Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Functional characterization of a gain-of-function mutant of AtMKK9 in Arabidopsis thaliana Cluis, Corinne Pamela 2006

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-ubc_2006-0022.pdf [ 11.39MB ]
Metadata
JSON: 831-1.0092459.json
JSON-LD: 831-1.0092459-ld.json
RDF/XML (Pretty): 831-1.0092459-rdf.xml
RDF/JSON: 831-1.0092459-rdf.json
Turtle: 831-1.0092459-turtle.txt
N-Triples: 831-1.0092459-rdf-ntriples.txt
Original Record: 831-1.0092459-source.json
Full Text
831-1.0092459-fulltext.txt
Citation
831-1.0092459.ris

Full Text

F U N C T I O N A L C H A R A C T E R I Z A T I O N O F A G A I N - O F - F U N C T I O N M U T A N T OF A T M K K 9 I N A R A B I D O P S I S T H A L I A N A B y Corinne Pamela Cluis B.Sc. , M c G i l l University, 2003 A T H E S I S S U B M I T T E D I N P A R T I A L F U L F I L M E N T OF T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F M A S T E R OF S C I E N C E in T H E F A C U L T Y OF G R A D U A T E S T U D I E S (Botany) T H E U N I V E R S I T Y OF B R I T I S H C O L U M B I A December 2005 © Corinne Pamela Cluis, 2005 Abstract The relatively small number of M A P K K s encoded in the Arabidopsis genome suggests that this particular class of kinases acts as a point of convergence within the plant's 'integration of external stimuli and their transduction to elicit biological responses. In an effort to gain information about the function of the M A P K K , A t M K K 9 , in Arabidopsis, I have characterized several aspects of the phenotype of D E X : C A - M K K 9 - F L A G transgenic plants, which express an inducible constitutively active version of A t M K K 9 , C A - M K K 9 . I have found that C A - M K K 9 expression can control the production of ethylene by activating a downstream M A P K , A t M P K 6 , which is known to promote the stabilization of ethylene biosynthesis enzymes. C A - M K K 9 induction was correlated with an increase in A t M P K 6 activity in planta, and was rapidly followed by production of a burst of ethylene in the induced plant tissues. I hypothesized that C A - M K K 9 directly activates A t M P K 6 , and demonstrated that a recombinant version of C A - M K K 9 was capable of phosphorylating A t M P K 6 in vitro. In addition, the production of the hormone was abolished when C A - M K K 9 was expressed in a mpk6 knock-out background, thus proving that A t M P K 6 is required for CA-MKK9-media ted ethylene biosynthesis. I have also confirmed preliminary data from our laboratory suggesting that C A - M K K 9 plays a role in oxidative programmed cell death. The necrotic lesions induced by C A - M K K 9 were still observed in the mpk6 background, indicating that programmed cell death was triggered by C A - M K K 9 activity independently of A t M P K 6 activity and of ethylene overproduction. In addition, in order to investigate short-term transcriptional events triggered by C A - M K K 9 , I attempted to capture the transcriptional profile of D E X : C A -M K K 9 - F L A G plants using two-channel oligonucleotide microarrays. The C A - M K K 9 -affected genes included a number of genes involved in the octadecanoid pathway, and their promoters were enriched in A B R E - l i k e elements. However, my attempts to validate the microarray results using additional biological replicates and quantitative real-time P C R revealed that the majority of these early-response microarray results were apparently false positives, indicating that the microarray experiment was probably inappropriately constructed for capturing early transcriptional responses to C A - M K K 9 using the dexamethasone-inducible system. i i Table of contents Abstract ii Table of contents iii List of Tables vii List of Figures viii Abbreviations ix Acknowledgments x Research contributions . x Material contributions x Support and guidance x i 1. General Introduction 1 1.1. Programmed cell death in plants 1 1.1.1. Morphological and biochemical characteristics of P C D 1 1.1.2. P C D in plant development 2 1.1.3. P C D in stress responses 3 1.2. Signaling events controlling P C D 4 1.2.1. Experimental models in the study of P C D signaling pathways 4 1.2.2. Reactive oxygen species and nitric oxide in P C D 5 1.2.3. Salicylic acid: a positive regulator of P C D 7 1.2.4. Jasmonates: negative regulators of P C D 8 1.3. Ethylene in stress responses and P C D 9 1.3.1. Ethylene is a major signaling molecule in stress responses 9 1.3.2. Ethylene biosynthesis and its regulation during stress responses 10 1.3.3. Ethylene crosstalk with other hormones during P C D 12 1.4. Mitogen-activated protein kinase cascades and signal transduction 13 i i i 1.4.1. M A P K cascades in plants 15 1.4.2. M A P K signaling in plant stress responses 18 1.4.3. M A P kinase substrates 22 1.4.4. Genetic approaches to the study of plant M A P K cascades 23 1.5. Problem statement and thesis objectives 26 2. C A - M K K 9 modulates ethylene biosynthesis 28 2.1. Introduction 28 2.2. Material and methods 31 2.2.1. Plant material 31 2.2.2. R T - P C R analysis 31 2.2.3. Ethylene measurements 32 2.2.4. Hydrogen peroxide detection 33 2.2.5. In vitro phosphorylation assay 33 2.2.6. Protein extraction and western blot analysis 34 2.2.7. Generation of D E X : C A - M K K 9 - F L A G / m p M transgenic plants 35 2.3. Results 36 2.3.1. The C A - M K K 9 transgene is rapidly activated following D E X induction 36 2.3.2. C A - M K K 9 activity results in the formation of lesions 41 2.3.3. CA-MKK9-med ia t ed cell death is associated with H 2 O 2 accumulation... 42 2.3.4. C A - M K K 9 causes a rapid increase in ethylene biosynthesis 45 2.3.5. C A - M K K 9 is similar to C A - M K K 4 in acting as a mediator of ethylene biosynthesis 46 2.3.6. C A - M K K 9 induction correlates with increased activity of A t M P K 6 in vivo 48 2.3.7. A t M K K 9 activates A t M P K 6 in vitro 49 2.3.8. A t M P K 6 is necessary for CA-MKK9-media ted ethylene production 51 iv 2.3.9. A t M P K 6 is not required for CA-MKK9-med ia t ed P C D 53 2.4. Discussion 54 3. Investigation of short-term transcriptional events induced by CA-MKK9 63 3.1. Introduction 63 3.2. Material and methods 65 3.2.1. Plant material and treatments 65 3.2.2. Microarrays 66 3.2.3. Functional and promoter analysis 69 3.2.4. Quantitative Real-time R T - P C R 69 3.3. Results 71 3.3.1. Experimental design for the study of CA-MKK9-med ia t ed transcriptional changes 71 3.3.2. C A - M K K 9 mediates significant short-term transcriptional changes 75 3.3.3. Functional analysis of transcriptional changes resulting from C A - M K K 9 activity 76 3.3.4. Promoter analysis of CA-MKK9-af fec ted genes 78 3.3.5. C A - M K K 9 induction results in the down-regulation of jasmonate biosynthesis genes 81 3.3.6. Validation of the most robust CA-MKK9-af fec ted genes 87 3.4. Discussion 89 4. Future directions 94 4.1. Role of A t M K K 9 in ethylene biosynthesis in vivo 94 4.2. Investigation into the CA-MKK9-ac t iva ted 42 kDa M A P K 95 4.3. Signaling events in CA-MKK9-media ted P C D 96 4.4. Mechanisms of C A - M K K 9 regulation of W R K Y 3 0 97 4.5. Concluding remarks 98 5. Bibl iography 99 6. Appendix 110 6.1. Microarray data analysis 110 6.2. CA-MKK9-acf fec ted genes 114 vi List of Tables Table 3.1. Primers used for Q R T - P C R 70 Table 3.2. Gene ontology summaries of CA-MKK9-af fec ted genes 76 Table 3.3. Over-represented cw-acting elements among C A - M K K 9 - u p - and down-regulated genes 79 Table 3.4. Octadecanoid pathway genes significantly altered by C A - M K K 9 activity 82 Table 3.5. Most significant CA-MKK9-af fec ted genes 87 Table 6.1. Factorial design of the microarray experiment 110 Table 6.2. CA-MKK9-af fec ted genes 114 List of Figures Figure 1.1. Proposed M A P K cascades in Arabidopsis and tobacco 17 Figure 2.1. Construction of C A - M K K 9 and insertion into the DEX-inducible vector to produce p T A 7 0 0 2 - C A - M K K 9 - F L A G 37 Figure 2.2. Time-course of C A - M K K 9 activation following D E X induction 40 Figure 2.3. Time-course of cell death progression following C A - M K K 9 activation 42 Figure 2.4. C A - M K K 9 activity results in H2O2 accumulation 44 Figure 2.5. Time-course of ethylene production in D E X : C A - M K K 9 - F L A G plants 46 Figure 2.6. Ethylene production in D E X : C A - M K K 9 - F L A G and D E X : C A - M K K 4 - F L A G plants 47 Figure 2.7. A t M P K 6 protein and activity levels following C A - M K K 9 activation 49 Figure 2.8. C A - M K K 9 kinase activity on A t M P K 6 and A t M P K 3 in vitro 51 Figure 2.9. Time-course of C A - M K K 9 induction in D E X : C A - M K K 9 - F L A G / m / ? & ( 5 plants 52 Figure 2.10. A t M P K 6 is required for CA-MKK9-media ted ethylene production 53 Figure 2.11. Lesion formation in D E X : C A - M K K 9 - F L A G / m p k 6 plant 54 Figure 2.12. Model of C A - M K K 9 participation in ethylene biosynthesis 57 Figure 3.1. G V G transcript levels in D E X : C A - M K K 9 - F L A G and E V transgenic lines. 73 Figure 3.2. Microarray experimental design 75 Figure 3.3. Putative M A P K phosphorylation sites in ABRE-b ind ing factors 80 Figure 3.4. The jasmonate biosynthesis pathway 81 Figure 3.5. Technical validation of J A biosynthesis genes 84 Figure 3.6. Q R T - P C R validation of down-regulated jasmonate biosynthesis genes predicted by microarray to be down-regulated by C A - M K K 9 86 Figure 3.7. Q R T - P C R validation of the most significant CA-MKK9-af fec ted genes 88 Figure 6.1. Effect of dye type, transgenic lines and time-point factors on gene expression differentials 113 V l l l Abbreviations A B R E abscissic acid response element M A P K K A C C 1 -aminocyclopropane-1 -carboxy lie acid M A P K K K A C O 1 -aminocyclopropane-1 -carboxy lie MeJA acid oxidase ACS 1 -aminocyclopropane-1 -carboxy lie mRNA acid synthase MS AGI Arabidopsis genome initiative A N O V A analysis of variance N A D P H A T P Adenosine triphosphate NahG BLASTn basic local alignment sequence tool (nucleotide) N O 0?" bp base pair ONOO" BSA bovine serum albumine PCD C(T) Threshold cycle PCR C A constitutively active ppm Pro cDNA Complementary deoxyribonucleic acid Cy3 cyanine 3 bihexanoic acid dye p-value Cy5 cyanine 5 bihexanoic acid dye PVDF D A B 3,3'-diaminobenzidine PVPP DEPC Diethylpyrocarbonate QRT D E X Dexamethasone RNA D N A deoxyribonucleic acid ROS dNTP deoxynucleotide triphosphate rpm D T T Dithiothreitol RT dTTP deoxythymidine triphosphate SA dUTP deoxyuracil triphosphate S-AdoMet E D T A ethylenediaminetetraacetic acid SAR E G T A ethyleneglycol-6w(P-aminoethyl)- SDS N,N,N',N'-tetraacetic Acid SDS-GO gene ontology P A G E GST glutathione s-transferase Ser H 2 0 2 Hydrogen peroxide SOD HEPES N-(2-hydroxyethyl)piperazine-N;-(2-ethanesulfonic acid) SSC HR hypersensitive response TBST JA jasmonic acid, jasmonate Thr kDa kiloDalton Tris log Logarithm Tyr M A P K mitogen-activated protein kinase V S N mitogen-activated protein kinase kinase mitogen-activated protein kinase kinase kinase methyljasmonate messenger ribonucleic acid Murashige and Skoog nicotamide adenine dinucleotide phosphate salicylate hydroxylase nitric oxide superoxide anion peroxynitrite programmed cell death polymerase chain reaction part per million proline probability value polyvinylidene fluoride polyvinylpolypyrrolidone quantitative real-time ribonucleic acid reactive oxygen species revolution per minute reverse transcription salicylic acid S-adenosyl-methionine systemic acquired resistance sodium dodecyl sulfate sodium dodecyl sulfate polyacrylamide gel electrophoresis serine superoxide dismutase sodium chloride-sodium citrate solution tris-buffered saline/Tween-20 threonine tris hydroxymethylaminoethane tyrosine variance stabilization normalization ix Acknowledgments Research contributions The research presented here was greatly enhanced by the contributions of several colleagues. In particular, Marcus Samuel has made this whole project possible by cloning A t M K K 9 , and by generating the D E X : C A - M K K 9 - F L A G plants. Marcus first identified the cell death and oxidative burst phenotypes induced by C A - M K K 9 , which I then explored and characterized in this thesis (Chapter 2, Figure 2.3). Marcus also generated the original D E X : C A - M K K 4 - F L A G plants (Chapter 2, Figure 2.6). I wish to acknowledge the work of Somrudee Sritubtim, who first screened and maintained the transgenic D E X : C A - M K K 9 - F L A G and E V lines used in this study, and whose careful observations first pointed to a possible role of A t M K K 9 in ethylene biosynthesis (Chapter 2, Figure 2.5). In addition, I thank Jin Suk Lee, who was entirely responsible for the cloning and conversion into kinase-inactive form of M P K 3 1 and M P K 6 1 , and for the purification of all the recombinant proteins used for my in vitro kinase assays (Chapter 2, Figure 2.8). Finally, Rick White, a statistical consultant for Genome B C , designed and conducted the four-way A N O V A used for the analysis of my microarray results (Chapter 3). Material contributions I would like to thank Dr Yuel in Zhang (Botany Department, U B C ) for the generous gift of the mpk6 knock-out seeds, which allowed me to generate the D E X : C A - M K K 9 -¥LAG/mpk6 double mutant (Chapter 2). I would also like to thank the Genome B C -funded Treenomix project for allowing me to use their Arabidopsis full genome 70-mer oligo microarrays, and their facilities for array hybridization and analysis. Nathalie Mattheus provided expert advice and technical support throughout those array experiments. x Support and guidance This thesis, which represents by far my greatest scientific achievement, would not be as it is today without the encouragement, guidance and inspiration offered by many people. In particular, sincere thanks are due to Brian El l is , my research supervisor, for sharing his knowledge and passion for biology, and for giving more resources and freedom than most Masters students get to have. I am indebted as well to my colleagues of the Ell is lab, Hardy Hal l , Greg Lampard, A lex Lane, Jin Suk Lee, Marcus Samuel, Somrudee Sritubtim and Anki t Walia, whose technical expertise and critical inputs have greatly improved my work. I would also like to acknowledge the Botany Department, which, through the approachability and kindness of its members, and the quality and originality of its curriculum, has opened my mind to new fields and ideas in science. On that same note, I wish to thank all the good friends I have made among the Botany graduate students and postdocs, without whom my time in Vancouver would certainly not have been so much fun. Many thanks as well to my close family and friends, who have accompanied and supported me throughout my Master's program: Vincent, my life partner, for always being there for me, and whose support and care in the difficult times were invaluable. I also thank my father, Daniel, for teaching me curiosity, perseverance and for passing on the science bug. Finally, I wish to dedicate this thesis to the memory of my mother, Paulette, who was an inspiration for me throughout my life, and whom I miss everyday. x i 1. General Introduction 1.1. Programmed cell death in plants 1.1.1. Morphological and biochemical characteristics of P C D Programmed cell death (PCD) is a genetically determined process that functions to remove harmful or redundant cells during development or in response to environmental cues. P C D is most well studied in animal systems, but is now well established as playing major roles in the plant life cycle as well . Plant P C D , both development- and stress-induced, shares physiological and metabolic features with apoptosis, a well-studied form of P C D in animals. A t the cellular level, apoptosis is characterized by condensation of the chromatin, fragmentation of the D N A , shrinkage of the cytoplasm, and blebbing of the plasma membrane into so called-apoptotic bodies. Likewise, plant P C D has been found to involve the condensation and shrinkage of the cytoplasm, as well as the degradation of genomic D N A by nucleases and the formation of DNA-containing bodies (Hoeberichts and Woltering, 2003). Moreover, in animals, apoptosis is mainly associated with the activation of highly specific proteases called caspases, which are responsible for the ordered dismantlement of the metabolism that eventually kil ls the cell (Lam et al., 2001; Hoeberichts and Woltering, 2003). Inhibitor studies have suggested that caspase activity may also be important for plant P C D (Hoeberichts and Woltering, 2003), and one recent study described the purification and characterization of a caspase-like protein involved in P C D in tobacco (Chichkova et 1 al., 2004). Finally, metazoan apoptosis is also associated with the release of so-called cell-death activators, triggered by disruption of mitochondrial outer membrane integrity, a process which is hypothesized to occur in plants as well , since plant P C D has been linked to changes in mitochondrial membrane permeability (Lam et al., 2001; Hoeberichts and Woltering, 2003). 1.1.2. P C D in plant development Several plant developmental processes involve the participation of P C D . For example, P C D is essential to the formation of the endosperm, the development of both phloem and xylem tissue, and the formation of epidermal trichomes, and is also involved in the processes of pollen tube elongation and plant senescence (Greenberg, 1996; Rao and Davis, 2001). Senescence is the most extensively studied form of developmental P C D . Defined as the final stage of leaf development, senescence is an active process where nutrients from the leaves are relocated to other actively growing parts of the plant, such as the reproductive organs (Lohman et al., 1994; Weaver et al., 1998). Senescence is characterized by a decline in photosynthetic activity in the affected tissue, and is accompanied by the loss of chlorophyll, and increased protein degradation and lipid peroxidation (Mil ler et al., 1999; Quirino et al., 2000). Under normal conditions, senescence is a constant and predictable process that occurs in an age-dependent manner. However, senescence can also be accelerated by a number of stimuli, including drought, darkness and acute ozone treatment (Weaver et a l , 1998; Mi l le r et al., 1999). 2 1.1.3. P C D in stress responses P C D can occur during plant responses to both abiotic and biotic stresses. For example, P C D has been reported in plants challenged with cold, U V , and ozone (Koukalova et al., 1997; Danon and Gallois, 1998; Rao and Davis, 1999) and is proposed to serve as a means of restricting the spread of pathogens such as bacteria, fungi and viruses (Greenberg, 1996). In particular, P C D is an important part of the hypersensitive response (HR), a form of rapidly induced disease resistance triggered by specific pathogens on a gene-for-gene basis. It involves the expression of a range of defense molecules, as well as the death of cells at and surrounding the infection site (Heath, 2000). The cell death component of the H R aims at preventing the pathogen from acquiring nutrients and water, thus resulting in its containment and subsequent death (Beers and McDowel l , 2001). P C D during the H R also serves a signaling purpose, triggering defense responses in neighboring tissue and activating systemic acquired resistance in the whole plant (Heath, 2000). The H R occurs during a plant's interaction with an incompatible or avirulent pathogen. This incompatibility usually implies that the plant possesses a dominant R gene whose activity allows the plant cell to recognize and respond to the presence o f an avr gene in the pathogen. This recognition event is most well studied in plants responding to biotrophic fungal pathogens, or bacterial pathogens, and these studies have shown that there is considerable variation in the timing, the signaling events and the phenotype of the H R , both among plant species, and depending on the pathogen or elicitor involved (Heath, 2000). 3 1.2. Signaling events controlling PCD 1.2.1. Experimental models in the study of P C D signaling pathways The onset and spread of P C D is controlled by complex signaling networks. In recent years, these signaling events have mainly been studied in the context of either senescence or the H R , where P C D is known to bear structural and genetic similarities in both processes (Quirino et al., 2000; Rao and Davis, 2001). Moreover, research in the P C D field has been facilitated by the use of simple experimental models, which combine the use of model plants such as Arabidopsis thaliana (Arabidopsis), Lycopersicon esculentum (tomato) and Nicotiana tabacum (tobacco), either as whole plants or as cell cultures, with simplified P C D contexts. Stresses commonly employed in these studies are ozone, fungal toxins and bacterial elicitors. In particular, ozone stress has emerged as a simple and efficient tool for studying P C D . It is believed that ozone reacts with various molecules in the cell 's extracellular space, generating an oxidative burst which mimics the initial events of pathogen infection leading to the H R (Rao and Davis, 2001; Overmyer et al., 2003). Moreover, it has been suggested that low levels of ozone trigger premature and accelerated senescence (Pell et al., 1997). Altogether, research on these models has defined many of the signaling events regulating P C D . It appears that reactive oxygen species, salicylic acid, jasmonates and ethylene, are particularly important global regulators in the complex network of events that orchestrate the onset, propagation and containment of P C D . 4 1.2.2. Reactive oxygen species and nitric oxide in P C D Although some exceptions exist, the inter- and intra-cellular production of reactive oxygen species (ROS), called the oxidative burst, is considered as a hallmark of plant P C D (Heath, 2000). During plant defense, the oxidative burst is typically biphasic. The first phase, which occurs early and transiently, and is believed to be nonspecific as to the type of pathogen, forms part of a general stress response of the plant. The second phase of R O S accumulation, on the other hand, occurs later, is long-lasting, and appears to be specific to the H R response (Lamb and Dixon, 1997). The most common forms of R O S associated with the stress-induced oxidative burst include hydrogen peroxide (H2O2) and superoxide anion radicals (O2""). The production of R O S during the oxidative burst is thought to arise from the activity of several enzymes, including plasma membrane NADPH-oxidases , cell wall peroxidases, apoplastic amine oxidase, as well as from the mitochondrial electron transport chain (Lam et al., 2001; Rao and Davis, 2001; Overmyer et al., 2003). In particular, the role of a membrane-bound N A D P H oxidase complex sharing homology to mammalian enzymes, is well established to participate in the conversion of O2 into O2'" during plant defense (Torres et al., 2002; Yoshioka et al., 2003). The expression of plant genes coding for respiratory burst oxidase homologues (RBOH) is induced by HR-inducing elicitors. Moreover, both in tobacco and Arabidopsis, the phenotype of plants that have reduced expression of R B O H genes reveals that these enzymes play crucial roles in the H R triggered by pathogens (Torres et al., 2002; Yoshioka et al., 2003). In Arabidopsis, an atrbohD/F double-mutant displayed a major reduction in H2O2 accumulation following 5 infection by avirulent bacteria, and reduced P C D (Torres et al., 2002). Interestingly, knocking out either AtrbohD or AtrbohF alone does not have the same effect on pathogen-induced H R , suggesting that these enzymes interact in a complex way to modulate the onset of P C D (Torres et al., 2002). A recent study indicates that AtrbohD can also negatively regulate P C D , as the atrbohD mutant accentuates the severity of P C D of the runaway cell death mutant, Isdl (Torres et al., 2005). There is controversy regarding the relative importance of CV" and H2O2 in plant P C D . H2O2 is the R O S most commonly found associated with plant cell responses to pathogens, while ozone studies rather point towards CV" as the key signal for induction of P C D (Levine et al., 1994; Overmyer et a l , 2003). CV" does not diffuse across membranes, so it is likely to accumulate only briefly in the apoplast before it is converted to the more stable and diffusible H2O2 (Lamb and Dixon, 1997). The conversion of CV" into H2O2 can either occur spontaneously or be catalyzed by a superoxide dismutase (SOD) (Torres et al., 2002). It is possible that differences in the detected levels of these R O S among plant species reflect the rate at which O2'" is dismutated into H2O2 (Lamb and Dixon, 1997). The biological effects of R O S can be potentiated by nitric oxide (NO), which also plays a crucial role in plant P C D (Heath, 2000). It has been suggested that P C D is driven not so much by the concentration of individual R O S , as by a fine balance between O2'", H2O2, and nitric oxide (NO) levels (Delledonne et al., 2001). According to this model, O2'" is formed first, and then dismutated spontaneously or catalytically into H2O2. N O , also 6 produced upon pathogen infection, can either react with O2" to form ONOO", which unlike in animals is harmless to plants, or it can interact with H2O2 to activate P C D . In this model, i f the N O / O2'" /H2O2 ratio favours either of the first two molecules, the formation of O N O O " is facilitated. However, the onset of H R is associated with an increase in S O D transcription, resulting in the increased conversion of superoxide anion to H2O2. This enhanced accumulation of H2O2 would promote the synergistic action of H 2 0 2 and N O on stimulating P C D (Delledonne et a l , 2001). The regulation of the oxidative burst itself involves multiple signaling events, including G-protein signaling, kinase activity and ion fluxes across the plasma membrane (Lamb and Dixon, 1997). In particular, Ca influxes are suggested to mediate the activity of the N A D P H oxidase complex, perhaps through the activity of a calcium-dependent protein kinase (Xing et al., 1997). In fact, kinase inhibitor studies have also revealed that kinase activity is necessary both for H2O2 accumulation and for the P C D subsequently induced during plant defense (Lamb and Dixon, 1997). In particular, the oxidative burst triggered by ozone is associated with increased activity of mitogen-activated protein kinases ( M A P K s ) (Samuel et al., 2000). 