Mecanismos moleculares y celulares que regulan la función del factor transcripcional CRzA en respuesta al estrés ambiental

199 p.-67 fig.-6 tab.1. Introduction. Communication with environment is an essential function for living organisms. In the case of microorganisms they are in direct contact with ambient and are immediately influenced by alterations into it (Serrano, 1996). These organisms have developed homeostatic systems that are intra and extracellular mechanisms to attenuate changes in environment. Abiotic stresses are caused by natural elements but not produced directly by other organisms. Among them, and more interesting for this work, are free mono or divalent cations, and pH, which is a direct effect of monovalent cation hydrogen (Serrano, 1996). Microorganisms are also classified accordingly to their capacity to tolerate cations, usually considered in their saline forms. Aspergillus nidulans, the model genetic organism used in this thesis, is able to tolerate moderate quantities of a wide range of mono and divalent cations, ranging from one molar concentration of sodium or potassium, half a molar of magnesium or lithium, and for others, like manganese or iron, only milli-molar concentrations are tolerated (Hagiwara et al. 2008; Spielvogel et al. 2008). Interestingly, calcium, an important cation because participates as secondary messenger in many intracellular regulatory and signalling pathways, is tolerated to a one molar concentration, condition that it is outstanding ((Clapham, 1995) and this work). The ability to tolerate calcium must rely on efficient mechanisms based on selective intracellular transport and storage to prevent undesirable effects of cytoplasmic free calcium (Berridge et al. 2000). An essential Ca2+-signaling mediator is calmodulin (CaM) (Clapham, 2007). Ca2+ binds and triggers structural rearrangements in CaM which allow the interaction of CaM with its effectors (Chin et al. 2000). A key target of CaM is the protein phosphatase calcineurin which is a heterodimer composed of a catalytic A subunit (CnA) and a regulatory B subunit (CnB) (Guerini, 1997). Among calcineurin substrates is the nuclear factor of activated T cells (NFAT) in higher eukaryotes and Crz transcription factors in fungi (Cyert, 2003; Feske et al. 2007). Although calcineurin is the only phosphatase involved in the dephosphorylation of these transcription factors, phosphorylation is carried out by different kinases. At least four different kinases are responsible for NFAT phosphorylation (Gsk3β, CK1, DYRK and PKA) while two kinases have been demonstrated to participate in Crz1p phosphorylation in Saccharomyces cerevisiae (Hrr25p and PKA) (Feske et al. 2007; Kafadar et al. 2003; Kafadar et al. 2004; Wu et al. 2007).The phosphorylation state of these proteins controls their nucleocitoplasmic trafficking and this determines their transcriptional activity. Many studies have focused on pathways regulated by calcineurin in fungal models due to the role of calcineurin signalling through Crz transcription factors in virulence and pathogenicity in both plants and animals (Rispail et al. 2009). 2. Objectives. A seminal work for this thesis was the discovery that in the filamentous fungus A. nidulans there are two regulatory systems mediating cation stress response (Spielvogel et al. 2008). One mediated by the zinc-finger transcription factor SltA and this is a novel pathway specific of filamentous fungi. SltA activity is required for generating a proper response to a wide range of mono and divalent cations, with the significant exception of calcium. The second pathway was predicted to be conserved among many organisms and is mediated by CrzA, a zinc-finger transcription factor. CrzA activity was associated with response to calcium and manganese but not to other cations. Tolerance to alkalinity probably requires the concerted regulatory activities of CrzA and SltA, in addition to PacC, the well-known regulator that mediates the ambient pH regulatory pathway (Penalva et al. 2002; Spielvogel et al. 2008). The main focus of this thesis was a functional study of CrzA and an in-deep analysis of the signalling pathway behind this transcription factor. To pursue this main objective three lines of research were proposed: 1. Analyse at both cellular and molecular levels the response mediated by CrzA to different abiotic stimuli, essentially calcium and alkalinity. Post-translational modifications and cell localization changes of this transcription factor will be determined. 