Steroids xxx (2013) xxx–xxx
Contents lists available at ScienceDirect
Steroids journal homepage: www.elsevier.com/locate/steroids
Rapid mineralocorticoid receptor trafficking M. Gekle, M. Bretschneider, S. Meinel, S. Ruhs, C. Grossmann ⇑ Julius Bernstein Institute of Physiology, Martin Luther University Halle-Wittenberg, Germany
a r t i c l e
i n f o
Article history: Available online xxxx Keywords: Mineralocorticoid receptor Trafficking Modulation Genomic signaling
a b s t r a c t The mineralocorticoid receptor (MR) is a ligand-dependent transcription factor that physiologically regulates water-electrolyte homeostasis and controls blood pressure. The MR can also elicit inflammatory and remodeling processes in the cardiovascular system and the kidneys, which require the presence of additional pathological factors like for example nitrosative stress. However, the underlying molecular mechanism(s) for pathophysiological MR effects remain(s) elusive. The inactive MR is located in the cytosol associated with chaperone molecules including HSP90. After ligand binding, the MR monomer rapidly translocates into the nucleus while still being associated to HSP90 and after dissociation from HSP90 binds to hormone-response-elements called glucocorticoid response elements (GREs) as a dimer. There are indications that rapid MR trafficking is modulated in the presence of high salt, oxidative or nitrosative stress, hypothetically by induction or posttranslational modifications. Additionally, glucocorticoids and the enzyme 11beta hydroxysteroid dehydrogenase may also influence MR activation. Because MR trafficking and its modulation by micro-milieu factors influence MR cellular localization, it is not only relevant for genomic but also for nongenomic MR effects. Ó 2013 Elsevier Inc. All rights reserved.
1. MR enigma The mineralocorticoid receptor (MR) with its endogenous ligand aldosterone is one of the main effectors in the renin-angiotensin-aldosterone-system (RAAS) and has a pivotal role in water-electrolyte homeostasis and regulation of blood pressure. It belongs to the steroid receptor superfamily that consists of the progesterone, the estrogen, the androgen and the glucocorticoid receptor. Steroid receptors possess a common structure comprising the domains A–F. The N-terminal A/B domain is the most variable among the receptors and is responsible for cofactor binding. The C domain of the MR is the DNA binding domain and possesses a 94% amino acid identity to the DNA binding domain of its closest relative, the glucocorticoid receptor (GR). After a short hinge region comes the C-terminal ligand binding domain that is also involved in dimerization. Of the steroid receptors, the mineralocorticoid receptor has been the least appreciated for a long time because of the more obvious clinical implications in for example cancer and immunological disease of its relatives. Consequently, many of the basic molecular observations concerning signaling and trafficking of the MR have been deduced from other steroid receptors, regardless of possible differences between them. The lack of interest changed after the importance of the MR for pathological ⇑ Corresponding author. Address: Julius-Bernstein-Institut für Physiologie, Universität Halle-Wittenberg, Magdeburger Strasse 6, 06097 Halle/Saale, Germany. Tel.: +49 345 557 1886; fax: +49 345 557 4019. E-mail address: [email protected] (C. Grossmann).
changes in the cardiovascular system and the kidneys became apparent. In two pivotal clinical studies that were followed by many others, the beneficial effect of MR antagonists like spironolactone and eplerenone for patients with cardiovascular disease was proven; however, without understanding the underlying mechanisms [1–3]. Since then, MR activation has been shown to be involved in different pathophysiological effects in the reno-cardiovascular system including endothelial dysfunction, inflammation, hypertrophy and fibrosis in both clinical studies and animal experiments [4–7]. It is well known, that the MR functions as a ligand-dependent transcription factor at hormone response elements called glucocorticoid response elements (GRE) that it shares with the GR. However, the GR acts in an anti-inflammatory and immunosuppressive way on the cardiovascular system, suggesting additional signaling mechanisms. The trigger that causes the MR to turn from a receptor regulating water-electrolyte homeostasis and not causing any harm into a receptor mediating pathological effects in the cardiovascular system is also an enigma. Of note, the MR needs to be inadequately activated to confer pathological effects as can be judged by the positive effects of MR antagonists. One way to achieve this is by having inappropriately high aldosterone levels in relation to salt status in an individual. Although such a scenario is likely in case of hyperaldosteronism caused by adrenal adenoma or hyperplasia, this does not seem to apply for the majority of patients benefitting from MR antagonists as in the above mentioned clinical studies, where aldosterone levels and salt status of participants were unremarkable. In animal studies, it is striking that aldosterone application only leads to
0039-128X/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.steroids.2013.10.016
Please cite this article in press as: Gekle M et al. Rapid mineralocorticoid receptor trafficking. Steroids (2013), http://dx.doi.org/10.1016/ j.steroids.2013.10.016
2
M. Gekle et al. / Steroids xxx (2013) xxx–xxx
pathological changes in the presence of additional permissive factors like salt, aging or oxidative stress, in other words a parainflammatory micro-milieu. Several mechanisms for differential action of MR and GR have been investigated. New MR specific DNA-binding elements have been postulated, protein–protein interactions in the cytosol explored and posttranscriptional regulatory mechanisms investigated without completely explaining MR actions. An additional regulatory option are ligand-dependent or independent mechanisms that affect MR trafficking and therefore subcellular localization and thereby MR interaction partners and activity. It has been already shown for other molecules like the EGFR that alternative subcellular distribution of the receptors can influence signaling and possibly progression of diseases. Consequently, studying rapid trafficking and its modulation seems relevant for understanding MR actions.
2. MR trafficking Although classical genomic signaling is the most investigated pathway of MR signaling, the steps leading up to transcriptional gene regulation are still not completely understood. With some cell-type specific exceptions, the MR seems to predominantly reside in the cytosol in its unliganded state [8–11]. There it is associated with a large heterocomplex of chaperone molecules including HSP90, HSP70, p23 and proteins with tetratricopeptide repeat sequences such as FKBP52, 52, HOP/p60, Cyp40, PP5 [10,12]. This complex enables the MR to stay in its high affinity state for ligands. Nevertheless, the localization of the MR is dynamic, meaning that there is an equilibrium between cytosolic and nuclear localization which can shift in either way, depending on the presence of ligands or other stimulating factors [13]. After binding of ligand, nucleocytoplasmic shuttling of the MR occurs and the equilibrium of MR localization is shifted to the nucleus. Previously, it was thought that dissociation of the chaperone molecules from the MR is a prerequisite for MR shuttling. Current studies suggest that there are two modes of nuclear MR trafficking. Besides a rapid mode with t1/2 = 4–10 min [8,14,15], a slower transport to the nucleus with t1/2 around 40–60 min has been described [8]. The rapid shifting seems to be a highly regulated process dependent on the presence of HSP90 because it can be inhibited by the HSP90 inhibitor geldanamycin. Geldanamycin leads to dissociation of MR from HSP90 and in some cases has also been found to be involved in MG132-inhibited degradation of MR, suggesting that HSP90 may protect MR from proteasomal degradation [8,15]. Importantly, HSP90 is located both in the cytoplasm and the nucleus and the HSP90-MR complex does not dissociate immediately upon steroid binding as postulated in the classical model [8,9,15,16]. Data point to the fact that HSP90 stays associated to MR for the first 10 min in the nucleus and then dissociates after facilitating MR binding to the insoluble chromatin fraction. Accumulation of MR in the nucleus is still possible in the presence of geldanamycin mediated by the slower transport mechanism within t1/2 40–60 min supposedly reflecting diffusion. Consequently, translocation per see does not seem to be impaired without HSP90 but rapid trafficking. Recent evidence suggests that more of the associated proteins of the cytosol are involved in MR transport and nuclear pore transition [15,17]. Coimmunoprecipitation experiments suggest that a sophisticated machinery of proteins is involved in MR trafficking to nuclear DNA, which besides HSP90 include the dynein/dynactin motor complex and FKBP52 [8,18]. The interaction between MR and HSP90 seems to influence the composition of the rest of the heterocomplex, with an inverse relationship between HSP90 and HSP70 content and an enrichment of dynein, FKBP52 and p23 in HSP90 containing complexes. Consequently, loss of HSP90 implies loss of the interacting acidic protein p23 and FKBP52 which leads
to dissociation from dynein/dynactin and impaired trafficking. Especially, the exchange between FKBP51 and FKBP52 seems to be of primary importance as they compete for the binding of HSP90 and dynein/dynactin can only bind to FKBP52 and not to FKBP51 [19]. The switch leading to the exchange of FKBP51 to FKBP52 in vivo seems to be binding of ligand, i.e. aldosterone. Accordingly, FKBP51 was shown to inhibit MR action [12]. The importance of the FKBP51/52 ratio for the nuclear cytoplasmic equilibrium of the MR was further emphasized by studies in FKBP52 knockout MEF cells, in which nuclear localization of MR is lower and trafficking is impaired. In line with these observations, the cardiomyocytes cell line HL-1 with predominantly nuclear localization of the MR possesses a low expression of HSP90. In agreement with GR trafficking, MR was found to be associated with tubulin via the HSP90, FKBP52 and dynein/dynactin interaction [8]. When the cytoskeleton was disrupted, rapid geldanamycin-sensitive MR transport was no longer possible, although the slower transport persists [8]. To facilitate the rapid transport, the MR possesses three nuclear localization sites (NLS). NLO is a serine/threonine-rich NLS that is located in the N-terminus and which mediates nuclear localization of unliganded as well as agonist-induced MR signaling. NLS1 is a bipartite basic motif localized at the border between the DNA-binding domain and the hinge region, which acts in concert with NLO and NL2 and stimulates nuclear uptake of agonist-treated receptor. NLS2 resides within the ligand binding domain and also depends on agonist or antagonist actions [20]. In unliganded cytoplasmic MR the NLS are supposedly masked by chaperones such as HSP90. Nuclear trafficking takes place via active transport through nuclear pore complexes and involves binding of MR to importin alpha, which translocates into the nucleus with the same kinetics as MR. For this transport intact MR NLS1 and especially aa675 and aa677 were necessary; with a NLSmutant neither MR nor importin alpha moved into the nucleus after treatment with corticosterone. Importin beta was located in the perinuclear region and was unaffected by the stimulation of MR with ligand [21]. Once in the nucleus, homo- or possibly heterodimerization of the MR occurs after dissociation from HSP90 [15,22,23]. For in vitro binding of MR no ligand is necessary [15,24]. The MR possesses a potential nuclear export signal but there is still a debate going on whether nuclear MR shuttling is uni- or bidirectional [8,20]. 3. Modulation of classical MR trafficking It is important to stress that aldosterone alone does not elicit pathological actions but requires the coincidence of a pathological milieu for its detrimental effects [6]. As components triggering inadequate MR activation, high salt load, RNS and ROS have been suggested (see Fig. 1). 3.1. Salt A current cochrane systematic view and metaanalysis reveals that a modest reduction in salt significantly lowers blood pressure in hypertensive and normotensive individuals, irrespective of sex and ethnic group, thus highlighting the importance of salt for hypertension without speculating about the mechanism. For the damaging effects of aldosterone in the renocardiovascular system likewise a high salt diet can function as the obligatory additional stressor [5,25,26]. Mechanistically, Rac-1, a member of the Rho family of small GTPases, was recently proposed as a modulator of MR activity. A ligand-independent activation of MR nuclear content and transactivation activity by Rac-1 was demonstrated as well as pathological relevance for kidney disease and possibly cardiac injury [27–30]. This mechanism was found to be involved in
Please cite this article in press as: Gekle M et al. Rapid mineralocorticoid receptor trafficking. Steroids (2013), http://dx.doi.org/10.1016/ j.steroids.2013.10.016
M. Gekle et al. / Steroids xxx (2013) xxx–xxx
3
Fig. 1. Model of rapid MR trafficking and its modulation.
