| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Physiology, University of Tuebingen, Tuebingen, Germany; and Departments of Medicine and Pharmacology, University of California San Diego and San Diego Veterans Affairs Healthcare System, San Diego, California
ABSTRACT I. INTRODUCTION II. REGULATION OF SERUM- AND GLUCOCORTICOID-INDUCIBLE PROTEIN KINASES A. Genomic Control of SGK Gene Transcription B. Regulation of SGK Kinase Activity C. Degradation of SGKs III. ROLE OF SERUM- AND GLUCOCORTICOID-INDUCIBLE PROTEIN KINASES IN THE REGULATION OF MOLECULAR AND CELLULAR FUNCTIONS A. Ion Channels 1. Epithelial Na+ channel ENaC 2. Renal outer medullary K+ channel ROMK1 3. Renal epithelial Ca2+ channel TRPV5 4. Renal and inner ear Cl– channel ClC-Ka 5. Ubiquitous Cl– channel ClC2 6. Cystic fibrosis transmembrane conductance regulator Cl– channel CFTR 7. Cardiac voltage-gated Na+ channel SCN5A 8. Cardiac and epithelial K+ channels KCNE1/KCNQ1 9. Inner ear K+ channels KCNQ4 10. Voltage-gated K+ channels Kv1.3, Kv1.5, and Kv4.3 11. Cation channels created by 4F2/LAT 12. Glutamate receptors B. Carriers and Pumps 1. Na+/H+ exchanger NHE3 2. Glucose transporters 3. Amino acid transporters 4. Na+-dicarboxylate cotransporter NaDC-1 5. Creatine transporter CreaT 6. Phosphate carrier 7. Na+-K+-ATPase C. Enzyme Activities D. Transcription Factors E. Release of Hormones F. Regulation of Cell Volume G. Cell Proliferation and Apoptosis H. Other Functions IV. ROLE OF SERUM- AND GLUCOCORTICOID-INDUCIBLE PROTEIN KINASES IN THE REGULATION OF INTEGRATED FUNCTIONS A. Metabolism B. Kidney Function and Salt Appetite C. Gastrointestinal Function D. Other Epithelia-Containing Tissues E. Cardiovascular Function F. Neuromuscular Function G. Other Functions H. Phenotype of Gene-Targeted Mice Lacking SGK1, SGK3, or Both V. PATHOPHYSIOLOGICAL SIGNIFICANCE OF THE SERUM- AND GLUCOCORTICOID-INDUCIBLE PROTEIN KINASES A. Tumor Growth B. Hypertension, Obesity, and the Metabolic Syndrome C. Neuronal Disease D. Diseases Involving Fibrosis E. Ischemia VI. CONCLUSIONS AND PERSPECTIVES GRANTS ACKNOWLEDGMENTS REFERENCES
 |
ABSTRACT |
|---|
 Top NextReferences |
|---|
-catenin, nuclear factor 
B). SGKs participate in the regulation of transport, hormone release, neuroexcitability, cell proliferation, and apoptosis. SGK1 contributes to Na+ retention and K+ elimination of the kidney, mineralocorticoid stimulation of salt appetite, glucocorticoid stimulation of intestinal Na+/H+ exchanger and nutrient transport, insulin-dependent salt sensitivity of blood pressure and salt sensitivity of peripheral glucose uptake, memory consolidation, and cardiac repolarization. A common (
5% prevalence) SGK1 gene variant is associated with increased blood pressure and body weight. SGK1 may thus contribute to metabolic syndrome. SGK1 may further participate in tumor growth, neurodegeneration, fibrosing disease, and the sequelae of ischemia. SGK3 is required for adequate hair growth and maintenance of intestinal nutrient transport and influences locomotive behavior. In conclusion, the SGKs cover a wide variety of physiological functions and may play an active role in a multitude of pathophysiological conditions. There is little doubt that further targets will be identified that are modulated by the SGK isoforms and that further SGK-dependent in vivo physiological functions and pathophysiological conditions will be defined. |
I. INTRODUCTION |
|---|
|
TopPreviousNextReferences |
|---|
The gene encoding human SGK1 has been localized to chromosome 6q23 (350), whereas the gene encoding SGK2 is localized to chromosome 20q12 (182). SGK3 is identical to the cytokine-independent survival kinase CISK (203). The "SGK-like" gene (SGKL) in chromosome 8q12.3 (81) encodes a protein whose predicted amino acid sequence is virtually identical to that of human SGK3.