1.2.3. Salicylic acid: a positive regulator of P C D The phenylpropanoid metabolite, salicylic acid ( S A ) , is a well-established player in plant defense against pathogens, and particularly during the H R . S A is considered a positive regulator of P C D , since a number of 'lesion mimic ' mutants, which display spontaneous lesion formation, can be rescued when cellular S A is depleted by ectopic expression of 7 the bacterial enzyme, salicylate hydroxylase (NahG) (Lorrain et al., 2003). The Arabidopsis ecotype, Cvi -0 , which is hypersensitive to ozone relative to the 'Columbia ' ecotype, also hyperaccumulates S A (Rao and Davis, 1999). Exogenous S A application has been shown to be associated with increased H2O2 production, possibly by stimulating SOD activity (Rao et a l , 1997). Therefore S A is thought to cooperate with R O S , perhaps in a positive feedback loop, to induce the onset of P C D . Furthermore, mutating the isochorismate synthase gene required for S A biosynthesis, EDS 16, suppresses the runaway cell death phenotype of the Isdl mutant, indicating that S A is also required for the propagation of P C D (Torres et al., 2005). 1.2.4. Jasmonates: negative regulators of P C D Jasmonic acid (JA) and its methyl-ester (MeJA) are the final products of the cyclization branch of the octadecanoid pathway. The jasmonates, as well other long-chain fatty acid-derived intermediates and derivatives, commonly referred to as octadecanoids, are well-characterized signals operating within plant responses to biotic and abiotic stresses, including wounding, insect herbivory, pathogen attack and elevated ozone levels (Devoto and Turner, 2005). Jasmonates are thought to act as negative regulators of P C D (Rao and Davis, 2001; Overmyer et al., 2003), since treatment with M e J A preceding ozone exposure blocks the spread of P C D lesions in both tobacco and Arabidopsis Cvi-0 plants (Orvar et al., 1997; Rao et al., 2000). Moreover, the JA-insensitive mutant jar-1 displays aberrant lesion spreading in response to ozone, consistent with a role for J A in the containment of P C D to a limited number of cells (Overmyer et al., 2000). 8 1.3. Ethylene in stress responses and PCD 1.3.1. Ethylene is a major signaling molecule in stress responses Ethylene is a gaseous two-carbon molecule that plays central roles in many developmental processes such seed germination, root hair formation, abscission, and senescence. Ethylene also participates in responses to abiotic stresses that include wounding and hypoxia, and in plant defenses against pathogens (Beers and McDowel l , 2001; Wang et al., 2002), and not surprisingly, is the hormone the most ubiquitously involved in plant P C D (Greenberg, 1996; Beers and McDowe l l , 2001). In Arabidopsis, the ethylene-overproducing mutants, etol and eto3, show increased lesion formation in response to ozone, and this effect can be blocked by inhibitors of ethylene biosynthesis (Rao et al., 2002). Furthermore, inhibitor studies in tomato have shown that both ethylene biosynthesis and ethylene perception are required for development of an ozone-induced oxidative burst and subsequent P C D (Moeder et al., 2002). Since ethylene production co-localizes with, and is required for, H2O2 accumulation, ethylene might be involved in directly regulating R O S production (Moeder et al., 2002). Our understanding of the role of ethylene in P C D has been further refined by the results of crossing an ozone-sensitive mutant, radical-induced cell death 1 (rcdl), with the ethylene-insensitive mutant, einl. The phenotype of the double mutant suggests that, while ethylene signaling is not required for the initiation of lesions during ozone treatment, it is necessary for the subsequent amplification of P C D once the ozone treatment ends (Overmyer et al., 2000). Altogether, those data indicate that ethylene is a 9 positive regulator of P C D and that it participates, along with R O S , in the propagation of P C D (Overmyer et al., 2003). 1.3.2. Ethylene biosynthesis and its regulation during stress responses The biochemical reactions leading to ethylene biosynthesis are well understood, as'are the enzymes catalyzing these reactions (Wang et al., 2002). The biosynthesis of ethylene starts with the conversion of S-adenosyl-methionine (S-AdoMet) into 1-aminocyclopropane-1-carboxylie acid ( A C C ) by the enzyme A C C synthase (ACS) . A C C is then oxidized by a second enzyme, A C C oxidase ( A C O ) , to form ethylene. The conversion of S-Ado-Met to A C C by A C S is considered the rate-limiting step in ethylene biosynthesis, due to the short half-life of the A C S protein and tight regulation of A C S activity (Tatsuki and M o r i , 2001; Wang et al., 2002; Tsuchisaka and Theologis, 2004). In Arabidopsis, the A C S gene family codes for nine A C S proteins, eight of which are functional. The A C S holoenzyme is also dimeric, which provides additional opportunities for functional diversity (Chae and Kieber, 2005). The A C O gene family, on the other hand, has only three putative members in Arabidopsis. The expression of the A C S genes in Arabidopsis is differentially regulated by developmental and stress stimuli (Tsuchisaka and Theologis, 2004). In tobacco, ozone treatment induces a biphasic increase in both A C S and A C O transcripts, which is correlated with a burst of ethylene (Moeder et al., 2002). Moreover, supporting the hypothesis that ethylene is closely involved in the production of R O S , the expression of A C O in tomato, as revealed in A C O promoter-P-glucuronidase fusion plants, co-localizes with the sites of accumulation of H2O2 (Moeder et al., 2002). In Arabidopsis, lesion 10 formation in the rcdl mutant is preceded by increased levels of A C S and A C O transcripts (ACS6 and A C O l , respectively) (Overmyer et al., 2000), a pattern that was also observed in tomato, where ozone treatment triggered a rapid increase in A C O m R N A , followed by a smaller increase in A C S 2 transcription (Tuomainen et al., 1997). However, the up-regulation of A C S and A C O genes cannot always fully account for the rapid ethylene burst triggered in response to certain stimuli ( K i m et al., 2003). The post-translational control of A C S enzymes, in particular, appears to also play an important role in stress-induced ethylene production. Several indirect lines of evidence have pointed specifically to a role for protein phosphorylation in modulating A C S activity. For instance, the addition of a protein kinase inhibitor rapidly blocked the increased activity of A C S triggered by fungal elicitors and by ozone in tomato, whereas the addition of a phosphatase inhibitor could, by itself, stimulate A C S activity (Spanu et al., 1994; Tuomainen et al., 1997). Interestingly, the phosphatase inhibitor had no effect on A C S activity when added simultaneously with a protein synthesis inhibitor, suggesting that protein phosphorylation and dephosphorylation influence the turnover of A C S rather than its catalytic activity (Spanu et al., 1994). A subsequent study directly demonstrated that tomato A C S 2 is phosphorylated at its C-terminus by a wound-induced calcium-dependent protein kinase (Tatsuki and M o r i , 2001). Further support for the idea that phosphorylation helps regulate A C S turnover came from the discovery that a protein called E T O l interacts with the A C S 5 C-terminus, thus blocking its modification and presumably promoting the degradation of A C S 5 by a proteasome-dependent pathway (Wang et al., 2004). Bringing these pieces of evidence together, L i u et al. (2004) established that a 11 specific M A P K , A t M P K 6 , could increase the stability, and thereby the activity, of two Arabidopsis A C S isoforms, A C S 2 and A C S 6 , by phosphorylating three conserved serine residues at their C-terminus. Based on these results, the authors proposed a model whereby biotic and abiotic stresses trigger the activation of a M A P K cascade composed of an as yet unknown M A P K K K , which acts upon A t M K K 4 and/or A t M K K 5 , which in turn act upon A t M P K 6 . Activated A t M P K 6 then phosphorylates A C S 6 , thereby preventing it from being targeted for degradation, and allowing increased conversion of S-AdoMet to A C C . The A C C would be further converted into ethylene by A C O , resulting in the onset of the ethylene response (Liu and Zhang, 2004). 1.3.3. Ethylene crosstalk with other hormones during P C D There is evidence for both synergistic and antagonistic interactions between ethylene and S A . For instance, the Arabidopsis mutants, etol and eto3, hyperaccumulate ethylene and have a higher S A content than the wild-type plants (Rao et al., 2002). Moreover, analyses of the S A signaling mutant, npr-1, as well as of the SA-depleted transgenic NahG plants, both revealed that the ethylene burst induced by ozone is S A signaling-dependent (Rao et al., 2002). This suggests that S A plays a role in both the initiation of lesions and also in their propagation, perhaps by controlling ethylene production (Rao et al., 2002). In contrast with this, ozone-treated transgenic tobacco plants over-expressing the A t M P K 6 orthologue, NtSIPK, were found to be blocked in ozone-induced S A accumulation, but to produce elevated amounts of ethylene (Samuel et al., 2005). These conflicting pieces of evidence may reflect species differences, or additional levels of complexity in the crosstalk between the two hormones. 12 Jasmonates also modulate the ethylene pathway. Analyses of jarl and ein2 mutants in Arabidopsis have revealed that the J A pathway negatively regulates ethylene signaling during ozone-induced P C D . Blocking ethylene perception in the ein2 mutant background, or by addition of the ethylene-receptor agonist norbornadiene, both reduced ozone-induced lesion formation in the jarl mutant (Tuominen et al., 2004). On the other hand, the jarl mutant did not display increased ethylene production in response to ozone, although it did show increased ethylene-mediated gene expression, indicating that jasmonates suppress the ethylene pathway downstream of ethylene biosynthesis (Tuominen et al., 2004). Since J A treatment induces the transcription of ethylene receptors, which themselves act as negative regulators until they bind the hormone, it has been suggested that J A could negatively regulate ethylene by decreasing the cell's sensitivity to the latter (Overmyer et al., 2003). It is thus possible that the lesion containment role of J A is executed by reducing the ethylene signal and the ethylene-dependent R O S production required for the propagation of P C D (Overmyer et al., 2003). 1.4. Mitogen-activated protein kinase cascades and signal transduction Eukaryotes have developed sophisticated signal transduction strategies to quickly and precisely transmit external stimuli to the interior of cells. One of the most efficient signal transduction methods involves the phosphorylation of proteins, mediated by kinases, and their dephosphorylation by phosphatases. 13 Mitogen-activated protein kinases ( M A P K ) cascades are ubiquitous players in eukaryotic signal transduction pathways, but they are best characterized in mammalian and yeast models. Typically, M A P K cascades are composed of a M A P K kinase kinase ( M A P K K K ) phosphorylating a M A P K kinase ( M A P K K ) , which in turn phosphorylates a M A P K . M A P K K K s are activated either by M A P K K K kinases ( M A P K K K K ) or through an interaction with GTP-binding proteins (Widmann et al., 1999). They activate M A P K K s by phosphorylating two serine/threonine residues in the M A P K K activation loop (Ser/Thr-(X)3. 5-Ser/Thr). M A P K K s , in turn, are dual-specificity kinases that recognize and dually phosphorylate a conserved threonine/tyrosine motif (Thr-X-Tyr) in the activation loop of M A P K s (Widmann et al., 1999; Zhang and Klessig, 2001). Finally, M A P K s are proline-directed serine/threonine kinases that target one or more sites distinguished by the consensus sequence Ser/Thr-Pro in their target proteins (Treisman, 1996; Widmann et al., 1999). In mammalian cells, M A P K cascades are involved in various cell processes, including differentiation, proliferation, stress responses and apoptosis. Mammalian M A P K s have been divided into three families: the E R K family, activated by growth factors such as E G F , the J N K / S A P K (Jun N-terminal kinase/ stress-activated protein kinase) family, which responds to stress and inflammatory cytokines, and the p38/Hog family, activated by osmotic stress, endotoxins and cytokines (Morris, 2001). These M A P K s activate a wide range of targets, which include transcription factors, protein kinases, phospholipases and cytoskeletal proteins (Widmann et al., 1999). 14 With multiple M A P K cascade components that each can interact with multiple targets, it is striking that M A P K cascades can achieve specificity in their downstream targets following activation by a given stimulus. In mammalian and yeast systems, specificity of M A P K modules is achieved through the formation of 'signalosomes', protein complexes that include the kinases themselves, as well as scaffold proteins (Morris, 2001). 1.4.1. M A P K cascades in plants Although they are far from being as well characterized as in yeast and mammalian systems, increasing numbers of studies indicate that M A P K signaling cascades also play critical roles in plants. For example, M A P K cascades have been implicated in cytokinesis, cell death, differentiation and stress responses (Tena et al., 2001; Jonak et al., 2002). In Arabidopsis, genomic analyses reveal that there are 20 M A P K s , 10 M A P K K s and 80 putative M A P K K K s . Based on structural motifs and sequence similarities, the plant M A P K s , which all belong to the E R K family, have been divided into four groups (A-D) . Except for members of the most distant group D , which carry a Thr-Asp-Tyr phosphorylation motif, all plant M A P K s are activated on a Thr-Glu-Tyr motif. Plant M A P K K s have also been divided into four groups (A-D) . In contrast to mammalian M A P K K s where only three amino acids separate the conserved serine/threonine residues of the activation motif, plant M A P K K s have the consensus sequence S e r / T h r - X X X X X -Ser/Thr. Plant M A P K K K s are divided into three main groups. Group A M A P K K K s are 15 most similar to characterized yeast and animal M A P K K K s , while groups B and C are more closely related to R A F kinases (Ichimura et al., 2002). Although the high numbers, as well as the diversity of plant M A P kinase cascade components suggest that they may be extensively involved in plant signal transduction, there is relatively little concrete information about functional interactions between the various M A P K K K s , M A P K K s and M A P K s . The difficulty in unravelling biologically relevant interactions between M A P K cascades components, in planta has a number of origins, including the transient nature of M A P K K K / M A P K K / M A P K interactions, as well as functional redundancy and crosstalk between M A P K modules. Evidence so far for plant M A P K K K / M A P K K / M A P K interactions comes mainly from in vitro, yeast two-hybrid and transgenic approaches. For example, interactions were detected between the Arabidopsis M A P K K K , A t M E K K l , the M A P K K , A t M K K 2 , and the M A P K , A t M P K 4 , using yeast two-hybrid and yeast M A P K mutant complementation (Ichimura et al., 1998; Mizoguchi et al., 1998). A similar approach was used in tobacco to demonstrate that the M A P K K K , N t N P K l , can activate the M A P K K , N t M E K l , which in turns phosphorylates the M A P K , NfNTF6 . The validity of these predicted interactions was further supported by in planta evidence for the co-activation of these M A P K cascade components, as well as by the analysis of immuno-complexes following the over-expression of N t M E K l - and NtNTF6-fusion proteins in B Y - 2 tobacco cells (Nishihama et al., 2001; Soyano et al., 2003). The latter approach, based on the expression of epitope-tagged proteins, was used to show that, in Arabidopsis protoplasts, A t M E K K l phosphorylates A t M K K 5 , and that A t M K K 4 and A t M K K 5 could both activate A t M P K 3 and A t M P K 6 (Asai et al., 2002). 16 The role for A t M K K 4 / A t M K K 5 as upstream activators of A t M P K 3 / A t M P K 6 is further supported by studies on their tobacco orthologues, the M A P K K , N f M E K 2 , and the M A P K s , NtSEPK and NtWTPK, respectively. N t M E K 2 was identified as acting upstream of N tSIPK and NtWTPK, based on the co-activation o f N t M E K 2 and NtSJPK/NtWTPK in suspension cells treated with a fungal elicitor, and the activation o f NtSIPK/NtWTPK in planta by a transiently expressed constitutively active N t M E K 2 (Yang et al., 2001b). M A P K cascades characterized so far in Arabidopsis and tobacco are summarized in Figure 1.1. j Arabidopsis tobacco ^ Arabidopsis tobacco Arabidopsis tobacco (^NtNPK?) MAPKK MAPK Target NtWIF (NtMPKj) ACS6/ ACS2 WRKY1 AtMKK6 ) ( NtMEKl ^ t M P M ^ (NtNTR^ Figure 1.1. Proposed MAPK cascades in Arabidopsis and tobacco. Each column corresponds to a characterized Arabidopsis MAPK cascade (black), and to the orthologous cascade in tobacco (grey). Full arrows indicate interactions supported by in planta evidence. Dotted arrows indicate interactions supported by in vitro and/or yeast expression evidence. 17 1.4.2. M A P K signaling in plant stress responses Inhibitor studies provided the earliest evidence of a role for kinase activity during plant stress responses (Viard et al., 1994; Suzuki and Shinshi, 1995). Numerous studies have since reported changes in the m R N A levels and/or in enzymatic activity of M A P K s in response to various stresses (Seo et al., 1995; Jonak et al., 1996; Bogre et al., 1997; Zhang and Klessig, 1998b). In the past few years, the identity of several of these M A P K s has been established, and the study of loss- and gain-of-function mutants of specific M A P K cascade components has increased our understanding of their roles in stress responses. Given the estimated number of M A P K s in Arabidopsis, it is striking that only a small number of M A P K s appear ubiquitously involved in plant stress responses. Three Arabidopsis M A P K s , as well as their respective tobacco homologues (Figure 1.1), appear to play particularly important roles in stress signaling: A t M P K 6 (NtSIPK), and A t M P K 3 (NtWIPK), which belong to the Group A M A P K s , and A t M P K 4 ( N t M P K 4 ) , which is a member of the Group B M A P K s (Ichimura et al., 2002). In Arabidopsis, A t M P K 6 was found to be activated by bacterial and fungal elicitors (Nuhse et al., 2000). Furthermore,, loss-of-function studies indicate that this particular M A P K is essential for resistance to virulent and avirulent pathogens (Menke et al., 2004). A s discussed above, there is now convincing evidence that one o f the functions of A t M P K 6 is to modulate ethylene production by stabilizing A C S isoforms during stress responses (L iu and Zhang, 2004). Interestingly, A t M P K 6 is often co-activated with 18 A t M P K 3 . For instance, oxidative and hyperosmotic stresses activate both A t M P K 6 and A t M P K 3 (Kovtun et al., 2000; Droillard et al., 2002; Rentel et al., 2004). A t M P K 6 and A t M P K 3 are also co-activated during ozone treatment, with A t M P K 6 displaying transient activity, while A t M P K 3 shows delayed but sustained activity (Ahlfors et al., 2004). Interestingly, ozone exposure results in the translocation of A t M P K 3 and A t M P K 6 into the nucleus, suggesting that they might modulate the activity of transcription factors (Ahlfors et al., 2004). The M A P K K s , A t M K K 4 and A t M K K 5 , can both activate A t M P K 3 and A t M P K 6 , suggesting a role for these M A P K K s in co-regulating the two kinases (Figure 1.1) (Asai et al., 2002). A t M P K 4 plays a negative role in the regulation of systemic acquired resistance (SAR) , perhaps by impacting S A and J A signaling. This model is supported by the phenotype of a knock-out mutant of the kinase, which displays constitutive S A R and resistance to virulent pathogens, as well as elevated S A levels (Petersen et al., 2000). A t M P K 4 , A t M P K 6 and A t M P K 3 are all reported to be co-activated by hypoosmolarity, as well as by the bacterial elicitor, flagellin (Droillard et al., 2004). There is also evidence for activation of A t M P K 4 solely with A t M P K 6 . For instance, both these M A P K s are transiently activated by the bacterial elicitor hairpin, independently of the subsequent oxidative burst (Desikan et al., 2001). Interestingly, the same study suggested that A t M P K 4 plays a particularly important role in hairpin-induced P C D . In a similar manner, another study demonstrated that these two same M A P K s are also activated by a number of abiotic stresses including low temperature, dehydration, touch, hyperosmolarity and wounding (Ichimura et al., 1998). A role for A t M P K 6 and A t M P K 4 in cold and salt 19 stress response is further supported by a more recent paper showing that their activation by A t M K K 2 results in increased freezing and salt tolerance (Teige et al., 2004), although the underlying mechanism remains unknown. The tobacco orthologue of A t M P K 6 is NtSIPK, which was originally identified as a S A -induced protein kinase (Zhang and Klessig, 1997). N t S I P K is rap idly activated in cell cultures and whole tobacco plants following treatment with the tobacco mosaic virus, fungal elicitor or mechanical stress (Zhang and Klessig, 1998a; Romeis et al., 1999). The orthologue of A t M P K 3 , N t W I P K , was first characterized as a wound-inducible protein kinase whose transient activity is required for wound-induced J A and M e J A accumulation (Seo et al., 1995; Seo et a l , 1999), although N t W I P K is also activated by certain fungal elicitors (Romeis et al., 1999; Zhang et al., 2000). However, in contrast to NtSIPK, whose increased activity appears solely due to post-translational mechanisms, N t W I P K activation is typically preceded by increases in both m R N A and protein levels (Zhang and Klessig, 1998b; Zhang et al., 2000). Interestingly, N t W I P K activation by tobacco mosaic virus requires the expression of the resistance gene N , suggesting that the kinase plays a role in the plant's defense strategy (Zhang and Klessig, 1998b). Ozone treatment, as well as external application of R O S such as H2O2, also leads to the rapid and sustained activation of NtSIPK, possibly along with N t W I P K (Samuel et al., 2000). A role for N t S I P K in ROS-mediated P C D is also indicated by the increased sensitivity to ozone conferred by both loss-of-function and gain-of-function mutants of this kinase (Samuel and El l i s , 2002). In fact, the over-expression of N t S I P K is sufficient to induce P C D (Zhang and L i u , 2001) and an increase in ethylene production ( K i m et al., 2003; 20 Samuel et al., 2005) similar to the behaviour of its Arabidopsis orthologue, A t M P K 6 , . The over-expression of N t M E K 2 , the upstream M A P K K of N t S I P K and N t W I P K (Figure 1.1) (Yang et al., 2001a), leads to P C D , independently of S A signaling (Yang et al., 2001b). Interestingly, this phenotype is dependent on the expression of NtrbohB, a gene encoding an N A D P H oxidase homologue, suggesting that N t M E K 2 and/or N t S I P K / N t W I P K might play a role in inducing or controlling R O S production (Yoshioka et al., 2003). The tobacco orthologue of A t M P K 4 , N t M P K 4 , is activated by wounding concurrently with both N t S I P K and N t W I P K (Gomi et al., 2005). N t M P K 4 is required for wound/JA-induced gene expression and for lesion containment after ozone treatment, suggesting that N t M P K 4 plays a positive role in J A signaling. Conversely, NtMPK4-suppressed lines accumulate high amounts of S A after ozone treatment, and N t M P K 4 expression has a negative effect on SA-induced gene expression, altogether suggesting that N t M P K 4 negatively regulates S A production. The M A P K K N t S I P K K activates N t M P K 4 , but is not able to activate N t S I P K / N t W I P K (Figure 1.1) (Gomi et al., 2005). Overall, the ubiquitous roles of A t M P K 3 , A t M P K 4 and A t M P K 6 , as well as of their tobacco orthologues, in both biotic and abiotic stress responses suggest that the timing of activation of the different kinases, and/or the co-activation of other pathways is responsible for the final response specificity. Moreover, it appears that the co-activation of these M A P K s is due, in some cases, to the pattern of joint activation by upstream M A P K K s , with A t M K K 4 / A t M K K 5 ( N t M E K 2 ) regulating the activation of A t M P K 3 and 21 A t M P K 6 (NtWIPK and NtSIPK) , while A t M K K 2 activity results in the co-activation of A t M P K 4 and A t M P K 6 (Figure 1.1). 1.4.3. M A P kinase substrates So far, only a few plant M A P K substrates have been established. One of these is A C S 6 (and plausibly the highly related A C S 2 ) , an A C C synthase isoform discussed previously. After the phosphorylation activity of A t M P K 6 on this protein was established in vitro, it was shown that the phosphorylation of A C S 6 by A t M P K 6 increases the stability of the enzyme and is necessary for AtMPK6-mediated ethylene accumulation in planta (Figure 1.1) (Liu and Zhang, 2004). Another confirmed M A P K substrate is M K S 1 , a protein of unknown function, that interacts with W R K Y transcription factors and whose homologues in other species appear to be involved in plant responses to pathogens. M K S 1 was shown to be phosphorylated by A t M P K 4 in vitro, a pattern which was corroborated in planta by demonstration of physical interaction between the two proteins, and by a marked reduction of M K S 1 phosphorylation observed in an A f M P K 4 knock-out background (Figure 1.1) (Andreasson et al., 2005). In Ozyza sativa (rice), a transcription factor, O s E R E B P l , is activated by a M A P K , B W M K 1 . The DNA-bind ing activity of O s E R E B P l was shown to increase following phosphorylation by B W M K 1 , both in vitro and in transient expression assays in Arabidopsis protoplasts (Cheong et al., 2003). In tobacco, a recent paper also reports that a novel transcription factor, N tWIF , is a substrate for N t W I P K (Figure 1.1). The phosphorylating activity of N t W I P K on NtWIF, established by in vitro kinase assays, was corroborated by the increased transcription stimulating activity of N t W I F when both proteins were transiently expressed in B Y 2 cells 22 (Yap et al., 2005). Also in tobacco, the transcription factor N f W R K Y l was shown by yeast two hybrid and in vitro assays to be a putative substrate for N t S I P K (Figure 1.1) (Menke et al., 2005). Moreover, using a high-throughput microarray-based proteomic approach, several putative in vitro substrates of A t M P K 3 and A t M P K 6 have recently been identified. They include transcription factors, transcription regulators, histones, ribosomal proteins and many others (Feilner et al., 2005). However, until these interactions are examined and confirmed in planta, it is unclear whether they are biologically significant. 