredoxin-like protein, AtTDX. Proc Natl Acad Sci U S A 106:5978–5983 Lemaire SD, Michelet L, Zaffagnini M et al (2007) Thioredoxins in chloroplasts. Curr Genet 51:343–365 Lennartz K, Plücken H, Seidler A et al (2001) HCF164 encodes a thioredoxin-like protein involved in the biogenesis of the cytochrome b6f complex in Arabidopsis. Plant Cell 13:2539–2551 Lepistö A, Kangasjarvi S, Luomala EM et al (2009) Chloroplast NADPH-thioredoxin reductase interacts with photoperiodic development in Arabidopsis. Plant Physiol 149:1261–1276 Lepistö A, Pakula E, Toivola J et al (2013) Deletion of chloroplast NADPH-dependent thioredoxin reductase results in inability to regulate starch synthesis and causes stunted growth under short- day photoperiods. J Exp Bot 64:3843–3854 Li X, Nield J, Hayman D et al (1996) A self-fertile mutant of Phalaris produces an S protein with reduced thioredoxin activity. Plant J 10:505–513 Li YB, Han LB, Wang HY et al (2016) The thioredoxin GbNRX1 plays a crucial role in homeostasis of apoplastic reactive oxygen species in response to Verticillium dahliae infection in cotton. Plant Physiol 170:2392–2406 Lichter A, Häberlein I (1998) A light-dependent redox signal participates in the regulation of ammonia fixation in chloroplasts of higher plants – ferredoxin: glutamate synthase is a thioredoxin-dependent enzyme. J Plant Physiol 153:83–90 Lorang J, Kidarsa T, Bradford CS et al (2012) Tricking the guard: exploiting plant defense for disease susceptibility. Science 338:659–662 I. Thormählen et al.

Luo T, Fan T, Liu Y et al (2012) Thioredoxin redox regulates ATPase activity of magnesium chelatase CHLI subunit and modulates redox-mediated signaling in tetrapyrrole biosynthesis and homeostasis of reactive oxygen species in pea plants. Plant Physiol 159:118–130 Machida T, Ishibashi A, Kirino A et al (2012) Chloroplast NADPH-dependent thioredoxin reduc- tase from Chlorella vulgaris alleviates environmental stresses in yeast together with 2-Cys peroxiredoxin. PLoS One 7:e45988 Marchal C, Delorme-Hinoux V, Bariat L et al (2014) NTR/NRX define a new thioredoxin system in the nucleus of Arabidopsis thaliana cells. Mol Plant 7:30–44 Marchand CH, Vanacker H, Collin V et al (2010) Thioredoxin targets in Arabidopsis roots. Proteomics 10:2418–2428 Marri L, Zaffagnini M, Collin V et al (2009) Prompt and easy activation by specific thioredoxins of calvin cycle enzymes of Arabidopsis thaliana associated in the GAPDH/CP12/PRK supramo- lecular complex. Mol Plant 2:259–269 Martí MC, Olmos E, Calvete JJ et al (2009) Mitochondrial and nuclear localization of a novel pea thioredoxin: identification of its mitochondrial target proteins. Plant Physiol 150:646–657 Marty L, Siala W, Schwarzländer M et al (2009) The NADPH-dependent thioredoxin system constitutes a functional backup for cytosolic glutathione reductase in Arabidopsis. Proc Natl Acad Sci U S A 106:9109–9114 McKinney DW, Buchanan BB, Wolosiuk RA (1978) Activation of chloroplast ATPase by reduced thioredoxin. Phytochemistry 17:794–795 Meng L, Wong JH, Feldman LJ et al (2010) A membrane-associated thioredoxin required for plant growth moves from cell to cell, suggestive of a role in intercellular communication. Proc Natl Acad Sci U S A 107:3900–3905 Meyer Y, Reichheld J-P, Vignols F (2005) Thioredoxins in Arabidopsis and other plants. Photosynth Res 86:419–433 Meyer Y, Riondet C, Constans L et al (2006) Evolution of redoxin genes in the green lineage. Photosynth Res 89:179–192 Meyer Y, Belin C, Delorme-Hinoux V et al (2012) Thioredoxin and glutaredoxin systems in plants: molecular mechanisms, crosstalks, and functional significance. Antioxid Redox Signal 17:1124–1160 Michalska J, Zauber H, Buchanan BB et al (2009) NTRC links built-in thioredoxin to light and sucrose in regulating starch synthesis in chloroplasts and amyloplasts. Proc Natl Acad Sci U S A 106:9908–9913 Mikkelsen R, Mutenda KE, Mant A et al (2005) α-Glucan, water dikinase (GWD): a plastidic enzyme with redox-regulated and coordinated catalytic activity and binding affinity. Proc Natl Acad Sci U S A 102:1785–1790 Moon JC, Jang HH, Chae HB et al (2006) The C-type Arabidopsis thioredoxin reductase ANTR-C acts as an electron donor to 2-Cys peroxiredoxins in chloroplasts. Biochem Biophys Res Commun 348:478–484 Motohashi K, Hisabori T (2006) HCF164 receives reducing equivalents from stromal thioredoxin across the thylakoid membrane and mediates reduction of target proteins in the thylakoid lumen. J Biol Chem 281:35039–35047 Najera VA, Gonzalez MC, Perez-Ruiz JM et al (2017) An event of alternative splicing affects the expression of the NTRC gene, encoding NADPH-thioredoxin reductase C, in seed plants. Plant Sci 258:21–28 Naranjo B, Diaz-Espejo A, Lindahl M et al (2016a) Type-f thioredoxins have a role in the short- term activation of carbon metabolism and their loss affects growth under short-day conditions in Arabidopsis thaliana. J Exp Bot 67:1951–1964 Naranjo B, Mignee C, Krieger-Liszkay A et al (2016b) The chloroplast NADPH thioredoxin reductase C, NTRC, controls non-photochemical quenching of light energy and photosynthetic electron transport in Arabidopsis. Plant Cell Environ 39:804–822 Navrot N, Collin V, Gualberto J et al (2006) Plant glutathione peroxidases are functional peroxiredoxins distributed in several subcellular compartments and regulated during biotic and abiotic stresses. Plant Physiol 142:1364–1379 On the Elaborate Network of Thioredoxins in Higher Plants

Née G, Zaffagnini M, Trost P et al (2009) Redox regulation of chloroplastic glucose-6-phosphate dehydrogenase: a new role for f-type thioredoxin. FEBS Lett 583:2827–2832 Nikkanen L, Toivola J, Rintamäki E (2016) Crosstalk between chloroplast thioredoxin systems in regulation of photosynthesis. Plant Cell Environ 39:804–822 Ojeda V, Perez-Ruiz JM, Gonzalez MC et al (2017) NADPH thioredoxin reductase C and thioredoxins act concertedly in seedling development. Plant Physiol 174:1436–1448 Okegawa Y, Motohashi K (2015) Chloroplastic thioredoxin m functions as a major regulator of Calvin cycle enzymes during photosynthesis in vivo. Plant J 84:900–913 Park SK, Jung YJ, Lee JR et al (2009) Heat-shock and redox-dependent functional switching of an h-type Arabidopsis thioredoxin from a disulfide reductase to a molecular chaperone. Plant Physiol 150:552–561 Pascual MB, Mata-Cabana A, Florencio FJ et al (2010) Overoxidation of 2-Cys peroxiredoxin in prokaryotes: cyanobacterial 2-Cys peroxiredoxins sensitive to oxidative stress. J Biol Chem 285:34485–34492 Pascual MB, Mata-Cabana A, Florencio FJ et al (2011) A comparative analysis of the NADPH thioredoxin reductase C-2-Cys peroxiredoxin system from plants and cyanobacteria. Plant Physiol 155:1806–1816 Perez-Ruiz JM, Cejudo FJ (2009) A proposed reaction mechanism for rice NADPH thioredoxin reductase C, an enzyme with protein disulfide reductase activity. FEBS Lett 583:1399–1402 Perez-Ruiz JM, Spinola MC, Kirchsteiger K et al (2006) Rice NTRC is a high-efficiency redox system for chloroplast protection against oxidative