SGK kinases are expressed in a wide variety of species including shark (348) and Caenorhabditis elegans (142). Yeast express two orthologs, Ypk1 and Ypk2, which are involved in endocytosis (87) and required for survival (60). Yeast lacking Ypk1 and Ypk2 can be rescued by mammalian SGK1 (60).
In mammals, SGK1 is expressed in virtually all tissues tested (349). However, transcript levels vary profoundly among different cell types of any given tissue, such as brain (127, 236, 304, 318, 359), eye (260, 261, 283), lung (360), kidney (65, 118, 147, 184, 206), liver (110), intestine (352), pancreas (170), and ovary (7). Moreover, typical expression patterns are found during embryonic development (73, 152, 193). The subcellular localization of SGK1 may depend on the functional state of the cell. Activation of SGK1 after exposure of cells to serum has been suggested to trigger importin-
-mediated entry of SGK1 into the nucleus (208), whereas activation by hyperosmotic shock or glucocorticoids enhances cytosolic localization of the kinase (112).
SGK3 is expressed in all tissues tested thus far (172) and is particularly high in the embryo (152, 193) and adult heart and spleen (172). Expression of SGK2 is most abundant in epithelial tissues including kidney, liver, pancreas, and presumably choroid plexus of the brain (172). The subcellular distribution may be nuclear and cytoplasmic, as SGK2 and SGK3 contain a similar nuclear localization signal sequence as SGK1 (112).
The functional significance of SGK1, SGK2, and SGK3 is still far from understood. Notably, all three kinases are potent regulators of ion channel activity, transport, and transcription (30, 111, 183, 186, 250, 305, 331, 378). Functional analysis of gene-targeted mice lacking SGK1 (368) and SGK3 (214) provided insight into the functional significance of SGK1- and SGK3-dependent regulation of physiological functions. Interestingly, neither knockout of SGK1 (368) or SGK3 (214), nor knockout of both SGK1 and SGK3 (133) leads to a severe phenotype, suggesting that neither SGK1 nor SGK3 is required for survival. Closer inspection of the physiology of those mice discloses, however, multiple physiological deficits pointing to the broad functional role of these kinases.
The following review attempts to delineate the current knowledge on the physiological and pathophysiological significance of these interesting kinases. Our knowledge on each of the kinases is far from complete. Particularly our knowledge on the physiological significance of SGK2 and SGK3 is presently, at best, fragmentary.
We first describe the mechanisms regulating expression, activity, and degradation of the SGKs (sect. II). Subsequently, we outline SGK-dependent regulation of molecular and cellular functions (sect. III). Where known, consequences on integrated function are outlined (sect. IV). Finally, we discuss the known and putative pathophysiological relevance of the SGKs (sect. V).
|
II. REGULATION OF SERUM- AND GLUCOCORTICOID-INDUCIBLE PROTEIN KINASES |
|---|
|
TopPreviousNextReferences |
|---|
Transcription of SGK1 is upregulated by both serum and glucocorticoids (49, 78, 96, 112, 162, 167, 221, 232, 289, 361, 362, 388). Several other hormones and mediators stimulate SGK1 transcription, including mineralocorticoids (29, 49, 65, 96, 135, 168, 197, 206, 231, 290), gonadotropins (68, 129, 263, 264, 278), 1,25-dihydroxyvitamin D3 [1,25(OH)2D3] (4), transforming growth factor- (TGF-) (178, 184, 352), interleukin-6 (216), fibroblast and platelet-derived growth factor (222), thrombin (23), endothelin (366), as well as other cytokines (80, 182, 330). Moreover, activation of peroxisome proliferator-activated receptor 
(PPAR) stimulates SGK1 gene transcription (146). The human isoform has been identified as a cell volume-regulated gene that is transcriptionally upregulated by cell shrinkage (349). In renal epithelial (A6) cells, SGK1 expression is stimulated by cell swelling rather than cell shrinkage (272). SGK1 transcription is further stimulated by excessive glucose concentrations (184, 275), heat shock, ultraviolet (UV) radiation, and oxidative stress (198).
SGK1 transcription is inhibited by heparin (90) and by mutations in the gene MECP2, which underlies Rett syndrome (RTT), a disorder with severe mental retardation (237). Intestinal SGK1 transcription is further regulated by dietary iron (212).