salt-sensitive hypertension as demonstrated in Dahl salt-sensitive rats as well as in high salt diet mice with activated RAAS [27,30]. Nagata et al. further showed that MR blockade is beneficial even in low-aldosterone salt-sensitive hypertension to prevent cardiac hypertrophy and failure [31]. These data show that salt-induced pathophysiological MR activation is a decisive mechanism and that aldosterone is mainly permissive. An additional mechanism for NaCl-induced pathophysiological MR activation with cardiovascular remodeling seems to be a shift in redox state with increased ROS production which can activate (occupied) MR. Rac-1 has been shown to play an important role in induction of NADPH oxidase and thereby ROS production [32]. Therefore, Rac-1 provides a nice link to enhanced MR activation by oxidative stress. Furthermore, high salt can also propagate inflammation which is again followed by enhanced ROS formation. 3.2. Oxidative stress The MR has been shown to lead to enhanced ROS formation itself, especially in collaboration with vasoactive mediators like angiotensin II [33,34]. Furthermore, many of its pathological effects require or are aggravated by oxidative stress. Nevertheless, the direct effect of oxidative stress on MR trafficking and classical GRE activation is still unclear. On the one hand, incubation of cells with H2O2 failed to affect basal or ligand-induced GRE transactivation activity [35]. On the other hand, injecting mice the glutathionedepleting agent L-buthionine-(SR)-sulfoximine inhibited the steroid binding capacity of the MR. This effect was mimicked in vitro by receptor oxidation. As underlying mechanism for reduced steroid binding capacity of MR impaired steroid binding due to oxidation of essential cysteine residues was postulated [36–38]. 3.3. Nitrosative stress Cell culture studies provide first indications for modulation of MR trafficking and therefore also MR activity by nitrosative stress
mimicked by incubation of cells with the peroxynitrite donor Sin-1 [39]. SIN-1 led to a dose-dependent increase in GRE-reporter gene activity even in the absence of ligand. With Sin-1 a maximum of about 80% of the activation level of 10 nM aldosterone was achieved and this effect could be inhibited by the peroxynitrite scavenger ebselen. Interestingly, the closely related glucocorticoid receptor did not show enhanced activity in response to Sin-1, giving rise to speculations that this mechanism may confer specificity to pathological MR effects. More detailed investigations revealed that peroxynitrite stimulates nuclear translocation of the MR in the absence of ligand. This trafficking takes place in a similar rapid fashion as that induced by aldosterone and it can be inhibited by the HSP90 inhibitor geldanamycin. Nevertheless, the shift of the equilibrium from cytosol into the nucleus is less complete than with classical MR ligands like aldosterone. With a truncated MR variant missing the n-terminal A/B domain, nuclear translocation still occurs in a similar fashion. We postulate that posttranslational modification of the MR by peroxynitrite could be responsible for the modulation of MR transport. Peroxynitrite is known to induce 3-nitrotyrosin modification, i.e. protein nitration. Alternatively, it can also react with carbon dioxide to nitrosoperoxocarbonate which decomposes into highly reactive radicals and then leads to oxidative protein modifications, i.e. carbonylation. In contrast NO is known to be vaso- and cardioprotective and to function as an attenuator of vascular tone, fibrosis, coagulation and remodelling [39]. It has been shown to attenuate MR and GR activation by inhibiting binding of corticosteroid receptors to the GRE binding element. No change in nuclear translocation could be detected. NO can be added to the cells by the NO donor SNAP and dose-dependently attenuates ligand-dependent activation of MR and GR in HEK cells or also in bovine aortic endothelial cells (GM7373). While other NO donors like NONO also elicit this effect, long lasting SNAP metabolites do not. Since db-cGMP failed to elicit the same response, direct modification of MR seems possible, thus suggesting s-nitrosylation of MR. Based on experiments with MRdeletion constructs, the domains C–F were identified as being
Please cite this article in press as: Gekle M et al. Rapid mineralocorticoid receptor trafficking. Steroids (2013), http://dx.doi.org/10.1016/ j.steroids.2013.10.016
4
M. Gekle et al. / Steroids xxx (2013) xxx–xxx
relevant. Conveniently, the DBDs of MR and GR possess a homology of 94% and both contain 10 cysteine residues as potential targets for s-nitrosylation which could well explain impaired DNA binding. No change in binding of aldosterone to MR was detectable. In the presence of additional oxidative stress, NO would be converted into more reactive radicals like peroxynitrite and thus would lose its attenuating effect on MR (and GR) and would simultaneously and ligand-independently inappropriately activated the MR. 3.4. Glucocorticoids and 11beta-HSD Another possibility to explain altered MR activity and/or trafficking without changes in aldosterone levels is binding of alternative ligands. For the MR it is well known that glucocorticoids like hydrocortisone bind to it with similar affinity as aldosterone and also lead to its activation. To prevent this, classical epithelial tissues posses an enzyme called 11beta-hydroxysteroid dehydrogenase (11beta-HSD) 2 that metabolizes active glucocorticoids into inactive derivatives that cannot activate the MR [40]. In the heart this enzyme is hardly expressed, so that permanent occupancy of the MR by hydrocortisone is postulated. One hypothesis is that depending on redox-state, glucocorticoids either attenuate or activate the MR perhaps mediated by the cofactor ratio NADH/NAD [40]. Furthermore, a shift in mRNA pattern from 11beta-HSD 2 towards the 11beta-HSD 1 isoform by inflammatory stress and aging was proposed [41]. Because 11beta-HSD 1 favors the reverse reaction, intracellular glucocorticoid activation, i.e. conversion from inactive cortisone to active hydrocortisone would be facilitated under these conditions, thereby enhancing MR trafficking and signaling. According to this scenario, MR antagonists could be effective in preventing end organ damage even under conditions of normal aldosterone concentrations in organs with 11beta-HSD 2 activity like the vasculature and the kidney. 3.5. MR posttranslational modifications One possible explanation how variations in extracellular micromilieu can affect MR trafficking and/or transcriptional activity is the introduction of posttranslational modifications. It has been assumed for a long time that aldosterone binding leads to posttranslational modification of the MR which can then influence nuclear translocation, gene expression and/or degradation. One of the best explored posttranslational modifications of the MR is phosphorylation. Early reports show that in a physiological milieu the MR is phosphorylated and that transformation to the active DNA binding form requires phosphatases [42,43]. A mutation of a potential phosphorylation site of the MR in brown Norway rats (Y73C) has been identified as a gain of function mutatin of MR with greater transactivation by aldosterone (and progesterone) [44]. The exact mechanism and effect on trafficking is unclear. Additionally, rapid MR serine-threonine phosphorylation has been described by PKC alpha [45]. Here, nongenomic PKC activation seems to increase genomic MR signaling. Again the exact mechanism and the effect on MR trafficking have not been investigated. For PKA an N-terminus- and ligand-dependent stimulation of GRE promoter activity was detected. Because no phosphorylation could be detected, an indirect effect is most likely responsible [46]. Furthermore, inhibition of nongenomic ERK1/2 phosphorylation was found to impair cytoplasmic-nuclear shuttling of MR and thereby transactivation of GRE. This effect depended on the N-terminal domain which was later shown to be directly phosphorylated by MR [47,48]. Because aldosterone/MR can rapidly activate ERK phosphorylation, this would suggest activation of genomic signaling by nongenomic MR effects. However, mutating the six predicted ERK1/2 sites in the A/B domain of the MR had no effect on trafficking but regulated the ubiquitination state of MR and therefore degradation [48]. In
its unliganded state, the MR was monoubiquitinated, a state which is stabilized by the tumor suppressor gene 101. Upon ligand binding TSG101 dissociated and triggers polyubiquitination which leads to destabilization of MR and degradation. Through this mechanism, ERK phosphorylation could make a nice negative feedback loop to end genomic MR actions. Obvious differences in the experimental setup that could explain these differences are that the one group inactivated 6 potential ERK phosphorylation sites in the Nterminus of the MR permanently while the other inhibited rapid ERK1/2 phosphorylation completely but for a shorter period of time with a MEK inhibitor [47]. An additional phosphorylation site of interest for trafficking lies in the nuclear localization signal NLS0 within the serine-threonine rich sequence at Serine 602. Mutations at this hypothetical phosphorylation site have been shown to influence nuclear trafficking of MR even though the biological relevance of this finding is unclear [9,20]. Furthermore, acetylation was reported for other steroid receptors. A matching acetylation motif is also present in the NLS1 of MR, suggesting an involvement in ligand-dependent shuttling or protein–protein-interaction and transcriptional regulation [49]. Other posttranslational modifications that have been discussed but not in the context of MR trafficking are ubiquitination and sumoylation. Several ubiquitination sites were predicted for MR with the PEST find algorithm [50]. Especially two strong potential ubiquitin acceptor lysates at K367 and K715 in the NTD and the hinge region respectively were suggested. Polyubiquitination was demonstrated by Tirard et al. and Faresse [48,51]. For sumoylation, four (five) highly conserved sumoylation sites were predicted and characterized by mutations in MR; three in the NTD and one in the ligand binding domain (K89, K399, (K428), K494, K953). Mutation of these sites led to a mutation-number dependent enhanced hormone-dependent transcription at GRE compound elements but not at MMTV elements. [52]. Additionally, nuclear mobility, i.e. the promoter occupancy was higher of aldosterone/MR with mutated sumoylation sites [51]. Inhibition of proteasome function by MG132 independent of sumoylation led to a predominantly nuclear localization of unliganded MR; however it was not sufficient to elicit transcriptional activation in the absence of hormone [51].
