Stomata open in response to light allowing CO2 uptake for photosynthesis and they close in response to abiotic stress, such as drought, to prevent transpirational water loss from the plant. A pair of guard cells surrounds each stoma and stomatal movements depend on K+ fluxes across the guard cell plasma membrane. These fluxes are mediated by inward and outward rectifying K+ channels (K+in and K+out). The SNARE SYP121 was originally identified in association with ion channel regulation in guard cells. SNARE proteins mediate vesicle fusion and facilitate delivery of membrane proteins to target membranes. They are also linked to a variety of physiological responses. In particular, the plasma membrane SNARE SYP121 has been attributed a role in immune response and K+ nutrition. I have used the Arabdopsis loss-of-function mutant syp121 and uncovered a set of mutant phenotypes associated with impaired stomatal opening. In the syp121 mutant stomatal reopening was delayed and incomplete following Ca2+-induced closure, and increase in stomatal transpiration was slowed in the light. Incomplete reopening was rescued by complementation with wild-type SYP121 and was not observed in the syp122 mutant, lacking the homologous gene product. Guard cell K+ in current, necessary for K+ uptake during stomatal reopening, was reduced in syp121 mutant guard cells. Analysis of current gating characteristics suggested an impaired delivery of K+in channels to the plasma membrane, which was consistent with inhibition of stomatal reopening by the trafficking inhibitor Brefeldin A in wild-type plants. Impaired stomatal reopening in the syp121 mutant was phenomenologically similar to a Ca2+-encoded form of ‘programmed closure’ and my results suggest that endocytosis and delayed recycling of K+in channels may underly this phenomenon. Impaired stomatal function manifested in a conditional syp121 mutant growth phenotype dependent on high light and low humidity, characterised by reduced stomatal conductance and photosynthetic CO2 assimilation. My results suggested the necessity for SYP121-dependent K+in channel traffic during stomatal reopening. My results revealed a novel syp121 stomatal phenotype that was linked to K+in channel recycling in guard cells and had consequences for whole-plant water use and biomass production.

The global success of plants depends largely on their ability to perceive and respond to light, mainly in two regions of the electromagnetic spectrum. Phytochromes are light sensors for the red and far-red wavelengths of light while cryptochromes, phototropins and members of the ZTL/ADO family respond to blue and UV-A wavelengths of light. Phototropins are UV-A/blue-light receptor kinases found ubiquitously in plants from the unicellular green alga Chalmydomonas reinhardtii through bryophytes and pteridophytes up to angiosperms. The model plant Arabidopsis possesses two phototropins (phot1 and phot2) and is the subject of the work presented in this thesis. The general structure of the phototropin protein comprises a photosensory region at the N-terminal that contains two LOV (light, oxygen and voltage sensing) domains and a C-terminal kinase domain belonging to the large AGC family of protein kinases. The LOV domains form a covalent adduct with the chromophore flavin mononucleotide (FMN) in response to illumination with blue light which in turn leads to structural changes throughout the protein resulting in autophosphorylation of the N-terminal region by the kinase domain. Phototropins function redundantly to mediate a number of physiological responses in planta which serve to promote plant fitness and maximise photosynthetic potential. Phototropism, chloroplast accumulation, blue light-induced stomatal opening, leaf expansion and leaf movements can be induced through the activation of both phot1 and phot2 in response to different intensities of light, with phot1 being more light sensitive than phot2. In addition to the functionally redundant responses, phot1 alone is responsible for destabilisation of certain mRNA transcripts and the rapid inhibition of hypocotyl elongation when etiolated seedlings are transferred to blue light, while phot2 is solely responsible for the high light induced chloroplast avoidance response. While much is known about the mechanisms of light perception by the phototropins at the molecular level, and the responses mediated by them