Signaling molecules involved in the transcriptional regulation of SGK1 include protein kinase C (184, 222), the protein kinase Raf (222), mammalian mitogen-activated protein kinase (BMK1) (137), mitogen-activated protein kinase (MKK1) (85, 222), stress-activated protein kinase-2 (SAPK2, p38 kinase) (24, 64, 351), phosphatidylinositol (PI) 3-kinase (129), cAMP (129, 170), and p53 (207, 209). Moreover, SGK1 transcription is stimulated by an increased cytosolic Ca2+ concentration (170) and by nitric oxide (321).
SGK1 transcript levels are increased by ischemia of brain (236) and kidney (108). SGK1 expression is decreased during rejection of transplanted kidneys (329). Leukocyte SGK1 transcript levels are enhanced by treatment with dialysis (116).
A striking increase of SGK1 expression is observed during wound healing (164) and in fibrosing tissue, such as diabetic nephropathy (178, 184), glomerulonephritis (118), liver cirrhosis (110), fibrosing pancreatitis (170), Crohn's disease (352), lung fibrosis (360), and cardiac fibrosis (323). SGK1 gene transcription is stimulated by DNA damage through p53 and activation of extracellular signal-regulated kinase (ERK1/2) (222, 375), and is also upregulated after neuronal injury (159), neuronal excitotoxicity (145), and neuronal challenge by exposure to microgravity (84).
The promoter of the rat SGK1 gene carries several putative and confirmed transcription factor binding sites including those for the glucocorticoid receptor (GR), the mineralocorticoid receptor (MR), the progesterone receptor (PR), the vitamin D receptor (VDR), the retinoid X receptor (RXR), the farnesoid X receptor (FXR), the sterol regulatory element binding protein (SREBP), PPAR, the cAMP response element binding protein (CREB), the p53 tumor suppressor protein, the Sp1 transcription factor, the activating protein 1 (AP1), the activating transcription factor 6 (ATF6), the heat shock factor (HSF), reticuloendotheliosis viral oncogene homolog (c-Rel) and nuclear factor B (NFB), signal transducers and activators of transcription (STAT), the TGF--dependent transcription factors SMAD3 and SMAD4, and forkhead activin signal transducer (FAST) (112).
The regulation of SGK1 transcript levels is fast; appearance and disappearance of SGK1 mRNA require <20 min (349). Little is known about genomic regulation of SGK2 and SGK3, which appear to be less sensitive to hormonal regulation than SGK1 (182).
B. Regulation of SGK Kinase Activity
To become functional, the SGK protein kinases require activation by phosphorylation, which is accomplished through a signaling cascade involving PI 3-kinase, the 3-phosphoinositide (PIP3)-dependent kinase PDK1, and a yet unidentified but also PIP3-dependent kinase that has been referred to as PDK2 or "hydrophobic motif" (H-motif) kinase (5, 31, 75, 119, 171, 224, 235, 249, 369). PIP3 is degraded by the phosphatase and tensin homolog PTEN (202, 240, 309), which thus disrupts PI 3-kinase-dependent activation of the SGKs.
Maximal stimulation of SGK1 activity requires the PDK1-dependent phosphorylation at 256Thr within the activation loop (T-loop) and phosphorylation at 422Ser in the hydrophobic motif at its COOH terminus by PDK2/H-motif kinase (171, 172, 249). The PDK1-mediated SGK1 phosphorylation is facilitated when 422Ser is already phosphorylated (Fig. 1). Phosphorylation of SGK1 at 422Ser promotes SGK1 binding to the PDK1 interacting fragment (PIF)-binding pocket and phosphorylation at 256Thr by PDK1 (31). An alternate mechanism of SGK1 activation by PDK1 involves the scaffold protein Na+/H+ exchanger regulating factor 2 (NHERF2). NHERF2 mediates the assembly of SGK1 and PDK1 via its PDZ domains and PIF consensus sequence (70). NHERF2 interacts with the PDZ binding motif of SGK1 through its first PDZ domain and with PIF-binding pocket of PDK1 through its PIF tail. The formation of the ternary complex facilitates the phosphorylation of SGK1 on 256Thr in its T-loop by PDK1 (70).
|
SGK2 and SGK3 may similarly be activated by PDK1 and PDK2/H-motif kinase. The equivalent phosphorylation sites for SGK2 and SGK3 are predicted to be at 193Thr/356Ser and 253Thr/419Ser, respectively, but this requires further investigation. The kinases are also regulated by WNK1, although to a lesser extent than SGK1 (371).