damage. Plant Cell 18:2356–2368 Perez-Ruiz JM, Guinea M, Puerto-Galan L et al (2014) NADPH thioredoxin reductase C is involved in redox regulation of the Mg-chelatase I subunit in Arabidopsis thaliana chloroplasts. Mol Plant 7:1252–1255 Perez-Ruiz JM, Naranjo B, Ojeda V et al (2017) NTRC-dependent redox balance of 2-Cys peroxiredoxins is needed for optimal function of the photosynthetic apparatus. Proc Natl Acad Sci U S A 114:12069–12074 Perkins A, Nelson KJ, Parsonage D et al (2015) Peroxiredoxins: guardians against oxidative stress and modulators of peroxide signaling. Trends Biochem Sci 40:435–445 Puerto-Galan L, Perez-Ruiz JM, Ferrandez J et al (2013) Overoxidation of chloroplast 2-Cys peroxiredoxins: balancing toxic and signaling activities of hydrogen peroxide. Front Plant Sci 4:310 Puerto-Galan L, Perez-Ruiz JM, Guinea M et al (2015) The contribution of NADPH thioredoxin reductase C (NTRC) and sulfiredoxin to 2-Cys peroxiredoxin overoxidation in Arabidopsis thaliana chloroplasts. J Exp Bot 66:2957–2966 Pulido P, Cazalis R, Cejudo FJ (2009) An antioxidant redox system in the nucleus of wheat seed cells suffering oxidative stress. Plant J 57:132–145 Pulido P, Spinola MC, Kirchsteiger K et al (2010) Functional analysis of the pathways for 2-Cys peroxiredoxin reduction in Arabidopsis thaliana chloroplasts. J Exp Bot 61:4043–4054 Qin Y, Leydon AR, Manziello A et al (2009) Penetration of the stigma and style elicits a novel transcriptome in pollen tubes, pointing to genes critical for growth in a pistil. PLoS Genet 8: e1000621 Reichheld J-P, Mestres-Ortega D, Laloi C et al (2002) The multigenic family of thioredoxin h in Arabidopsis thaliana: specific expression and stress response. Plant Physiol Biochem 40:685–690 Reichheld J-P, Meyer E, Khafif M et al (2005) AtNTRB is the major mitochondrial thioredoxin reductase in Arabidopsis thaliana. FEBS Lett 579:337–342 Reichheld J-P, Khafif M, Riondet C et al (2007) Inactivation of thioredoxin reductases reveals a complex interplay between thioredoxin and glutathione pathways in Arabidopsis development. Plant Cell 19:1851–1865 Rey P, Pruvot G, Becuwe N et al (1998) A novel thioredoxin-like protein located in the chloroplast is induced by water deficit in Solanum tuberosum L. plants. Plant J 13:97–107 I. Thormählen et al.

Rey P, Cuine S, Eymery F et al (2005) Analysis of the proteins targeted by CDSP32, a plastidic thioredoxin participating in oxidative stress responses. Plant J 41:31–42 Richter AS, Peter E, Rothbart M et al (2013) Posttranslational influence of NADPH-dependent thioredoxin reductase C on enzymes in tetrapyrrole synthesis. Plant Physiol 162:63–73 Rivera-Madrid R, Mestres D, Marinho P et al (1995) Evidence for five divergent thioredoxin h sequences in Arabidopsis thaliana. Proc Natl Acad Sci U S A 92:5620–5624 Sahrawy M, Hecht V, Lopez-Jaramillo J et al (1996) Intron position as an evolutionary marker of thioredoxins and thioredoxin domains. J Mol Evol 42:422–431 Sanchez-Riego AM, Mata-Cabana A, Galmozzi CV et al (2016) NADPH-thioredoxin reductase C mediates the response to oxidative stress and thermotolerance in the cyanobacterium Anabaena sp. PCC7120. Front Microbiol 7:1283 Sanz-Barrio R, Corral-Martinez P, Ancin M et al (2013) Overexpression of plastidial thioredoxin f leads to enhanced starch accumulation in tobacco leaves. Plant Biotechnol J 11:618–627 Sarkar N, Lemaire S, Wu-Scharf D et al (2005) Functional specialization of Chlamydomonas reinhardtii cytosolic thioredoxin h1 in the response to alkylation