2. Identify and characterise cis- and trans-acting elements in the activation and deactivation processes of CrzA function. 3. Determine the regulatory mechanisms of gene expression in the adaptation to ambient stress which are related to the transcriptional activity of CrzA.3. Results. The results section is divided into three chapters. In the first chapter diverse studies approach how signalling of CrzA is exerted at the molecular level. A deletion analysis allowed the identification of functional regions. As expected the DNA binding motif is essential for CrzA function but requires the N-terminal region for providing adequate response to calcium and alkalinity. Interestingly, the C-terminal region absent from yeast Crz proteins, also allows functionality of a truncated CrzA version when over expressed but not if expressed from crzA promoter ((Spielvogel et al. 2008) and this work).Calcineurin is required for CrzA function and over-expression of this transcription factor (TF) does not suppress null-calcineurin derived phenotypes such as the strong defect on colony radial growth or sensitivity to certain cations and alkalinity. CrzA is a phospho-protein and calcineurin activity influences its modification stages. Absence of calcineurin allows detection of a hyperphosphorylated form that shows a strong reduced mobility in denaturing polyacrylamide-gel electrophoresis. Calcineurin acts on CrzA through the functional N-terminal region present in this TF. Two calcineurin docking domains (CDD) have been studied in CrzA. CDD1, a putative double binding site for calcineurin, was identified because of similarity to CDD described in other Crz proteins (Roy et al. 2007). However, CDD1 is not functional and a CDD2 showing sufficient similarity to consensus calcineurin binding sites was detected downstream of CDD1 and analysed. CDD2 is a functional site and calcineurin binds to CrzA through this motif, however a remnant activity of calcineurin is still detectable on a protein lacking CDD2. The opposed activity of calcineurin on CrzA is likely to be exerted by several kinases. The role of two kinases on CrzA has been studied using partial or complete loss of function mutants. Mutants in casein kinase I homologue in A. nidulans, CkiA have been analysed for their effects on CrzA electrophoretic mobility. CkiA is required for proper modification of CrzA but its absence does not affect functionality of the transcription factor. In this thesis the homologue of glycogen synthase kinase 3β has been identified and characterised in A. nidulans, and named GskA. Absence of GskA activity greatly affects colonial growth, conidia germination and polarised growth of hyphae is also altered. When analysing posttraductional modifications in CrzA, the absence of GskA increases electrophoretic mobility of CrzA under non stressing conditions. However the functionality of CrzA is not completely abolished.In the second chapter the cellular signalling of CrzA is analysed. Nuclear entry of CrzA is a major regulatory step. Using fluorescent tagged versions of CrzA this work demonstrates that this TF is accumulated intranuclearly in response to a selected number of stresses. Divalent cations as calcium and manganese induced nuclear accumulation of CrzA but not the addition of large amounts of alkali cations such as sodium or potassium, and the divalent cation magnesium. Sensitivity of null crzA strains to alkalinity is a phenotype consequence of need of CrzA into nuclei when extracellular pH is alkaline, however medium acidification does not modify the cellular localisation of this TF. Calcineurin activity is responsible of this signalling and in the absence of this phosphatase CrzA remains cytoplasmic and excluded from nuclei at any ambient condition. The role of CDDs were studied by generating mutant versions lacking these sites, and, in agreement with functional analyses, CDD2 is required for nuclear migration of CrzA in response to both calcium and alkaline pH stimuli. Nuclear accumulation of CrzA is based on the activity of at least two nuclear localisation signals (NLS). Calcineurin must regulate both NLSs when N-terminal domain is present. An NLS, denoted as NLS1, has been identified by deletion analysis in the proximity of CrzA DNA binding domain between residues 487 and 492. A nuclear export sequence (NES) is also present in CrzA and responsible of nuclear exclusion in the absence of a proper stimulus as indicated above. NES locates in the