have been well described, little is known about their methods of signalling to induce these physiological responses upon photoactivation by blue light. Therefore, the aims of this study were to identify novel phot-interacting proteins and to investigate the modes of phot1 signalling by structure/function analyses in order to better understand the way phototropins elicit signal transduction to downstream components in order to bring about the responses described above. Initially, a yeast two-hybrid screen was carried out to try to identify immediate interacting partners for phot1. The results of the yeast two-hybrid screen are described in Chapter 3. One hundred and thirty yeast colonies containing putative phot1-interacting proteins were identified from the screen and preliminary characterisation of six of these proteins are described in this chapter. Two of the proteins investigated are members of the ADP-ribosylation family which is involved in the regulation of membrane trafficking. The ARF proteins identified show a blue-light-sensitive interaction with phot1 and also interact with phot2. These proteins are of interest given the subcellular movement of phototropins from the plasma membrane after exposure to blue light. The C-terminal kinase domain of phot1 was found to interact with p-glycoprotein 19 (PGP19), a protein involved in polar auxin transport. The interaction between these proteins is interesting because of the role auxin plays in phot1-mediated responses such as phototropism and leaf expansion, and preliminary characterisation of the interaction in vitro is shown in Chapter 3. The implications of a direct link between phototropins and the proteins involved in auxin transport are discussed. A further two proteins identified from the screen are members of the NPH3/RPT2-Like (NRL) family. RPT2 has already been identified as a phot1-interacting protein and identification of this protein increased confidence in the efficacy of the screen to identify genuine interacting proteins. A novel member of the NRL family, designated NPH3-L, was also identified from the screen. Chapter 4 describes the tissue specific and subcellular localisation of NPH3-L and contains results of preliminary investigations into the function of NPH3-L in planta. 14-3-3λ was identified from the screen using full-length phot1 as bait. A 14-3-3 protein has been shown previously to bind to autophosphorylated phototropin in Vicia faba (Kinoshita et al., 2003). Chapter 5 details the localisation of 14-3-3λ at tissue and subcellular levels and shows that 14-3-3λ binding to plant-derived phot1-GFP is both light dependent and induced by receptor autophosphorylation. Creation of GFP-14-3-3λ overexpressing lines in wild-type and phot1-5phot2-1 backgrounds allowed investigation into the roles that light and phototropins play in regulating the subcellular localisation of 14-3-3λ. It is shown that light-induced movement of 14-3-3λ at the plasma membrane is dependent on the presence of endogenous phototropins. Physiological implications of this interaction are discussed. Finally, in order to determine the modes of phototropin signalling, structure/function studies were carried out by expressing different regions of phot1 in a variety of Arabidopsis backgrounds. The results of the structure/function studies are described in Chapter 6. Known phot1-mediated responses were investigated in the transgenic plants to determine the effects of individual domains of phot1. Particular attention was paid to the role of receptor autophosphorylation in phot1-mediated responses to light. A transgenic line overexpressing the LOV2-kinase region of phot1 demonstrates that phot1 autophosphorylation is not the primary signalling event involved in phot1-mediated responses to light and shows that the truncated version of phot1 is sufficient to complement most phot1-mediated responses. This also shows that the LOV1 domain is dispensable and suggests phot1 may signal through phosphorylation of substrates. Comparisons are drawn between phot1 kinase overexpressing lines and inactive phot1 kinase overexpressing lines. Preliminary observations of a transgenic line overexpressing phot1 in a wild-type background indicate that overexpression of phot1 may alter polar auxin transport. Together these studies provide new insights into possible mechanisms of phot1 signalling and the function of major domains of phot1.