induced DNA damage. Eukaryot Cell 4:262–273 Sasaki Y, Kozaki A, Hatano M (1997) Link between light and fatty acid synthesis: thioredoxin- linked reductive activation of plastidic acetyl-CoA carboxylase. Proc Natl Acad Sci U S A 94:11096–11101 Scheibe R, Anderson LE (1981) Dark modulation of NADP-dependent malate dehydrogenase and glucose-6-phosphate dehydrogenase in the chloroplast. Biochim Biophys Acta 636:58–64 Schmidtmann E, König AC, Orwat A et al (2014) Redox regulation of Arabidopsis mitochondrial citrate synthase. Mol Plant 7:156–169 Schürmann P, Buchanan BB (2008) The ferredoxin/thioredoxin system of oxygenic photosynthesis. Antioxid Redox Signal 10:1235–1274 Serrato AJ, Cejudo FJ (2003) Type-h thioredoxins accumulate in the nucleus of developing wheat seed tissues suffering oxidative stress. Planta 217:392–399 Serrato AJ, Crespo JL, Florencio FJ et al (2001) Characterization of two thioredoxins h with predominant localization in the nucleus of aleurone and scutellum cells of germinating wheat seeds. Plant Mol Biol 46:361–371 Serrato AJ, Perez-Ruiz JM, Cejudo FJ (2002) Cloning of thioredoxin h reductase and characteri- zation of the thioredoxin reductase-thioredoxin h system from wheat. Biochem J 367:491–497 Serrato AJ, Perez-Ruiz JM, Spinola MC et al (2004) A novel NADPH thioredoxin reductase, localized in the chloroplast, which deficiency causes hypersensitivity to abiotic stress in Arabidopsis thaliana. J Biol Chem 279:43821–43827 Serrato AJ, Guilleminot J, Meyer Y et al (2008) AtCXXS: atypical members of the Arabidopsis thaliana thioredoxin h family with a remarkably high disulfide isomerase activity. Physiol Plant 133:611–622 Seung D, Thalmann M, Sparla F et al (2013) Arabidopsis thaliana AMY3 is a unique redox- regulated chloroplastic α-amylase. J Biol Chem 288:33620–33633 Silver DM, Silva LP, Issakidis-Bourguet E et al (2013) Insight into the redox regulation of the phosphoglucan phosphatase SEX4 involved in starch degradation. FEBS J 280:538–548 Skryhan K, Cuesta-Seijo JA, Nielsen MM et al (2015) The role of cysteine residues in redox regulation and protein stability of Arabidopsis thaliana starch synthase 1. PLoS One 10: e0136997 Sparla F, Costa A, Lo Schiavo F et al (2006) Redox regulation of a novel plastid-targeted beta- amylase of Arabidopsis. Plant Physiol 141:840–850 Sweat TA, Wolpert TJ (2007) Thioredoxin h5 is required for victorin sensitivity mediated by a CC- NBS-LRR gene in Arabidopsis. Plant Cell 19:673–687 Tada Y, Spoel SH, Pajerowska-Mukhtar K et al (2008) Plant immunity requires conformational changes of NPR1 via S-nitrosylation and thioredoxins. Science 321:952–956 Tarrago L, Laugier E, Zaffagnini M et al (2010) Plant thioredoxin CDSP32 regenerates 1-Cys methionine sulfoxide reductase B activity through the direct reduction of sulfenic acid. J Biol Chem 285:14964–14972 On the Elaborate Network of Thioredoxins in Higher Plants

Thormählen I, Ruber J, von Roepenack-Lahaye E et al (2013) Inactivation of thioredoxin f1 leads to decreased light activation of ADP-glucose pyrophosphorylase and altered diurnal starch turn- over in leaves of Arabidopsis thaliana. Plant Cell Environ 36:16–29 Thormählen I, Meitzel T, Groysman J et al (2015) Thioredoxin f1 and NADPH-dependent thioredoxin reductase C have overlapping functions in regulating photosynthetic metabolism and plant growth in response to varying light conditions. Plant Physiol 169:1766–1786 Thormählen I, Zupok A, Rescher J et al (2017) Thioredoxins play a crucial role in dynamic acclimation