4. Conclusion Overall, the effectiveness of MR antagonists in the treatment of cardiovascular patients is still mechanistically not understood. While in some reports a high incidence of hyperaldosteronism has been reported in the general population and is held responsible for the beneficial effects of MR antagonists, other studies did not find any abnormal aldosterone concentrations or salt status alterations to explain an inappropriate MR activation. Furthermore, animal studies show that development of pathophysiological MR effects in the cardiovascular system and the kidneys requires additional permissive factors like high salt load, ROS or RNS. One way to enhance MR activity is by facilitating its trafficking into the nucleus. Recently, the prerequisites and interaction partners for cytosolic-nuclear trafficking of the MR have been investigated in more detail. Modulators have been identified, although the findings are not always consistent and there is still much research to be done. Nevertheless, modulation of MR trafficking by parainflammatory factors which mediate inadequate MR activation and thereby pathological MR effects is an attractive hypothesis. Direct posttranslational modifications but also indirect posttranslational modulation of MR interaction partners and cofactors may be responsible for these observations. Because activated MR itself is known to enhance inflammation and ROS generation by NADPH oxidase induction, such a mechanism could initiate a vicious circle with enhanced expression of genes involved in vascular remodeling
Please cite this article in press as: Gekle M et al. Rapid mineralocorticoid receptor trafficking. Steroids (2013), http://dx.doi.org/10.1016/ j.steroids.2013.10.016
M. Gekle et al. / Steroids xxx (2013) xxx–xxx
like MCP1, NFKB and EGFR and with additional non-genomic effect further supporting the genomic effects through transactivation of EGFR, PKC and Src [53–55]. Under these conditions, the MR may turn from a homeostasis regulator into a harmful mediator of remodeling processes.
References [1] Pitt B, Zannad F, Remme WJ, Cody R, Castaigne A, Perez A, et al. The effect of spironolactone on morbidity and mortality in patients with severe heart failure. Randomized aldactone evaluation study investigators. N Engl J Med 1999;341:709–17. [2] Pitt B, Remme W, Zannad F, Neaton J, Martinez F, Roniker B, et al. Eplerenone, a selective aldosterone blocker, in patients with left ventricular dysfunction after myocardial infarction. N Engl J Med 2003;348:1309–21. [3] Pitt B, Reichek N, Willenbrock R, Zannad F, Phillips RA, Roniker B, et al. Effects of eplerenone, enalapril, and eplerenone/enalapril in patients with essential hypertension and left ventricular hypertrophy: the 4E-left ventricular hypertrophy study. Circulation 2003;108:1831–8. [4] Sartorio CL, Fraccarollo D, Galuppo P, Leutke M, Ertl G, Stefanon I, et al. Mineralocorticoid receptor blockade improves vasomotor dysfunction and vascular oxidative stress early after myocardial infarction. Hypertension 2007;50:919–25. [5] Blasi ER, Rocha R, Rudolph AE, Blomme EA, Polly ML, McMahon EG. Aldosterone/salt induces renal inflammation and fibrosis in hypertensive rats. Kidney Int 2003;63:1791–800. [6] Brilla CG, Weber KT. Mineralocorticoid excess, dietary sodium, and myocardial fibrosis. J Lab Clin Med 1992;120:893–901. [7] Qin W, Rudolph AE, Bond BR, Rocha R, Blomme EAG, Goellner JJ, et al. Transgenic model of aldosterone-driven cardiac hypertrophy and heart failure. Circ Res 2003;93:69–76. [8] Galigniana MD, Erlejman AG, Monte M, Gomez-Sanchez C, Piwien-Pilipuk G. The hsp90-FKBP52 complex links the mineralocorticoid receptor to motor proteins and persists bound to the receptor in early nuclear events. Mol Cell Biol 2010;30:1285–98. [9] Hernandez-Diaz I, Giraldez T, Arnau MR, Smits VAJ, Jaisser F, Farman N, et al. The mineralocorticoid receptor is a constitutive nuclear factor in cardiomyocytes due to hyperactive nuclear localization signals. Endocrinology 2010;151:3888–99. [10] Nishi M, Ogawa H, Ito T, Matsuda Ki, Kawata M. Dynamic changes in subcellular localization of mineralocorticoid receptor in living cells. In comparison with glucocorticoid receptor using dual-color labeling with green fluorescent protein spectral variants. Mol Endocrinol 2001;15:1077–92. [11] Pearce D, Naray-Fejes-Toth A, Fejes-Toth G. Determinants of subnuclear organization of mineralocorticoid receptor characterized through analysis of wild type and mutant receptors. J Biol Chem 2002;277:1451–6. [12] Gallo LI, Ghini AA, Piwien Pilipuk G, Galigniana MD. Differential recruitment of tetratricorpeptide repeat domain immunophilins to the mineralocorticoid receptor influences both heat-shock protein 90-dependent retrotransport and hormone-dependent transcriptional activity. Biochemistry 2007;46:14044–57. [13] Galigniana MD, Piwien Pilipuk G, Kanelakis KC, Buton G, Lantos CP. Molecular mechanism of activation and nuclear translocation of the mineralocorticoid receptor upon binding of pregnanesteroids. Mol Cell Endocrinol 2004;217:167–79. [14] Bamberger CM, Wald M, Bamberger AM, Schulte HM. Inhibition of mineralocorticoid and glucocorticoid receptor function by the heat shock protein 90-binding agent geldanamycin. Mol Cell Endocrinol 1997; 131:233–40. [15] Grossmann C, Ruhs S, Langenbruch L, Mildenberger S, Str+ñtz N, Schumann K. Nuclear shuttling precedes dimerization in mineralocorticoid receptor signaling. Chem Biol 2012;19:742–51. [16] Csermely P, Schnaider T, Soti C, Prohaszka Z, Nardai G. The 90-kDa molecular chaperone family: structure, function, and clinical applications. Compr Rev Pharmacol Ther 1998;79:129–68. [17] Galigniana MD, Scruggs JL, Herrington J, Welsh MJ, Carter-Su C, Housley PR, et al. Heat shock protein 90-dependent (geldanamycin-inhibited) movement of the glucocorticoid receptor through the cytoplasm to the nucleus requires intact cytoskeleton. Mol Endocrinol 1998;12:1903–13. [18] Galigniana MD, Echeverria PC, Erlejman AG, Piwien-Pilipuk G. Role of molecular chaperones and TPR-domain proteins in the cytoplasmic transport of steroid receptors and their passage through the nuclear pore. Nucleus 2010;1:299–308. [19] Wochnik GM, Rüegg JL, Abel GA, Schmidt U, Holsboer F, Rein T. FK506-binding proteins 51 and 52 differentially regulate dynein interaction and nuclear translocation of the glucocorticoid receptor in mammalian cells. J Biol Chem 2005;280:4609–16. [20] Walther RF, Atlas E, Carrigan A, Rouleau Y, Edgecombe A, Visentin L, et al. A serine/threonine-rich motif is one of three nuclear localization signals that determine unidirectional transport of the mineralocorticoid receptor to the nucleus. J Biol Chem 2005;280:17549–61.