UV-B is an integral component of the daylight spectrum that regulates plant gene expression and development, but very little is known about how plants perceive UV-B. Although UV-B-induced damage and repair have been extensively investigated, the mechanisms by which UV-B is perceived as a signal, which mediates physiological and protective responses is not yet clearly understood neither in mammals, nor in higher plants. Low fluence rates of UV-B induce the expression of genes involved in UV-protective responses such as flavonoid biosynthesis and promote plant survival in UV-B. The aim of this study is to contribute to the elucidation of the signal transduction events that lead to the acclimation of plants in response to non-damaging levels of UV-B (< 3.5 μmol m-2 s-1). In particular, the characterisation of UVR8 (UV-RESISTANCE LOCUS 8), a UV-B specific signalling component, is carried out at the protein level. The function of UVR8 involves the orchestration of the expression of a range of genes mediating vital UV-protective responses, including those encoding light-regulated transcription factors HY5 and HYH, enzymes involved in the phenylpropanoid pathway, antioxidant and stress proteins (Brown et al., 2005). UVR8 shows 30% sequence identity to the human regulator of chromatin condensation (RCC1) but differs both in activity and function. The phenotype of uvr8 mutant plants is characterised by an increased susceptibility to UV-B and the lack of the UV-B-specific induction of genes involved in UV-protection, such as CHS (encoding the flavonoid biosynthetic enzyme chalcone synthase) and the transcription factor HY5. The UVR8-mediated regulation of transcription in response to UV-B seems to occur via the association of UVR8 with chromatin via histones in the promoter region of HY5 (Brown et al., 2005) and other genes involved in light signalling. In this study, further investigation of the mechanism by which UVR8 acts as a UV-B specific signalling component is performed by employing a number of approaches including: spatial, temporal protein analysis, subcellular localisation studies, structure-function analyses, and the yeast-two-hybrid assay for the identification of UVR8 interacting proteins. To study spatial, temporal and wavelength specific UVR8 protein abundance anti-UVR8 peptide antibodies were generated. Western blot analyses showed that UVR8 is ubiquitously expressed in all plant tissues from the very early stages of development and at every light treatment tested (dark, white light, UV-B). The subcellular localisation of UVR8 analysed by confocal fluorescence microscopy revealed that a fusion of UVR8 with green fluorescent protein (GFP) is localised in the cytoplasm and the nucleus of various plant tissues (leaf, hypocotyl, root, flower) and under various light fluence rates and qualities (white, red, UV-A, UV-B). Interestingly, a treatment of low fluence rates of UV-B led to an increase of GFP-UVR8 protein accumulation in the nucleus, which was confirmed by western blot analysis based on protein fractionation studies in wild-type plants. The wavelength specificity, the kinetics and the fluence-rate sensitivity of GFP-UVR8 nuclear accumulation suggest that this response is UV-B specific, rapid (10 min UV-B) and very sensitive to very low fluence rates of UV-B (0.1 μmol m-2 s-1). Protein synthesis does not seem to be involved in this process, as there is no change in the protein levels before and after a UV-B irradiation. To assess the importance of the presence of UVR8 in the nucleus and the cytoplasm of the plant cell, uvr8-1 transgenic plants were produced expressing either constitutively nuclear localised GFP-UVR8 fused to a nuclear localisation signal (NLS), or cytosolically retained GFP-UVR8 fused to a nuclear export signal (NES). Nuclear exclusion of NES-GFP-UVR8 fusion protein was sustained under most light conditions apart from UV-B, which induced nuclear import of the protein. This indicates that the mechanism involved in the nuclear accumulation of UVR8 can overcome an export signal either by masking it or by simply superseding it. Furthermore, the NES-GFP-UVR8 construct was functional after UV-B treatment, since it rescued the mutant uvr8 phenotype. None of the inhibitor treatments tested (staurosporine, cycloheximide, cantharidin) was successful in blocking the UV-B induced nuclear import of NES-GFP-UVR8, although they impaired the UVR8 regulated induction of CHS expression. Thus, no evidence is presented for a specific protein modification, which could control this response. Constitutive nuclear localisation of NLS-GFP-UVR8 had no effect on the function of the protein according to complementation analyses. Furthermore, no change in localisation, fluorescence intensity or protein abundance was observed in response to white light or after a UV-B