Replacement of the serine at position 422 by aspartate in the human SGK1 leads to the constitutively active S422DSGK1 (172), whereas replacement of lysine at position 127, within the ATP-binding region required for enzymatic activity, with asparagine leads to the inactive K127NSGK1 (172). Analogous mutations in the human SGK2 and SGK3 lead to the constitutively active S356DSGK2 and S419DSGK3 and the constitutively inactive K64NSGK2 and K191NSGK3 (41).
In part through the PI 3-kinase pathway, SGK1 is activated by insulin (171, 254), IGF-I (137, 171, 179), hepatic growth factor (HGF) (287), and follicle stimulating hormone (FSH) (265).
SGK1 can be activated by bone marrow kinase/extracellular signal-regulated kinase 5 (BK/ERK5) or by p38. The kinases do not phosphorylate SGK1 at 256Thr but at 78Ser, which is outside the catalytic domain (137, 216). How this phosphorylation activates SGK1 is not known. SGK1 can also be activated by an increase of cytosolic Ca2+ activity, an effect presumably mediated by calmodulin-dependent protein kinase kinase (CaMKK) (158). Moreover, the small G protein Rac1 activates SGK1 via a PI 3-kinase-independent pathway (287). Additional activators of SGK1 include neuronal depolarization (179), cAMP (179, 254, 315), lithium (179), oxidation (171, 256), and adhesion to fibronectin (287).
Similar to SGK1, SGK2 and SGK3 are activated by oxidation, insulin, and IGF-I through a signaling cascade involving PI 3-kinase as well as PDK1 and PDK2/H-motif kinase (171, 335).
SGK1 is rapidly degraded, with a half-life of 30 min (50). Ubiquitination of SGK1 labels the kinase for degradation by the proteasome (50). SGK1 degradation may be mediated by the ubiquitin ligase Nedd4–2 (neuronal precursor cells expressed developmentally downregulated) (386). Nedd4–2 contains a series of tryptophan-rich sequences (WW motifs) that interact with a proline-tyrosine PY motif present in its target proteins. SGK1 bears such a PY motif. Overexpression of Nedd4–2 decreases steady-state levels of SGK1 in a dose-dependent manner by increasing SGK1 ubiquitination (presumably within the first 60 NH2-terminal amino acids) and subsequent degradation in the 26S proteasome. Conversely, silencing of Nedd4–2 by RNA interference, or loss of the NH2-terminal amino acids, abrogates the ubiquitination and thus increases the half-life of SGK1. The effect of Nedd4–2 apparently requires phosphorylation of the ubiquitin ligase by SGK1, as SGK1 degradation is reduced by a phosphorylation site-deficient Nedd4–2 mutant (Nedd4–2S/T-A) or by SGK1 inhibition (Fig. 1). Accordingly, active SGK1 favors its own degradation, thus contributing to the limitation of its action (386).
|
III. ROLE OF SERUM- AND GLUCOCORTICOID-INDUCIBLE PROTEIN KINASES IN THE REGULATION OF MOLECULAR AND CELLULAR FUNCTIONS |
|---|
|
TopPreviousNextReferences |
|---|
, where X stands for any amino acid, R for arginine, and indicates a hydrophobic amino acid (31, 75, 119, 171, 205, 224, 235, 249). It shares the consensus sequence with PKB/Akt and with the SGK2 and SGK3 isoforms. By direct phosphorylation of effector molecules or by regulators thereof, SGK1 modifies a variety of cellular functions in virtually every organ. It should be kept in mind that most experiments demonstrating SGK1-dependent phosphorylation, or regulation of target proteins, have employed overexpression of the kinase, and the results do not necessarily reflect the physiological function of the kinase. As a matter of fact, the only targets hitherto shown to be exclusively phosphorylated by SGK1 are N-myc downregulated genes NDRG1 and NDRG2 (227, 229), proteins with ill-defined function. Thus the regulation of most physiological functions listed below involves but is not exclusively accomplished by the SGK isoforms. SGK1-dependent regulation may become particularly important following genomic upregulation of the kinase, e.g., under glucocorticoid or mineralocorticoid excess, an assumption supported by observations in SGK1 knockout animals.