1.4.4. Genetic approaches to the study of plant M A P K cascades 1.4.4.1. Loss-of-function mutants The large number of M A P K cascade components in plants suggests that there might be redundancies in the functions of several of these proteins. A s a result, the loss of a given M A P K may not lead to a visible phenotype. Nevertheless, loss-of-function approaches have, in a few cases, provided insight into M A P K functions in plants. For instance, examination of a knock-out mutant of AtMPK4, and of an RNAi-s i lenced atmpk6 mutant, revealed that both kinases play a role in disease resistance (Petersen et al., 2000; Menke et al., 2004). Moreover, loss-of-function mutants have shed light on the reciprocal influence of certain M A P K s on each other's activity. For instance, silencing of NtSIPK leads to the prolonged hyperactivation of N t W I P K under ozone stress (Samuel and Ell is , 2002). Similarly, in ozone-treated Arabidopsis plants, silencing AtMPK6 results in the prolonged activity of A t M P K 3 , while a knock-out mutant of the latter displays prolonged activation o f A t M P K 6 (Miles et al., 2005). It is also noteworthy that loss-of-function 23 approaches are a powerful tool for revealing genetic interactions. Confirmation of M A P K control over ethylene biosynthesis, for instance, relied critically on the use of an atmpk6 knock-out mutant (L iu and Zhang, 2004). 1.4.4.2. Gain-of-function mutants A s M A P K cascade components are, in their native form, in an inactive state, their ectopic expression does not necessarily make it possible to study their activity in vivo. However, protein modifications by deletions or site-directed mutagenesis can mimic the activation state of certain kinases. M A P K K K s , for instance, can be made constitutively active by deletion of their regulatory domain, which then allows investigation of the identity of their downstream M A P K K targets during transgenic expression in plants (Kovtun et al., 2000; Asai et al., 2002). Likewise, M A P K K s , normally activated by phosphorylation of a pair of serine/threonine residues in a conserved Ser/Thr-X(3-5)-Ser/Thr region of their catalytic domain, are routinely made constitutively active by converting these serine/threonine residues by site-directed mutagenesis to the acidic amino acids glutamate and aspartate, which mimics the conformation of phosphorylated M A P K K s (Alessi et al., 1994; Mansour et a l , 1994; Wurgler-Murphy et al., 1997). This gain-of-function approach has been used in the study of plant M A P K cascades, either to gain insight about the M A P K s activated by a M A P K K of interest, or to study downstream and phenotypic impacts of M A P K cascades (Desikan et al., 2001; Yang et al., 2001b; Asai et al., 2002; K i m et al., 2003; Yoshioka et al., 2003; L i u and Zhang, 2004). Unfortunately, M A P K s themselves cannot be made constitutively active, presumably because changes in the 24 kinase three-dimensional structure brought about by phosphorylation of the Thr-X-Tyr motif are too complex to be mimicked by simple amino acid changes. 1.4.4.3. Inducible gene expression systems The over-expression of kinases using constitutive promoters can sometimes lead to severe growth defects, as well as cell death, making it difficult to study these proteins in a transgenic plant context (Chen et al., 2003). To overcome these difficulties, a number of regulated gene-expression systems have been developed for plant transformation. These include the dexamethasone-inducible system, the ethanol-inducible ale expression system and the estrogen receptor-based chemical-inducible system (Aoyama and Chua, 1997; Zuo et al., 2000; Roslan et al., 2001). The dexamethasone-inducible system is based on the activity of a chimeric transcription factor, G V G , comprising the DNA-binding domain of the yeast transcription factor G A L 4 , the transactivating domain of the herpes viral protein V P 16, and the hormone-binding domain of the rat glucocorticoid receptor, which results in nuclear translocation of the transcription factor upon binding to a glucocorticoid. When a transgenic plant expressing the appropriate construct is treated with the synthetic glucocorticoid dexamethasone ( D E X ) , the D E X - G V G complex is transported into the nucleus. There, it binds to and activates a G A L 4 promoter. When the latter has been situated upstream of a transgene of interest, this activation induces transcription of the transgene (Aoyama and Chua, 1997). The dexamethasone-inducible system is now commonly used to study the 25 activity of M A P K s in plants ( K i m et al., 2003; Soyano et al., 2003; L i u and Zhang, 2004). 1.5. Problem statement and thesis objectives The relatively small number of M A P K K s encoded in the Arabidopsis genome suggests that this particular class of kinases acts as a point of convergence within the plant's integration of external stimuli and their transduction to elicit biological responses. Moreover, our ability to convert these enzymes to constitutively active forms through site-directed mutagenesis makes M A P K K s the tools of choice for studying the downstream targets and phenotypes of M A P K cascades. Use of this gain-of-function approach as well as other strategies, has made it possible to gain insight into the biological function of a number of Arabidopsis M A P K K s (Asai et a l , 2002; Ren et al., 2002; Teige et al., 2004). Nevertheless, several other M A P K K s remain to be characterized. In particular, the Group D M A P K K s , which includes Arabidopsis A t M K K 7 , A t M K O , A t M K K 9 and A t M K K l O , are largely unstudied so far (Ichimura et a l , 2002). Our laboratory has successfully cloned A t M K K 9 and created a constitutively active (CA) form of the protein. The C A version of A t M K K 9 ( C A - M K K 9 ) , was placed under the control of the DEX-inducible gene expression system, and stable transgenic Arabidopsis lines expressing D E X : C A - M K K 9 - F L A G were obtained. A preliminary analysis of these 26 plants revealed that C A - M K K 9 could trigger lesion formation in plants, and it was thought that it might be somehow involved in ethylene biosynthesis. Based on these preliminary results, the overall aim of my M . S c . research was to gain a better understanding of the function of A t M K K 9 in Arabidopsis. More specifically, the following objectives were pursued: 1. To characterize the induction pattern of C A - M K K 9 in D E X : C A - M K K 9 - F L A G plants, both at the m R N A and protein levels. 2. To characterize the cell death phenotype induced by C A - M K K 9 , in the context of P C D . 3. To investigate the role of C A - M K K 9 in controlling ethylene production, and establish the relationship between my findings and the existing model of M A P K control of the ethylene biosynthesis pathway. 4. To investigate short-term transcriptional events resulting from C A - M K K 9 activity, using 70-mer oligomer microarrays. 27 2. CA-MKK9 modulates ethylene biosynthesis 2.1. Introduction A n expression survey of all Arabidopsis M A P K K s identified a so-far uncharacterized M A P K K , A t M K K 9 , whose expression profile was correlated with aging of the rosette leaves, as well as with some stress responses (Ellis laboratory, unpublished data). Moreover, the El l i s laboratory recently generated transgenic Arabidopsis plants expressing an inducible gain-of-function mutant form of A t M K K 9 ( C A - M K K 9 ) . A preliminary phenotypic analysis of those transgenic plants revealed that the ectopic expression of this kinase leads to the formation of necrotic lesions in tobacco and Arabidopsis plants. These data have led me to explore the possible role of A t M K K 9 programmed cell death (PCD), a process playing critical roles in the plant's life cycle, including senescence and responses to abiotic and biotic stresses (Rao and Davis, 2001). The onset of P C D , which has been mainly studied in the context of ozone and pathogen responses, is associated with a defined set of genetic and metabolic events. Among those, it is well established that reactive oxygen species (ROS) play a central role. The ROS most commonly associated with P C D in plants include hydrogen peroxide (H2O2) and superoxide anion radicals (O2'"), both of which accumulate in the apoplast in response to challenge by incompatible pathogens or by ozone. 28 Several studies have demonstrated a role for M A P K cascades in regulating P C D . In tobacco, elicitor-induced P C D is associated with the prolonged activation of NtSIPK, as well as with the somewhat later activation of N t W I P K (Zhang and Klessig, 1998b; Suzuki et al., 1999; Zhang et al., 2000). Ozone treatment was similarly shown to result in the activation of N t S I P K in tobacco, and of A t M P K 3 and A t M P K 6 in Arabidopsis (Samuel et al., 2000; Ahlfors et a l , 2004). M A P K cascades act both upstream and downstream of R O S production. Evidence for an upstream role includes the participation of N t M E K 2 in upregulating a respiratory burst oxidase gene involved in R O S production, as well as a role for A t M K K 4 and A t M K K 5 in inducing H2O2 accumulation and promoting P C D in Arabidopsis (Ren et al., 2002; Yoshioka et al., 2003). On the other hand, the fact that several M A P K are activated by oxidative stress indicate that M A P K cascades also act downstream of R O S (Ichimura et al., 2000; Samuel et al., 2000; Desikan et al., 2001; Yuasa et al., 2001). P C D is also tightly regulated by a complex interplay of plant hormones, as reviewed earlier in Chapter 1. Among those, ethylene is proposed to control the propagation of P C D (Overmyer et al., 2003). Ethylene production, perception and/or signaling are required for pathogen-induced P C D in tomato, and for both R O S accumulation and lesion formation following ozone treatment in Arabidopsis (Lund et al., 1998; Overmyer et al., 2000). Moreover, high levels of ethylene biosynthesis are correlated with more extensive P C D following ozone treatment (Rao et al., 2002). 29 A biochemical link between M A P K signaling and ethylene biosynthesis was provided by a recent study that demonstrated that the stability of rate-limiting enzymes in the ethylene biosynthesis pathway is regulated by a M A P kinase cascade (L iu and Zhang, 2004). According to this model, specific biotic and abiotic stresses trigger the activation of a M A P K cascade composed of an as yet unknown M A P K K K , which acts upon A t M K K 4 and/or A t M K K 5 , which in turn act upon A t M P K 6 . Activated A t M P K 6 then phosphorylates the A C S isoform A C S 6 , thereby preventing it from being targeted for degradation, and allowing increased conversion of S-AdoMet to A C C , which would be further converted into ethylene, resulting in the onset of the ethylene response (Liu and Zhang, 2004). Notably, the same group had previously reported that the inducible overexpression of constitutively active (CA) versions of A t M K K 4 and A t M K K 5 results in H2O2 accumulation and necrotic lesions in tobacco and Arabidopsis, supporting the hypothesis that high ethylene production could lead to uncontrolled propagation of P C D (Ren et al., 2002). In this chapter, I have characterized Arabidopsis transgenic plants expressing the gain-of-function version of A t M K K 9 , C A - M K K 9 , under the control of the dexamethasone (DEX)-inducible promoter. I have found that, as observed for C A - M K K 4 and C A - M K K 5 (Ren et al., 2002), C A - M K K 9 triggers a form of P C D . In addition, using a combination of protein analyses, biochemical assays, and genetic approaches, I have demonstrated that C A - M K K 9 activates A t M P K 6 , causing a burst of ethylene release in DEX-induced plants. M y results also demonstrate that CA-MKK9-media ted P C D occurs independently of A t M P K 6 activity, and of the observed ethylene burst. 30 2.2. Material and methods 2.2.1. Plant material Arabidopsis seeds were vernalized for three days, sown on Ready-Earth planting medium, and grown under a 16:8 hours lightdark regime for nineteen to twenty-two days. DEX-inducible transgenic plants were induced by spraying the foliage uniformly with a solution 2 5 p M D E X and 0.015% Silwet-77. 2.2.2. R T - P C R analysis R N A was extracted from frozen leaf tissue using an Rneasy Plant K i t (Quiagen). Total R N A (1 ug) was reverse-transcribed into c D N A with OligodT (Invitrogen) using R T Superscript II (Invitrogen), according to the manufacturer's instruction. The polymerase chain reaction (PCR) was performed in a 20pl reaction containing I X Jumpstart™ REDTaq™ ReadyMix™ P C R Reaction M i x (Sigma), l p l c D N A (the equivalent of 50 ng R N A ) and 0.75 u M of each forward (F) and reverse (R) primers. A 502 bp region of the C A - M K K 9 - F L A G transcript was amplified using the following primers: I n M K K 9 F (5' C G C C G G A T T C G C T A A A C A G A T 3') and I n F L A G R (5' C T T G T C A T C G T C G T C C T T G T A 3'). A 415 bp region of A C T 8 (Atlg49240) was amplified using the following primers: A C T 8 F (5' A T T A A G G T C G T G G C A 3') and A C T 8 R (5' T T A T C C G A G T T T G A A G A G G C T A C 3'). The P C R reactions were performed in a Biometra T-gradient thermocycler, using the following program: initial 31 denaturation for 2 minutes at 94°C, followed by 20 ( C A - M K K 9 - F L A G ) or 25 (ACT8) amplification cycles consisting of denaturation (30 seconds at 94°C), annealing (30 seconds at 55°C) and extension (1 minutes at 72°C), and terminated by a final extension (5 minutes at 72°C). P C R products were analyzed by 1% agarose gel electrophoresis. 2.2.3. Ethylene measurements Three week-old plants were induced with D E X . After the tested induction time, four or five rosettes were cut free from their root system and placed in the barrel of a 10 ml syringe. The plunger was pushed in to leave 2 ml of headspace in the syringe, and the needle was sealed with parafilm. One hour later, 1 ml headspace was analyzed by gas chromatography. Analyses presented in Figure 2.5 were performed using a gas chromatograph (Hewlett Packard Series II 5890) equipped with a CarbonPLOT column ( J & W Scientific) and a flame ionization detector. The samples were injected, with helium as the carrier, under the following temperature program: column temperature 275°C, injection temperature 200°C, detection temperature 250°C. Analyses presented in Figure 2.6 and Figure 2.10 were performed using a different gas chromatograph (Varian 3400), equipped with a Porapak Q column and a flame ionization detector. The samples were injected, with helium as the carrier, under the following temperature program: column temperature 70°C, injection temperature 2°C, detection temperature 220°C. In each case the amount of ethylene detected was quantified in terms of peak area, and converted to parts per mil l ion (ppm) units through a standard curve established with a 100% Ethylene Standard (Alltech). 32 2.2.4. Hydrogen peroxide detection Hydrogen peroxide was detected by 3,3'-diaminobenzidine ( D A B ) staining. A t chosen times after induction, whole rosettes were transferred to a side-armed Erlenmeyer flask containing 50 ml of lmg/ml D A B (Sigma)-HCl, p H 4. The rosettes were vacuum-infiltrated with the solution, then transferred to wet paper towels and incubated under bright light for five hours. The tisssue was then cleared by gentle shaking in a 3:1:1 ethanol:lactic acid:glycerol solution, in darkness, for 48 hours. Representative leaves were mounted on microscope slides and photographed. The quantification of the D A B precipitate was done using ImageJ version 1.34s. The pictures were converted to 8-bit grayscale images, and all pixels within the leaf area with brightness values greater than 8 and smaller than 155 were quantified with help of the Analyze Particules tool. 2.2.5. In vitro phosphorylation assay The recombinant proteins used in the in vitro assays were produced in Escherichia coli, using the Glutathione S-tran sferase (GST) Gene Fusion System, with G S T fused to their N-terminus. They had been previously purified from bacterial lysates by affinity chromatography. M P K 6 1 and M P K 3 1 had also been cleaved from G S T using the thrombin protease, and provided to me as such by Jin Suk Lee. M P K 6 1 / M P K 3 1 (1 ug) was added to 0.5 pg C A - M K K 9 - G S T in 40 pi kinase buffer [25 m M Tris pH7.5, 5 m M p-glycerophosphate, 2 m M D T T , 0.1 m M N a 3 V 0 4 , 10 m M M g C l 2 , 200 p M A T P ] . The samples were incubated for forty-five minutes at 30°C, and the reactions terminated by addition of 5 X SDS sample buffer [0.25M Tr i s -HCl p H 6.8, 9% SDS, 40% glycerol, 33 0.125% bromophenol blue and 20% v/v [3-mercaptoethanol]. After being boiled for five minutes, the proteins were separated by S D S - P A G E and analyzed by a western blot with an anti-phospho-p44/42 M A P kinase (Thr202/Tyr204) antibody (Cell Signaling). 2.2.6. Protein extraction and western blot analysis Leaves were ground in liquid nitrogen and the powder was mixed with protein extraction buffer [lOOmM H E P E S pH7.5, 5 m M E D T A , 5 m M E G T A , I m M N a 3 V 0 4 , 10mM NaF, 10%) glycerol, 7.5% P V P P , 5ul [3-mercaptoethanol and one protease inhibitor tablet (Roche)]. The extract was incubated at 4°C on a reciprocal shaker for ten minutes, and centrifuged for thirty minutes at 4°C (14,000 rpm). The amount of protein in the supernatant was quantified using the Bio-Rad protein assay kit with bovine serum albumin ( B S A ) as the standard, and the supernatant was either assayed directly or flash-frozen and stored at -80°C. For analysis, 100 u.g total protein was concentrated by acetone precipitation. Briefly, total proteins were mixed with five volumes of acetone, incubated ten minutes at -80°C, then overnight at -20°C. The precipitated proteins were then centrifuged at 4°C for fifteen minutes (14,000 rpm). The pellet was air-dried, resuspended in 40 pi I X SDS protein buffer [50 m M Tr i s -HCl pH=6.8, 1.8 % SDS, 8% glycerol, 0.025%) bromophenol blue and 4% v/v [3-mercaptomethanol], and boiled for 5 minutes. Samples were subsequently separated on a 12% S D S - P A G E gel, and the proteins were transferred to a P V D F membrane (Millipore). For a n t i - F L A G analysis, the membrane was blocked for two hours with 5% no-fat dried milk (Carnation) in Tris-buffered saline/Tween-20 (TBST) [20mM Tr i s -HCl p H 7.5, 100 34 m M N a C l , 0.05% Tween-20], and then incubated for two hours in a 1:10,000 dilution of an t i -FLAG antibody M 2 (Sigma) in 5% no-fat dried milk and T B S T . For anti-phospho-E R K analysis, the membrane was blocked for two hours in 3% B S A , then incubated for two hours with 1:1000 phospho-p44/42 Map kinase (Thr202/Tyr204) antibody (Cell Signaling) in 3% B S A and T B S T . For ant i -MPK6 analysis, the membrane was blocked two hours in 5% B S A in T B S T , then incubated two hours in 1:5000 ant i -MPK6 (provided as a beta product by Sigma) in 5% B S A and T B S T . After incubation with the primary antibody, the membrane was washed four times for five minutes with T B S T , and then incubated with a 1:7500 dilution of peroxidase-conjugated goat anti-mouse antibody (Dako) in T B S T for one hour. The membrane was washed again four times five minutes in T B S T , and then visualized by using an enhanced chemiluminescence detection kit (Amersham). 2.2.7. Generation of D E X : C A - M K K 9 - F L A G / m / ? A : 6 ' transgenic plants Agrobacterium tumefaciens GV3101 carrying the construct D E X : : C A M K K 9 - F L A G in the pTA7002 plasmid (Aoyama and Chua, 1997) was grown overnight in a 300 ml culture containing 50 pg/ml rifamycin, 25 pg/ml gentamycin and 50 pg/ml kanamycin. The mpk6 knock-out mutant was transformed using the floral dip method, i.e. by dipping the immature flowers into the Agrobacterium cells that had been resuspended in 5% sucrose and 0.05% Silwet-L77. T l transformants were selected by germinating the resulting seeds on M S plates containing 50 pg/ml hygromycin. 35 2.3. Results 2.3.1. The C A - M K K 9 transgene is rapidly activated following D E X induction In the C A - M K K 9 protein, the conserved serine residues present in the activation loop of the kinase have been converted into glutamate residues, a modification which structurally mimics the acidic character created by post-translational phosphorylation of serine residues, thus bypassing the requirement for a M A P K K K - d r i v e n phosphorylation for its activity. Furthermore, a F L A G epitope has been added to the C-terminus of C A - M K K 9 (Figure 2.1 A ) . This construct was put into the pTA7002 vector (Aoyama and Chua, 1997), to generate the p T A 7 0 0 2 - C A - M K K 9 - F L A G construct (Figure 2 . IB) which, once inserted into a plant genome, allows induced expression of C A - M K K 9 - F L A G by treating the transformed plants with the steroid hormone analogue, dexamethasone ( D E X ) (Figure 2.1C). 36 MALVRERRQLNLRLPLPPISDRRFSTSSSSATTTTVAGCNGISACDLEKLNVLGCG NGGIWKVRHKTTSEIYALKTVNGDMDPIFTRQLMREMEILRRTDSPYWKCHGIF EKPWGEVSILMEYMDGGTLESLRGGVTEQKLAGFAKQILKGLSYLHALKIVHRDIK PANLLLNSKNEVKIADFGVSKILVRELDSCNEYVGTCAYMSPERFDSESSGGSSDI YAGDIWSFGLMMLELLVGHFPLLPPGQRPDWATLMCAVCFGEPPRAPEGCSEEF RSFVECCLRKDSSKRWTAPQLLAHPFLREDLDYKDDDDK B RB 35S)| G V G [ E 9 ^ N O ^ H P T |-|3A CA-MKK9-FI_AG 6 X U A S GAL4 - LB NUCI FUS Figure 2 .1 . Construction of CA-MKK9 and insertion into the DEX-inducible vector to produce pTA7002-CA-MKK9-FLAG. A) Amino acid sequence of CA-MKK9. In red are the serine residues converted to glutamate. In blue is the C-terminus FLAG tag B) Schematic representation of the T-DNA region of the pTA7002-CA-MKK9-FLAG construct used for transformation. RB, T-DNA right border; 35S, CaMV 35S promoter GVG, chimeric GAL4-VP16-GR dexamethasone inducible transcription factor; E9, pea rbcS-E9 polyadenylation sequence; Nos, nopaline synthase promoter; HPT, hygromycin phosphotransferase; N,, nopaline synthase polyadenylation sequence, 6XUAS g ai4, GVG binding site; LB, left T-DNA border (Aoyama and Chua, 1997; McNellis et al., 1998) C) Schematic diagram of the DEX-inducible system. Adapted from McNellis e ta l . (1998). 37 The p T A 7 0 0 2 - C A - M K K 9 - F L A G construct was used to transform Arabidopsis, and stable, homozygous D E X : C A - M K K 9 - F L A G transgenic lines were obtained. In the same manner, in order to detect potential artefactual effects due to the pTA7002 vector alone, Arabidopsis plants were also transformed with the "empty" pTA7002 vector, to establish several Empty Vector (EV) lines. A s groundwork for future investigation into the downstream effects of expressing C A -M K K 9 , it was important to define the time-course of C A - M K K 9 expression after induction by D E X . Therefore, I used R T - P C R to assess the rise in C A - M K K 9 m R N A levels following the transgene induction. In order to detect solely the C A - M K K 9 transgene, and not the endogenous A t M K K 9 , primers were designed to encompass the F L A G tag fused to the C-terminus of the C A - M K K 9 transgene. A s shown by an R T - P C R analysis (Figure 2.2A), D E X : C A - M K K 9 - F L A G plants already showed a marked rise in C A - M K K 9 m R N A only two hours after D E X treatment. Furthermore, the levels of C A -M K K 9 m R N A appeared to remain elevated for at least twelve hours after the initial induction. I also analyzed the increase in C A - M K K 9 protein levels following D E X induction. A s previously mentioned, a F L A G tag had been fused to the C-terminus C A - M K K 9 , which should allow the translated transgene product to be detected using the a n t i - F L A G M 2 antibody. This strategy has been previously used in plant systems to detect transgenic fusion proteins (Samuel et al., 2000; Samuel and El l is , 2002; K i m et al., 2003; L i u et al., 38 2003). However, due to nonspecific binding of this antibody to other proteins in the total Arabidopsis protein extracts, it proved difficult to unambiguously identify C A - M K K 9 by using the a n t i - F L A G antibody. The mass predicted for the C A - M K K 9 - F L A G protein is 36.3 kDa, but I could only detect an increase in a -40 k D a protein, starting two hours after D E X induction in the D E X : C A - M K K 9 - F L A G plants. A faint band of similar size can also be seen in the E V plants, probably reflecting the nonspecific binding of the antibody (Figure 2.2B). Interestingly, aberrant molecular weight of C A - M A P K K s was also reported by Ren et al. (2002), who described up-shifts in the mobilities of C A -A t M K K 4 and C A - A t M K K 5 proteins on P A G E gels compared to wild-type proteins. Nevertheless, in order to localize more confidently the C A - M K K 9 protein, our laboratory has recently generated an a n t i - A t M K K 9 antibody. Once optimized, western blots using this new antibody should allow us to clarify the expression pattern of the C A - M K K 9 protein. 39 D EX: CA-M KK9-F LAG L2 DEX:CA-MKK9-FI_AG L12 Time after DEX (hrs): CA-MKK9-FLAG ACT8 0 2 4 8 12 0 2 4 8 12 — — — Time after DEX (hrs): CA-MKK9-FLAG ACT8 DEX:CA-MKK9-FLAG L13 8 12 EVL1 8 12 B Time after DEX (hrs): CA-MKK9- J FLAG DEX:CA-MKK9-FLAG L12 0 2 4 8 12 EVL1 4 8 12 Figure 2.2. Time-course of CA-MKK9 activation following DEX induction Three-week-old DEX:CA-MKK9-FLAG (L2, L12 and L13) and EV plants (L1) were induced with DEX and harvested at the indicated time. A) Total RNA was extracted from plant tissue, and CA-MKK9-FLAG mRNA levels were estimated by RT-PCR using AtMKK9- and FLAG-specrfic primers. Equal loading is shown by comparable ACT8 levels. B) Total proteins extracted from plant tissue (100ug) were separated by SDS-PAGE. CA-MKK9-FLAG was detected by immunoblot with an anti-FLAG antibody. Equal loading is shown by Coomassie Blue staining of the membrane. 40 Overall, the analysis of C A - M K K 9 - F L A G expression suggests that the transgene is expressed two hours after the D E X induction, and that its expression stays high up to twelve hours, and possibly longer. 2.3.2. C A - M K K 9 activity results in the formation of lesions Since it had been observed in a preliminary experiment that the induction of C A - M K K 9 causes a cell death phenotype in the transgenic plants, I carried out a more detailed characterization of the timing and pattern of the formation of these necrotic lesions in D E X : C A - M K K 9 - F L A G plants. A s shown in Figure 2.3, leaves began to yellow two days after D E X induction, starting from the center of the rosette, including the meristematic region. Scattered chlorotic lesions became clearly visible on the leaves three days after induction, and by the fourth day these had taken over the entire plant (Figure 2.3). The induced D E X : C A - M K K 9 - F L A G plants did not recover, and eventually died (data not shown). 41 Days after DEX: * CM 5 —I Ul G I £ 5 X u-Ul G ft* 1 E Figure 2.3. Time-course of cell death progression following CA-MKK9 activation. Three-week-old DEX:CA-MKK9-FI_AG (L12 and L13) and EV (L1) plants were induced with DEX. Pictures show the same plant at the indicated time after DEX induction. 