N-terminal region probably in the proximity of NLS2, however the identity of both signals remains to be determined. Following the regulatory model of yeast Crz1p the role of both homologues for nuclear transport in A. nidulans have been studied by using the collection of strains carrying single null alleles in karyopherins (Cyert, 2003; Markina-Inarrairaegui et al. 2011). Apparently, CrzA nuclear import and export systems are not conserved with those described for Crz1p. Calcineurin must regulate the activity of the export sequence, most probably by inhibiting its activity because absence of calcineurin prevents nuclear entry but not export of CrzA. A strong phosphorylated state of CrzA must be responsible of maintaining NES activity. A reduction of CkiA activity does not prevent nuclear export of CrzA neither affects nuclear entry in response to stress, and this suggests that nucleocytoplasmic transport systems are not altered in CkiA mutants. Absence of GskA activity prevents nuclear accumulation of CrzA in response to alkalinity, but neither nuclear accumulation in response to calcium stress nor nuclear export in the absence of stress are affected. These results clearly state that a precise, GskA mediated, phosphorylation step is required for nuclear accumulation of CrzA at alkaline pH.4. Discussion. The starting point of this work was the analysis of CrzA using reverse genetic techniques. CrzA locus was identified by in silico searches and a null crzA strain was generated and characterised. This thesis is focused in the functional analysis of CrzA and the description of its signalling pathway. Three essential topics are discussed. I) Conservation and divergence of calcineurin/transcription factor signalling pathways. This work fully demonstrates that CrzA is a member of the calcineurin responsive zinc-finger family of transcription factors (Cyert, 2003). This is a fungal family but calcineurin also signals other transcription factors in higher eukaryotes such as NFATs (Feske et al. 2007). This study identifies similarities among signalling elements. Calcineurin is, perhaps, the main element in these pathways and this activity is always positive and targeted to the transcription factor.As a phosphatase, the TF must become dephosphorylated to become active. Opposed to calcineurin activity is the phosphorylating activity due to a number of kinases. Here, casein kinase I (CKI) and glycogen synthase kinase 3β (GSK3β) activities are shown to be related to CrzA signalling. Interestingly, in the model of CrzA phosphorylation CKI activity is in common with signalling of Crz1p in S. cerevisiae and GSK3β activity is with the NFAT pathway (Beals et al. 1997; Kafadar et al. 2003). Phosphatase and kinase activities modify the transcription factor and at least five different phosphorylation states are described for CrzA, which must be modulating the cellular location of this TF in response to a restricted number of stresses. Although much effort has been made to understand how phosphorylation state restricts nuclear or cytoplasmic localisations, a direct correlation has not been established between these two cellular processes. This fact is in agreement with recent studies in the other calcineurin mediated pathways. When compared CrzA and yeast Crz1p signalling pathways, nuclear import and export pathways are not conserved. CrzA, as Crz1p or NFAT, locates in the cytoplasm in the absence of stimulus and translocates in the nucleus in response to a signal (Cyert, 2003; Feske et al. 2007). A nuclear export signal is present in addition to, as in the other systems, alternative nuclear localisation signals. The nuclear export pathway acting on Crz1p depends on the Msn5p karyopherin, but its homologue, KapL, in A. nidulans seems not to be the nuclear carrier mediating nuclear export of CrzA. Similarly, the putative homologous pathway of Nmd5p, KapD in A. nidulans, is not responsible of nuclear transport of CrzA. It is possible that nucleocytoplasmic trafficking of CrzA might be more similar to those described for NFATs where general transport pathways are involved. Conservation between CrzA and Crz1p signalling pathways is evident, however differences found in CrzA pathway strongly support a broader study of these pathways in order to fully understand this important regulatory mechanism of transcription in other fungi.II) The transcription factor is a bottle neck of signalling and response. Cellular-localisation studies of the different signalling elements