irradiation. These results indicate that the constitutive nuclear localisation of UVR8 is not sufficient for constitutive activation of UVR8 regulated gene expression and that a UV-B stimulus is still necessary to trigger these responses. Unfortunately, based on the current data it cannot be concluded whether the UV-B signal perception occurs in the nucleus or in the cytosol of the plant cell. To investigate the structure-function relationship within the UVR8 protein, deletion analyses followed by complementation studies in transgenic plants were performed. More specifically, deletion of the first 23 amino acids at the N-terminus of UVR8 impaired its nuclear accumulation in response to UV-B. Deletion of a 27 amino acid region near the C-terminus had no effect on the UV-B dependent re-localisation of the protein, but abolished UVR8 regulated gene expression. In addition, a highly basic sequence at the extreme C-terminal of UVR8, resembling a putative monopartite nuclear localisation signal, was deleted. Subcellular localisation and complementation analyses suggest that this sequence does not serve as a nuclear localisation signal, it is not involved in the UV-B induced nuclear accumulation and its absence does not affect UVR8 protein function. Chromatin immunoprecipitation assays show that none of the regions deleted is required for chromatin association and none of the deletions affects subcellular localisation in white light. In order to identify interacting partners for UVR8, the yeast-two hybrid system was used. Unfortunately no interacting proteins have been identified, neither from a screen, nor by directed-interaction studies. A different approach could be employed in the future involving size exclusion chromatography of protein extracts from plants in order to establish whether UVR8 functions as part of a complex in vivo.

Stomatins belong to the band-7 (or SPFH domain) family (short for Stomatin, Prohibitin, Flotillin HflC/K) of diverse membrane proteins. This protein family is evolutionary conserved with members found in all sequenced eukaryotes and in most prokaryotes. Band-7 family proteins have the ability to oligomerise and generally aid in the assembly and regulation of large membrane-bound protein complexes. In animals, stomatins have been demonstrated to regulate ion channels by direct protein interactions. Additionally, they localise to membrane microdomains where they actively contribute to their assembly by binding sterols, and they also associate with the actin cytoskeleton. The Arabidopsis genome encodes for two structurally similar stomatin-like proteins that are functionally completely unknown yet. They will be referred to as AtSlp1 (for Arabidopsis thaliana stomatin-like protein) and AtSlp2. The aim of this thesis was to provide a detailed characterisation of these two genes on a molecular and functional level. Both proteins are expressed ubiquitously throughout plant development, but they accumulate at particularly high levels in pollen and other metabolically active cells. Phylogenetic analysis reveals that AtSlps are homologous to stomatin-like proteins of type 2. Amongst these, the human stomatin-like protein 2 (HsSlp2) is localised to mitochondria where it participates in large membrane-bound protein complexes and is also involved in the proliferation of cancer cells. Evidence is provided here that demonstrates mitochondrial localisation of both Arabidopsis Slp proteins in vitro and in vivo. On a functional level, mitochondria from an slp1 knockout mutant plant have a decreased mitochondrial membrane potential and increased oxygen consumption rates. This is interpreted as a defect in coupling efficiency and an impairment of the mitochondrial inner membrane integrity. This defect results in a variety of other growth phenotypes that are related to metabolically active tissues and cell types. Knockout plants are delayed in overall growth of shoots and roots and have decreased seed germination rates. Additionally, these plants are less resistant to conditions of high salinity and are less fertile. Overexpression of a protein acting as a putative dominant-negative Slp fragment results in plants with a dwarf phenotype and early onset of leaf senescence. This phenotype correlates with increased levels of reactive oxygen species and altered organelle ultrastructure. Guard cells from these plants in particular have enlarged chloroplasts and are impaired in transpirational control. It is concluded that also in plants, stomatins act together with other band-7 family proteins as parts of large protein complexes that have regulatory roles important for development and stress responses. Their main role is probably to provide membrane scaffolds that affect mitochondrial function and morphology during cell division and in situations of mitochondrial stress.