2.3.3. CA-MKK9-media ted cell death is associated with H2O2 accumulation The appearance o f the chlorotic lesions observed following C A - M K K 9 activation suggested that they could be the result o f P C D . In plants, P C D , in particular in the context o f the H R and in response to ozone stress, is associated with the accumulation o f R O S , which include H2O2 (Lamb and Dixon, 1997). In order to assess whether the phenotype of D E X : C A - M K K 9 - F L A G was consistent with this type o f P C D , I examined the correlation between lesion formation and accumulation of H2O2 in the induced transgenic plants. A 3,3'-diaminobenzidine (DAB)-based histochemical assay revealed that H2O2 accumulation can be detected one day after D E X induction; i.e. immediately 42 preceding the appearance of the first visible lesions (Figure 2.4). However, no significant H2O2 accumulation could be detected two or four hours after D E X induction, indicating that the R O S response is probably not an early consequence of ectopic C A - M K K 9 activity (Figure 2.4). Nevertheless, the pattern and timing of H2O2 accumulation suggest that R O S are closely associated with the death of the leaf tissue observed in the D E X : C A - M K K 9 - F L A G transgenic plants, consistent with the hypothesis that C A - M K K 9 activity is inducing HR- l ike P C D . 43 Hours after DEX: 0 2 4 24 Figure 2.4. CA-MKK9 activity results in H 20 2 accumulation. Three week-old DEX:CA-MKK9-FLAG (L12 and L13) and EV (L1) plants were induced with DEX. At the indicated time-points, H2O2 generation was assessed by infiltrating a whole plant rosette with a DAB solution. Infiltrated plants were left under light for five hours and then cleared for two days. The amount of DAB precipitate was quantified using ImageJ, and indicated on each picture as the percentage of the total leaf area. 44 2.3.4. C A - M K K 9 causes a rapid increase in ethylene biosynthesis As R O S production and P C D both require the action of ethylene (Moeder et al., 2002; Rao et al., 2002), the above results suggested that that C A - M K K 9 might affect the biosynthesis of this hormone. It has been previously reported that a M A P K cascade involving A t M P K 6 controls ethylene production (L iu and Zhang, 2004). It thus seemed possible that C A - M K K 9 also triggers that signaling cascade, resulting in a rise of ethylene production that eventually leads to cell death in the transgenic plants. A s a first step to test that hypothesis, I asked whether the activation of the C A - M K K 9 transgene was temporally correlated with an increase in ethylene production. To answer that question, the time-course of ethylene production was monitored in D E X : C A - M K K 9 -F L A G plants after D E X induction. The induction of C A - M K K 9 was indeed quickly followed by a rise in ethylene production (Figure 2.5). This increase was detectable four hours after the D E X treatment, and reached up to six times wild-type levels after twelve hours. The timing of this response suggests that C A - M K K 9 rapidly modulates ethylene biosynthesis, possibly through a post-translational mechanism. 45 Figure 2.5. Time-course of ethylene production in DEX:CA-MKK9-FI_AG plants. DEX:CA-MKK9-FLAG (L2 and L12) and EV (L3 and L10) plants were induced with DEX. At the indicated time-points, four rosettes were collected and placed into a syringe. One hour after, the ethylene contained in the headspace was quantified by gas chromatography. Each time-point assay was done in triplicate. Error bars represent standard errors. 2.3.5. C A - M K K 9 is similar to C A - M K K 4 in acting as a mediator o f ethylene biosynthesis The current model of M A P K regulation of ethylene production proposes that A t M K K 4 and/or A t M K K 5 act upstream of A t M P K 6 , and that the latter's activity leads to stabilization o f A C S isoforms (Liu and Zhang, 2004). However, no experimental data has been published regarding the participation o f these specific M A P K K s in the A C S 2 / 6 stabilization process, since L i u and Zhang's (2004) model is based on the ectopic expression o f A t M K K 4 / A t M K K 5 ' s tobacco orthologue, N t M E K 2 , in Arabidopsis. Since 46 my evidence suggested that C A - M K K 9 might also participate in the same pathway, I compared the extent of the ethylene burst in D E X : C A - M K K 9 - F L A G plants with that obtained in D E X : C A - M K K 4 - F L A G plants. A s shown by Figure 2.6, two D E X : C A -M K K 4 - F L A G transgenic lines, L I 5 and L I 7 , produced approximately twice as much ethylene as the tested D E X : C A - M K K 9 - F L A G lines, twelve hours after induction. This confirms the role of A t M K K 4 as acting upstream of A t M P K 6 in the control of ethylene biosynthesis, but also shows that A t M K K 9 can participate in an analogous fashion. 60 , 50 g 40 Q. a. £= o o -o 30 o I— Q_ CD C a >. 10 0 i i I i i . i---- _ _ i t : : : J i. = DEX:CA-MKK9-FLAG DEX:CA-MKK9-FLAG DEX:CA-MKK4-FLAG DEX:CA-MKK4-FLAG EV L1 L12 L13 L15 L17 Figure 2.6. Ethylene production in DEX:CA-MKK9-FI_AG and DEX:CA-MKK4-FLAG plants. Three week-old DEX:CA-MKK9-FLAG (L12 and L13), DEX:CA-MKK4-FLAG (L15 and L17) and EV (L1) were induced with DEX and collected twelve hours after. Four or five rosettes were placed into a syringe. One hour after, the ethylene contained in the headspace was quantified by gas chromatography. Each genotype was tested in triplicate. Error bars represent standard errors. 47 2.3.6. C A - M K K 9 induction correlates with increased activity of A t M P K 6 in vivo The rapidity with which the ethylene burst is detected following C A - M K K 9 activation implied that C A - M K K 9 must quickly activate A t M P K 6 following its own activation, i f it is to be responsible for the observed increase in ethylene production. In order to determine whether C A - M K K 9 induction correlates with increased A t M P K 6 activity in D E X : C A - M K K 9 - F L A G plants, I monitored protein and activation levels of A t M P K 6 at different time-points after D E X induction. While protein levels were estimated using a specific an t i -AtMPK6 antibody, the activity of the kinase was assessed with a commercial anti-phospho-ERK antibody. This antibody was raised against the conserved phosphorylated tyrosine and threonine residues present in the activated form of the mammalian M A P K , E R K , and has been shown to also detect a number of phosphorylated plant M A P K s , including N t S I P K / A t M P K 6 (Samuel et al., 2000; Ahlfors et al., 2004). It is well established that the phosphorylation of the conserved -Tyr-X-Thr- motif present in the activation loop of M A P K s correlates with increased kinase activity (Ray and Sturgill, 1988), which makes the anti-phospho-ERK antibody a useful tool for assessing the presence of activated M A P K s . I found that D E X : C A - M K K 9 - F L A G plants, following induction, displayed increased levels of a phosphorylated -45 kDa M A P K , which the an t i -AtMPK6 antibody confirmed to be A t M P K 6 (Figure 2.7). Activated A t M P K 6 , which was almost absent before D E X induction, was clearly detectable after two hours, and increased to a maximum between eight (LI3) and twelve (LI2) hours (Figure 2.7). This early increase in A t M P K 6 activity is concurrent with the activation of C A - M K K 9 after D E X induction (Figure 2.2), again 48 consistent with the hypothesis that the former could be a direct target o f the latter. On the other hand, the levels o f A t M P K 6 protein remained relatively constant throughout the time-course, indicating that the increased presence o f phosphorylated A t M P K 6 is likely due to a post-translational modification rather than to an increase in the A t M P K 6 pool. Interestingly, the anti-phospho-ERK blot revealed the concurrent activation of a second M A P K o f lower molecular weight than A t M P K 6 . We are currently investigating the identity of this second M A P K . DEX:CA-MKK9-FLAG L12 DEX:CA-MKK9-FLAG L13 EV L1 Time after DEX (hrs): 0 2 4 8 12 0 2 4 8 12 0 2 4 8 12 a-p-ERK • Figure 2.7. AtMPK6 protein and activity levels following CA-MKK9 activation. DEX:CA-MKK9-FLAG (L12 and L13) and EV (L1) plants were collected at the indicated time-points. Total proteins (100 pg) were separated by SDS-PAGE. Membranes were first blotted with anti-phospho-ERK, then stripped and re-blotted with anti-AtMPK6. Equal loading is shown by Coomassie Blue staining of the membrane. 2.3.7. A t M K K 9 activates A t M P K 6 in vitro To further test the model o f an A t M K K 9 - A t M P K 6 signaling cascade, I examined the ability o f A t M K K 9 to phosphorylate recombinant A t M P K 6 in an in vitro kinase assay. 49 Moreover, in order to establish whether this assay was stringent enough to reveal specific substrate preferences, I also tested A t M K K 9 ' s ability to phosphorylate A t M P K 3 , the Arabidopsis M A P K most closely related to A t M P K 6 and also one that is known to play a role in stress responses (Kovtun et al., 2000; Asai et al., 2002; Ahlfors et al., 2004). In order to by-pass the requirement for an upstream M A P K K K for A t M K K 9 activation, I used a recombinant C A - M K K 9 - G S T protein in the assay. Because M A P K s often display auto-phosphorylating activity, the assayed recombinant M A P K s had previously been converted to kinase-inactive forms ( M A P K 1 ) by site-directed mutagenesis, whereby the lysine residue of the kinase domain was replaced with a catalytically ineffective arginine residue. Following co-incubation of the C A - M K K 9 with either M P K 3 ' or M P K 6 ' , the phosphorylation state of the tested M A P K s was assessed by immunoblot using anti-phospho-ERK. A s shown in Figure 2.8, C A - M K K 9 strongly increases the phosphorylation state of M P K 6 ' . In contrast, it does not appear to have any activity against M P K 3 1 . Therefore, I conclude that, at least in vitro, C A - M K K 9 is capable of activating A t M P K 6 . Together with the rapid rise of A t M P K 6 activity observed in D E X : C A - M K K 9 - F L A G plants, this supports my hypothesis that A t M P K 6 is activated by C A - M K K 9 in transgenic plants upon D E X induction. 50 MPK6j MPK3j CA-MKK9 + - + + - + MAPK - + + - + + mm* mm _ CA-MKK9-GST a-p-ERK ^-p-MPKS1 . -CA-MKK9-GST Total ( proteins — ~I\/IPK3' «- —MPK6' Figure 2.8. CA-MKK9 kinase activity on AtMPK6 and AtMPK3 in vitro. Recombinant CA-MKK9-GST (0.5 ug), MPK6' (1 ug) and MPK3j (1 ug) proteins were subjected to an in vitro kinase assay. Proteins were then separated by SDS-PAGE and analyzed with anti-phospho-ERK. Equal loading is shown by Coomassie Blue staining of the membrane. 2.3.8. A t M P K 6 is necessary for CA-MKK9-media ted ethylene production M y results so far suggested that, following D E X induction, C A - M K K 9 directly activates A t M P K 6 . However, in order to establish a causal relationship between A t M P K 6 activity and the observed CA-MKK9-med ia t ed phenotypes, i.e. ethylene overproduction and P C D , I wanted to find genetic evidence of such dependency. To do this, I transformed a mpk6 knock-out mutant line with the D E X : C A - M K K 9 - F L A G construct. After T3 generation plants were obtained, I screened the transgenic lines to isolate those where the level o f expression o f C A - M K K 9 following D E X induction was comparable to that observed in the original D E X : C A - M K K 9 - F L A G lines (L2, L 1 2 and L13) (Figure 2.2A). A n R T - P C R analysis o f the time-course o f C A - M K K 9 induction in the double-mutant 51 lines revealed that three independent DEX:CA-MKK9-FLAG//w/?£r5 lines, L I , L 4 and L 7 , displayed strong induction of C A - M K K 9 two hours after induction, with sustained transgene expression up to twelve hours after expression (Figure 2.9). A s this was very similar to what I had observed in D E X : C A - M K K 9 - F L A G plants (Figure 2.2A), it seemed reasonable to assume that these double-mutant lines would behave the same way as the single C A - M K K 9 mutant, and I therefore used these three lines in subsequent analyses. DEX:CA-MKK9-FLAG/mpft6 L1 DEX:CA-MKK9-FLAG/mpft6 L4 DEX;CA-MKK9-FLAG/mp/f6 L7 Time after ~ " " " " ~ - — ~ ~ ~ ~ ~ ~ DEX (hrs): 0 2 4 8 12 0 2 4 8 12 0 2 4 8 12 Figure 2.9. Time-course of CA-MKK9 induction in DEX:CA-MKK9-FLAG/mp/c6 plants. Three-week-old DEX:CA-MKK9-FLAG/mp/c6 plants (L1, L4 and L7) were induced with DEX and harvested at the indicated time. Total RNA was extracted from plant tissue, and CA-MKK9-FLAG mRNA levels were estimated by RT-PCR using AtMKK9- and FLAG-specific primers. Equal loading is shown by comparable ACT8 levels. In contrast to the pattern I observed with DEX-induced D E X : C A - M K K 9 - F L A G plants, I found that treating the DEX:CA-MKK9-FLAG//w/?jfc<5 plants with D E X did not induce increased ethylene production following D E X treatment (Figure 2.10). This provides compelling genetic evidence that A t M P K 6 is necessary for CA-MKK9-media ted ethylene production in the transgenic plants. 52 35 DEX.CA-MKK9-FLAGL12 DEX:CA-MKK9-FLAGL13 DEX:CA-MKK9-FLAG/mpk6 L1 DEX:CA-MKK9-FLAG/mpk6 L4 DEX:CA-MKK9-FLAG/mpk6 L7 EVL1 Figure 2.10. AtMPK6 is required for CA-MKK9-mediated ethylene production. DEX:CA-MKK9-FI_AG (L12 and L13), DEX:CA-MKK9-FLAG/mp/c6 (L1, L4 and L7) and EV (L1) plants were treated with DEX. Twelve hours later, four rosettes per transgenic line were placed into a syringe and the headspace was analyzed by gas chromatography. Samples were done in triplicate. Error bars represent standard error. 2.3.9. A t M P K 6 is not required for CA-MKK9-media ted P C D The development of ROS-associated lesions in the D E X : C A - M K K 9 - F L A G plants (Figure 2.3, Figure 2.4) had led me to believe that AtMPK6-mediated ethylene overproduction might be the direct cause of the CA-MKK9-media ted P C D . In order to test that hypothesis, I compared the phenotype of the D E X : C A - M K K 9 - F L A G / / w p & ( 5 plants to that of D E X : C A - M K K 9 - F L A G plants four days after D E X induction. If the AtMPK6-mediated ethylene burst indeed caused P C D , I expected that the latter process would not be observed in the double mutant, where A t M P K 6 activity and ethylene 5 3 overproduction are absent (Figure 2.10). However, as shown in Figure 2.11, D E X : C A -MKK9-FLAG7/wp£6 plants developed lesions comparable to those o f D E X C A - M K K 9 -F L A G plants following D E X induction. Therefore, CA-MKK9-media ted P C D is triggered independently o f A t M P K 6 , and is not directly caused by the overproduction o f ethylene. DEX:CA-MKK9-FLAG L12 DEX:CA-MKK9-FLAG/mp/c6 L1 Figure 2.11. Lesion formation in DEX:CA-MKK9-FLAG/mpk6 plant. DEX:CA-MKK9-FLAG (L12) and DEX:CA-MKK9-FLAG/mpk6 (L1) plants were induced with DEX, and were photographed four days after. A comparable phenotype was observed for two additional DEX:CA-MKK9-FLAG/mp/f6 lines, L4 and L7. 2.4. Discussion The gaseous hormone ethylene is involved in several stages of plant development, as well as in responses to biotic and abiotic stresses. The production of ethylene is controlled by two family o f enzymes, A C C synthases ( A C S ) and A C C oxidases (Wang et al., 2002). 54 The rate of ethylene biosynthesis can be modulated by the differential expression of both A C S and A C O genes (Tuomainen et al., 1997; Overmyer et al., 2000; Moeder et al., 2002). Moreover, several lines of evidence indicate that the stability of A C S enzymes is controlled by post-translational mechanisms, allowing rapid changes in the rate of ethylene biosynthesis. For instance, A C S activity in tomato depends upon intact kinase activity (Spanu et al., 1994). In addition, in Arabidopsis, an ethylene regulatory protein ( E T O l ) has been shown to interact with the A C S 5 C-terminus, thus blocking its activity and presumably promoting its degradation by a proteasome-dependent pathway (Wang et al., 2004). More recently, it was demonstrated that activated A t M P K 6 phosphorylates two Arabidopsis A C S isoforms, A C S 2 and A C S 6 , thereby stabilizing them and causing increased production of ethylene (Liu and Zhang, 2004). The authors of that study, using a DEX-inducible/gain-of-function system, described a rapid increase in ethylene production following the activation of N t M E K 2 , both in tobacco and Arabidopsis (K im et al., 2003; L i u and Zhang, 2004). In this Chapter, I report that another Arabidopsis M A P K K , A t M K K 9 , is also capable of controlling ethylene biosynthesis when overexpressed in a C A form, under the control of the DEX-inducible promoter. When I examined ethylene production in DEX-induced C A - M K K 9 plants, I found that ethylene production starts to increase only four hours after D E X induction (Figure 2.5). This was preceded by an increase in A t M P K 6 ' s activity (Figure 2.7), which mirrored the pattern of C A - M K K 9 induction (Figure 2.2). Since A t M P K 6 ' s activity was practically absent before D E X induction, as well as in the control line, E V L I , it appears that C A - M K K 9 ' s activity is responsible for activation of the 55 M A P K . Nevertheless, it could be argued that the rise in A t M P K 6 activity is not due directly to C A - M K K 9 , but is an indirect effect o f the stress caused by the transgene activation. For instance, A t M P K 6 is known to be activated upon oxidative stress, and more particularly by H2O2 (Ichimura et a l , 2000; Desikan et al., 2001; Yuasa et al., 2001), and D E X : C A - M K K 9 - F L A G plants indeed show increased accumulation of H2O2 (Figure 2.4). However, this H2O2 accumulation had not yet begun two hours after D E X induction (Figure 2.4), at which time A t M P K 6 activity was already increasing (Figure 2.7). I therefore conclude that the observed A t M P K 6 activity is not an indirect consequence of oxidative stress. Supporting this idea, I also showed by an in vitro kinase assay that A t M P K 6 is a suitable substrate for C A - M K K 9 (Figure 2.8). In addition to these lines of biochemical evidence, I was able to genetically demonstrate that the D E X : C A - M K K 9 - F L A G / m p & f 5 double-mutant did not accumulate ethylene upon D E X induction (Figure 2.10), despite comparable levels of expression of the C A - M K K 9 transgene (Figure 2.9). Altogether, this evidence supports my hypothesis that, upon activation in the D E X : C A - M K K 9 - F L A G plants, C A - M K K 9 rapidly activates A t M P K 6 , which in turn is known to stabilize the A C C synthases, A C S 2 and/or A C S 6 through phosphorylation o f specific sites on the A C S protein. This stabilization would then lead to the observed burst of ethylene in induced C A - M K K 9 plants (Figure 2.12). 56 Figure 2.12. Model of CA-MKK9 participation in ethylene biosynthesis. CA-MKK9, similarly to AtMKK4/AtMKK5, activates AtMPK6, which in turn stabilized ACS6, leading to a burst of ethylene (Liu and Zhang, 2004). CA-MKK9 also activates the onset of PCD, independently of AtMPK6 and of ethylene biosynthesis. According to Liu and Zhang (2004), the upstream MAPKKs responsible for AtMPK6 are either AtMKK4 and/or AtMKK5. However, the experimental data presented by those authors was based on the over-expression in Arabidopsis of a gain-of-function mutant of the tobacco MAPK kinase, NtMEK2, which they showed was able to activate AtMPK6 and induce ethylene production. Because the orthologues of NtMEK2 in Arabidopsis are AtMKK4 and AtMKK5, and because these two MAPKKs have also been shown to activate AtMPK6 in protoplast studies (Asai et al., 2002), Liu et al. (2004) speculated that AtMKK4 and/or AtMKK5 act upstream of AtMPK6 in controlling ethylene production in vivo. The same group had previously shown that CA-AtMKK4/AtMKK5 induces H2O2 accumulation and PCD in Arabidopsis (Ren et al., 2002), in a manner similar to what I 57 observed in D E X : C A - M K K 9 - F L A G plants (Figure 2.3). When I compared ethylene production between D E X : C A - M K K 9 - F L A G and D E X : C A - M K K 4 - F L A G plants, both genotypes displayed a marked increase in ethylene biosynthesis twelve hours after D E X induction (Figure 2.6). This confirms L i u et al. (2004)'s model, where A t M K K 4 acts upstream of A t M P K 6 in controlling the stability of A C S isoforms. Although the two D E X : C A - M K K 4 transgenic lines tested produced approximately twice as much ethylene as the D E X : C A - M K K 9 - F L A G lines (Figure 2.6), it is not clear that this difference reflects the relative participation rates of the two M K K s in ethylene biosynthesis in planta. M y experimental approach is solely based on ectopic transgene expression, and does not take into account the normal activity levels of the endogenous M K K s , including their spatial or temporal distribution. Nevertheless, the similarity between A t M K K 4 / A t M K K 5 - and AtMKK9-media ted ethylene production and cell death raises questions about the specificity of M A P K K s in vivo. A t M K K 4 and A t M K K 5 are part of Group C M A P K K s , whose members are well-established upstream effectors of Group A M A P K s (e.g. A t M P K 6 ) in stress signaling (Ichimura et al., 2002). A t M K K 9 , on the other hand, is part of group D M A P K K s , and only bears approximately 50% amino acid sequence similarity to Group C M A P K K s . Whether the Group C and D M A P K K s are functionally redundant, or whether their specificity depends on other factors which cannot be assessed by the current gain-of-function studies, remains unclear. The only study published so far on a Group D M A P K K describes the tomato M A P K K , L e M K K 4 . Interestingly, the over-expression of L e M K K 4 causes cell death and correlates with the activation of L e M P K 2 and L e M P K 3 , which are putative orthologues of 58 A t M P K 6 and A t M P K 3 , respectively (Pedley and Martin, 2004). Moreover, the authors of that study demonstrated that L e M K K 4 strongly activates L e M P K 2 , while its capacity to activate L e M P K 3 , although detectable, is much less important (Pedley and Martin, 2004). A l l together, these data closely mirror our finding on C A - M K K 9 , suggesting that L e M K K 4 and A t M K K 9 could be functional orthologues. Interestingly, the same study reveals similar findings for L e M K K 2 , a Group C M A P K K , strengthening the idea that Group C and D M A P K K s may have at least partly overlapping functions in plants (Pedley and Martin, 2004). M y results also revealed that a second M A P K of smaller size is co-activated with A t M P K 6 following D E X induction of the D E X : C A - M K K 9 - F L A G plants (Figure 2.7). Its apparent molecular weight, -42 kDa, points toward several Arabidopsis M A P K candidates: A f M P K l (42.6 kDa), A t M P K 3 (42.7 kDa), A t M P K 4 (42.9 kDa), A t M P K 5 (42.9 kDa), A f M P K l 1 (42.4 kDa), A t M P K 1 3 (42.2 kDa). However, only a few of these M A P K s are reported to be co-activated with A t M P K 6 . A t M P K 3 , in particular, is the M A P K whose activity is the most frequently associated with that of A t M P K 6 , under several biotic and abiotic stress conditions. However, since I found that A t M P K 3 cannot be directly activated by C A - M K K 9 (Figure 2.8), it appears unlikely that this M A P K could be activated so early after C A - M K K 9 induction. A t M P K 4 has also been reported, in a few instances, to be co-activated with A t M P K 6 (Ichimura et al., 2000; Desikan et al., 2001; Teige et al., 2004). In fact, I have obtained preliminary data suggesting that C A -M K K 9 may be able to activate A t M P K 4 in vitro, although this w i l l require confirmation 59 (data not shown). Future experiments should allow me to confirm whether the 42 kDa M A P K co-activated with A t M P K 6 by C A - M K K 9 is indeed A t M P K 4 . I have also found that expression of C A - M K K 9 triggers a form of P C D in Arabidopsis plants. Induced D E X : C A - M K K 9 - F L A G transgenic plants display necrotic lesions that become clearly visible two to three days after activation of the transgene (Figure 2.3). Moreover, the appearance of visible lesions in D E X : C A - M K K 9 - F L A G plants was preceded by the accumulation of H2O2 (Figure 2.4). It is now well established that R O S , including H2O2, play a central role as signaling molecules orchestrating the onset and the progression of HR- l ike P C D (Lamb and Dixon, 1997). Moreover, a complex network of interactions between these R O S and the hormones salicylic acid (SA), ethylene and jasmonic acid (JA) regulates the initiation, progression and containment of P C D (Overmyer et al., 2000; Rao et al., 2002; Overmyer et a l , 2003). Ethylene has been associated with multiple forms of P C D , and its participation in H R and ozone-induced cell death is well-established. Ethylene insensitivity is associated with increased resistance to ozone damage (Overmyer et al., 2000). In addition, ethylene stimulates cell death, and both its biosynthesis and perception are required for H2O2 accumulation and subsequent P C D in plant tissues in response to certain abiotic stimuli (de Jong et al., 2002; Moeder et al., 2002). It thus appeared possible that C A - M K K 9 might trigger P C D through its activity on A t M P K 6 (Figure 2.8, Figure 2.7), and the resulting overproduction of ethylene (Figure 2.5). However, I found that the DEX-induced D E X : C A - M K K 9 -¥LAGImpk6 plants developed lesions similar to those seen on D E X : C A - M K K 9 - F L A G plants following D E X induction (Figure 2.11), despite the wild-type levels of ethylene, 60 and the absence of A t M P K 6 in the double mutant genotype (Figure 2.10). On one hand, this tells me that CA-MKK9-med ia t ed P C D is not caused by the ethylene burst observed in D E X : C A - M K K 9 - F L A G plants (Figure 2.12). In fact, although ethylene is reported to promote the spreading of lesions during P C D , there is no evidence that it can, by itself, trigger the onset of P C D . For example, the ethylene-overproducing mutants etol and eto3 display increased damage in response to ozone, but do not spontaneously form lesions (Rao et al., 2002). Therefore, although I may hypothesize that D E X : C A - M K K 9 - F L A G plants should indeed be oversensitive to PCD-triggering stimuli due to high ethylene production, the latter is not sufficient to explain the spontaneous lesion formation observed following D E X induction. The cell death phenotype of DEX:CA-MKK9-FLAG//w/?