of CrzA state that the transcription factor is the only element with distinctive mobility in response to calcium or alkalinity. This modification of cellular localisation is mediated by changing the post-translational state of this transcription factor. The absence of conserved regions between Crz1p and CrzA, excluding the DNA binding domain, precludes the identification of functional regions by a mere comparison of sequences (Spielvogel et al. 2008). Functional analyses demonstrate the importance of the N-terminal part of CrzA for signalling where the presence of a NES and two NLSs must regulate its localisation. Possibly, calcineurin docks to a sequence located among such signals in CrzA. A CDD similar to consensus low affinity sites for S. cerevisiae calcineurin has been identified in CrzA, suggesting that similar but alternative molecular mechanisms differentiate both fungal phosphatases (Roy et al. 2007).III) Regulatory models for alkaline pH and cation stress responses based on CrzA modulation. Because CrzA is a zinc-finger transcription factor its final activity must be closely related to transcriptional regulation. CrzA is needed for response to alkalinity and calcium. Meanwhile a prediction of genes involved in calcium regulation is immediate, ie. calcium transporters or channels. Meanwhile, those genes mediating response to alkalinity are not of easy recognition. A role of chromatin in alkaline-pH response through CrzA has become a fact in this study. The negative effect of CrzA DNA binding domain when expressed alone suggested an unregulated function for this domain through its interference in gene expression. In our model, nuclear CrzA might bind to promoters of genes whose products are required for tolerance to alkalinity. Absence of CrzA reduces the expression of these genes and we observed that the absence of a fully functional RcoA protein, a known transcriptional repressor (Hicks et al. 2001), might compensate for the loss of this positive action of CrzA. However, our mutant analysis gives a broader view of RcoA activity and a new model for this chromatin remodelator is open in A. nidulans. Finally, identification of vacuolar calcium transporters as targets of calcium regulation has been successful. Vacuolar influx but not efflux must be transcriptionally regulated by calcium and CrzA. An interesting dual effect of calcium is observed. Low concentrations are acting in a positive manner, meanwhile high concentrations repress expression of AN5821 and vcxA.5. Conclusions. 1. CrzA is modified post-translationally through, at least, phosphorylation of residues located at the N-terminal region to the DNA binding domain. Five different modified states of this TF have been detected by electrophoretic techniques in protein extracts from mycelia grown in different media conditions and of diverse mutant backgrounds. 2. CrzA is a target for the phosphatase calcineurin and CnaA is the catalytic subunit in A. nidulans. CnaA recognizes a motif present in the TF which is necessary for the complete dephosphorylation activity. 3. A. nidulans casein kinase I, CkiA, and GskA, the A.nidulans Gsk3β homologue, are involved in CrzA post translational modifications. 4. High concentrations of calcium and manganese, and alkalinization of the extracellular medium are signals that induce nuclear accumulation of CrzA. CnaA activity is required for this transcriptional factor to respond to these stresses and GskA activity is required for proper nuclear accumulation at alkaline pH stress. 5. Calmodulin, calcineurin (catalytic subunit), and kinases CkiA and GskA maintain their cellular localizations in those environmental conditions that induce nuclear accumulation of CrzA. Therefore, the transcription factor is the only element in this pathway that modifies its cellular localization in response to stress. 6. CrzA displays two opposed cellular locations, nuclear accumulation in the presence of stress and nuclear exclusion under non-stressing conditions. At least one nuclear export signal and two nuclear localization signals are present in the primary structure of CrzA and this functional organization is conserved with Crz and NFAT transcription factors from yeast and higher eukaryotes, respectively. Nevertheless, nuclear transport systems are likely not conserved between S. cerevisiae and A. nidulans.Proyectos BFU(2009)-08701 y BFU(2012)-33142 Beca del Programa Junta para Ampliación de Estudios e Investigaciones Científicas (JAE) del CSICPeer reviewe