of photosynthesis in fluctuating light. Mol Plant 10:168–182 Toivola J, Nikkanen L, Dahlstrom KM et al (2013) Overexpression of chloroplast NADPH- dependent thioredoxin reductase in Arabidopsis enhances leaf growth and elucidates in vivo function of reductase and thioredoxin domains. Front Plant Sci 4:389 Traverso JA, Micalella C, Martinez A et al (2013) Roles of N-terminal fatty acid acylations in membrane compartment partitioning: Arabidopsis h-type thioredoxins as a case study. Plant Cell 25:1056–1077 Valerio C, Costa A, Marri L et al (2011) Thioredoxin-regulated β-amylase (BAM1) triggers diurnal starch degradation in guard cells, and in mesophyll cells under osmotic stress. J Exp Bot 62:545–555 Vieira Dos Santos C, Laugier E, Tarrago L et al (2007) Specificity of thioredoxins and glutaredoxins as electron donors to two distinct classes of Arabidopsis plastidial methionine sulfoxide reductases B. FEBS Lett 581:4371–4376 Vignols F, Mouaheb N, Thomas D et al (2003) Redox control of Hsp70-Co-chaperone interaction revealed by expression of a thioredoxin-like Arabidopsis protein. J Biol Chem 278:4516–4523 Wang P, Liu J, Liu B et al (2013) Evidence for a role of chloroplastic m-type thioredoxins in the biogenesis of photosystem II in Arabidopsis. Plant Physiol 163:1710–1728 Williams CH, Arscott LD, Muller S et al (2000) Thioredoxin reductase two modes of catalysis have evolved. Eur J Biochem 267:6110–6117 Wimmelbacher M, Börnke F (2014) Redox activity of thioredoxin z and fructokinase-like protein 1 is dispensable for autotrophic growth of Arabidopsis thaliana. J Exp Bot 65:2405–2413 Wolosiuk RA, Buchanan BB (1977) Thioredoxin and glutathione regulate photosynthesis in chloroplasts. Nature 266:565–567 Wolosiuk RA, Buchanan BB, Crawford NA (1977) Regulation of NADP-malate dehydrogenase by the light-actuated ferredoxin/thioredoxin system of chloroplasts. FEBS Lett 81:253–258 Wood ZA, Poole LB, Karplus PA (2003) Peroxiredoxin evolution and the regulation of hydrogen peroxide signaling. Science 300:650–653 Wulff RP, Lundqvist J, Rutsdottir G et al (2011) The activity of barley NADPH-dependent thioredoxin reductase C is independent of the oligomeric state of the protein: tetrameric structure determined by cryo-electron microscopy. Biochemist 50:3713–3723 Yamamoto M, Nasrallah JB (2009) In planta assessment of the role of thioredoxin h proteins in the regulation of S-locus receptor kinase signaling in transgenic Arabidopsis. Plant Physiol 163:1387–1395 Yamaryo Y, Motohashi K, Takamiya K et al (2006) In vitro reconstitution of monogalacto- syldiacylglycerol (MGDG) synthase regulation by thioredoxin. FEBS Lett 580:4086–4090 Yoshida K, Hisabori T (2014) Mitochondrial isocitrate dehydrogenase is inactivated upon oxidation and reactivated by thioredoxin-dependent reduction in Arabidopsis. Front Environ Sci 2:38 Yoshida K, Hisabori T (2016a) Adenine nucleotide-dependent and redox-independent control of mitochondrial malate dehydrogenase activity in Arabidopsis thaliana. Biochim Biophys Acta 1857:810–818 Yoshida K, Hisabori T (2016b) Two distinct redox cascades cooperatively regulate chloroplast functions and sustain plant viability. Proc Natl Acad Sci U S A 113:E3967–E3976 Yoshida K, Noguchi K, Motohashi K et al (2013) Systematic exploration of thioredoxin target proteins in plant mitochondria. Plant Cell Physiol 54:875–892 Yoshida K, Hara S, Hisabori T (2015) Thioredoxin selectivity for thiol-based redox regulation of target proteins in chloroplasts. J Biol Chem 290:14278–14288