£f5 plants also demonstrates that CA-MKK9-media ted P C D does not require the activity of A t M P K 6 (Figure 2.12). This suggests that A t M P K 6 , although it is activated by several biotic stresses associated with P C D , plays more of a parallel role in plant defense and is not directly responsible for activating the cell death program (Nuhse et a l , 2000; Menke et al., 2004). This conclusion contrasts with the interpretation of studies on A t M P K 6 ' s tobacco orthologue (NtSIPK) where the authors concluded that N t S I P K overexpression is sufficient in itself to trigger P C D (Zhang and L i u , 2001). In light of my findings, it appears that an AtMPK6-independent pathway activated by C A - M K K 9 triggers cell death in the D E X : C A - M K K 9 - F L A G plants. Interestingly, the inhibition of A t M P K 4 activity, which plausibly corresponds to the 42 kDa M A P K activated by C A - M K K 9 (Figure 2.7), has been correlated with the inhibition of P C D in Arabidopsis cell cultures (Desikan et al., 2001). Moreover, in tobacco, the M A P K K N t M E K 2 has been hypothesized to trigger 61 P C D by upregulating the expression of a gene coding for a respiratory burst oxidase homologue, thereby increasing R O S production (Yoshioka et al., 2003). These observations and models provide many interesting avenues for further research into the mechanism of CA-MKK9-med ia t ed P C D . The gain-of-function system used in this study, as well as others (Zhang and L i u , 2001; K i m et al., 2003; L i u and Zhang, 2004; Pedley and Martin, 2004), is a useful tool for defining the M A P K s and metabolic events acting downstream of a M A P K K of interest. However, this system is somewhat artificial in that it presumably does not exactly reflect the usual biological context in which a M A P K K is active, and may lack modifying factors that control the specificity of that M A P K K . Therefore, my model, while fully consistent with my experimental results undoubtedly presents a simplistic view of what is likely a complex signaling cascade. In order to better understand the role of A t M K K 9 in vivo, as well as to address the issue of overlapping function and specificity between different groups of M A P K K s , it w i l l be necessary to monitor variation in the catalytic activity of these M A P K K s in planta, in their endogenous form. Such experiments would permit us to gain a fuller understanding of A t M K K 9 ' s role in mediating both ethylene biosynthesis and P C D , and could also provide a template for analogous studies of the roles of other, as yet uncharacterized, M A P K K s in plant biology. 62 3. Investigation of short-term transcriptional events induced by CA-MKK9 3.1. Introduction M A P K cascades have been identified as mediators of transcriptional changes in several organisms. In yeast, humans, Drosophila melanogaster and Caenorhabditis elegans, transcription factors serve as targets of M A P K signaling. The phosphorylation of transcription factors by M A P K s can modulate their interaction with the basal transcription machinery, their D N A binding kinetics, their stability or their subcellular localization (Treisman, 1996). Although little is known about the substrates of M A P K s in plants, there is growing evidence that transcription factors can be such phosphorylation targets. In rice, O s E R E B P l , an A P 2 / E R E B P family transcription factor, displayed enhanced DNA-bind ing after phosphorylation by the nuclear-localized M A P K , B W M K 1 (Cheong et al., 2003). More recently, in tobacco, N t W I P K was found to increase the transcription modifying activity of NtWIF , a transcription factor with homology to the A R F family (Yap et al., 2005), while N t S I P K has been reported to phosphorylate a N t W R K Y l transcription factor and thereby increase its DNA-bind ing activity (Menke et a l , 2005). In addition, several studies report the translocation of plant M A P K s into the nucleus, suggesting that they could modulate the activity of transcription factors (Ligterink et a l , 1997; Kroj et al., 2003; Ahlfors et al., 2004; Lee et al., 2004). 63 Genome-wide expression profiling by microarrays provides an efficient way to capture the transcriptional re-modeling triggered by given stimuli. These high-throughput experiments can provide insight into the metabolic pathways activated or silenced by a stimulus of interest. Moreover, they can also be used as tools to study co-regulated transcriptional modules. It is well-accepted that the transcription of a gene is mediated by the interaction of a transcription factor and specific D N A sequences (c/s-acting elements) in the promoter of that gene. B y retrieving common cz's-acting elements in the upstream sequence of co-regulated genes, one can generate hypotheses about the identity of the transcription factors regulating the activity of those genes (Yeung et al., 2004). This approach has been pioneered and explored mainly in the yeast model (Wyrick and Young, 2002), but appears applicable to a wide range of organisms (Ohler and Niemann, 2001). The inducible gain-of-function system described in Chapter 2 allowed me to trigger the activity of A t M K K 9 , as well as of its downstream signaling targets, in a temporally-controlled manner. It therefore presents an opportunity to investigate the transcriptional changes triggered by the M A P K cascade(s) to which A t M K K 9 contributes. In this chapter, I present the results of the transcriptional profiling of D E X : C A - M K K 9 - F L A G plants using two-channel oligonucleotide microarrays. The goal of this experiment was the identification of metabolic pathways that are affected by C A - M K K 9 activity, and that might be contributing to the phenotypes of the transgenic plants. In addition, I wished to analyze the common c/s-acting elements found in the promoters of CA-MKK9-af fec ted 64 genes to create hypotheses about transcription factors that might be operating downstream of an A t M K K 9 / M A P K cascade. The statistical analysis of my microarray results showed that significant transcriptional changes could be detected two hours after C A - M K K 9 activation. The C A - M K K 9 -affected genes included a number of genes involved in the octadecanoid pathway. Moreover, I found that the promoters of genes down-regulated in the short-term by C A -M K K 9 were enriched in A B R E - l i k e elements, pointing towards A B F proteins as putative targets of MKK9-act ivated M A P K s . However, my attempts to validate the microarray results using additional biological replicates and quantitative real-time (QRT)-PCR revealed that the majority of these early-response microarray results were apparently false positives. I therefore conclude that the microarray experimental design I used was probably inappropriate to study early downstream targets of C A - M K K 9 using the DEX-inducible system. 3.2. Material and methods 3.2.1. Plant material and treatments For each genotype and treatment, twenty to thirty plants were germinated and grown on soil. Seventeen days after germination, plants were sprayed with either a D E X solution (25 u M D E X and 0.015% Silwet-L77) or a control solution (0.015% Silwet-77). Rosettes 65 were collected and flash-frozen in liquid nitrogen, at the indicated times after the treatment. 3.2.2. Microarrays 3.2.2.1. RNA isolation Tissue (0.8-1.0 g) from the pooled frozen plants were ground in liquid nitrogen, and mixed with 15 ml TRIzol reagent (Invitrogen). The samples were incubated for five minutes at 65°C with regular mixing, and centrifuged for twenty minutes at 4°C (12,000 rpm, Beckman J A 25.50). Three ml chloroform was then added and vortexed with the supernatant. The samples were spun for 30 minutes at 4°C (4,000 rpm, Sorvall R T H -750), and the aqueous phase transferred to new DEPC-treated Corex tubes. The extraction was repeated a second time with an equal volume of chloroform. The samples were then precipitated with a half volume of isopropanol and a half volume of 0 .8M Na3-citrate, gently mixed by inversion, and centrifuged for thirty minutes at 15°C (10,000 rpm, Beckman J A 25.50). The pellets were washed with 15 ml 70% ethanol, centrifuged for 20 minutes at 4°C (10,000 rpm, Beckman J A 25.50), air dried for five minutes and resuspended in 200-350 pi RNAse-free water. The R N A was quantified using a 2100 Bioanalyzer (Agilent Technologies) and re-precipitated overnight at -20°C with 0.1 volume of 3 M N a O A c and 2.5 volumes of 100% ethanol. Samples were then centrifuged for 15 minutes at 4°C (12,000g, microcentrifuge) and the pellets resuspended to a final R N A concentration of 5 pg/pl. 66 3.2.2.2. cDNA flurorescent probe preparation The reverse transcription reactions were performed using 70u.g total R N A into a 40 ul volume containing I X First-Strand buffer (Invitrogen), 3 .75uM Anchor-T primer, 500uM dNTPs-dTTp, 5 0 u M dTTP, l O m M D T T and 0.025 m M Cy3- or Cy5-dUTP (Amersham), which were incubated for 5 minutes at 65°C and for 5 minutes at 42°C before the addition of 40 U R N A s e O U T (Invitrogen) and of 400 U Superscript II reverse transcriptase (Invitrogen). After incubation for two and a half hours at 42°C, the reactions were stopped with 8ul 1 M N a O H , incubated for 15 minutes at 65°, and neutralized by adding 8 ul 1 M HC1, 4 p i 1 M Tr i s -HCl pH7.5 and 40 pi water. The labeled c D N A s were then purified using the Qiaquick P C R purification kit (Qiagen) according to the manufacturer's instructions and eluted twice in 50 pi . Cy3- and Cy5-labeled samples were combined, 1 pi C y 5 - G F P was added, and the product was precipitated overnight at --20°C using 0.1 volume 3 M N a O A c and 2.5 volumes 100% ethanol. 3.2.2.3. Hybridization, scanning and image processing We used microarray slides printed with the Arabidopsis Genome Oligo Set Version 1.0 (26,090 70mer oligonucleotides). To prepare them for hybridization, the slides were shaken (80 rpm) for one hour at 48°C in prehybridization solution (5X SSC, 0.1%> SDS, 0.2%) B S A ) . The slides were then rinsed twice in water, dipped five times in isopropanol, and dried by a three minute centrifugation (2000 rpm, Haerreus) in 50 ml Falcon tubes. Meanwhile, the probes were spun down at 4°C for 30 minutes (14,000 rpm), briefly washed with 500 pi 70% ethanol and centrifuged for 15 minutes at 4°C (14,000 rpm), air-dried, and resuspended in 3.5 pi l O m M E D T A . The probes were then denatured at 95°C 67 for two minutes, mixed with 50ul hybridization solution (Ambion) and kept at 65°C until used. The microarray slides were placed into hybridization chambers in which small slots were filled with 20 pi water to help prevent dehydration. On each array, the probe was applied lengthwise as small drops and gently covered with an untreated glass cover slip (Fisher Scientific). The hybridization chambers were sealed and incubated in a shaking water bath (40 rpm) at 42°C for approximately eighteen hours. After the hybridization, the cover slips were removed and the slides washed for fifteen minutes once in 2 X SSC, 0.5% SDS, and twice in 0.5X SSC, 0.5% SDS, briefly rinsed in 0.1 X SSC and dried by a three minute centrifugation (2,000 rpm, Haereus) in 50ml Falcon tubes. Microarrays were scanned with a ScanArray Express (Perkin-Elmer) with laser power set at 90%) and photomultiplier tube set to 63-66%. The spots were then identified and quantified using the ImaGene software (BioDiscovery). Spot grids were placed manually and adjusted by using the Auto Adjust function three times. Raw spot and background intensities were saved as text files. Median spot and background pixel intensities were used for further analysis. 3.2.2.4. Data analysis The background was calculated as the mean of the 10 %> weakest spots for every subgrid, and subtracted from each spot signal value. The background-corrected intensities were then normalized for each channel using variance stabilization normalization (VSN) . This normalization method was chosen under the assumption that the variance observed between the two channels was independent of the signal intensities (Huber et al., 2002). Therefore, for each slide, the variance was equalized across intensities. 68 A four-way A N O V A was performed on the normalized data. This statistical procedure was entirely designed and carried out by Rick White, a statistical consultant for Genome B C . A summary of the A N O V A design and results is presented in the Appendix. For filtering, p-values for the differential effect of C A - M K K 9 at T=2 and T=0 were considered. The fold-change between T=2 and T=0 was obtained by exponentiating (10 ) the absolute value of the difference between l o g 2 ( D E X : C A - M K K 9 - F L A G -F L A G T = 2 / E V T = 2 ) and l o g 2 ( D E X : C A - M K K 9 - F L A G - F L A G T = 0 / E V T = 0 ) . Ratios smaller than one were converted to negative ratios by raising to the power -1 (X" 1 ) . 3.2.3. Functional and promoter analysis The in silico promoter analysis was performed using the web-based software package, "Athena" (O'Connor et al., 2005). The A G I numbers of the down- and up-regulated genes identified in the filtered gene lists from the microarray analysis were entered in the Analysis Suite application (http://www.bioinformatics2.wsu.edu/cgi-bin/Athena/cgi/analysis_select.pl). The promoter region of these genes was retrieved as the 1000 base-pair sequence upstream of the transcription start, cutting off at adjacent upstream genes as necessary. The software calculated the transcription factor binding frequency and enrichment of the tested promoters. 3.2.4. Quantitative Real-time R T - P C R Total R N A was extracted from new plant samples with the RNeasy Plant M i n i kit (Qiagen). Total R N A (2 pg) was used for reverse transcription with OligodT (Invitrogen) 69 using Superscript II reverse transcriptase (Invitrogen) according to the manufacturer's instructions. The resulting c D N A samples were diluted by a factor of ten in water. Primers were designed using 'Primer3' (http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi) to amplify a 140 to 160 bp unique region of the tested genes, i f possible spanning an intron. The specificity of each primer pair was tested using B L A S T ( n ) on the Arabidopsis genome. Table 3.1. Primers used for QRT-PCR Gene Description AGI number Primer name Sequence Actin (ACT8) At1g49240 ACT8 (QRT)F ACT8 R 5' TCTAAGGAGGAGCAGGTTTGA 3' 5' TTATCCGAGTTTGAAGAGGCTAC 3' Lipoxygenase (LOX3) At1g17420 LOX3 F LOX3 R 5' AAGAGGTTCCTTACCCTAGACGTT 3' 5' AAGTGTCCTGCTTCGACTCTTC 3' Lipoxygenase (LOX4) At1g72520 LOX4 2 F LOX4R 5' CCGGGTGTTACGTGTAGAGG 3' 5' TCGGCAAATAAACCATACTGC 3' 2-oxophytodienoate (OPR3) At2g06050 OPR3 F OPR3 R 5' GGAGTGGTCCGTTGAGCATA 3' 5' GGCACAAGGGAACTCTAACG 3' 2-oxophytodienoate (OPR1) At1g76680 OPR1 F OPR1 R 5' AGACGGCTTGGTATCGAAGA 3' 5' CCGTATCCTTCATGAACTGG 3' Jasmonic acid carboxyl methyltransferase (JMT) At1g 19640 JMT F JMT R 5' GCTTATTTTGGTGAAACCTTGC 3' 5' TCCTTGACGCTCAATACAGAAA 3' WRKY transcription factor (WRKY30) At5g24110 WRKY F WRKY R 5' AACTACTCCGGCGAACTTGA 3' 5' GGGGCAATTCTGAC I I I IGA 3' Universal stress protein At1g48960 UPS F UPS R 5' GCCGTCACGGATACAATCTT 3' 5' AGAAGCATCGAAGCACCAAT 3' Zinc finger (C2H2 type) family protein At3g45260 ZnFgF ZnFg R 5' CATCATCCCCTCTCATTTCC 3' 5' CGATGACTTCTGCATCTGGA 3' Stress-inducible protein At1g 12270 StP F StP R 5' GGACTTTGAAACTGCTATTCAGC 3' 5' CCCTTTCCACAGCCTTGTTA 3' Glycosyl hydrolase family 1 proteins At5g28510 GlyH F GlyH R 5' AATGCAATGGCGATAATGGT 3' 5' GAGGAAAAATTCTTGTCCATGAG 3' For Q R T - P C R , 2 pi of diluted c D N A (the equivalent of 20ng R N A ) was mixed with 0.5 u M of each primer and I X QuantiTect S Y B R Green P C R Master M i x (Qiagen) in a total volume of 20 pi . The reactions were ran in a D N A Engine Opt icon™ ( M J Research), using the following program: 95°C for 15 minutes, followed by 40 cycles of amplification (94°C for 15 seconds, 55°C for 30 seconds, 68°C for 45 seconds, and a fluorescence reading). After a final elongation step of 5 minutes at 68°C, a melting curve was 70 performed from 60 to 90°C, with a reading followed by a 1 second hold every 0.2°C, in order to assess the specificity of the primers. The results of each Q R T - P C R run were analyzed using Opticon Moni tor™ Analysis Software version 1.07 ( M J Research). The threshold level was adjusted manually above background fluorescence. Each P C R run included a set of A C T 8 standards (10 1 to 10 7 molecules), used to generate a standard curve that allowed me to convert threshold cycle (C(T)) units into molecule units. The standards were obtained by serial dilution of a purified A C T 8 P C R product whose concentration was determined independently by spectrophotometry. The number of molecules for each gene of interest was averaged from two technical replicates and divided by the number of A C T 8 molecules amplified from the same biological sample. 3.3. Results 3.3.1. Experimental design for the study of CA-MKK9-med ia t ed transcriptional changes In order to study transcriptional changes resulting from A t M K K 9 activity, I assessed the global transcriptional profile of multiple D E X : C A - M K K 9 - F L A G transgenic lines, using Arabidopsis oligomer microarrays. Two factors were considered with particular care when choosing a time-point for monitoring CA-MKK9-media ted transcriptional changes. First, I wished to choose a time-point after D E X induction when the C A - M K K 9 transgene was strongly induced. I have previously described the induction pattern of C A -M K K 9 in D E X : C A - M K K 9 - F L A G transgenic lines (L2, L I 2 and L I 3 ) and shown that the transgene is transcribed at high levels two hours after D E X induction (Chapter 2;Figure 2.2). A t this same time, I also observed increased activity of two M A P K s , suggesting that 71 C A - M K K 9 is active in the plants (Chapter 2; Figure 2.7). Second, the time-point had to be early enough to avoid capturing gene expression changes caused by secondary signaling pathways that might be triggered by early transcriptional events resulting directly from C A - M K K 9 activity. I had observed that, following the transgene activation, D E X : C A - M K K 9 - F L A G plants started overproducing ethylene four hours after D E X treatment (Chapter2; Figure 2.5), and treating plants with ethylene is known to induce transcriptional changes (Solano et al., 1998; Guo and Ecker, 2004). A s the goal of my experiment was not to study ethylene-mediated transcriptional changes, but rather AtMKK9-mediated changes, it appeared necessary to capture the transcriptional profile of the plants before they showed detectable production of ethylene. Taken together, these data indicated that two hours after D E X induction (T=2) was an appropriate time-point to monitor global transcriptional activity in D E X : C A - M K K 9 - F L A G plants. There are several caveats to the use of the DEX-inducible system for a global transcriptional study. Although the DEX-inducible system has been repeatedly used in plant research, the potential influence of the chimeric transcription factor G V G on endogenous gene expression, independently or not of D E X treatment, remains largely uncharacterized. It has been reported that high levels of G V G in transgenic lines expressing the empty vector are correlated with growth defects, as well as with the expression of defense-related genes such as PDF1.2 and PR-5 (Kang et al., 1999). This is of particular concern in the context of my study, since C A - M K K 9 appears to be involved in stress responses (Chapter 2). Furthermore, as the aim of this microarray study was to capture early changes resulting from C A - M K K 9 activity, I reasoned that the presence of 72 undefined noise background solely due to G V G could potentially hinder my ability to successfully capture these changes. In order to minimize the impact o f that hypothetical artefactual noise on the experiment, two controls were used. First, the microarrays were performed on three independent D E X : C A - M K K 9 - F L A G transgenic lines (L2, L12 and L13). Second, each o f these transgenic lines was paired with an Empty Vector (EV) transgenic line expressing comparable levels o f G V G . Figure 3.1 shows that D E X : C A -M K K 9 - F L A G L 2 had G V G levels similar to those in E V L 3 , whereas D E X : C A - M K K 9 -F L A G L12 was most similar to E V L 1 0 and D E X : C A - M K K 9 - F L A G L13 to E V L l . Based on this information, each o f these D E X : C A - M K K 9 - F L A G lines was paired to the appropriate E V line, and both were subjected to the same treatment in the microarray experiments. ACT8 Figure 3.1. GVG transcript levels in DEX:CA-MKK9-FLAG and EV transgenic lines. DEX: CA-MKK9-FLAG plants (L2, L12 and L13) and EV plants (L1, L3 and L10) were grown for 18 days. Total RNA was extracted and reverse-transcribed. The levels of GVG transcript were determined by RT-PCR. Equal loading in each lane is indicated by the comparable ACT8 levels detected in each sample. 73 Another point to consider in the experimental design of the microarray was the basal expression levels of the C A - M K K 9 transgene before D E X induction. M y time-course study of C A - M K K 9 induction had revealed that in D E X : C A - M K K 9 - F L A G L I 3 , low levels of C A - M K K 9 expression could be detected even before the D E X treatment (Figure 2.2). Replicates of this experiment, in which higher number of P C R cycles were performed to amplify the C A - M K K 9 c D N A , showed that the transgene displays similar basal expression levels in the two other lines (L2 and L I 2 ) as well (data not shown). This indicates that G V G might have some activity in the nucleus before the D E X treatment, which, in addition to the basal and induced expression of C A - M K K 9 , might constitutively influence the transcription of other genes. Altogether, I hypothesized that the basal activity of G V G and of C A - M K K 9 might create a background of responses that would contribute an unknown component to the expression profiles of the D E X : C A -M K K 9 - F L A G / E V plants two hours after D E X induction. In order to be able to subtract that background from the two-hour post-DEX transcriptional changes, I therefore assessed the transcription profile of the same D E X : C A - M K K 9 - F L A G / E V lines pairs before the D E X treatment (T=0). The experimental design of the microarray experiment is pictured in Figure 3.2. For the L13/L1 and the L12/L12 pairs, the arrays were performed on two biological replicates. For the L2/L3 pair, arrays were done on one biological replicate. 74 DEX:CA-MKK9-FLAG L13 Vs EVL1 T=0 T=2 DEX:CA-MKK9-FLAG L2 Vs E V L3 1 T -0 1 J DEX:CA-MKK9-FLAG L12 Vs EVL10 T=0 T=2 B-1 B-2 B-1 B-2 T-2 T-4 T-3 T-2 BE] B-1 B-2 T-2 T-2 * B-1 T- 2 T-2 Total number of slides = 27 Figure 3.2. Microarray experimental design. DEX:CA-MKK9-FLAG L13 was hybridized against EV L1, DEX:CA-MKK9-FLAG L12 against EV L10, and DEX:CA-MKK9-FLAG L2 against EV L3. Plants were treated either with DEX (T=2) or with a mock spray (T=0) and collected two hours thereafter. The number of biological replicates (B-X) and technical replicates (T-X) is indicated for each experiment. (*) indicates that for one of the technical replicates, EV L3 RNA from B-1 was used. 3.3.2. C A - M K K 9 mediates significant short-term transcriptional changes After background subtraction and normalization of the data, the statistical significance o f the expression differentials was determined using a multifactorial A N O V A (Appendix). The A N O V A revealed that there were indeed a significant number o f genes whose expression was altered due to the C A - M K K 9 transgene, between T=2 and T=0, suggesting that the microarray experiment was successful in revealing short-term C A -MKK9-mediated transcriptional changes. I generated a list o f the most significantly altered genes by filtering the normalized data according to the following criteria: the p-value should be <0.05 for the C A - M K K 9 effect between T=2 and T=0 data, and the observed fold-change should be >2 between T=2 and 75 T=0 data. Based on these criteria, I obtained a list of 191 genes, among which only 11 were up-regulated while 180 were down-regulated (Appendix, Table 5.2). The large number of down-regulated genes suggested that C A - M K K 9 might generally mediate transcriptional repression. Notably, the up-regulated genes included A t M K K 9 / C A -M K K 9 , indicating that our analysis method was at least able to accurately detect the transgene induction. 3.3.3. Functional analysis of transcriptional changes resulting from C A - M K K 9 activity Having retrieved the 191 genes whose expression was most altered by C A - M K K 9 induction, I wished to determine whether that group shared any functional similarities. To do so, I regrouped these genes based on their common gene ontology (GO) terms. The relative abundance of each G O term assignment among these genes was compared with the abundance of the given term across a random set of 3000 genes (Table 3.2). GO Slim Term CA-MKK9 Total # % Random set % Cellular component other membranes 61 17.4 23.2 other cellular components 53 15.1 10.7 cellular component unknown 42 12.0 19.2 other intracellular components 42 12.0 9.2 chloroplast 39 11.1 9.6 Nucleus 35 10.0 8.1 mitochondria 23 6.6 7.6 other cytoplasmic components 19 5.4 4.5 ribosome 8 2.3 2.5 extracellular 8 2.3 1.5 plasma membrane 8 2.3 1.1 ER 5 1.4 0.7 cytosol 3 0.9 0.7 plastid 2 0.6 0.8 Cell wall 1 0.3 0.7 Golgi apparatus 1 0.3 0.3 Molecular function 76 GO Slim Term CA-MKK9 Total # % Random set % Hydrolase activity 63 13.7 9.4 other enzyme activity 47 10.2 14.1 Molecular function unknown 41 8.9 15.8 Nucleotide binding 40 8.7 5.4 DNA or RNA binding 37 8.0 7.1 protein binding 32 7.0 3.9 kinase activity 29 6.3 6.0 transferase activity 29 6.3 9.8 transporter activity 28 6.1 7.2 other binding 27 5.9 6.7 other molecular functions 27 5.9 3.0 structural molecule activity 23 5.0 2.8 nucleic acid binding 19 4.1 3.2 transcription factor activity 15 3.3 4.2 receptor binding or activity 3 0.7 1.0 Biological process other physiological processes 183 22.4 21.2 other metabolic processes 149 18.2 19.1 other cellular processes 148 18.1 18.1 protein metabolism 68 8.3 6.7 transport 53 6.5 5.8 biological process unknown 46 5.6 10.1 other biological processes 31 3.8 3.8 cell organization and biogenesis 29 3.5 1.9 response to stress 27 3.3 1.7 transcription 23 2.8 3.2 response to abiotic or biotic stimulus 17 2.1 1.5 signal transduction 13 1.6 1.3 DNA or RNA metabolism 13 1.6 1.6 developmental processes 11 1.3 1.2 Electron transport or energy pathways 7 0.9 2.1 The "cellular component" ontology of C A - M K K 9 - a l e r e d genes indicated that these genes contain a large proportion of extracellular- (2.3%), plasma membrane- (2.3%) and endoplasmic reticulum- (1.4%) targeted proteins, as compared to the average percentages of these proteins in the random set (1.5%), 1.1% and 0.7%>, respectively). As for the "molecular function" ontology of CA-MKK9-af fec ted genes, it appeared that "protein binding" (7%>) and "structural molecules" (5%) were the most overrepresented, as compared to the average representation (3.9% and 2.8%, respectively). 