5
[21] Tanaka M, Nishi M, Morimoto M, Sugimoto T, Kawata M. Imaging analysis of mineralocorticoid receptor and importins in single living cells by using GFP color variants. Cell Tissue Res 2005;320:447–53. [22] Nishi M, Tanaka M, Matsuda Ki, Sunaguchi M, Kawata M. Visualization of glucocorticoid receptor and mineralocorticoid receptor interactions in living cells with GFP-based fluorescence resonance energy transfer. J Neurosci 2004;24:4918–27. [23] Trapp T, Holsboer F. Heterodimerization between mineralocorticoid and glucocorticoid receptors increases the functional diversity of corticosteroid action. Trends Pharmacol Sci 1996;17:145–9. [24] Savory JG, Prefontaine GG, Lamprecht C, Liao M, Walther RF, Lefebvre YA, et al. Glucocorticoid receptor homodimers and glucocorticoid–mineralocorticoid receptor heterodimers form in the cytoplasm through alternative dimerization interfaces. Mol Cell Biol 2001;21:781–93. [25] Sun Y, Zhang J, Lu L, Chen SS, Quinn MT, Weber KT. Aldosterone-induced inflammation in the rat heart – role of oxidative stress. Am J Pathol 2002; 161:1773–81. [26] Park JB, Schiffrin EL. Cardiac and vascular fibrosis and hypertrophy in aldosterone-infused rats: role of endothelin-1. Am J Hypertens 2002; 15:164–9. [27] Shibata S, Mu S, Kawarazaki H, Muraoka K, Ishizawa Ki, Yoshida S, et al. Rac1 GTPase in rodent kidneys is essential for salt-sensitive hypertension via a mineralocorticoid receptor-dependent pathway. J Clin Investig 2011; 121:3233–43. [28] Shibata S, Nagase M, Yoshida S, Kawarazaki W, Kurihara H, Tanaka H, et al. Modification of mineralocorticoid receptor function by Rac1 GTPase: implication in proteinuric kidney disease. Nat Med 2008;14:1370–6. [29] Nagase M, Ayuzawa N, Kawarazaki W, Ishizawa K, Ueda K, Yoshida S, et al. Oxidative stress causes mineralocorticoid receptor activation in rat cardiomyocytes: role of small GTPase Rac1. Hypertension 2012;59:500–6. [30] Kawarazaki W, Nagase M, Yoshida S, Takeuchi M, Ishizawa K, Ayuzawa N, et al. Angiotensin II- and salt-induced kidney injury through Rac1-mediated mineralocorticoid receptor activation. J Am Soc Nephrol 2012;23:997–1007. [31] Nagata K, Obata K, Xu J, Ichihara S, Noda A, Kimata H, et al. Mineralocorticoid receptor antagonism attenuates cardiac hypertrophy and failure in lowaldosterone hypertensive rats. Hypertension 2006;47:656–64. [32] Hordijk PL. Regulation of NADPH oxidases: the role of rac proteins. Circ Res 2006;98:453–62. [33] Ojeda-Cervantes M, Barrera-Chimal J, Alberu J, Perez-Villalva R, MoralesBuenrostro L, Bobadilla NA. Mineralocorticoid receptor blockade reduced oxidative stress in renal transplant recipients: a double-blind, randomized pilot study. Am J Nephrol 2013;37:481–90. [34] Di Zhang A, Nguyen Dinh CA, Soukaseum C, Escoubet B, Cherfa A, Messaoudi S, et al. Cross-talk between mineralocorticoid and angiotensin II signaling for cardiac remodeling. Hypertension 2008;52:1060–7. [35] Pfau A, Grossmann C, Freudinger R, Mildenberger S, Benesic A, Gekle M. Ca2+ but not H2O2 modulates GRE-element activation by the human mineralocorticoid receptor in HEK cells. Mol Cell Endocrinol 2007;264:35–43. [36] Piwien-Pilipuk G, Ayala A, Machado A, Galigniana MD. Impairment of mineralocorticoid receptor (MR)-dependent biological response by oxidative stress and aging. Correlation with post-translational modification of MR and decreased ADP-ribosylatable level of elongation factor 2 in kidney cells. J Biol Chem 2002;277:11896–903. [37] Piwien-Pilipuk G, Galigniana MD. Oxidative stress induced by L-buthionine-(S, R)-sulfoximine, a selective inhibitor of glutathione metabolism, abrogates mouse kidney mineralocorticoid receptor function. Biochim Biophys Acta (BBA) – Mol Cell Res 2000;1495:263–80. [38] Souque A, Fagart J, Couette B, Rafestin-Oblin M. Sulfhydryl groups are involved in the binding of agonists and antagonists to the human mineralocorticoid receptor. J Steroid Biochem Mol Biol 1996;57:315–21. [39] Ruhs S, Strätz N, Schlör K, Meinel S, Mildenberger S, Rabe S, et al. Modulation of transcriptional mineralocorticoid receptor activity by nitrosative stress. Free Radical Biol Med 2012;53:1088–100. [40] Funder JW. RALES, EPHESUS and redox. J Steroid Biochem Mol Biol 2005;93:121–5. [41] Tsugita M, Iwasaki Y, Nishiyama M, Taguchi T, Shinahara M, Taniguchi Y, et al. Differential regulation of 11beta-hydroxysteroid dehydrogenase type-1 and -2 gene transcription by proinflammatory cytokines in vascular smooth muscle cells. Life Sci 2008;83:426–32. [42] Alnemri ES, Maksymowych AB, Robertson NM, Litwack G. Overexpression and characterization of the human mineralocorticoid receptor. J Biol Chem 1991;266:18072–81. [43] Galigniana MD. Native rat kidney mineralocorticoid receptor is a phosphoprotein whose transformation to a DNA-binding form is induced by phosphatases. Biochem J 1998;333:555–63. [44] Marissal-Arvy N, Lombes M, Petterson J, Moisan MP, Mormede P. Gain of function mutation in the mineralocorticoid receptor of the brown Norway rat. J Biol Chem 2004;279:39232–9. [45] Le Moellic C, Ouvrard-Pascaud A, Capurro C, Cluzeaud F, Fay M, Jaisser F, et al. Early nongenomic events in aldosterone action in renal collecting duct cells: PKCalpha activation, mineralocorticoid receptor phosphorylation, and crosstalk with the genomic response. J Am Soc Nephrol 2004;15:1145–60. [46] Massaad C, Houard N, Lombes M, Barouki R. Modulation of human mineralocorticoid receptor function by protein kinase A. Mol Endocrinol 1999;13:57–65.
Please cite this article in press as: Gekle M et al. Rapid mineralocorticoid receptor trafficking. Steroids (2013), http://dx.doi.org/10.1016/ j.steroids.2013.10.016
6
M. Gekle et al. / Steroids xxx (2013) xxx–xxx
[47] Grossmann C, Benesic A, Krug AW, Freudinger R, Mildenberger S, Gassner B. Human mineralocorticoid receptor expression renders cells responsive for nongenotropic aldosterone actions. Mol Endocrinol 2005;19:1697–710. [48] Faresse N, Vitagliano JJ, Staub O. Differential ubiquitylation of the mineralocorticoid receptor is regulated by phosphorylation. FASEB J 2012;26:4373–82. [49] Vienchareun S, Le Menuet D, Martinerie L, Munier M, Pascual-Le Tallec L, Lombes M. The mineralocorticoid receptor: insights into its molecular and (patho)physiological biology. Nucl Recept Signal 2007;30:e012. [50] Pascual-Le Tallec L, Lombes M. The mineralocorticoid receptor: a journey exploring its diversity and specificity of action. Mol Endocrinol 2005;19:2211–21. [51] Tirard M, Almeida OFX, Hutzler P, Melchior F, Michaelidis TM. Sumoylation and proteasomal activity determine the transactivation properties of the mineralocorticoid receptor. Mol Cell Endocrinol 2007;268:20–9. [52] Pascual-Le Tallec L, Kirsh O, Lecomte MC, Viengchareun S, Zennaro MC, Dejean A, et al. Protein inhibitor of activated signal transducer and activator of
transcription 1 interacts with the N-terminal domain of mineralocorticoid receptor and represses its transcriptional activity: implication of small ubiquitin-related modifier 1 modification. Mol Endocrinol 2003;17:2529–42. [53] Nagata D, Takahashi M, Sawai K, Tagami T, Usui T, Shimatsu A, et al. Molecular mechanism of the inhibitory effect of aldosterone on endothelial NO synthase activity. Hypertension 2006;48:165–71. [54] Nakano S, Kobayashi N, Yoshida K, Ohno T, Matsuoka H. Cardioprotective mechanisms of spironolactone associated with the angiotensin-converting enzyme/epidermal growth factor receptor/extracellular signal-regulated kinases, NAD(P)H oxidase/lectin-like oxidized low-density lipoprotein receptor-1, and Rho-kinase pathways in aldosterone/salt-induced hypertensive rats. Hypertens Res 2005;28:925–36. [55] Stas S, Whaley-Connell A, Habibi J, Appesh L, Hayden MR, Karuparthi PR, et al. Mineralocorticoid receptor blockade attenuates chronic overexpression of the renin–angiotensin–aldosterone system stimulation of reduced nicotinamide adenine dinucleotide phosphate oxidase and cardiac remodeling. Endocrinology 2007;148:3773–80.