1. Epithelial Na+ channel ENaC
The potent transcriptional regulation of SGK1 by mineralocorticoids (29, 49, 65, 95, 128, 206, 213, 226, 231, 290, 316) suggests a role of this kinase in the control of epithelial Na+ transport. A key channel in the mineralocorticoid regulation of renal tubular Na+ reabsorption is epithelial Na+ channel (ENaC) (59, 206) (Fig. 1). As described elsewhere in more detail (250, 331), SGK1 interacts with ENaC (355), and coexpression of SGK1 indeed markedly enhances the activity of ENaC heterologously expressed in Xenopus oocytes (10, 44, 65, 67, 184, 231, 341, 345), cortical collecting duct cells (138, 233), and A6 cells (11, 13, 105).
SGK1 has been suggested to modulate ENaC activity directly by channel phosphorylation (91). On the other hand, overexpression of constitutively active Xenopus S425DSGK1 in A6 cells did not significantly increase the phosphorylation of any of the three subunits of ENaC (13). Thus the effect of SGK1 on ENaC phosphorylation requires further experimental support. In any case, SGK1 may regulate ENaC indirectly by phosphorylating the ubiquitin ligase Nedd4–2 (17, 86, 113, 243, 298, 299, 341, 386), which otherwise ubiquitinates ENaC thus preparing the channel protein for clearance from the cell membrane and subsequent degradation (302) (Fig. 1). The phosphorylation of Nedd4–2 decreases the ability of Nedd4–2 to ubiquitinate ENaC, and SGK1 thus increases the abundance of ENaC channel protein within the cell membrane (10, 165, 184, 345). Mutations of Nedd4–2 affecting the PY motif of ENaC may disrupt Nedd4–2 dependent degradation of the channel and thus lead to excessive renal salt reabsorption and arterial hypertension (3, 94, 131, 166, 281, 303). The naturally occurring human P355LNedd4–2 variant, found in patients with end-stage renal disease (ESRD) that suffered from arterial hypertension, was shown to be more sensitive to phosphorylation by SGK1 and, accordingly, to exert a weaker negative effect on ENaC (114).
SGK1-mediated phosphorylation of Nedd4–2 induces its interaction with members of the 14–3-3 family of regulatory proteins. Interaction of Nedd4–2 with 14–3-3 inhibits Nedd4–2-dependent ubiquitination of its targets (28, 156) (Fig. 1). Overexpression of the chaperone 14–3-3 decreases, whereas expression of a transdominant inhibitory mutant of 14–3-3 increases the Nedd4–2-dependent ubiquitination of ENaC (28, 156). The prominent expression of 14–3-3 proteins in collecting duct principal cells suggests a physiological role for these proteins in the regulation of distal nephron Na+ transport (28). It should be pointed out that phosphorylation of Nedd4–2 is not an exclusive function of SGK1, as it is observed even in the absence of the kinase (113) and upon coexpression of SGK3 in Xenopus oocytes (243). Although SGK1 shares the consensus sequence with its isoforms SGK2, SGK3, and the related kinase PKB/Akt, the phosphorylation efficiency differs between the kinases. This was shown in expression studies in Xenopus oocytes, which revealed that SGK1 and SGK3 phosphorylate Nedd4–2 more efficiently than SGK2 (243). Nedd4–2 is likewise phosphorylated and thereby inhibited by cAMP-dependent protein kinase (PKA) (298). The SGK1 consensus sequence overlaps with a PKA consensus sequence (R-X-X-S/T) and, in fact, the Nedd4–2 residues found to be phosphorylated by PKA (221Ser, 246Thr, and 327Ser) are also phosphorylated by SGK1. Nedd4–2 contains additional PKA motifs that might be phosphorylated but are not responsible for the inhibition of Nedd4–2 by PKA (298). The relative functional contribution of the three phosphorylation sites to PKA- and SGK1-dependent regulation of Nedd4–2 activity is not identical (298). Similar to SGK1, PKA altered ENaC surface expression by modulating the function of Nedd4–2 (298). Thus PKA and SGK1 might regulate ENaC in part through convergent phosphorylation of Nedd4–2 (298).
Beyond its effect on ENaC activity and Nedd4–2 phosphorylation, SGK1 stimulates ENaC transcription (47) (Fig. 1). Moreover, SGK1 may inhibit inducible nitric oxide synthase, and the decreased formation of nitric oxide may then disinhibit ENaC (139).