77 Most importantly, the analysis of the "biological function" ontology group revealed that 3.3% of CA-MKK9-af fec ted genes were involved in "response to stress", which represents a 94% over-representation compared to the 1.7% of the random set that fell into this G O category. CA-MKK9-af fec ted genes also contained more "cell organization and biogenesis" genes (3.5%) compared to the average (1.9%). It is notable that the C A -MKK9-affected gene group contained a disproportionately low number (0.9%) of "electron transport and energy pathway" genes, as compared to the random group (2.1%). 3.3.4. Promoter analysis of CA-MKK9-af fec ted genes I hypothesized that genes that had a similar early expression pattern following C A -M K K 9 activation would share a common set of cw-acting elements in their promoter region. Using the web-based application 'Athena', I retrieved the promoter region of down- and up-regulated genes of the CA-MKK9-af fec ted gene list, and identified common binding factors motifs among these promoters. The frequency of each motif among the promoter subset was compared to its frequency in the entire Arabidopsis genome. Motifs that were significantly overrepresented (p-value<0.05) among the up- or down-regulated genes subsets are presented in Table 3.3. The CA-MKK9-down-regulated genes were mostly enriched with A B R E - l i k e motifs (pO.OOl). The CA-MKK9-up-regulated genes, on the other hand, were most significantly enriched with the C C A 1 binding site motif (Table 3.3). However, because the latter gene set contains only ten genes, it is unclear whether meaningful biological conclusions can be derived from its analysis. 78 Table 3.3. Over-represented c/'s-acting elements among CA-MKK9-up- and down-regulated genes. Motif Name Consensus sequence 3 Subset % Subset # Genome % Genome # P-value CA-MKK9-down-regulated genes (180) ABRE-l ike binding site motif B A C G T G K M 28% 50 18% 5493 0.001 L E A F Y A T A G C C A A T G T 13% 24 8% 2490 0.01 MYB1 binding site motif M T C C W A C C 9% 16 4% 1438 0.01 A R F binding site motif T G T C T C 38% 68 30% 9245 0.013 Z-box promoter motif A T A C G T G T 5% 9 2% 674 0.017 UPRMOTIFIIAT C C ( N ) i 2 C C A C G 5% 10 2% 888 0.035 RAV1-B binding site motif C A C C T G 13% 24 9% 2808 0.036 D R E core motif R C C G A C 24% 42 18% 5550 0.039 CA-MKK9-up-regulated genes (11) CCA1 binding site motif A A M A A T C T 54% 6 23% 7075 0.025 L1-box promoter motif T A A A T G Y A 36% 4 13% 3965 0.046 C A C G T G M O T I F C A C G T G 36% 4 13% 4045 0.049 a A, adenine; C, cytosine; G, guanine ; T, thymine ; R= A or G; B=C, G or T; D= A, G or T; H= A, C or T; V= A, C or G; N= A, Y= Cor T; W C, Gor T. = A o r T ; S= C or G; M,=A or C; K= G or T; In order to provide groundwork for future studies on potential A t M K K 9 - d r i v e n M A P K cascades leading to transcriptional changes, I explored the hypothesis that C A - M K K 9 might alter the activity of a ABRE-b ind ing factor ( A B F ) , leading to the down-regulation of the 50 genes under the bearing these A B R E motifs. There are four A B F s characterized so far in Arabidopsis. I retrieved the amino acid sequences of the A B F s , and identified putative M A P K phosphorylation sites, as well as M A P K docking and recognition sites in those sequences. M A P K s are known to phosphorylate serine and threonine residues that are followed by a proline residue (Cohen, 1997). Moreover, transcription factors targeted by the mammalian M A P K s E R K and p38 have a docking site known as the ' D domain', which is necessary for efficient phosphorylation by M A P K s . The D domain is typically upstream of the phosphorylation sites, composed of a stretch of basic amino acids, followed by a L e u - X - L e u motif and a hydrophobic region (Tanoue and Nishida, 2003). Some M A P K substrates also have a M A P K recognition site called the Phe-X-Phe-Pro 79 motif, located downstream of the phosphorylation sites (Tanoue and Nishida, 2003). A B F 2 and A B F 3 , which contained six and four, respectively, potential phosphorylation sites, stood out as most likely targets o f a C A - M K K 9 - d r i v e n M A P K cascade (Figure 3.3), although they had neither a D domain, nor a Phe-X-Phe-Pro motif However, as Phe-X-Phe-Pro motifs are present in none o f the known plant M A P K substrates, and putative D domains are found in some but not all o f the latter, these two transcription factors could still be considered as M A P K substrate candidates in future investigations. ABF1 - At1g49720 MGTHIDINNLGGDTSRGNESKPU^RQSSLYSLTFDELQSTLGEPGKDFGSMNMDELU<NIWTAEDTQAFMTTTSS VAAPGF«GFVPGGNGLQRQGSLTLPRTLSQKTVDEVWKYLNSKEGSNGNTGTDALERQQTLGEMTLEDFLLRAG WKEDNTQQNENSSSGFYANNGAAGLEFGFGQPNQNSISFNGNNSSMIMNQAPGLGLKVGGTMQQQQQPHQQ QLQQPHQRLPPTIFPKQANWFAAPVNMWIRG VFGRGRRSNTGLEKWERRQKRMIKNRESAARSRARKQAYTLELEAEIESLKLVNQDLQKKQAEIMKTHNSELKEF SKQ PPLLAKRQC LR RT LTGPW ABF2 - At1g45249 MDGSMNLGNEPPGDGGGGGGLTRQGSIYSLTFDEFQSSVGKDFGSMNMDELLKNIWSAEETQAMASGWPVLG GGQEGLQLQRQGSLTLPRTLSQKTVDQVWKDLSKVGSSGVGGSNLSQVAQAQSQSQSQRQQTLGEVTLEEFLV RAGVVREEAQVAARAQIAENNKGGYFGNDANTGFSVEFQQPSPRWAAGVMGNLGAETANSLQVQGSSLPLNV NGARTTYQQSQQQQPIMPKQPGFGYGTQMGQLNSPGIRGGGLVGLGDQSLTNNVGFVQGASAAIPGALGVGAV SPVTPLSSEGIGKSNGDSSSLSPSPYMFNGGVRGRKSGTVEKWERRQRRMIKNRESAARSRARKQAYTVELEA EVAKLKEENDELQRKQARIMEMQKNQETEMRNLLQGGPKKKLRRTESGPW ABF3 - At4g34000 MGSRLNFKSFVtXBVSEQQPTVGTSIPLTRQNSWSLTFDEFQNSWGGGIGKDFGSMNMDElXKNIWTAEESHSM MGNNTSYTNISNGNSGNTVINGGGNNIGGI^VGVGGESGGFFTGGSLQRQGSLTLPRTISQKRVDDVWKELMKE DDIGNGWNGGTSGIPQRQQTLGEMTLEEFLVRAGWREEPQPVESVTNFNGGFYGFGSNGGLGTASNGFVANQ PQDLSGNGVAVRQDLLTAQTQPLQMQQPQMVQQPQMVQQPQQLIQTQERPFPKQTTIAFSNTVDWNRSQPAT QCQEVKPSILGIHNHPMNNNLLQAVDFKTGVTVAAVSPGSQMSPDLTPKSALDASLSPVPYMFGRVRKTGAVLEK VIERRQKRMIKNRESAARSRARKQAYTMELEAEIAQLKELNEELQKKQVEIMEKQKNQLLEPLRQPWGMGCKRQC LRRTLTGPW ABF4 - At3g19290 MGTHINFNNLGGGGHPGGEGSSNQMKPTGSVMPLARQSSVYSLTFDELQNTLGGPGKDFGSMNMDELLKSIWT AEEAQAMAMTSAPAATAVAQPGAGIPPPGGNLQRQGSLTLPRTISQKTVDEVWKCLITKDGNMEGSSGGGGES NVPPGRQQTLGEMTLEEFLFRAGWREDNCVQQMGQVNGNNNNGFYGNSTAAGGLGFGFGQPNQNSITFNGT NDSMILNQPPGLGLKMGGTMQQQQQQQQLLQQQQQQMQQLNQPHPQQRLPQTIFPKQANVAFSAPVNITNKG FAGAANNSINNNNGLASYGGTGVTVAATSPGTSSAENNSLSPVPYVLNRGRRSNTGLEKVIERRQRRMIKNRESA ARSRARKQAYTLELEAEIEKLKKTNQELQKKQAEMVEMQKNELKETSKRPWGSKRQCLRRTLTGPW Figure 3.3. Putative MAPK phosphorylation sites in ABRE-binding factors. Serine or threonine residues followed by a proline are shown in bold red. 80 3.3.5. C A - M K K 9 induction results in the down-regulation of jasmonate biosynthesis genes. Within the CA-MKK9-af fec ted gene group were found a number of genes coding for enzymes of the octadecanoid pathway, which leads to the biosynthesis of octadecanoid signaling metabolites, including jasmonic acid (JA). The first committed step in that pathway is the conversion of the a-linolenic acid to 13-hydroperoxylinolenic acid, catalyzed by lipoxygenase enzymes (Figure 3.4). a-linolenic acid | Lipoxygenase (LOX) 13-hydroperoxy linolenic acid | Allene oxide synthase (AOS) Allene oxide | Allene oxide cyclase (AOC) 12-oxo-phytodienoic acid | 12-oxo-phytodienoic acid reductase (OPR) 3-oxo-2(2 '-pentenyl)-cyclopentane-1 octanoic acid I | B-oxidation i (+)-7-iso-jasmonic acid | Jasmonic acid methyl transferase (JMT) (+)-7-iso-methyljasmonate Figure 3.4. The jasmonate biosynthesis pathway. Adapted from Devoto and Turner (2005). 81 According to my microarray data, L O X 2 , L O X 3 and L 0 X 4 , which all code for lipoxygenases, were down-regulated upon C A - M K K 9 activation. L O X 2 and L O X 4 , which displayed a -2.8 and -2.7 fold-change, respectively, between T=0 and T=2, were the most strongly down-regulated, while L O X 3 showed a -2.1 fold-change (Table 3.4). Since it appeared that this might represent a significant aspect of C A - M K K 9 - d r i v e n transcriptional activity, I asked whether other genes coding for enzymes of the octadecanoid pathway might also have been affected by C A - M K K 9 activity, but not passed the initial filtering, i.e. they might not have shown an expression difference of more than two-fold between T=0 and T=2. Keeping the p<0.05 criterion, I found that four additional genes involved in J A biosynthesis were also modestly down-regulated in the microarray data set as a result of C A - M K K 9 activity. L O X 5 , which also codes for a lipoxygenase, was 1.4-fold downregulated, while OPR1 and OPR3 , coding for O P D A reductases, were down-regulated 1.4- and 1.7-fold, respectively. Finally, J M T , coding for a S-adenosyl-L-methionine:jasmonic acid carboxyl methyltransferase, was down-regulated 1.3-fold (Table 3.4). Table 3.4. Octadecanoid pathway genes significantly altered by CA -MKK9 activity. AGI number Oligo ID Name Description A009087_01 A025792_01 A004372_01 A011266_01 A025313_01 A007905_01 A019988 01 At3g45140 LOX2 At1g17420 LOX3 At1g72520 LOX4 At3g22400 LOX5 At1g76680 OPR1 At2g06050 OPR3 At1g 19640 JMT Chloroplast lipoxygenase Lipoxygenase Lipoxygenase, putative Lipoxygenase, putative 12-oxophytodienoic acid reductase 12-oxophytodienoate reductase S-adenosyl-L-methionine:jasmonic acid carboxyl methyltransferase Fold-change P-value 0.0437 -2.80 -2.09 -2.73 -1.36 -1.42 -1.72 -1.34 0.0252 0.0189 0.0188 0.0444 0.0004 0.0290 82 Since these were not major shifts in transcriptional intensity, and microarray data inevitably contain considerable noise, I re-evaluated the expression of these genes by mean of an independent and more reliable method: quantitative real-time (QRT-) P C R . First, in order to see whether the microarrays results could be technically replicated, I used the same R N A samples that I had used for some of the microarray replicates and measured the expression of L O X 2 , L O X 3 and OPR3 relative to the housekeeping gene A C T 8 . As shown by Figure 3.5, Q R T - P C R analysis revealed that L O X 2 and L O X 3 were indeed down-regulated in D E X : C A - M K K 9 - F L A G L12 and L13 , while OPR3 expression was essentially unchanged. However, I also observed that these genes all appeared to be up-regulated in the E V L10 line at T=2. This suggests that the down-regulation detected in the microarray analysis might be partly due to this EV-dr iven shift in ratio of gene expression between D E X : C A - M K K 9 - F L A G L12 and E V L10. Moreover, the presence of gene expression changes in the E V L10 genotype upon D E X induction indicates that this particular empty vector line was an inappropriate choice as a control, in the context of my microarray experiment. 83 L0X2 At3g45140 fold change -2.80, p=0.0437 LOX3 At1g17420 fold-change -2.1, p=0.0252 30 -25 eculc 20 -o E o 15 • > 10 -B DEX:CA-MKK9-FLAG L12 DEX:CA-MKK9-FLAGL13 OPR3 At2g06050 fold-change-1.72, p=0.0004 T 1 ... I DEX:CA-MKK9- DEX:CA-MKK9-FLAGL12 FLAG L13 3.5 3 I 2 5 o I 2 o * 1.5 o > 1 1 0.5 0 0EX:CA-MKK9- DEX:CA-MKK9-FLAGL12 FLAG L13 Figure 3.5. Technical validation of JA biosynthesis genes Total RNA of DEX:CA-MKK9-FLAG (L12 and L13) and EV (L1 and L10) used for the microarrays was used for QRT-RT analysis, to determine the abundance of LOX2, LOX3, OPR3 and ACT8 transcripts. Transcript levels are expressed as the ratio of molecules between the tested gene and ACT8 for a given samples. Values represent the average of two replicates. Error bars represent the standard error. M y inability to validate the microarray results for OPR3, despite the apparent statistical robustness of my original results (p-value=0.004) led me to suspect that the putative down-regulation of the suite of J A biosynthesis genes might not be real. I therefore prepared new plant samples to more fully examine the validation of this gene set by Q R T - P C R . Tissue samples from D E X . C A - M K K 9 - F L A G L I 2 , L13 and E V L I plants were collected in triplicate at zero and two hours after D E X induction, in the same manner as for the microarray experiments. Moreover, because I suspected that the two-hour time-point might be so early that plant-to-plant variation in the induction time would obscure meaningful transcriptional events, I also collected samples from additional plants 84 eight hours after the D E X induction. The E V L10 control line was not tested in this experiment, as it appeared from my previous results that it displayed severe mis-regulation of the J A biosynthesis genes upon D E X treatment. Figure 3.6 shows the expression pattern of L O X 2 , L O X 3 , L O X 4 , O P R 1 , OPR3 and J M T in the new biological replicates of DEX-induced D E X : C A - M K K 9 - F L A G (L I2 and LI3 ) and E V L I plants. Only L O X 2 and J M T appeared significantly down-regulated in both D E X : C A - M K K 9 - F L A G lines, as compared to the E V line (Figure 3.6). Conversely, L O X 4 and OPR1 were up-regulated at T=2, while OPR3 again appeared essentially unchanged (Figure 3.6). Most importantly, it appeared that the expression patterns of these six genes displayed a roughly similar trend in the empty vector line ( E V L I ) as in the D E X : C A - M K K 9 - F L A G lines (Figure 3.6). Overall, this suggests that the down-regulation of the octadecanoid pathway predicted by my microarray data may be more a reflection of the effect of D E X treatment than an effect of C A - M K K 9 induction on the plants metabolism. 85 LOX2 At3g45140 fold-change -2.80, p=0.0437 LOX3 At1g17420 fold-change -2.09, p=0.0252 DEX :CA-MKK9-FLAG L12 DEXCA-MKK9-FLAG L13 DEX:CA-MKK9-FLAG L12 DEX:CA-MKK9-FLAG L13 LOX4 At1g72520 fold-change -2.73, p=0.0189 OPR1 At1g76680 fold-change -1.42, p=0.0444 T 1 DEX;CA-MKK9-FLAG L12 DEX: CA-M KK9-FIAG L13 DEX:CA-MKK9-FLAG L12 DEXCA-MKK9-FLAG L13 EVL1 OPR3 At2g06050 fold-change -1.72, p=0,0004 JMT At1g19640 fold-change -1.33, p=0.0290 DEX:CA-MKK9-FLAG U12 DEX:CA-MKK9-FLAG L13 DEXCA-MKK9-FLAG DEXCA-MKK9-FLAG L12 L13 Figure 3.6. QRT-PCR validation of down-regulated jasmonate biosynthesis genes predicted by microarray to be down-regulated by CA-MKK9. Independent tissue samples of DEX:CA-MKK9-FLAG (L12 and L13) and EV (L1) plants were collected 0, 2, and 8 hours after DEX treatment. The abundance of LOX2, LOX3, LOX4, OPR1, OPR3, JMT and ACT8 transcripts was determined by QRT-PCR. Transcript levels are expressed as the ratio of molecules between the tested gene and ACT8 for a given samples. Values represent the average of three independent samples, for which QRT-PCR was performed in duplicates. Error bars represent the standard error. The response of each gene as previously assessed by microarray analysis is shown in smaller font in each graph header. 86 3.3.6. Validation of the most robust CA-MKK9-af fec ted genes The results described in 3.3.5 suggested that the CA-MKK9-af fec ted gene set, as determined by the microarray experiment, contained a large proportion of false positives. In order to determine whether this was due to insufficient stringency in our filtering strategy, or to a general problem of the microarray results, I examined the expression pattern of a subset of those genes that displayed the most statistically robust expression differentials in the microarray experiments. Seeing this analysis as a potential preliminary study for further investigation, I chose to investigate only those genes for which some functional annotation was available. The five chosen genes had the lowest p-values amongst the annotated genes of the CA-MKK9-af fec ted gene set (Table 3.5). Table 3.5. Most significant CA-MKK9-affected genes Oligo ID AGI number Description Fold-change P-value A016008. A002005. .01 .01 At5g24110 At1g48960 WRKY 30, member of WRKY Transcription Factor; Group III universal stress protein (USP) family protein 3.25 3.34 0.000054 0.000056 A024355. .01 At3g45260 zinc finger (C2H2 type) family protein -2.21 0.000084 A004610. .01 At1g 12270 stress-inducible protein, putative -2.05 0.000417 A019910. .01 At5g28510 glycosyl hydrolase family 1 protein -2.02 0.000487 Figure 3.7 shows the Q R T - P C R analysis of the expression of these five genes, in independent samples of D E X : C A - M K K 9 - F L A G plants (L12 and L13) and E V L I plants, measured at zero and two hours after D E X induction. Despite the statistical robustness of the expression data for these genes in the microarrays, only W R K Y 3 0 could be consistently confirmed as being up-regulated between T=0 and T=2 (Figure 3.7). 87 WRKY30 Atg24110 fold-change 3.25, p=0.000054 Universal stress protein At1g48960 fold-change 3.34, p=0.000056 0.4 0.2 0.5 0.45 0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05 0 DEX:CA-MKK9-FLAG L12 DEX:CA-MKK9-FLAG L13 DEX:CA-MKK9-FLAG L12 DEX:CA-MKK9-FLAG L13 EVL1 Zinc finger protein At3g45260 fold-change -2.21, p=0.000084 Stress-inducible protein At1g12270 fold-change-2.05, p=0.000417 0.9 0.6 0.7 "5 0.6 o E 0.5 o % 0.4 01 > 0.3 0.2 0.1 0 1.6 1.4 1.2 a 1 o E o 0.8 n stive 0.6 S 0.4 0.2 0 T T 1 • — t — - f -DEX:CA-MKK9-FLAG L12 DEX:CA-MKK9-FLAG L13 DEX:CA-MKK9-FLAG L12 DEX:CA-MKK9-FLAG L13 EVL1 0.006 0.005 -8 0.004 aj o E ? 0.003 -Glycosyl hydrolase At5g28510 fold-change -2.02, p=0.000487 'I 0.002 -E 0.001 -0 -D£X:CA-MKK9-FLAG L12 DEX:CA-MKK9-FLAG L13 Figure 3.7. QRT-PCR validation of the most significant CA-MKK9-affected genes. Independent tissue samples of DEX:CA-MKK9-FLAG (L12 and L13) and EV (L1) plants were collected 0, 2 hours after DEX treatment. The abundance of At5g24110, At1g48960, At3g45260, At1g12270, At5g28510 and ACT8 transcripts was determined by QRT-PCR. Transcript levels are expressed as the ratio of molecules between the tested gene and ACT8 for a given samples. Values represent the average of three independent samples, for which QRT-PCR was performed in duplicates. Error bars represent the standard error. The response of each gene as previously assessed by microarray analysis is shown in smaller font in each graph header. 88 3.4. Discussion In animal systems, M A P K cascades play a crucial role in relaying the information conveyed by an extracellular stimulus to the nucleus (Treisman, 1996). In plants, although the number and diversity of M A P K modules suggests that M A P K cascades play important roles in the plant life cycle, only a few studies have explored the downstream molecular events specifically resulting from M A P K activity. Crosstalk and overlapping functions between different M A P K modules, as well as the transient nature of their activation, can make it difficult to analyze the downstream targets of a single M A P K . Asai et al. (2002) successfully dissected a single M A P K cascade and its downstream targets by re-constructing the cascade in Arabidopsis protoplasts, using a transient expression system. They showed that challenge with a flagellin elicitor, which is recognized by a specific membrane-bound receptor-like kinase, F L S 2 , can activate a M A P K cascade involving A t M E K K l , A t M K K 4 / A t M K K 5 and A t M P K 3 / A t M P K 6 , ultimately leading to the transcription of W R K Y 2 9 and W R K Y 1 (Asai et al., 2002). The inducible expression in planta of a gain-of-function mutant of a M A P K K , A t M K K 9 , should, in principle, provide another good experimental system to study the downstream targets of a single M A P K cascade. A s shown in the previous chapter, D E X : C A - M K K 9 -F L A G transgenic plants rapidly express the C A - M K K 9 transgene upon D E X induction (Chapter 2, Figure 2.2). Moreover, I have found that the transgene activation results in the activation of A t M P K 6 . A s this M A P K can be translocated into the nucleus (Ahlfors et a l , 2004), it appeared possible that its activity would cause rapid transcriptional changes. 89 In this Chapter, I conducted a microarray experiment that aimed at capturing genome-wide transcriptional events resulting from C A - M K K 9 activity. A multifactorial A N O V A allowed me to identify genes that were predicted to be significantly altered as a result of the C A - M K K 9 transgene induction, between zero and two hours after the D E X treatment (Appendix). In order to gain clues as to the identity of cellular processes regulated by C A - M K K 9 activity, I characterized the predicted CA-MKK9-af fec ted gene set using gene ontology (Table 3.2). This analysis revealed that several genes of the octadecanoid pathway appeared down-regulated in the microarrays upon C A - M K K 9 induction (Table 3.4). Since J A and ethylene have been proposed to play antagonistic roles in-the control of pathogen- and ozone-induced programmed cell death (Overmyer et al., 2003), these results seemed especially interesting in the light o f our previous finding regarding C A -M K K 9 ' s role in ethylene biosynthesis (Chapter 2). In addition, I was able to detect a significant over-representation of A B R E - l i k e elements in the promoter region of the down-regulated gene set, which suggested a link between A t M K K 9 and A B A and G A signaling, as well as A B F proteins as putative targets of M A P K s (Skriver et al., 1991; Choi et al., 2000) (Table 3.3, Figure 3.3). However, the gene expression patterns found by the microarrays in the CA-MKK9-af fec ted gene set could not be validated by Q R T -P C R (Figure 3.6, Figure 3.7), suggesting that the aforementioned analyses do not reflect real biological processes. Overall, therefore, the microarray experiment was not successful in detecting early downstream transcriptional events triggered by C A - M K K 9 , despite the scale and level of replication employed. 90 It is possible that the failure of the microarray experiment is due to flaws in its experimental design. The hybridizations were designed to obtain a direct comparison between a given D E X : C A - M K K 9 - F L A G line and a given E V line. This model was used under the assumption that a D E X : C A - M K K 9 - F L A G line and a E V line of comparable G V G level should behave similarly in every respect, apart from the CA-MKK9-media ted transcriptional events (Figure 3.1). Moreover, I assumed that the D E X treatment would have only minor effects on the E V lines. However, I found that the E V L10 line displayed marked mis-regulation of J A biosynthesis genes upon D E X induction (Figure 3.5), making it quite plausible that this line also shows mis-regulation of other genes. The results of the multi-factorial A N O V A support this hypothesis, as they demonstrated that a relatively high proportion of genes showed altered expression in the D E X : C A - M K K 9 -F L A G L12/ E V L10 pair compared to the other two combinations of transgenic lines (Appendix; Figure 5.1C, D). A re-analysis of the G V G levels in the E V L10 plants revealed much higher G V G levels than had been originally observed (data not shown), indicating that the use of this E V line was inappropriate in the first place. Moreover, the Q R T - P C R analysis of J A biosynthesis gene expression in another empty vector line, E V L l , showed that those genes were also down-regulated in this line, suggesting that a general pattern exists where D E X application influences the metabolism of plants in which the G V G transcription factor is being constitutively expressed (Figure 3.6). Taken together, these observations suggest that the behavior of pTA7002 E V lines could explain the high number of false-positive in the microarrays. Since I was comparing gene expression between D E X : C A - M K K 9 - F L A G plants at T=2 and T=0 in an indirect manner, by looking at the ratio between D E X : C A - M K K 9 - F L A G and E V lines at both 91 time-points, any artefactual gene expression in the E V lines is l ikely to have been falsely interpreted as a C A - M K K 9 - d r i v e n process. Moreover, indirect comparisons have less statistical power than direct ones. Therefore, true C A - M K K 9 - d r i v e n expression differential may have been missed because of insufficient statistical power in the design. A s we were most interested in differences between D E X . C A - M K K 9 - F L A G plants at T=2 and T=0, it would have been more appropriate to directly co-hybridize the aforementioned samples together, instead of running each against a E V line. Despite these challenges, I was able to identify and verify W R K Y 3 0 as a transcriptionally up-regulated gene responding to C A - M K K 9 activation (Figure 3.7). The W R K Y family of transcription factors is unique to the plant kingdom and contains 74 members (Ulker and Somssich, 2004). There is growing evidence that W R K Y transcription factors are downstream players in M A P K cascade. The promoter of W R K Y 2 9 was shown to be activated upon flagellin-dependent activation of the A t M E K K l - A t M K K 4 / 5 - A t M P K 3 / 6 cascade (Asai et al., 2002). Moreover, W R K Y 3 3 was recently identified as an in vitro substrate of A t M P K 4 , and knock-out mutants of both W R K Y 3 3 and A t M P K 4 showed the same SA-related phenotype (Andreasson et al., 2005). Tobacco W R K Y 1 was also recently shown to be a direct substrate for S IPK (the tobacco orthologue o f A t M P K 6 ) (Menke et al., 2005). There is only limited information available concerning the biological role of W R K Y 3 0 in Arabidopsis. It has been reported that the transcription of W R K Y 3 0 is activated by challenge to the plant tissue from both host and non-host types of pathogen, in a SA-independent manner (Kalde et al., 2003). It would therefore be 92 interesting to examine whether A t M K K 9 influences susceptibility to pathogen attack through its participation in the regulation of W R K Y 3 0 . 93 4. Future directions In an effort to gain information about the function of the M A P K K A t M K K 9 in Arabidopsis, I have characterized several aspects of D E X : C A - M K K 9 - F L A G transgenic plants, which express an inducible constitutively active version of A t M K K 9 . The results of this study confirm preliminary data from our laboratory suggesting roles for A t M K K 9 in programmed cell death (PCD) and in regulating ethylene biosynthesis. Moreover, I have established that A t M K K 9 could act as an upstream effector of the M A P K A t M P K 6 , and that its activity results in the up-regulation of a specific transcription factor, W R K Y 3 0 . To conclude this thesis, I wish to propose future directions for the presented research. 