Please cite this article in press as: Gekle M et al. Rapid mineralocorticoid receptor trafficking. Steroids (2013), http://dx.doi.org/10.1016/ j.steroids.2013.10.016
Contents lists available at ScienceDirect
Steroids journal homepage: www.elsevier.com/locate/steroids
Rapid mineralocorticoid receptor trafficking M. Gekle, M. Bretschneider, S. Meinel, S. Ruhs, C. Grossmann ⇑ Julius Bernstein Institute of Physiology, Martin Luther University Halle-Wittenberg, Germany
a r t i c l e
i n f o
Article history: Available online xxxx Keywords: Mineralocorticoid receptor Trafficking Modulation Genomic signaling
a b s t r a c t The mineralocorticoid receptor (MR) is a ligand-dependent transcription factor that physiologically regulates water-electrolyte homeostasis and controls blood pressure. The MR can also elicit inflammatory and remodeling processes in the cardiovascular system and the kidneys, which require the presence of additional pathological factors like for example nitrosative stress. However, the underlying molecular mechanism(s) for pathophysiological MR effects remain(s) elusive. The inactive MR is located in the cytosol associated with chaperone molecules including HSP90. After ligand binding, the MR monomer rapidly translocates into the nucleus while still being associated to HSP90 and after dissociation from HSP90 binds to hormone-response-elements called glucocorticoid response elements (GREs) as a dimer. There are indications that rapid MR trafficking is modulated in the presence of high salt, oxidative or nitrosative stress, hypothetically by induction or posttranslational modifications. Additionally, glucocorticoids and the enzyme 11beta hydroxysteroid dehydrogenase may also influence MR activation. Because MR trafficking and its modulation by micro-milieu factors influence MR cellular localization, it is not only relevant for genomic but also for nongenomic MR effects. Ó 2013 Elsevier Inc. All rights reserved.
1. MR enigma The mineralocorticoid receptor (MR) with its endogenous ligand aldosterone is one of the main effectors in the renin-angiotensin-aldosterone-system (RAAS) and has a pivotal role in water-electrolyte homeostasis and regulation of blood pressure. It belongs to the steroid receptor superfamily that consists of the progesterone, the estrogen, the androgen and the glucocorticoid receptor. Steroid receptors possess a common structure comprising the domains A–F. The N-terminal A/B domain is the most variable among the receptors and is responsible for cofactor binding. The C domain of the MR is the DNA binding domain and possesses a 94% amino acid identity to the DNA binding domain of its closest relative, the glucocorticoid receptor (GR). After a short hinge region comes the C-terminal ligand binding domain that is also involved in dimerization. Of the steroid receptors, the mineralocorticoid receptor has been the least appreciated for a long time because of the more obvious clinical implications in for example cancer and immunological disease of its relatives. Consequently, many of the basic molecular observations concerning signaling and trafficking of the MR have been deduced from other steroid receptors, regardless of possible differences between them. The lack of interest changed after the importance of the MR for pathological ⇑ Corresponding author. Address: Julius-Bernstein-Institut für Physiologie, Universität Halle-Wittenberg, Magdeburger Strasse 6, 06097 Halle/Saale, Germany. Tel.: +49 345 557 1886; fax: +49 345 557 4019. E-mail address: [email protected] (C. Grossmann).
changes in the cardiovascular system and the kidneys became apparent. In two pivotal clinical studies that were followed by many others, the beneficial effect of MR antagonists like spironolactone and eplerenone for patients with cardiovascular disease was proven; however, without understanding the underlying mechanisms [1–3]. Since then, MR activation has been shown to be involved in different pathophysiological effects in the reno-cardiovascular system including endothelial dysfunction, inflammation, hypertrophy and fibrosis in both clinical studies and animal experiments [4–7]. It is well known, that the MR functions as a ligand-dependent transcription factor at hormone response elements called glucocorticoid response elements (GRE) that it shares with the GR. However, the GR acts in an anti-inflammatory and immunosuppressive way on the cardiovascular system, suggesting additional signaling mechanisms. The trigger that causes the MR to turn from a receptor regulating water-electrolyte homeostasis and not causing any harm into a receptor mediating pathological effects in the cardiovascular system is also an enigma. Of note, the MR needs to be inadequately activated to confer pathological effects as can be judged by the positive effects of MR antagonists. One way to achieve this is by having inappropriately high aldosterone levels in relation to salt status in an individual. Although such a scenario is likely in case of hyperaldosteronism caused by adrenal adenoma or hyperplasia, this does not seem to apply for the majority of patients benefitting from MR antagonists as in the above mentioned clinical studies, where aldosterone levels and salt status of participants were unremarkable. In animal studies, it is striking that aldosterone application only leads to
0039-128X/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.steroids.2013.10.016
Please cite this article in press as: Gekle M et al. Rapid mineralocorticoid receptor trafficking. Steroids (2013), http://dx.doi.org/10.1016/ j.steroids.2013.10.016
2
M. Gekle et al. / Steroids xxx (2013) xxx–xxx
pathological changes in the presence of additional permissive factors like salt, aging or oxidative stress, in other words a parainflammatory micro-milieu. Several mechanisms for differential action of MR and GR have been investigated. New MR specific DNA-binding elements have been postulated, protein–protein interactions in the cytosol explored and posttranscriptional regulatory mechanisms investigated without completely explaining MR actions. An additional regulatory option are ligand-dependent or independent mechanisms that affect MR trafficking and therefore subcellular localization and thereby MR interaction partners and activity. It has been already shown for other molecules like the EGFR that alternative subcellular distribution of the receptors can influence signaling and possibly progression of diseases. Consequently, studying rapid trafficking and its modulation seems relevant for understanding MR actions.
2. MR trafficking Although classical genomic signaling is the most investigated pathway of MR signaling, the steps leading up to transcriptional gene regulation are still not completely understood. With some cell-type specific exceptions, the MR seems to predominantly reside in the cytosol in its unliganded state [8–11]. There it is associated with a large heterocomplex of chaperone molecules including HSP90, HSP70, p23 and proteins with tetratricopeptide repeat sequences such as FKBP52, 52, HOP/p60, Cyp40, PP5 [10,12]. This complex enables the MR to stay in its high affinity state for ligands. Nevertheless, the localization of the MR is dynamic, meaning that there is an equilibrium between cytosolic and nuclear localization which can shift in either way, depending on the presence of ligands or other stimulating factors [13]. After binding of ligand, nucleocytoplasmic shuttling of the MR occurs and the equilibrium of MR localization is shifted to the nucleus. Previously, it was thought that dissociation of the chaperone molecules from the MR is a prerequisite for MR shuttling. Current studies suggest that there are two modes of nuclear MR trafficking. Besides a rapid mode with t1/2 = 4–10 min [8,14,15], a slower transport to the nucleus with t1/2 around 40–60 min has been described [8]. The rapid shifting seems to be a highly regulated process dependent on the presence of HSP90 because it can be inhibited by the HSP90 inhibitor geldanamycin. Geldanamycin leads to dissociation of MR from HSP90 and in some cases has also been found to be involved in MG132-inhibited degradation of MR, suggesting that HSP90 may protect MR from proteasomal degradation [8,15]. Importantly, HSP90 is located both in the cytoplasm and the nucleus and the HSP90-MR complex does not dissociate immediately upon steroid binding as postulated in the classical model [8,9,15,16]. Data point to the fact that HSP90 stays associated to MR for the first 10 min in the nucleus and then dissociates after facilitating MR binding to the insoluble chromatin fraction. Accumulation of MR in the nucleus is still possible in the presence of geldanamycin mediated by the slower transport mechanism within t1/2 40–60 min supposedly reflecting diffusion. Consequently, translocation per see does not seem to be impaired without HSP90 but rapid trafficking. Recent evidence suggests that more of the associated proteins of the cytosol are involved in MR transport and nuclear pore transition [15,17]. Coimmunoprecipitation experiments suggest that a sophisticated machinery of proteins is involved in MR trafficking to nuclear DNA, which besides HSP90 include the dynein/dynactin motor complex and FKBP52 [8,18]. The interaction between MR and HSP90 seems to influence the composition of the rest of the heterocomplex, with an inverse relationship between HSP90 and HSP70 content and an enrichment of dynein, FKBP52 and p23 in HSP90 containing complexes. Consequently, loss of HSP90 implies loss of the interacting acidic protein p23 and FKBP52 which leads
to dissociation from dynein/dynactin and impaired trafficking. Especially, the exchange between FKBP51 and FKBP52 seems to be of primary importance as they compete for the binding of HSP90 and dynein/dynactin can only bind to FKBP52 and not to FKBP51 [19]. The switch leading to the exchange of FKBP51 to FKBP52 in vivo seems to be binding of ligand, i.e. aldosterone. Accordingly, FKBP51 was shown to inhibit MR action [12]. The importance of the FKBP51/52 ratio for the nuclear cytoplasmic equilibrium of the MR was further emphasized by studies in FKBP52 knockout MEF cells, in which nuclear localization of MR is lower and trafficking is impaired. In line with these observations, the cardiomyocytes cell line HL-1 with predominantly nuclear localization of the MR possesses a low expression of HSP90. In agreement with GR trafficking, MR was found to be associated with tubulin via the HSP90, FKBP52 and dynein/dynactin interaction [8]. When the cytoskeleton was disrupted, rapid geldanamycin-sensitive MR transport was no longer possible, although the slower transport persists [8]. To facilitate the rapid transport, the MR possesses three nuclear localization sites (NLS). NLO is a serine/threonine-rich NLS that is located in the N-terminus and which mediates nuclear localization of unliganded as well as agonist-induced MR signaling. NLS1 is a bipartite basic motif localized at the border between the DNA-binding domain and the hinge region, which acts in concert with NLO and NL2 and stimulates nuclear uptake of agonist-treated receptor. NLS2 resides within the ligand binding domain and also depends on agonist or antagonist actions [20]. In unliganded cytoplasmic MR the NLS are supposedly masked by chaperones such as HSP90. Nuclear trafficking takes place via active transport through nuclear pore complexes and involves binding of MR to importin alpha, which translocates into the nucleus with the same kinetics as MR. For this transport intact MR NLS1 and especially aa675 and aa677 were necessary; with a NLSmutant neither MR nor importin alpha moved into the nucleus after treatment with corticosterone. Importin beta was located in the perinuclear region and was unaffected by the stimulation of MR with ligand [21]. Once in the nucleus, homo- or possibly heterodimerization of the MR occurs after dissociation from HSP90 [15,22,23]. For in vitro binding of MR no ligand is necessary [15,24]. The MR possesses a potential nuclear export signal but there is still a debate going on whether nuclear MR shuttling is uni- or bidirectional [8,20]. 3. Modulation of classical MR trafficking It is important to stress that aldosterone alone does not elicit pathological actions but requires the coincidence of a pathological milieu for its detrimental effects [6]. As components triggering inadequate MR activation, high salt load, RNS and ROS have been suggested (see Fig. 1). 3.1. Salt A current cochrane systematic view and metaanalysis reveals that a modest reduction in salt significantly lowers blood pressure in hypertensive and normotensive individuals, irrespective of sex and ethnic group, thus highlighting the importance of salt for hypertension without speculating about the mechanism. For the damaging effects of aldosterone in the renocardiovascular system likewise a high salt diet can function as the obligatory additional stressor [5,25,26]. Mechanistically, Rac-1, a member of the Rho family of small GTPases, was recently proposed as a modulator of MR activity. A ligand-independent activation of MR nuclear content and transactivation activity by Rac-1 was demonstrated as well as pathological relevance for kidney disease and possibly cardiac injury [27–30]. This mechanism was found to be involved in
Please cite this article in press as: Gekle M et al. Rapid mineralocorticoid receptor trafficking. Steroids (2013), http://dx.doi.org/10.1016/ j.steroids.2013.10.016
M. Gekle et al. / Steroids xxx (2013) xxx–xxx
3
Fig. 1. Model of rapid MR trafficking and its modulation.