4.1. Role of AtMKK9 in ethylene biosynthesis in vivo The results presented in Chapter 2 strongly suggest a role for A t M K K 9 in modulating the AtMPK6-control of A C S stability, and thereby ethylene biosynthesis. However, as the evidence obtained so far relies solely on the ectopic expression of C A - M K K 9 , the function of A t M K K 9 in its endogenous form remains purely hypothetical. In order to gain insights into the biological context in which A t M K K 9 stimulated ethylene production, several approaches can be used. First, it is necessary to determine when and where A t M K K 9 is active. Publicly available gene expression databases, as well as R T -P C R surveys performed in the El l i s laboratory, suggest that the transcription of A t M K K 9 is increased under a number of conditions associated with ethylene production, including senescence, cold stress, and pathogen infection (data not shown). These can be used as a starting point to investigate how changes in the activity of A t M K K 9 might be correlated 94 with specific biological processes. Moreover, since a complex pattern of cross-talk exists between the hormones ethylene, salicylic acid (SA) and jasmonates, it would also be interesting to investigate whether the exogenous application of these hormones can affect A t M K K 9 activity. While the activity of M A P K s in response to different stresses can be easily assessed by subjecting proteins to in-gel kinase assays using the myelin basic protein substrate, or indirectly, as I did, using anti-phospho-ERK antibodies, measuring the activity of M A P K K s directly remains a technical challenge. To do so, one might have to immunoprecipitate either the endogenous M A P K K using a specific antibody, or an epitope-tagged version of the M A P K K , and test its ability to activate in vitro a known cognate M A P K substrate. For instance, one could immunoprecipitate A t M K K 9 from proteins extracted from tissue treated under an appropriate test condition, and measure its ability in vitro to activate recombinant A t M P K 6 , which I have already established to be an A t M K K 9 substrate. Once the biological contexts of A t M K K 9 activity have been defined, it would be possible to explore whether reducing the activity of A t M K K 9 , using transgenic plants expressing a D E X : A t M K K 9 - R N A i construct (which the El l i s laboratory has obtained), alters ethylene production in that particular context. 4.2. Investigation into the CA-MKK9-activated 42 kDa MAPK I have also found that the induction of C A - M K K 9 is associated with the activation of a 42 kDa M A P K in vivo (Chapter 2, Figure 2.7). A s I have done for A t M P K 6 , a combination of crude protein extract analysis with a specific antibody, in vitro kinase assay, and epistatic analysis using a knock-out mutant of the appropriate M A P K , would 95 allow us to characterize that second CA-MKK9-ac t iva ted M A P K cascade. A s mentioned in Chapter 2, A t M P K 4 stands out as the most likely candidate corresponding to that 42 kDa M A P K . Interestingly A t M P K 4 and A t M P K 6 are co-activated by a bacterial elicitor, harpin, independently of R O S production. Moreover, A t M P K 4 activity is correlated with harpin-induced P C D (Desikan et al., 2001). If we can obtain data showing that C A -M K K 9 indeed activates A t M P K 4 , it would be interesting to investigate whether knocking-out A t M P K 4 can suppress or reduce CA-MKK9-med ia t ed P C D . This might require the use of an inducible R N A i suppression approach, given the dwarf phenotype of the characterized transposon-tagged mpk4 mutant (Petersen et al., 2000), 4.3. Signaling events in CA-MKK9-mediated PCD The analysis of the DEX:CA-MKK9-FLAG/m/?A; r5 double-mutant revealed that the P C D phenotype induced by C A - M K K 9 does not require A t M P K 6 activity or increased ethylene production (Chapter 2, Figure 2.11). In order to investigate whether C A -MKK9-mediated P C D results from the mis-regulation of another pathway, we could assess the status of other major P C D regulators, S A and jasmonates. Quantitatively measuring these hormones by gas chromatography is possible, but can be technically challenging. A s a first step, or an alternative to directly measuring S A and jasmonate levels, the transcriptional activity of marker genes for these hormones could easily be assessed by R T - P C R . These markers include, for instance, PR1 for S A and PDF 1.2 for jasmonates (Rao et al., 2002). In addition, it has been previously reported that M A P K K activity can trigger the onset of P C D by up-regulating respiratory burst oxidase homologue genes (Yoshioka et al., 2003). 96 In Arabidopsis, the activity of AtrbohD and AtrbohF during pathogen attack modulates R O S production and subsequent P C D (Torres et al., 2002). It would therefore be interesting to assess the expression of the genes coding for these enzymes in induced D E X : C A - M K K 9 - F L A G plants. If we find that one or more Atrboh genes is indeed up-regulated following C A - M K K 9 induction, combining the D E X : C A - M K K 9 - F L A G construct with the corresponding Atrboh knock-out mutant would allow us to assess whether R B O H activity is required for CA-MKK9-media ted P C D . 4.4. Mechanisms of CA-MKK9 regulation of WRKY30 The microarray experiment and associated Q R T - P C R presented in Chapter 3 has revealed that C A - M K K 9 induces a significant increase in the expression of W R K Y 3 0 , a gene coding for a group III W R K Y transcription factor. Several reports have proposed a link between W R K Y transcription factors and M A P K signaling (Asai et al., 2002; Andreasson et al., 2005). I have established A t M P K 6 as a substrate of A t M K K 9 . In order to test whether the up-regulation of W R K Y 3 0 is due to A t M P K 6 activity, as opposed to another A t M K K 9 - d r i v e n pathway, we could use Q R T - P C R to compare W R K Y 3 0 transcript levels in D E X : C A - M K K 9 - F L A G plants and the D E X : C A - M K K 9 - F L A G / m / ? & f 5 double-mutant plants. Moreover, certain W R K Y transcription factors can bind their own promoter, thereby promoting their transcription in a positive feedback loop (Asai et al., 2002). Therefore, it is possible that W R K Y 3 0 is, in fact, a direct substrate of a CA-MKK9-ac t iva ted M A P K . We could test the ability of CA-MKK9-ac t iva ted M A P K s , such as A t M P K 6 , to phosphorylate W R K Y 3 0 by in vitro kinase assays using 3 2 P-labelled A T P . Moreover, i f 97 we find that W R K Y 3 0 is indeed an A t M K K 9 - M A P K substrate, we could assess changes in its ability to activate a reporter gene, such as P-glucuronidase, upon induction of C A -M K K 9 in transgenic plants. 4.5. Concluding remarks Overall, the results presented in this thesis represent essential groundwork that enables further exploration of the biological and molecular functions of A t M K K 9 . In particular, as shown by the proposed experiments of this concluding chapter, I believe that we now have exciting opportunities to investigate in detail the role of endogenous A t M K K 9 , particularly in the context of stress responses involving ethylene. The data and resources I have developed w i l l also make it possible to define, biochemically and genetically, the pattern of interactions between C A - M K K 9 and its downstream targets, and to place those interactions in their biological context. 98 5. Bibliography Ahlfors, R., Macioszek, V., Rudd, J., Brosche, M., Schlichting, R., Scheel, D., and Kangasjarvi, J. (2004). Stress hormone-independent activation and nuclear translocation of mitogen-activated protein kinases in Arabidopsis thaliana during ozone exposure. Plant J 40, 512-522. Alessi, D.R., Saito, Y., Campbell, D.G., Cohen, P., Sithanandam, G., Rapp, U., Ashworth, A., Marshall, C.J., and Cowley, S. (1994). Identification of the sites in M A P kinase kinase-1 phosphorylated by p74raf-l. Embo J 13, 1610-1619. Andreasson, E., Jenkins, T., Brodersen, P., Thorgrimsen, S., Petersen, N.H., Zhu, S., Qiu, J.L., Micheelsen, P., Rocher, A., Petersen, M., Newman, M.A., Bjorn Nielsen, H., Hirt, H., Somssich, I., Mattsson, O., and Mundy, J. (2005). The M A P kinase substrate M K S 1 is a regulator of plant defense responses. Embo J 24, 2579-2589. Aoyama, T., and Chua, N.H. (1997). A glucocorticoid-mediated transcriptional induction system in transgenic plants. Plant J 11, 605-612. Asai, T., Tena, G., Plotnikova, J., Willmann, M.R., Chiu, W.L., Gomez-Gomez, L., Boiler, T., Ausubel, F.M., and Sheen, J. (2002). M A P kinase signalling cascade in Arabidopsis innate immunity. Nature 415, 977-983. Beers, E.P., and McDowell, J.M. (2001). Regulation and execution of programmed cell death in response to pathogens, stress and developmental cues. Curr Opin Plant B i o l 4,561-567. Bogre, L., Ligterink, W., Meskiene, I., Barker, P.J., Heberle-Bors, E., Huskisson, N.S., and Hirt, H. (1997). Wounding induces the rapid and transient activation of a specific M A P kinase pathway. Plant Cel l 9, 75-83. Chae, H.S., and Kieber, J.J. (2005). Eto Brute? Role of A C S turnover in regulating ethylene biosynthesis. Trends Plant Sci 10, 291-296. Chen, K., Du, L., and Chen, Z. (2003). Sensitization of defense responses and activation of programmed cell death by a pathogen-induced receptor-like protein kinase in Arabidopsis. Plant M o l B i o l 53, 61-74. Cheong, Y.H., Moon, B.C., Kim, J.K., Kim, C.Y., Kim, M.C., Kim, I.H., Park, C.Y., Kim, J.C., Park, B.O., Koo, S.C., Yoon, H.W., Chung, W.S., Lim, CO., Lee, S.Y., and Cho, M.J. (2003). B W M K 1 , a rice mitogen-activated protein kinase, locates in the nucleus and mediates pathogenesis-related gene expression by activation of a transcription factor. Plant Physiol. 132, 1961-1972. Chichkova, N.V., Kim, S.H., Titova, E.S., Kalkum, M., Morozov, V.S., Rubtsov, Y.P., Kalinina, N.O., Taliansky, M.E., and Vartapetian, A.B. (2004). A plant 99 caspase-like protease activated during the hypersensitive response. Plant Cel l 16, 157-171. Choi, H., Hong, J., Ha, J., Kang, J., and Kim, S.Y. (2000). A B F s , a family of A B A -responsive element binding factors. J B i o l Chem 275,1723-1730. Cohen, P. (1997). The search for physiological substrates of M A P and S A P kinases in mammalian cells. Trends in Cel l Biology 7, 353-361. Danon, A . , and Gallois, P. (1998). U V - C radiation induces apoptotic-like changes in Arabidopsis thaliana. F E B S Lett 437, 131-136. de Jong, A . J . , Yakimova, E.T., Kapchina, V.M., and Woltering, E.J. (2002). A critical role for ethylene in hydrogen peroxide release during programmed cell death in tomato suspension cells. Planta 214, 537-545. Delledonne, M., Zeier, J., Marocco, A . , and Lamb, C. (2001). Signal interactions between nitric oxide and reactive oxygen intermediates in the plant hypersensitive disease resistance response. Proc Natl Acad Sci U S A 98, 13454-13459. Desikan, R., Hancock, J.T., Ichimura, K., Shinozaki, K., and Neill, S.J. (2001). Harpin induces activation of the Arabidopsis mitogen-activated protein kinases A t M P K 4 and A t M P K 6 . Plant Physiol 126, 1579-1587. Devoto, A . , and Turner, J.G. (2005), Jasmonate-regulated Arabidopsis stress signalling network. Physiologia Plantarum 123, 161-172. Droillard, M., Boudsocq, M., Barbier-Brygoo, H., and Lauriere, C. (2002). Different protein kinase families are activated by osmotic stresses in Arabidopsis thaliana cell suspensions. Involvement of the M A P kinases A t M P K 3 and A t M P K 6 . F E B S Lett 527, 43-50. Droillard, M.-J., Boudsocq, M., Barbier-Brygoo, H., and Lauriere, C. (2004). Involvement of M P K 4 in osmotic stress response pathways in cell suspensions and plantlets of Arabidopsis thaliana: activation by hypoosmolarity and negative role in hyperosmolarity tolerance. F E B S Letters 574, 42-48. Feilner, T., Hultschig, C , Lee, J., Meyer, S., Immink, R.G., Koenig, A . , Possling, A . , Seitz, H., Beveridge, A . , Scheel, D., Cahill, D.J., Lehrach, H., Kreutzberger, J., and Kersten, B. (2005). High throughput identification of potential Arabidopsis mitogen-activated protein kinases substrates. M o l Cel l Proteomics 4, 1558-1568. Gomi, K., Ogawa, D., Katou, S., Kamada, H., Nakajima, N., Saji, H., Soyano, T., Sasabe, M., Machida, Y., Mitsuhara, I., Ohashi, Y., and Seo, S. (2005). A mitogen-activated protein kinase N t M P K 4 activated by S I P K K is required for jasmonic acid signaling and involved in ozone tolerance v ia stomatal movement in tobacco. Plant Ce l l Physiol. 100 Greenberg, J.T. (1996). Programmed cell death: a way of life for plants. Proc Natl Acad Sci U S A 93, 12094-12097. Guo, H., and Ecker, J.R. (2004). The ethylene signaling pathway: new insights. Current Opinion in Plant Biology 7, 40-49. Heath, M.C. (2000). Hypersensitive response-related death. Plant M o l B i o l 44, 321-334. Hoeberichts, F .A. , and Woltering, E.J. (2003). Multiple mediators of plant programmed cell death: interplay of conserved cell death mechanisms and plant-specific regulators. Bioessays 25, 47-57. Huber, W., von Heydebreck, A . , Sultmann, H., Poustka, A . , and Vingron, M. (2002). Variance stabilization applied to microarray data calibration and to the quantification of differential expression. Bioinformatics 18, 96S-104. Ichimura, K., Mizoguchi, T., Yoshida, R., Yuasa, T., and Shinozaki, K. (2000). Various abiotic stresses rapidly activate Arabidopsis M A P kinases A T M P K 4 and A T M P K 6 . Plant J 24, 655-665. Ichimura, K., Mizoguchi, T., Irie, K., Morris, P., Giraudat, J., Matsumoto, K., and Shinozaki, K. (1998). Isolation of A T M E K K 1 (a M A P kinase kinase kinase)-interacting proteins and analysis of a M A P kinase cascade in Arabidopsis. Biochem Biophys Res Commun 253, 532-543. Ichimura, K., Shinozaki, K., Tena, G., Sheen, J., Henry, Y., Champion, A . , Kreis, M., Zhang, S., and Hirt, H. (2002). Mitogen-activated protein kinase cascades in plants: a new nomenclature. Trends in Plant Science 7, 301-308. Jonak, C , Okresz, L., Bogre, L., and Hirt, H. (2002). Complexity, cross talk and integration of plant M A P kinase signalling. Curr Opin Plant B i o l 5, 415-424. Jonak, C , Kiegerl, S., Ligterink, W., Barker, P.J., Huskisson, N.S., and Hirt, H. (1996). Stress signaling in plants: a mitogen-activated protein kinase pathway is activated by cold and drought. Proc Natl Acad Sci U S A 93, 11274-11279. Kalde, M., Barth, M., Somssich, I.E., and Lippok, B. (2003). Members of the Arabidopsis W R K Y group III transcription factors are part of different plant defense signaling pathways. Molecular Plant-Microbe Interactions 16, 295-305. Kang, H.G., Fang, Y., and Singh, K.B. (1999). A glucocorticoid-inducible transcription system causes severe growth defects in Arabidopsis and induces defense-related genes. Plant J 20,127-133. Kim, C.Y., Liu, Y., Thorne, E.T., Yang, H., Fukushige, H., Gassmann, W., Hildebrand, D., Sharp, R.E., and Zhang, S. (2003). Activation of a stress-responsive mitogen-activated protein kinase cascade induces the biosynthesis of ethylene in plants. Plant Cel l 15, 2707-2718. 101 Koukalova, B., Kovarik, A., Fajkus, J., and Siroky, J. (1997). Chromatin fragmentation associated with apoptotic changes in tobacco cells exposed to cold stress. F E B S Lett 414, 289-292. Kovtun, Y., Chiu, W.L., Tena, G., and Sheen, J. (2000). Functional analysis of oxidative stress-activated mitogen-activated protein kinase cascade in plants. Proc Natl Acad Sci U S A 97, 2940-2945. Kroj, T., Rudd, J.J., Nurnberger, T., Gabler, Y., Lee, J., and Scheel, D. (2003). Mitogen-activated protein kinases play an essential role in oxidative burst-independent expression of pathogenesis-related genes in parsley. J B i o l Chem 278, 2256-2264. Lam, E., Kato, N., and Lawton, M. (2001). Programmed cell death, mitochondria and the plant hypersensitive response. Nature 411, 848-853. Lamb, C , and Dixon, R.A. (1997). The oxidative burst in plant disease resistance. Annu Rev Plant Physiol Plant M o l B i o l 48, 251-275. Lee, J., Rudd, J.J., Macioszek, V.K., and Scheel, D. (2004). Dynamic changes in the localization of M A P K cascade components controlling pathogenesis-related (PR) gene expression during innate immunity in parsley. J B i o l Chem 279, 22440-22448. Levine, A., Tenhaken, R., Dixon, R., and Lamb, C. (1994). H 2 0 2 from the oxidative burst orchestrates the plant hypersensitive disease resistance response. Cel l 79, 583-593. Ligterink, W., Kroj, T., zur Nieden, U., Hirt, H., and Scheel, D. (1997). Receptor-mediated activation of a M A P kinase in pathogen defense o f plants. Science 276, 2054-2057. Liu, Y., and Zhang, S. (2004). Phosphorylation of 1-aminocyclopropane-l-carboxylic acid synthase by M P K 6 , a stress-responsive mitogen-activated protein kinase, induces ethylene biosynthesis in Arabidopsis. Plant Cel l 16, 3386-3399. Liu, Y., Jin, H., Yang, K.Y., Kim, C.Y., Baker, B., and Zhang, S. (2003). Interaction between two mitogen-activated protein kinases during tobacco defense signaling. Plant J 34, 149-160. Lohman, K.N., Gann, S., John, M.C., and Amaniso, R.M. (1994). Molecular analysis of natural leaf senescence in Arabidopsis thaliana. physiologia plantarum 92, 322-328. Lorrain, S., Vailleau, F., Balague, C , and Roby, D. (2003). Lesion mimic mutants: keys for deciphering cell death and defense pathways in plants? Trends Plant Sci 8, 263-271. 102 Lund, S.T., Stall, R.E., and Klee, H.J. (1998). Ethylene regulates the susceptible response to pathogen infection in tomato. Plant Cel l 10, 371-382. Mansour, S.J., Matten, W.T., Hermann, A.S., Candia, J.M., Rong, S., Fukasawa, K., Vande Woude, G.F., and Ahn, N.G. (1994). Transformation of mammalian cells by constitutively active M A P kinase kinase. Science 265, 966-970. McNellis, T.W., Mudgett, M.B., Li, K., Aoyama, T., Horvath, D., Chua, N.H., and Staskawicz, B.J. (1998). Glucocorticoid-inducible expression of a bacterial avirulence gene in transgenic Arabidopsis induces hypersensitive cell death. Plant J 14, 247-257. Menke, F.L., Kang, H.G., Chen, Z., Park, J.M., Kumar, D., and Klessig, D.F. (2005). Tobacco transcription factor W R K Y 1 is phosphorylated by the M A P kinase SIPK and mediates HR- l ike cell death in tobacco. M o l Plant Microbe Interact 18, 1027-1034. Menke, F.L.H., van Pelt, J.A., Pieterse, C.M.J., and Klessig, D.F. (2004). Silencing of the mitogen-activated protein kinase M P K 6 compromises disease resistance in Arabidopsis. Plant Ce l l 16, 897-907. Miles, G.P., Samuel, M.A., Zhang, Y., and Ellis, B.E. (2005). R N A interference-based ( R N A i ) suppression of A t M P K 6 , an Arabidopsis mitogen-activated protein kinase, results in hypersensitivity to ozone and misregulation of A t M P K 3 . Environ Pollut 138, 230-237. Miller, J.D., Arteca, R.N., and Pell, E.J. (1999). Senescence-associated gene expression during ozone-induced leaf senescence in Arabidopsis. Plant Physiol 120, 1015-1023. Mizoguchi, T., Ichimura, K., Irie, K., Morris, P., Giraudat, J., Matsumoto, K., and Shinozaki, K. (1998). Identification of a possible M A P kinase cascade in Arabidopsis thaliana based on pairwise yeast two-hybrid analysis and functional complementation tests of yeast mutants. F E B S Lett 437, 56-60. Moeder, W., Barry, C.S., Tauriainen, A.A., Betz, C , Tuomainen, J., Utriainen, M., Grierson, D., Sandermann, H., Langebartels, C , and Kangasjarvi, J. (2002). Ethylene synthesis regulated by biphasic induction of 1-aminocyclopropane-l-carboxylic acid synthase and 1-aminocyclopropane-l-carboxylic acid oxidase genes is required for hydrogen peroxide accumulation and cell death in ozone-exposed tomato. Plant Physiol 130, 1918-1926. Morris, P.C. (2001). M A P kinase signal transduction pathways in plants. N e w Phytologist 151, 67-89. Nishihama, R., Ishikawa, M., Araki, S., Soyano, T., Asada, T., and Machida, Y. (2001). The N P K 1 mitogen-activated protein kinase kinase kinase is a regulator of cell-plate formation in plant cytokinesis. Genes Dev 15, 352-363. 103 Nuhse, T.S., Peck, S.C., Hirt, H., and Boiler, T. (2000). Microbial elicitors induce activation and dual phosphorylation of the Arabidopsis thaliana M A P K 6. J B i o l Chem 275, 7521-7526. O'Connor, T.R., Dyreson, C , and Wyrick, J.J. (2005). Athena: a resource for rapid visualization and systematic analysis of Arabidopsis promoter sequences. Bioinformatics In Press. Ohler, U . , and Niemann, H. (2001). Identification and analysis of eukaryotic promoters: recent computational approaches. Trends in Genetics 17, 56-60. Orvar, B.L., McPherson, J., and Ellis, B.E. (1997). Pre-activating wounding response in tobacco prior to high-level ozone exposure prevents necrotic injury. Plant J 11, 203-212. Overmyer, K., Brosche, M., and Kangasjarvi, J. (2003). Reactive oxygen species and hormonal control of cell death. Trends Plant Sci 8, 335-342. Overmyer, K., Tuominen, H., Kettunen, R., Betz, C , Langebartels, C , Sandermann, H., Jr., and Kangasjarvi, J. (2000). Ozone-sensitive arabidopsis rcd l mutant reveals opposite roles for ethylene and jasmonate signaling pathways in regulating superoxide-dependent cell death. Plant Cel l 12, 1849-1862. Pedley, K.F., and Martin, G.B. (2004). Identification of M A P K s and their possible M A P K kinase activators involved in the Pto-mediated defense response of tomato. J B i o l Chem 279, 49229-49235. Pell, E.J., Schlagnhaufer, CD., and Arteca, R.N. (1997). Ozone induced oxidative stress: Mechanisms of action and reaction. Physiologia Plantarum 100, 264-273. Petersen, M., Brodersen, P., Naested, H., Andreasson, E., Lindhart, U., Johansen, B., Nielsen, H.B., Lacy, M., Austin, M.J., Parker, J.E., Sharma, S.B., Klessig, D.F., Martienssen, R., Mattsson, O., Jensen, A.B., and Mundy, J. (2000). Arabidopsis map kinase 4 negatively regulates systemic acquired resistance. Cel l 103, 1111-1120. Quirino, B.F., Noh, Y.S., Himelblau, E., and Amasino, R.M. (2000). Molecular aspects of leaf senescence. Trends Plant Sci 5, 278-282. Rao, M.V., and Davis, K.R. (1999). Ozone-induced cell death occurs via two distinct mechanisms in Arabidopsis: the role of salicylic acid. Plant J 17, 603-614. Rao, M.V., and Davis, K.R. (2001). The physiology of ozone induced cell death. Planta 213, 682-690. Rao, M.V., Lee, H.I., and Davis, K.R. (2002). Ozone-induced ethylene production is dependent on salicylic acid, and both salicylic acid and ethylene act in concert to regulate ozone-induced cell death. Plant J 32, 447-456. 104 Rao, M.V., Paliyath, G., Ormrod, D.P., Murr, D.P., and Watkins, C.B. (1997). Influence of salicylic acid on H 2 0 2 production, oxidative stress, and H 2 0 2 -metabolizing enzymes. Salicylic acid-mediated oxidative damage requires H 2 0 2 . Plant Physiol 115, 137-149. Rao, M.V., Lee, H., Creelman, R.A., Mullet, J.E., and Davis, K.R. (2000). Jasmonic acid signaling modulates ozone-induced hypersensitive cell death. Plant Cel l 12, 1633-1646. Ray, L., and Sturgill, T. (1988). Characterization of insulin-stimulated microtubule-associated protein kinase. Rapid isolation and stabilization of a novel serine/threonine kinase from 3T3-L1 cells. J. B i o l . Chem. 263, 12721-12727. Ren, D., Yang, H., and Zhang, S. (2002). Cel l death mediated by M A P K is associated with hydrogen peroxide production in Arabidopsis. J B i o l Chem 277, 559-565. Rentel, M.C, Lecourieux, D., Ouaked, F., Usher, S.L., Petersen, L., Okamoto, H., Knight, H., Peck, S.C., Grierson, C.S., Hirt, H., and Knight, M.R. (2004). OXI1 kinase is necessary for oxidative burst-mediated signalling in Arabidopsis 427,858-861. Romeis, T., Piedras, P., Zhang, S., Klessig, D.F., Hirt, H., and Jones, J.D. (1999). Rapid Avr9- and Cf-9 -dependent activation of M A P kinases in tobacco cell cultures and leaves: convergence of resistance gene, elicitor, wound, and salicylate responses. Plant Cel l 11, 273-287. Roslan, H.A., Salter, M.G., Wood, CD., White, M.R., Croft, K.P., Robson, F., Coupland, G., Doonan, J., Laufs, P., Tomsett, A.B., and Caddick, M.X. (2001). Characterization of the ethanol-inducible ale gene-expression system in Arabidopsis thaliana. Plant J 28, 225-235. Samuel, M.A., and Ellis, B.E. (2002). Double jeopardy: both overexpression and suppression of a redox-activated plant mitogen-activated protein kinase render tobacco plants ozone sensitive. Plant Ce l l 14, 2059-2069. Samuel, M.A., Miles, G.P., and Ellis, B.E. (2000). Ozone treatment rapidly activates M A P kinase signalling in plants. Plant J 22, 367-376. Samuel, M.A., Walia, A., Mansfield, S.D., and Ellis, B.E. (2005). Overexpression of S IPK in tobacco enhances ozone-induced ethylene formation and blocks ozone-induced S A accumulation. J Exp Bot 56, 2195-2201. Seo, S., Sano, H., and Ohashi, Y. (1999). Jasmonate-based wound signal transduction requires activation of W I P K , a tobacco mitogen-activated protein kinase. Plant Cel l 11,289-298. 105 Seo, S., Okamoto, M., Seto, H., Ishizuka, K., Sano, H., and Ohashi, Y. (1995). Tobacco M A P kinase: a possible mediator in wound signal transduction pathways. Science 270, 1988-1992. Skriver, K., Olsen, F.L., Rogers, J.C., and Mundy, J. (1991). cis-acting D N A elements responsive to gibberellin and its antagonist abscisic acid. Proc Natl Acad Sci U S A 88, 7266-7270. Solano, R., Stepanova, A., Chao, Q., and Ecker, J.R. (1998). Nuclear events in ethylene signaling: a transcriptional cascade mediated by E T H Y L E N E -I N S E N S I T I V E 3 and E T H Y L E N E - R E S P O N S E - F A C T O R 1 . Genes Dev. 12, 3703-3714. Soyano, T., Nishihama, R., Morikiyo, K., Ishikawa, M., and Machida, Y. (2003). N Q K 1 / N t M E K l is a M A P K K that acts in the N P K 1 M A P K K K - m e d i a t e d M A P K cascade and is required for plant cytokinesis. Genes Dev 17, 1055-1067. Spanu, P., Grosskopf, D.G., Felix, G., and Boiler, T. (1994). The Apparent Turnover of 1 -Aminocyclopropane-1-Carboxylate Synthase in Tomato Cells Is Regulated by Protein Phosphorylation and Dephosphorylation. Plant Physiol 106, 529-535. Suzuki, K., and Shinshi, H. (1995). Transient activation and tyrosine phosphorylation of a protein kinase in tobacco cells treated with a fungal elicitor. Plant Cel l 7, 639-647. Suzuki, K., Yano, A., and Shinshi, H. (1999). Slow and prolonged activation of the p47 protein kinase during hypersensitive cell death in a culture of tobacco cells. Plant Physiol 119, 1465-1472. Tanoue, T., and Nishida, E. (2003). Molecular recognitions in the M A P kinase cascades. Cel l Signal 15, 455-462. Tatsuki, M., and Mori, H. (2001). Phosphorylation of tomato 1-aminocyclopropane-1-carboxylic acid synthase, L E - A C S 2 , at the C-terminal region. J B i o l Chem 276, 28051-28057. Teige, M., Scheikl, E., Eulgem, T., Doczi, R., Ichimura, K., Shinozaki, K., DangI, J.L., and Hirt, H. (2004). The M K K 2 pathway mediates cold and salt stress signaling in Arabidopsis. Molecular Cel l 15,141-152. Tena, G., Asai, T., Chiu, W.L., and Sheen, J. (2001). Plant mitogen-activated protein kinase signaling cascades. Curr Opin Plant B i o l 4, 392-400. Torres, M.A., Dangl, J.L., and Jones, J.D. (2002). Arabidopsis gp91phox homologues AtrbohD and AtrbohF are required for accumulation of reactive oxygen intermediates in the plant defense response. Proc Natl Acad Sci U S A 99, 517-522. 106 Torres, M.A., Jones, J.D., and Dangl, J.L. (2005). Pathogen-induced, N A D P H oxidase-derived reactive oxygen intermediates suppress spread of cell death in Arabidopsis thaliana. Nat Genet 37,1130-1134. Treisman, R. (1996). Regulation of transcription by M A P kinase cascades. Curr Opin Cel l B i o l 8, 205-215. Tsuchisaka, A., and Theologis, A. (2004). Unique and overlapping expression patterns among the Arabidopsis 1-amino-cyclopropane-l-carboxylate synthase gene family members. Plant Physiol 136, 2982-3000. Tuomainen, J., Betz, C , Kangasjarvi, J., Ernst, D., Yin, Z.