salt-sensitive hypertension as demonstrated in Dahl salt-sensitive rats as well as in high salt diet mice with activated RAAS [27,30]. Nagata et al. further showed that MR blockade is beneficial even in low-aldosterone salt-sensitive hypertension to prevent cardiac hypertrophy and failure [31]. These data show that salt-induced pathophysiological MR activation is a decisive mechanism and that aldosterone is mainly permissive. An additional mechanism for NaCl-induced pathophysiological MR activation with cardiovascular remodeling seems to be a shift in redox state with increased ROS production which can activate (occupied) MR. Rac-1 has been shown to play an important role in induction of NADPH oxidase and thereby ROS production [32]. Therefore, Rac-1 provides a nice link to enhanced MR activation by oxidative stress. Furthermore, high salt can also propagate inflammation which is again followed by enhanced ROS formation. 3.2. Oxidative stress The MR has been shown to lead to enhanced ROS formation itself, especially in collaboration with vasoactive mediators like angiotensin II [33,34]. Furthermore, many of its pathological effects require or are aggravated by oxidative stress. Nevertheless, the direct effect of oxidative stress on MR trafficking and classical GRE activation is still unclear. On the one hand, incubation of cells with H2O2 failed to affect basal or ligand-induced GRE transactivation activity [35]. On the other hand, injecting mice the glutathionedepleting agent L-buthionine-(SR)-sulfoximine inhibited the steroid binding capacity of the MR. This effect was mimicked in vitro by receptor oxidation. As underlying mechanism for reduced steroid binding capacity of MR impaired steroid binding due to oxidation of essential cysteine residues was postulated [36–38]. 3.3. Nitrosative stress Cell culture studies provide first indications for modulation of MR trafficking and therefore also MR activity by nitrosative stress
mimicked by incubation of cells with the peroxynitrite donor Sin-1 [39]. SIN-1 led to a dose-dependent increase in GRE-reporter gene activity even in the absence of ligand. With Sin-1 a maximum of about 80% of the activation level of 10 nM aldosterone was achieved and this effect could be inhibited by the peroxynitrite scavenger ebselen. Interestingly, the closely related glucocorticoid receptor did not show enhanced activity in response to Sin-1, giving rise to speculations that this mechanism may confer specificity to pathological MR effects. More detailed investigations revealed that peroxynitrite stimulates nuclear translocation of the MR in the absence of ligand. This trafficking takes place in a similar rapid fashion as that induced by aldosterone and it can be inhibited by the HSP90 inhibitor geldanamycin. Nevertheless, the shift of the equilibrium from cytosol into the nucleus is less complete than with classical MR ligands like aldosterone. With a truncated MR variant missing the n-terminal A/B domain, nuclear translocation still occurs in a similar fashion. We postulate that posttranslational modification of the MR by peroxynitrite could be responsible for the modulation of MR transport. Peroxynitrite is known to induce 3-nitrotyrosin modification, i.e. protein nitration. Alternatively, it can also react with carbon dioxide to nitrosoperoxocarbonate which decomposes into highly reactive radicals and then leads to oxidative protein modifications, i.e. carbonylation. In contrast NO is known to be vaso- and cardioprotective and to function as an attenuator of vascular tone, fibrosis, coagulation and remodelling [39]. It has been shown to attenuate MR and GR activation by inhibiting binding of corticosteroid receptors to the GRE binding element. No change in nuclear translocation could be detected. NO can be added to the cells by the NO donor SNAP and dose-dependently attenuates ligand-dependent activation of MR and GR in HEK cells or also in bovine aortic endothelial cells (GM7373). While other NO donors like NONO also elicit this effect, long lasting SNAP metabolites do not. Since db-cGMP failed to elicit the same response, direct modification of MR seems possible, thus suggesting s-nitrosylation of MR. Based on experiments with MRdeletion constructs, the domains C–F were identified as being
Please cite this article in press as: Gekle M et al. Rapid mineralocorticoid receptor trafficking. Steroids (2013), http://dx.doi.org/10.1016/ j.steroids.2013.10.016
4
M. Gekle et al. / Steroids xxx (2013) xxx–xxx
relevant. Conveniently, the DBDs of MR and GR possess a homology of 94% and both contain 10 cysteine residues as potential targets for s-nitrosylation which could well explain impaired DNA binding. No change in binding of aldosterone to MR was detectable. In the presence of additional oxidative stress, NO would be converted into more reactive radicals like peroxynitrite and thus would lose its attenuating effect on MR (and GR) and would simultaneously and ligand-independently inappropriately activated the MR. 3.4. Glucocorticoids and 11beta-HSD Another possibility to explain altered MR activity and/or trafficking without changes in aldosterone levels is binding of alternative ligands. For the MR it is well known that glucocorticoids like hydrocortisone bind to it with similar affinity as aldosterone and also lead to its activation. To prevent this, classical epithelial tissues posses an enzyme called 11beta-hydroxysteroid dehydrogenase (11beta-HSD) 2 that metabolizes active glucocorticoids into inactive derivatives that cannot activate the MR [40]. In the heart this enzyme is hardly expressed, so that permanent occupancy of the MR by hydrocortisone is postulated. One hypothesis is that depending on redox-state, glucocorticoids either attenuate or activate the MR perhaps mediated by the cofactor ratio NADH/NAD [40]. Furthermore, a shift in mRNA pattern from 11beta-HSD 2 towards the 11beta-HSD 1 isoform by inflammatory stress and aging was proposed [41]. Because 11beta-HSD 1 favors the reverse reaction, intracellular glucocorticoid activation, i.e. conversion from inactive cortisone to active hydrocortisone would be facilitated under these conditions, thereby enhancing MR trafficking and signaling. According to this scenario, MR antagonists could be effective in preventing end organ damage even under conditions of normal aldosterone concentrations in organs with 11beta-HSD 2 activity like the vasculature and the kidney. 3.5. MR posttranslational modifications One possible explanation how variations in extracellular micromilieu can affect MR trafficking and/or transcriptional activity is the introduction of posttranslational modifications. It has been assumed for a long time that aldosterone binding leads to posttranslational modification of the MR which can then influence nuclear translocation, gene expression and/or degradation. One of the best explored posttranslational modifications of the MR is phosphorylation. Early reports show that in a physiological milieu the MR is phosphorylated and that transformation to the active DNA binding form requires phosphatases [42,43]. A mutation of a potential phosphorylation site of the MR in brown Norway rats (Y73C) has been identified as a gain of function mutatin of MR with greater transactivation by aldosterone (and progesterone) [44]. The exact mechanism and effect on trafficking is unclear. Additionally, rapid MR serine-threonine phosphorylation has been described by PKC alpha [45]. Here, nongenomic PKC activation seems to increase genomic MR signaling. Again the exact mechanism and the effect on MR trafficking have not been investigated. For PKA an N-terminus- and ligand-dependent stimulation of GRE promoter activity was detected. Because no phosphorylation could be detected, an indirect effect is most likely responsible [46]. Furthermore, inhibition of nongenomic ERK1/2 phosphorylation was found to impair cytoplasmic-nuclear shuttling of MR and thereby transactivation of GRE. This effect depended on the N-terminal domain which was later shown to be directly phosphorylated by MR [47,48]. Because aldosterone/MR can rapidly activate ERK phosphorylation, this would suggest activation of genomic signaling by nongenomic MR effects. However, mutating the six predicted ERK1/2 sites in the A/B domain of the MR had no effect on trafficking but regulated the ubiquitination state of MR and therefore degradation [48]. In
its unliganded state, the MR was monoubiquitinated, a state which is stabilized by the tumor suppressor gene 101. Upon ligand binding TSG101 dissociated and triggers polyubiquitination which leads to destabilization of MR and degradation. Through this mechanism, ERK phosphorylation could make a nice negative feedback loop to end genomic MR actions. Obvious differences in the experimental setup that could explain these differences are that the one group inactivated 6 potential ERK phosphorylation sites in the Nterminus of the MR permanently while the other inhibited rapid ERK1/2 phosphorylation completely but for a shorter period of time with a MEK inhibitor [47]. An additional phosphorylation site of interest for trafficking lies in the nuclear localization signal NLS0 within the serine-threonine rich sequence at Serine 602. Mutations at this hypothetical phosphorylation site have been shown to influence nuclear trafficking of MR even though the biological relevance of this finding is unclear [9,20]. Furthermore, acetylation was reported for other steroid receptors. A matching acetylation motif is also present in the NLS1 of MR, suggesting an involvement in ligand-dependent shuttling or protein–protein-interaction and transcriptional regulation [49]. Other posttranslational modifications that have been discussed but not in the context of MR trafficking are ubiquitination and sumoylation. Several ubiquitination sites were predicted for MR with the PEST find algorithm [50]. Especially two strong potential ubiquitin acceptor lysates at K367 and K715 in the NTD and the hinge region respectively were suggested. Polyubiquitination was demonstrated by Tirard et al. and Faresse [48,51]. For sumoylation, four (five) highly conserved sumoylation sites were predicted and characterized by mutations in MR; three in the NTD and one in the ligand binding domain (K89, K399, (K428), K494, K953). Mutation of these sites led to a mutation-number dependent enhanced hormone-dependent transcription at GRE compound elements but not at MMTV elements. [52]. Additionally, nuclear mobility, i.e. the promoter occupancy was higher of aldosterone/MR with mutated sumoylation sites [51]. Inhibition of proteasome function by MG132 independent of sumoylation led to a predominantly nuclear localization of unliganded MR; however it was not sufficient to elicit transcriptional activation in the absence of hormone [51].