-H., Langebartels, C , and Sandermann, H. (1997). Ozone induction of ethylene emission in tomato plants: regulation by differential accumulation of transcripts for the biosynthetic enzymes. The Plant Journal 12, 1151-1162. Tuominen, H., Overmyer, K., Keinanen, M., Kollist, H., and Kangasjarvi, J. (2004). Mutual antagonism of ethylene and jasmonic acid regulates ozone-induced spreading cell death in Arabidopsis. Plant J 39, 59-69. Ulker, B., and Somssich, I.E. (2004). W R K Y transcription factors: from D N A binding towards biological function. Curr Opin Plant B i o l 7, 491-498. Viard, M.P., Martin, F., Pugin, A., Ricci, P., and Blein, J.P. (1994). Protein phosphorylation is induced in tobacco cells by the elicitor cryptogein. Plant Physiol 104, 1245-1249. Wang, K.L., Li, H., and Ecker, J.R. (2002). Ethylene biosynthesis and signaling networks. Plant Ce l l 14 Suppl, S131-151. Wang, K.L., Yoshida, H., Lurin, C , and Ecker, J.R. (2004). Regulation of ethylene gas biosynthesis by the Arabidopsis E T O l protein. Nature 428, 945-950. Weaver, L.M., Gan, S., Quirino, B., and Amasino, R.M. (1998). A comparison of the expression patterns of several senescence-associated genes in response to stress and hormone treatment. Plant M o l B i o l 37, 455-469. Widmann, C , Gibson, S., Japre, M.B., and Johnson, G.L. (1999). Mitogen-activated protein kinase: conservation of a three-kinase module from yeast to human. Physiol. Rev. 79, 143-180. Wurgler-Murphy, S.M., Maeda, T., Witten, E.A., and Saito, H. (1997). Regulation of the Saccharomyces cerevisiae HOG1 mitogen-activated protein kinase by the PTP2 and PTP3 protein tyrosine phosphatases. M o l Ce l l B i o l 17, 1289-1297. Wyrick, J.J., and Young, R.A. (2002). Deciphering gene expression regulatory networks. Current Opinion in Genetics & Development 12, 130-136. 107 Xing, T., Higgins, V.J., and Blumwald, E. (1997). Race-specific elicitors of Cladosporium fulvum promote translocation of cytosolic components of N A D P H oxidase to the plasma membrane of tomato cells. Plant Cel l 9, 249-259. Yang, K.Y., Liu, Y., and Zhang, S. (2001a). Activation of a mitogen-activated protein kinase pathway is involved in disease resistance in tobacco. Proc Natl Acad Sci U S A 98, 741-746. Yang, K.-Y., Liu, Y., and Zhang, S. (2001b). Activation of a mitogen-activated protein kinase pathway is involved in disease resistance in tobacco. P N A S 98, 741-746. Yap, Y.-K., Kodama, Y., Waller, F., Chung, K.M., Ueda, H., Nakamura, K., Oldsen, M., Yoda, H., Yamaguchi, Y., and Sano, H. (2005). Activation of a Novel Transcription Factor through Phosphorylation by W I P K , a Wound-Induced Mitogen-Activated Protein Kinase in Tobacco Plants. Plant Physiol. 139, 127-137. Yeung, K.Y., Medvedovic, M., and Bumgarner, R.E. (2004). From co-expression to co-regulation: how many microarray experiments do we need? Genome B i o l 5, R48. Yoshioka, H., Numata, N., Nakajima, K., Katou, S., Kawakita, K., Rowland, O., Jones, J.D., and Doke, N. (2003). Nicotiana benthamiana gp91phox homologs NbrbohA and NbrbohB participate in H 2 0 2 accumulation and resistance to Phytophthora infestans. Plant Cel l 15, 706-718. Yuasa, T., Ichimura, K., Mizoguchi, T., and Shinozaki, K. (2001). Oxidative stress activates A T M P K 6 , an Arabidopsis homologue of M A P kinase. Plant Cel l Physiol 42, 1012-1016. Zhang, S., and Klessig, D.F. (1997). Salicylic acid activates a 48-kD M A P kinase in tobacco. Plant Ce l l 9, 809-824. Zhang, S., and Klessig, D.F. (1998a). The tobacco wounding-activated mitogen-activated protein kinase is encoded by SIPK. Proc Natl Acad Sci U S A 95, 7225-7230. Zhang, S., and Klessig, D.F. (1998b). Resistance gene N-mediated de novo synthesis and activation of a tobacco mitogen-activated protein kinase by tobacco mosaic virus infection. Proc Natl Acad Sci U S A 95, 7433-7438. Zhang, S., and Klessig, D.F. (2001). M A P K cascades in plant defense signaling. Trends Plant Sci 6, 520-527. Zhang, S., and Liu, Y. (2001). Activation of salicylic acid-induced protein kinase, a mitogen-activated protein kinase, induces multiple defense responses in tobacco. Plant Cel l 13, 1877-1889. 108 Zhang, S., L i u , Y . , and Kless ig , D.F. (2000). Multiple levels of tobacco W I P K activation during the induction of cell death by fungal elicitins. Plant J 23, 339-347. Zuo, J., N i u , Q .W., and C h u a , N .H. (2000). Technical advance: A n estrogen receptor-based transactivator X V E mediates highly inducible gene expression in transgenic plants. Plant J 24, 265-273. 109 6. Appendix 6.1. Microarray data analysis The data from the microarray experiment described in Chapter 3 was subjected to a four-way A N O V A . This statistical procedure was entirely designed and conducted by a Genome B C statistical consultant, Rick White. The A N O V A tested the effect of four independent factors on gene expression. Table 6.1 shows the factors tested, as well as the levels within each factor. For the sake, of simplicity the variable numbers of biological replicates were ignored, and all replicates were considered as technical replicates. Table 6.1. Factorial design of the microarray experiment Factor Levels Transgene 1. CA-MKK9 2. EV Transgenic line 1. DEX;CA-MKK9-FLAG L2/ EV L3 2. DEX;CA-MKK9-FLAG L12/EV L10 3. DEX;CA-MKK9-FLAG L13/ EV L1 Dye 1. Cy3 2. Cy5 Time-point 1. T=0 2. T=2 This statistical test assessed importance of the different variables accounted for in the microarrays, i.e. the two C y - dyes, the three pairs of transgenic lines, the presence or absence of the C A - M K K 9 transgene and the two time-points, as well as possible interactions between these factors (Table 6.1). B y looking at the percentage of genes altered due to each of the factors, it was possible to estimate whether differentials recorded by the microarray experiment were more likely to be artefactual, or to reflect changes induced by the C A - M K K 9 transgene two hours after its induction. 110 The main effect of each factor was determined, as well as the interaction effect of all factors combinations. M a i n and interaction effects were quantified by an estimate value and their significance by a p-value and q-value. We found that, except for the dye factor, significant gene expression differences were mainly due to interactions between factors, rather to each factor taken independently. For instance, differences due to the transgenic lines could be detected when that factor was assessed in combination with the C A - M K K 9 transgene and the T=2 time-point. Figure 6.1 presents the most relevant results of the A N O V A , i.e. the dye effect, the transgenic line effect (with interactions with the C A -M K K 9 transgene and the T=2 time-point), the T=0 time-point (with an interaction with the C A - M K K 9 transgene) and the difference between T=2 and T=0 (with an interaction with the C A - M K K 9 transgene). The percentage above each graph estimates the number of differentially expressed genes. It represents the number of genes falling above the graph's blue bar,.i.e. above threshold of false discovery, and therefore encompasses both some true positives (p-value <0.05) as well as some false negatives (p-value>0.05) (Figure 6.1). Surprisingly, more than 50% of the observed difference can be attributed solely due to the type of dye used to label the two c D N A preparations, which shows the importance of 'dye balance' in microarray experiments (Figure 6.1 A ) . However, almost 45%> of the tested genes showed differences in expression due to factors other than the dye (Figure 6.IB). Although I assumed that C A - M K K 9 expression would induce a common set of transcriptional changes in each of the three D E X : C A - M K K 9 - F L A G transgenic lines (L2, L12 and L13), it was also possible that each line would display a somewhat different set of transcriptionally altered genes. Both at T=0 (Figure 6.1C) and T=2 (Figure 6.ID), D E X : C A - M K K 9 - F L A G L 1 3 / E V L l and D E X : C A - M K K 9 - F L A G 111 L 2 / E V L3 had only - 1 0 % genes whose differential expression could be attributed solely to the transgenic background, whereas D E X : C A - M K K 9 - F L A G L 1 2 / E V L10 displayed more than 20 % of the detected gene expressions altered because of that factor. This indicated that either D E X : C A - M K K 9 - F L A G L12 or E V L10 had significantly increased levels of background transcription variation. Ultimately, the multi-factorial A N O V A procedure permitted me to estimate the impact of C A - M K K 9 expression after two hours induction. A s shown by Figure 6. I E , expression of <7% of the responding genes was altered by C A - M K K 9 before the D E X treatment (T=0), indicating that the basal expression of the transgene was probably insignificant. B y contrast, expression of more than 35% of the genes was altered by C A - M K K 9 at T=2, as compared to T=0 (Figure 6.IF). Overall, this indicated that the T=2 time-point was sufficiently late to detect above-background changes. 112 53.87 % Dif fExp B t 44.63 % Di f fExp, 111 Dye parametric p-values Whole-model parametric p-values c s cu o ti c ' IT 27.17 % Di f fExp 10.6 % Diff Exp D C13/P1_CA-MKK9_T=0 p-values 13.51 % Di f fExp C13/P1 _CA-M KK9_T=2 p-values O 8 C «? CU C12/P10_CA-MKK9_T= p-values . Diff Exp C12/P10_CA-MKK9_T=2 p-values CU 3 s C2/P3_CA-MKK9_T=0 p-values 13.49 % Dif fExp C2/P3_CA-MKK9_T=2 p-values 3 8 0> 6.89 % Diff Exp CA-MKK9_T=0 p-values cu CT K cu "• 36.6? % Diff Exp CA-MKK9_(T=2)-(T=0) p-values Figure 6.1. Effect of dye type, transgenic lines and time-point factors on gene expression differentials. Normalized intensities were analyzed by ANOVA. A) Histogram of p-values for genes showing differential expression due to the dye factor B) Histogram of p-values for genes showing differential expression due to factors other than the dye 113 C) Histogram of p-values for genes showing differential expression due to CA-MKK9 in the L13/L1, L12/L10 and L2/L3 backgrounds at T=0 D) Histogram of p-values for genes showing differential expression due to the L13/L1, L12/L10 and L2/L3 backgrounds at 1=2 E) Histogram of p-values for genes showing differential expression due to CA-MKK9 at T=0 F) Histogram of p-values for genes showing differential expression between T=2 and T=0 due CA-MKK9. The blue line on each graph represents the threshold for false discoveries. The percentage is the estimated the number of differentially expressed genes. 6.2. CA-MKK9-acffected genes From the results obtained from the A N O V A , I filtered the genes that had a p-value smaller than 0.05 for the C A - M K K 9 effect between T=2 and T=0, and a fold-change greater than 2 between T=2 and T=0. Based on these criteria, I obtained a list of 191 genes, among which only 11 were up-regulated while 180 were down-regulated (Table 6.1). These "CA-MKK9-a f fec ted genes", were used for further analysis. Table 6.2. CA-MKK9-affected genes OligoJD AGI number Annotation Fold-change P-value A017580 01 At5g 14370 expressed protein 3.59 0.00000004 A022201 01 At3g60430 hypothetical protein -2.08 0.00004081 A011628 01 At3g21320 hypothetical protein -2.17 0.00004334 A016008 01 At5g24110 member of WRKY Transcription Factor; Group III 3.25 0.00005380 A002005 01 At1g48960 universal stress protein (USP) family protein 3.34 0.00005592 A024355 01 At3g45260 zinc finger (C2H2 type) family protein -2.21 0.00008371 A020698 01 At1g73500 member of MAP Kinase Kinase 3.62 0.00009900 A001407_01 At1g 14780 At1g 14790 expressed protein -2.23 0.00022182 A003480 01 At1g56660 expressed protein -2.38 0.00027813 A004610 01 At1g 12270 stress-inducible protein, putative -2.05 0.00041654 A019910 01 At5g28510 glycosyl hydrolase family 1 protein -2.02 0.00048739 A024327 01 AM g21730 kinesin-related protein (MKRP1) -2.38 0.00055993 A022148_01 At3g57830 leucine-rich repeat transmembrane protein kinase, putative, several receptor-like protein kinases -2.53 0.00057055 A002523_01 At1g76810 eukaryotic translation initiation factor 2 family protein -2.11 0.00065006 A007441 01 At2g 16650 expressed protein -2.37 0.00067833 A004372 01 At1g72520 lipoxygenase, putative -2.73 0.00070827 A022092 01 At2g 12700 hypothetical protein, -2.45 0.00072686 A024399_01 At3g 14270 phosphatidylinositol-4-phosphate 5-kinase family protein -2.32 0.00076290 A017268 01 At5g66210 calcium dependent protein kinase -2.07 0.00101666 114 OligoJD AG I number Annotation Fold-change P-value A019068_01 At5g 10570 basic helix-loop-helix (bHLH) transcription factor family protein 3.19 0.00117855 A011361 01 At3g08530 clathrin heavy chain, putative -2.66 0.00119467 A006305 01 At2g34930 disease resistance family protein -3.11 0.00121429 A024172 01 At5g17440 LUC7 N terminus domain-containing protein -2.02 0.00124474 A014851 01 At4g34260 expressed protein -2.04 0.00141990 A015780 01 At5g56420 F-box family protein 2.08 0.00153535 A024227 01 At1g18620 expressed protein -2.41 0.00185639 A025843_01 At1g53430 leucine-rich repeat family protein / protein kinase family protein -2.05 0.00190918 A024619_01 At3g25500 formin homology 2 domain-containing protein / FH2 domain-containing protein -2.08 0.00191832 A017280 01 At5g35980 expressed protein -2.03 0.00215544 A024527_01 At1g75750 GA-responsive GAST1 protein homolog regulated by BR and GA antagonistically. Possibly involved in cell elongation based on expression data 2.16 0.00222445 A020705 01 At5g46210 cullin, putative -2.15 0.00223091 A001349_01 At1g32060 phosphoribulokinase (PRK) / phosphopentokinase -2.70 0.00226019 A000484_01 At1g29870, tRNA synthetase class II (G, H, P and S) family "protein -2.53 0.00234548 A002332 01 At1g21170 expressed protein -2.48 0.00240396 A009035_01 At3g42670 SNF2 domain-containing protein / helicase domain-containing protein -2.73 0.00248381 A007751 01 At2g34080 cysteine proteinase, putative -2.53 0.00260725 A007745 01 At2g21440 RNA recognition motif (RRM)-containing protein -2.62 0.00296144 A010624 01 At3g14230 AP2 domain-containing protein RAP2.2 -2.00 0.00302789 A013282_01 At4g27500 interacts with H+-ATPase, and regulates its activity -2.36 0.00324237 A023593_01 At2g 16480 SWIB complex BAF60b domain-containing protein / plus-3 domain-containing protein -2.09 0.00362380 A019768_01 At1g70320 At1g55860 encodes a ubiquitin-protein ligase-like protein containing a HECT domain -2.39 0.00367355 A015694_01 At5g38710 proline oxidase, putative / osmotic stress-responsive proline dehydrogenase -2.02 0.00376199 A017036_01 At5g58290 26S proteasome AAA-ATPase subunit RPT3 (RPT3) -2.38 0.00381498 A006948 01 At2g07360 hypothetical protein -2.09 0.00406205 A008938 01 At3g57410 villin 3 (VLN3) -2.05 0.00406751 A006035 01 At2g22125 C2 domain-containing protein -2.38 0.00410772 A015257_01 At4g34450 coatomer gamma-2 subunit, putative / gamma-2 coat protein, putative / gamma-2 COP -2.28 0.00421507 A025445_01 At2g 17930 FAT domain-containing protein / phosphatidylinositol 3- and 4-kinase family protein -2.42 0.00424829 A007228_01 At2g45530 At2g45540 zinc finger (C3HC4-type RING finger) family protein -2.05 0.00426283 A003052 01 At1g 18040 cell division protein kinase, putative -3.17 0.00433710 A005267_01 At3g 13870 required for regulated cell expansion and normal root hair development. Encodes an evolutionarily conserved protein with putative GTP-binding motifs -2.53 0.00460739 A005086 01 At1g22930 T-complex protein 11 -2.39 0.00472917 A013882_01 At4g26760 microtubule associated protein (MAP65/ASE1) family protein -2.22 0.00483178 A006891_01 At2g42270 U5 small nuclear ribonucleoprotein helicase, putative -2.13 0.00484676 115 OligoJD AGI number Annotation Fold-change P-value A021782 01 At3g09500 60S ribosomal protein L35 (RPL35A) 2.06 0.00489962 A020511_01 At3g 11450 DNAJ heat shock N-terminal domain-containing protein / cell division protein-related -2.78 0.00492631 A010510_01 At3g04000 short-chain dehydrogenase/reductase (SDR) family protein -2.35 0.00494051 A007454_01 At2g29580 zinc finger (CCCH-type) family protein / RNA recognition motif (RRM)-containing protein -2.59 0.00505960 A011306_01 At3g12280 encodes a retinoblastoma homologue in Arabidopsis. This protein is involved in G1/S cell cycle transition in fungi, animals, and plants. -2.11 0.00510689 A023758_01 At1g20510 4-coumarate—CoA ligase family protein / 4-coumaroyl-CoA synthase family protein -2.04 0.00513408 A005126 01 At5g20990 homologous to E. coli MogA -2.32 0.00517218 A016205 01 At5g05940 expressed protein -2.07 0.00518187 A014038 01 At4g28910 expressed protein -2.26 0.00519377 A012702 01 At3g22860 member of elF3c - eukaryotic initiation factor 3c -2.74 0.00521121 A002217 01 At1g79820 hexose transporter, putative -2.06 0.00522019 A017468 01 At5g25280 serine-rich protein-related -2.54 0.00527032 A003654 01 At1g63700 member of MEKK subfamily (YODA) -2.07 0.00528430 A006384_01 At2g47410 teransducin family protein / WD-40 repeat family protein -2.07 0.00532912 A025177 01 At3g 13470 chaperonin, putative -2.78 0.00544900 A006954 01 At2g39340 SAC3/GANP family protein -2.08 0.00556886 A025264 01 At5g08680 ATP synthase beta chain, mitochondrial, putative -2.20 0.00567328 A011866_01 At3g47900 ubiquitin carboxyl-terminal hydrolase family protein -2.09 0 :00573976 A011688 01 At3g 19055 hypothetical protein -2.42 0.00576834 A011639 01 At3g13300 required for leaf development -2.59 0.00583690 A004615 01 At1g61690 tetratricopeptide repeat (TPR)-containing protein -2.12 0.00583922 A012703 01 At3g52250 myb family transcription factor -2.34 0.00588580 A019449 01 At1g22770 late flowering protein -2.85 0.00594350 A023293 01 At1g73020 expressed protein -2.69 0.00600035 A024622_01 At2g 18960 ATPase 1, plasma membrane-type, putative / proton pump 1, putative / proton-exporting ATPase, putative -2.23 0.00641275 A020036 01 At5g42220 ubiquitin family protein -2.01 0.00651547 A008112 01 At2g33435 RNA recognition motif (RRM)-containing protein -2.40 0.00666955 A011694 01 At3g 18490 aspartyl protease family protein -2.10 0.00668733 A000981 01 At1g35220 expressed protein -2.03 0.00679655 A016574_01 At5g05730 anthranilate synthase, alpha subunit, component 1-1 (ASA1) -2.17 0.00680408 A007934 01 At2g40540 putative potassium transporter AtKT2p (AtKT2) -2.31 0.00712780 A018889 01 At5g65180 expressed protein -2.23 0.00717650 A001355 01 At1g30795 hydroxyproline-rich glycoprotein family protein -2.35 0.00729700 A022994_01 At5g38670 F-box family protein, similar to SKP1 interacting partner 6 (Arabidopsis thaliana) -2.25 0.00737730 A016129 01 At5g45510 leucine-rich repeat family protein -2.04 0.00763283 A025923 01 At4g25860 oxysterol-binding family protein -3.18 0.00781124 A015007_01 At4g 17330 gene of unknown function expressed in seedlings, flower buds and stems -2.01 0.00783359 A001250_01 At1g03090 MCCA is the biotinylated subunit of the dimer MCCase, which is involved in leucine degradation. -2.19 0.00792468 A000336_01 At1g63830 proline-rich family protein, contains one predicted transmembrane domain 2.05 0.00798142 A000052 01 At1g04810 26S proteasome regulatory subunit -2.07 0.00798700 A005416 01 At5g27030 WD-40 repeat family protein -2.32 0.00817355 A010358 01 transcription activation domain-interacting -2.13 0.00833219 116 OligoJD AGI number Annotation Fold-change P-value At3g21480 protein-related A019846 01 At2g38440 putative WAVE homolog -2.37 0.00840097 A017482 01 At5g65520 expressed protein -2.20 0.00840525 A002163_01 At1g 13980 homologous to Sec7p and YEC2 from yeast. Involved in the specification of apical-basal pattern formation. -2.11 0.00844582 A020665 01 At4g32330 expressed protein -2.13 0.00844932 A016135 01 At5g24710 WD-40 repeat family protein -2.01 0.00883518 A024969 01 At4g08410 proline-rich extensin-like family protein -2.08 0.00907410 A016311 01 At5g42100 glycosyl hydrolase family 17 protein -2.69 0.00933898 A001896 01 At1g31730 epsilon-adaptin, putative -2.37 0.00944185 A024871 01 At5g65770 nuclear matrix constituent protein-related -2.67 0.00963027 A001286_01 At1g31780 conserved oligomeric Golgi complex component-related / COG complex component-related -2.03 0.00985685 A000214 01 At1g79280 expressed protein -2.01 0.00988185 A018082 01 At5g54230 putative transcription factor (MYB49) -2.30 0.00991258 A017224 01 At5g38370 hypothetical protein -2.92 0.00992506 A008949 01 At3g26560 ATP-dependent RNA helicase -2.19 0.00993022 A020631 01 At5g54720 ankyrin repeat family protein -2.17 0.01003557 A005346 01 At1g 10290 dynamin-like protein 6 (ADL6) -2.23 0.01007926 A013832 01 At4g29060 elongation factor Ts family protein -2.13 0.01117506 A003446 01 At1g80070 -2.32 0.01121576 A002003 01 At1g21610 wound-responsive family protein -3.63 0.01121903 A013575 01 At4g 16660 heat shock protein 70, putative / HSP70, putative -2.69 0.01156464 A002462 01 At1g06560 NOL1/NOP2/sun family protein -2.01 0.01182247 A013433 01 At4g01530 hypothetical protein, -4.52 0.01216392 A006925_01 At2g34210 KOW domain-containing transcription factor family protein -2.04 0.01222729 A011741_01 At3g23430 mutant is deficient in the transfer of phosphate from root epidermal and cortical cells to the xylem -2.02 0.01231944 A003159 01 At1g50770 hypothetical protein -2.10 0.01256179 A007459 01 At2g44480 glycosyl hydrolase family 1 protein -2.16 0.01319754 A006978_01 At2g24050 MIF4G domain-containing protein / MA3 domain-containing protein -2.36 0.01324092 A021895_01 At5g21274 encodes a calmodulin isoform. Expressed in leaves. 2.26 0.01356201 A023935 01 At1g60030 xanthine/uracil permease family protein -2.18 0.01360638 A025950 01 At3g54580 proline-rich extensin-like family protein -2.03 0.01375120 A001320 01 At1g61210 expressed protein -2.84 0.01385491 A023141 01 At3g52870 calmodulin-binding family protein -2.11 0.01416477 A005998 01 At1g29790 expressed protein -2.03 0.01446589 A018921_01 At5g49920 octicosapeptide/Phox/Bemlp (PB1) domain-containing protein -2.85 0.01497564 A021021_01 At5g42020 At5g28540 luminal binding protein (BiP) -2.93 0.01521725 A019610 01 At5g02220 expressed protein -2.10 0.01624952 A025517_01 At1g34410 At1g35520 At1g35540 At1g43950 At1g35240 At1g34310 At1g34390 transcriptional factor B3 family protein / auxin-responsive factor AUX/IAA-related -2.24 0.01658463 A004668 01 At1g 14970 expressed protein -2.16 0.01692409 A012453_01 At3g23670 phragmoplast-associated kinesin-related protein, putative -2.16 0.01705467 A011166 01 At3g45450 Clp amino terminal domain-containing protein -2.11 0.01711266 117 OligoJD AGI number Annotation Fold-change P-value A014080 01 At4g00450 mutant has dwarf; late flowering phenotypes -2.64 0.01730766 A014743 01 At4g20850 subtilase family protein -2.25 0.01746549 A018414 01 At5g09840 expressed protein -2.30 0.01754231 A008021 01 At2g36910 member of MDR subfamily -2.21 0.01762966 A002117 01 At1g79350 DNA-binding protein, putative -2.12 0.01810830 A001296_01 At1g48850 At1g48840 chorismate synthase, putative / 5-enolpyruvylshikimate-3-phosphate phospholyase, putative -2.16 0.01816726 A000710 01 At1g48090 C2 domain-containing protein -2.06 0.01836100 A001238 01 At1g 16270 protein kinase family protein -2.23 0.01844679 A008647_01 At2g40680 At5g52065 hypothetical protein -2.15 0.01890859 A003023 01 At1g62740 stress-inducible protein, putative -2.17 0.01897061 A020650 01 At1g 15340 methyl-CpG-binding domain-containing protein -2.11 0.01925895 A000449 01 At1g 15770 expressed protein -2.37 0.01944278 A025319 01 At1g05660 polygalacturonase, putative / pectinase, putative -2.30 0.01955440 A024330 01 At3g09840 member of AAA-type ATPases -2.16 0.01958703 A003280_01 At1g28060 small nuclear ribonucleoprotein family protein / snRNP family protein -2.41 0.02082370 A002139_01 At1g09770 member of MYB3R- and R2R3- type MYB-encoding genes -2.25 0.02108034 A002953 01 At1g15130 hydroxyproline-rich glycoprotein family protein -2.22 0.02302878 A012158 01 At3g01370 expressed protein -2.08 0.02307892 A008463 01 At2g 16050 DC1 domain-containing protein -2.25 0.02326059 A017880_01 At5g67470 formin homology 2 domain-containing protein / FH2 domain-containing protein -2.05 0.02346571 A020961 01 At1g21630 calcium-binding EF hand family protein -2.78 0.02356713 A019880_01 At2g42600 encodes one of four Arabidopsis phosphoenolpyruvate carboxylase proteins. -2.01 0.02432634 A014493 01 At4g32410 cellulose synthase catalytic subunit -2.10 0.02457772 A025792 01 At1g 17420 Lipoxygenase -2.09 0.02519402 A015984 01 At5g40450 expressed protein -2.47 0.02534816 A020779 01 At5g42530 expressed protein 2.24 0.02549915 A013669_01 At4g24190 encodes an ortholog of GRP94, an ER-resident HSP90-like protein and is involved in regulation of meristem size and organization. Single and double mutant analyses suggest that SHD may be required for the correct folding and/or complex formation of CLV pro -2.05 0.02582019 A023370 01 At2g47090 zinc finger (C2H2 type) family protein -2.06 0.02634684 A019158_01 At5g 15720 GDSL-motif lipase/hydrolase family protein, similar to family II lipase EXL3 and EXL2 -2.05 0.02634738 A008967 01 At3g54670 Cohesion -2.20 0.02642654 A017561 01 At5g01840 ovate family protein -2.10 0.02734042 A020581_01 At1g06670 nuclear DEIH-box helicase (NIH) encoding a putative RNA and/or DNA helicase homologous to a group of nucleic acid helicases from the DEAD/H family -2.25 0.02735415 A025862 01 At1g79920 heat shock protein 70, putative / HSP70, putative -2.20 0.02756319 A001141 01 At1g06950 chloroplast inner envelope protein-related -2.22 0.02799841 A006250 01 At2g30690 expressed protein -2.10 0.02946237 A007327_01 At2g26080 glycine dehydrogenase (decarboxylating), putative / glycine decarboxylase, putative / glycine cleavage system P-protein, putative -2.12 0.02952559 A016937 01 At5g64570 beta-xylosidase -2.35 0.03045088 A014105_01 At4g26690 glycerophosphoryl diester phosphodiesterase family protein -2.17 0.03049098 A024149 01 At2g34910 expressed protein -2.99 0.03110226 118 OligoJD AGI number Annotation Fold-change P-value A002067_01 At1g56560 beta-fructofuranosidase, putative / invertase, putative / saccharase, putative / beta-fructosidase, putative -2.07 0.03150710 A002569 01 At1g70220 hypothetical protein -3.32 0.03158312 A018932_01 At5g04140 encodes a gene whose sequence is similar to ferredoxin dependent glutamate synthase (Fd-GOGAT). Expression in leaves is induced by light and sucrose. Proposed to be involved in photorespiration and nitrogen assimilation -2.05 0.03427149 A022473 01 At4g 12770 auxilin-related -2.03 0.03490292 A016210_01 At5g 14780 NAD-dependent formate dehydrogenase 1B (FDH1B) -2.13 0.03499392 A001488 01 At1g62750 elongation factor Tu family protein -2.10 0.03579065 A025933 01 At1g59610 dynamin-like protein, putative (ADL3) -2.27 0.03632609 A020824_01 At2g48080 oxidoreductase, 20G-Fe(ll) oxygenase family protein -2.03 0.03641010 A010371 01 At3g14010 hydroxyproline-rich glycoprotein family protein -2.05 0.03670238 A018064_01 At5g50920 CIpC mRNA, nuclear gene encoding chloroplast protein -2.12 0.04055739 A016207_01 At5g 13630 encodes magnesium chelatase involved in plastid-to-nucleus signal transduction -2.36 0.04093862 A005168_01 At1g53540 17.6 kDa class I small heat shock protein (HSP17.6C-CI) -2.08 0.04261357 A009087_01 At3g45140 chloroplast lipoxygenase required for wound-induced jasmonic acid accumulation in Arabidopsis -2.80 0.04374736 A013257 01 At4g35800 RNA polymerase II large subunit -2.70 0.04459444 A025632_01 At3g61830 transcriptional factor B3 family protein / auxin-responsive factor AUX/IAA-related -2.04 0.04571103 119 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0092459/manifest

Comment

Related Items