4. Conclusion Overall, the effectiveness of MR antagonists in the treatment of cardiovascular patients is still mechanistically not understood. While in some reports a high incidence of hyperaldosteronism has been reported in the general population and is held responsible for the beneficial effects of MR antagonists, other studies did not find any abnormal aldosterone concentrations or salt status alterations to explain an inappropriate MR activation. Furthermore, animal studies show that development of pathophysiological MR effects in the cardiovascular system and the kidneys requires additional permissive factors like high salt load, ROS or RNS. One way to enhance MR activity is by facilitating its trafficking into the nucleus. Recently, the prerequisites and interaction partners for cytosolic-nuclear trafficking of the MR have been investigated in more detail. Modulators have been identified, although the findings are not always consistent and there is still much research to be done. Nevertheless, modulation of MR trafficking by parainflammatory factors which mediate inadequate MR activation and thereby pathological MR effects is an attractive hypothesis. Direct posttranslational modifications but also indirect posttranslational modulation of MR interaction partners and cofactors may be responsible for these observations. Because activated MR itself is known to enhance inflammation and ROS generation by NADPH oxidase induction, such a mechanism could initiate a vicious circle with enhanced expression of genes involved in vascular remodeling
Please cite this article in press as: Gekle M et al. Rapid mineralocorticoid receptor trafficking. Steroids (2013), http://dx.doi.org/10.1016/ j.steroids.2013.10.016
M. Gekle et al. / Steroids xxx (2013) xxx–xxx
like MCP1, NFKB and EGFR and with additional non-genomic effect further supporting the genomic effects through transactivation of EGFR, PKC and Src [53–55]. Under these conditions, the MR may turn from a homeostasis regulator into a harmful mediator of remodeling processes.
References [1] Pitt B, Zannad F, Remme WJ, Cody R, Castaigne A, Perez A, et al. The effect of spironolactone on morbidity and mortality in patients with severe heart failure. Randomized aldactone evaluation study investigators. N Engl J Med 1999;341:709–17. [2] Pitt B, Remme W, Zannad F, Neaton J, Martinez F, Roniker B, et al. Eplerenone, a selective aldosterone blocker, in patients with left ventricular dysfunction after myocardial infarction. N Engl J Med 2003;348:1309–21. [3] Pitt B, Reichek N, Willenbrock R, Zannad F, Phillips RA, Roniker B, et al. Effects of eplerenone, enalapril, and eplerenone/enalapril in patients with essential hypertension and left ventricular hypertrophy: the 4E-left ventricular hypertrophy study. Circulation 2003;108:1831–8. [4] Sartorio CL, Fraccarollo D, Galuppo P, Leutke M, Ertl G, Stefanon I, et al. Mineralocorticoid receptor blockade improves vasomotor dysfunction and vascular oxidative stress early after myocardial infarction. Hypertension 2007;50:919–25. [5] Blasi ER, Rocha R, Rudolph AE, Blomme EA, Polly ML, McMahon EG. Aldosterone/salt induces renal inflammation and fibrosis in hypertensive rats. Kidney Int 2003;63:1791–800. [6] Brilla CG, Weber KT. Mineralocorticoid excess, dietary sodium, and myocardial fibrosis. J Lab Clin Med 1992;120:893–901. [7] Qin W, Rudolph AE, Bond BR, Rocha R, Blomme EAG, Goellner JJ, et al. Transgenic model of aldosterone-driven cardiac hypertrophy and heart failure. Circ Res 2003;93:69–76. [8] Galigniana MD, Erlejman AG, Monte M, Gomez-Sanchez C, Piwien-Pilipuk G. The hsp90-FKBP52 complex links the mineralocorticoid receptor to motor proteins and persists bound to the receptor in early nuclear events. Mol Cell Biol 2010;30:1285–98. [9] Hernandez-Diaz I, Giraldez T, Arnau MR, Smits VAJ, Jaisser F, Farman N, et al. The mineralocorticoid receptor is a constitutive nuclear factor in cardiomyocytes due to hyperactive nuclear localization signals. Endocrinology 2010;151:3888–99. [10] Nishi M, Ogawa H, Ito T, Matsuda Ki, Kawata M. Dynamic changes in subcellular localization of mineralocorticoid receptor in living cells. In comparison with glucocorticoid receptor using dual-color labeling with green fluorescent protein spectral variants. Mol Endocrinol 2001;15:1077–92. [11] Pearce D, Naray-Fejes-Toth A, Fejes-Toth G. Determinants of subnuclear organization of mineralocorticoid receptor characterized through analysis of wild type and mutant receptors. J Biol Chem 2002;277:1451–6. [12] Gallo LI, Ghini AA, Piwien Pilipuk G, Galigniana MD. Differential recruitment of tetratricorpeptide repeat domain immunophilins to the mineralocorticoid receptor influences both heat-shock protein 90-dependent retrotransport and hormone-dependent transcriptional activity. Biochemistry 2007;46:14044–57. [13] Galigniana MD, Piwien Pilipuk G, Kanelakis KC, Buton G, Lantos CP. Molecular mechanism of activation and nuclear translocation of the mineralocorticoid receptor upon binding of pregnanesteroids. Mol Cell Endocrinol 2004;217:167–79. [14] Bamberger CM, Wald M, Bamberger AM, Schulte HM. Inhibition of mineralocorticoid and glucocorticoid receptor function by the heat shock protein 90-binding agent geldanamycin. Mol Cell Endocrinol 1997; 131:233–40. [15] Grossmann C, Ruhs S, Langenbruch L, Mildenberger S, Str+ñtz N, Schumann K. Nuclear shuttling precedes dimerization in mineralocorticoid receptor signaling. Chem Biol 2012;19:742–51. [16] Csermely P, Schnaider T, Soti C, Prohaszka Z, Nardai G. The 90-kDa molecular chaperone family: structure, function, and clinical applications. Compr Rev Pharmacol Ther 1998;79:129–68. [17] Galigniana MD, Scruggs JL, Herrington J, Welsh MJ, Carter-Su C, Housley PR, et al. Heat shock protein 90-dependent (geldanamycin-inhibited) movement of the glucocorticoid receptor through the cytoplasm to the nucleus requires intact cytoskeleton. Mol Endocrinol 1998;12:1903–13. [18] Galigniana MD, Echeverria PC, Erlejman AG, Piwien-Pilipuk G. Role of molecular chaperones and TPR-domain proteins in the cytoplasmic transport of steroid receptors and their passage through the nuclear pore. Nucleus 2010;1:299–308. [19] Wochnik GM, Rüegg JL, Abel GA, Schmidt U, Holsboer F, Rein T. FK506-binding proteins 51 and 52 differentially regulate dynein interaction and nuclear translocation of the glucocorticoid receptor in mammalian cells. J Biol Chem 2005;280:4609–16. [20] Walther RF, Atlas E, Carrigan A, Rouleau Y, Edgecombe A, Visentin L, et al. A serine/threonine-rich motif is one of three nuclear localization signals that determine unidirectional transport of the mineralocorticoid receptor to the nucleus. J Biol Chem 2005;280:17549–61.
5
[21] Tanaka M, Nishi M, Morimoto M, Sugimoto T, Kawata M. Imaging analysis of mineralocorticoid receptor and importins in single living cells by using GFP color variants. Cell Tissue Res 2005;320:447–53. [22] Nishi M, Tanaka M, Matsuda Ki, Sunaguchi M, Kawata M. Visualization of glucocorticoid receptor and mineralocorticoid receptor interactions in living cells with GFP-based fluorescence resonance energy transfer. J Neurosci 2004;24:4918–27. [23] Trapp T, Holsboer F. Heterodimerization between mineralocorticoid and glucocorticoid receptors increases the functional diversity of corticosteroid action. Trends Pharmacol Sci 1996;17:145–9. [24] Savory JG, Prefontaine GG, Lamprecht C, Liao M, Walther RF, Lefebvre YA, et al. Glucocorticoid receptor homodimers and glucocorticoid–mineralocorticoid receptor heterodimers form in the cytoplasm through alternative dimerization interfaces. Mol Cell Biol 2001;21:781–93. [25] Sun Y, Zhang J, Lu L, Chen SS, Quinn MT, Weber KT. Aldosterone-induced inflammation in the rat heart – role of oxidative stress. Am J Pathol 2002; 161:1773–81. [26] Park JB, Schiffrin EL. Cardiac and vascular fibrosis and hypertrophy in aldosterone-infused rats: role of endothelin-1. Am J Hypertens 2002; 15:164–9. [27] Shibata S, Mu S, Kawarazaki H, Muraoka K, Ishizawa Ki, Yoshida S, et al. Rac1 GTPase in rodent kidneys is essential for salt-sensitive hypertension via a mineralocorticoid receptor-dependent pathway. J Clin Investig 2011; 121:3233–43. [28] Shibata S, Nagase M, Yoshida S, Kawarazaki W, Kurihara H, Tanaka H, et al. Modification of mineralocorticoid receptor function by Rac1 GTPase: implication in proteinuric kidney disease. Nat Med 2008;14:1370–6. [29] Nagase M, Ayuzawa N, Kawarazaki W, Ishizawa K, Ueda K, Yoshida S, et al. Oxidative stress causes mineralocorticoid receptor activation in rat cardiomyocytes: role of small GTPase Rac1. Hypertension 2012;59:500–6. [30] Kawarazaki W, Nagase M, Yoshida S, Takeuchi M, Ishizawa K, Ayuzawa N, et al. Angiotensin II- and salt-induced kidney injury through Rac1-mediated mineralocorticoid receptor activation. J Am Soc Nephrol 2012;23:997–1007. [31] Nagata K, Obata K, Xu J, Ichihara S, Noda A, Kimata H, et al. Mineralocorticoid receptor antagonism attenuates cardiac hypertrophy and failure in lowaldosterone hypertensive rats. Hypertension 2006;47:656–64. [32] Hordijk PL. Regulation of NADPH oxidases: the role of rac proteins. Circ Res 2006;98:453–62. [33] Ojeda-Cervantes M, Barrera-Chimal J, Alberu J, Perez-Villalva R, MoralesBuenrostro L, Bobadilla NA. Mineralocorticoid receptor blockade reduced oxidative stress in renal transplant recipients: a double-blind, randomized pilot study. Am J Nephrol 2013;37:481–90. [34] Di Zhang A, Nguyen Dinh CA, Soukaseum C, Escoubet B, Cherfa A, Messaoudi S, et al. Cross-talk between mineralocorticoid and angiotensin II signaling for cardiac remodeling. Hypertension 2008;52:1060–7. [35] Pfau A, Grossmann C, Freudinger R, Mildenberger S, Benesic A, Gekle M. Ca2+ but not H2O2 modulates GRE-element activation by the human mineralocorticoid receptor in HEK cells. Mol Cell Endocrinol 2007;264:35–43. [36] Piwien-Pilipuk G, Ayala A, Machado A, Galigniana MD. Impairment of mineralocorticoid receptor (MR)-dependent biological response by oxidative stress and aging. Correlation with post-translational modification of MR and decreased ADP-ribosylatable level of elongation factor 2 in kidney cells. J Biol Chem 2002;277:11896–903. [37] Piwien-Pilipuk G, Galigniana MD. Oxidative stress induced by L-buthionine-(S, R)-sulfoximine, a selective inhibitor of glutathione metabolism, abrogates mouse kidney mineralocorticoid receptor function. Biochim Biophys Acta (BBA) – Mol Cell Res 2000;1495:263–80. [38] Souque A, Fagart J, Couette B, Rafestin-Oblin M. Sulfhydryl groups are involved in the binding of agonists and antagonists to the human mineralocorticoid receptor. J Steroid Biochem Mol Biol 1996;57:315–21. [39] Ruhs S, Strätz N, Schlör K, Meinel S, Mildenberger S, Rabe S, et al. Modulation of transcriptional mineralocorticoid receptor activity by nitrosative stress. Free Radical Biol Med 2012;53:1088–100. [40] Funder JW. RALES, EPHESUS and redox. J Steroid Biochem Mol Biol 2005;93:121–5. [41] Tsugita M, Iwasaki Y, Nishiyama M, Taguchi T, Shinahara M, Taniguchi Y, et al. Differential regulation of 11beta-hydroxysteroid dehydrogenase type-1 and -2 gene transcription by proinflammatory cytokines in vascular smooth muscle cells. Life Sci 2008;83:426–32. [42] Alnemri ES, Maksymowych AB, Robertson NM, Litwack G. Overexpression and characterization of the human mineralocorticoid receptor. J Biol Chem 1991;266:18072–81. [43] Galigniana MD. Native rat kidney mineralocorticoid receptor is a phosphoprotein whose transformation to a DNA-binding form is induced by phosphatases. Biochem J 1998;333:555–63. [44] Marissal-Arvy N, Lombes M, Petterson J, Moisan MP, Mormede P. Gain of function mutation in the mineralocorticoid receptor of the brown Norway rat. J Biol Chem 2004;279:39232–9. [45] Le Moellic C, Ouvrard-Pascaud A, Capurro C, Cluzeaud F, Fay M, Jaisser F, et al. Early nongenomic events in aldosterone action in renal collecting duct cells: PKCalpha activation, mineralocorticoid receptor phosphorylation, and crosstalk with the genomic response. J Am Soc Nephrol 2004;15:1145–60. [46] Massaad C, Houard N, Lombes M, Barouki R. Modulation of human mineralocorticoid receptor function by protein kinase A. Mol Endocrinol 1999;13:57–65.
Please cite this article in press as: Gekle M et al. Rapid mineralocorticoid receptor trafficking. Steroids (2013), http://dx.doi.org/10.1016/ j.steroids.2013.10.016
6
M. Gekle et al. / Steroids xxx (2013) xxx–xxx
[47] Grossmann C, Benesic A, Krug AW, Freudinger R, Mildenberger S, Gassner B. Human mineralocorticoid receptor expression renders cells responsive for nongenotropic aldosterone actions. Mol Endocrinol 2005;19:1697–710. [48] Faresse N, Vitagliano JJ, Staub O. Differential ubiquitylation of the mineralocorticoid receptor is regulated by phosphorylation. FASEB J 2012;26:4373–82. [49] Vienchareun S, Le Menuet D, Martinerie L, Munier M, Pascual-Le Tallec L, Lombes M. The mineralocorticoid receptor: insights into its molecular and (patho)physiological biology. Nucl Recept Signal 2007;30:e012. [50] Pascual-Le Tallec L, Lombes M. The mineralocorticoid receptor: a journey exploring its diversity and specificity of action. Mol Endocrinol 2005;19:2211–21. [51] Tirard M, Almeida OFX, Hutzler P, Melchior F, Michaelidis TM. Sumoylation and proteasomal activity determine the transactivation properties of the mineralocorticoid receptor. Mol Cell Endocrinol 2007;268:20–9. [52] Pascual-Le Tallec L, Kirsh O, Lecomte MC, Viengchareun S, Zennaro MC, Dejean A, et al. Protein inhibitor of activated signal transducer and activator of
transcription 1 interacts with the N-terminal domain of mineralocorticoid receptor and represses its transcriptional activity: implication of small ubiquitin-related modifier 1 modification. Mol Endocrinol 2003;17:2529–42. [53] Nagata D, Takahashi M, Sawai K, Tagami T, Usui T, Shimatsu A, et al. Molecular mechanism of the inhibitory effect of aldosterone on endothelial NO synthase activity. Hypertension 2006;48:165–71. [54] Nakano S, Kobayashi N, Yoshida K, Ohno T, Matsuoka H. Cardioprotective mechanisms of spironolactone associated with the angiotensin-converting enzyme/epidermal growth factor receptor/extracellular signal-regulated kinases, NAD(P)H oxidase/lectin-like oxidized low-density lipoprotein receptor-1, and Rho-kinase pathways in aldosterone/salt-induced hypertensive rats. Hypertens Res 2005;28:925–36. [55] Stas S, Whaley-Connell A, Habibi J, Appesh L, Hayden MR, Karuparthi PR, et al. Mineralocorticoid receptor blockade attenuates chronic overexpression of the renin–angiotensin–aldosterone system stimulation of reduced nicotinamide adenine dinucleotide phosphate oxidase and cardiac remodeling. Endocrinology 2007;148:3773–80.
Please cite this article in press as: Gekle M et al. Rapid mineralocorticoid receptor trafficking. Steroids (2013), http://dx.doi.org/10.1016/ j.steroids.2013.10.016