In the Xenopus oocyte system, the effect of SGK1 expression on ENaC is shared by SGK2 and SGK3 (117), which may, however, not be relevant for regulation of renal salt excretion (see below).
2. Renal outer medullary K+ channel ROMK1
Aldosterone stimulates not only the renal reabsorption of Na+, but also the renal excretion of K+ (347). Renal tubular K+ secretion has been thought to be accomplished by ROMK1 (6, 126, 347, 357), which is, similar to ENaC and SGK1, expressed in the principal cells of the aldosterone-sensitive distal nephron (342) (Fig. 1). Thus ROMK1 appeared to be an obvious target for regulation by SGK1. Indeed, ROMK1 is upregulated by coexpression of SGK1 in Xenopus oocytes, an effect in part due to enhanced channel protein abundance at the plasma membrane (245, 374, 380). However, full effect of SGK1 on ROMK1 requires the NHERF2 (245, 380), a protein that allows trafficking of carriers and channels to the cell membrane (288, 378) and that is similarly expressed in principal cells of the aldosterone-sensitive distal nephron (288). Coexpression of ROMK1 together with NHERF2, but neither SGK1 nor NHERF2 alone, leads to profound upregulation of ROMK1 channel activity (380). Presumably, the binding of ROMK1 to NHERF2 involves a postsynaptic density 95, disk large, zona occludens-1 (PDZ)-binding motif at the COOH terminus of ROMK1 (245). The ENaC sequence does not include a PDZ-binding motif, suggesting that NHERF2 cannot directly interact with ENaC. On the other hand, the sequence of ROMK1 does not include a PY motif that is thought to be required for binding of Nedd4–2 to its targets (86). Besides its effect on ROMK1 surface abundance, SGK1 directly phosphorylates ROMK1 (244, 374, 380). Phosphorylation introduces a negative charge and leads to a marked shift of ROMK1's pH sensitivity, resulting in enhanced activity of the channel at the usual cytosolic pH (244).
3. Renal epithelial Ca2+ channel TRPV5
As shown in Xenopus oocytes, SGK1 and SGK3 activate the renal epithelial Ca2+ channel TRPV5 by enhancing channel abundance in the plasma membrane, an effect again requiring cooperation with NHERF2 (101, 246). The TRPV5 C-tail interacts in a Ca2+-independent manner with NHERF2. Deletion of the second, but not the first, PDZ domain in NHERF2 abrogates the stimulating effect of SGK1 on TRPV5 protein abundance (246).
4. Renal and inner ear Cl– channel ClC-Ka
In Xenopus oocytes SGK1, SGK2, and SGK3 upregulate the Cl– channel complex ClC-Ka/barttin (99). Similar to the mechanism involved in SGK1-dependent regulation of ENaC, SGK1 enhances ClC-Ka/barttin protein abundance and activity in part by phosphorylation of Nedd4–2. The ubiquitin ligase downregulates ClC-Ka/barttin, an effect abrogated by a mutation of barttin (Y98Abarttin) that eliminates the PY motif. However, the Y98Abarttin mutation blunts, but does not abrogate, the regulation of ClC-Ka by SGK1, pointing to Nedd4–2-independent mechanisms of ClC-Ka/barttin regulation by SGK1.
5. Ubiquitous Cl– channel ClC2
In Xenopus oocytes SGK1, SGK2, and SGK3 enhance the activity and surface abundance of the cell volume-regulated Cl– channel ClC2 (242). The kinases do not modulate the activity of the channel directly, as the deletion of the SGK1 phosphorylation site in the ClC2 protein does not abrogate SGK1-dependent stimulation of the channel. Instead, SGKs stimulate ClC2 by inhibiting Nedd4–2, leading to reduced clearance of ClC2 protein from the cell membrane.
6. Cystic fibrosis transmembrane conductance regulator Cl– channel CFTR
In Xenopus oocytes, SGK1 activates the Cl– currents induced by the cystic fibrosis transmembrane conductance regulator (CFTR) (345). Given the marked upregulation of SGK1 in inflammatory lung disease (360), this effect could participate in the stimulation of bronchial Cl– secretion and/or reabsorption during lung infection.
7. Cardiac voltage-gated Na+ channel SCN5A
In Xenopus oocytes, both SGK1 and SGK3 stimulate the cardiac voltage-gated Na+ channel SCN5A (42). SCN5A has previously been shown to be downregulated by Nedd4 (2). Accordingly, inhibition of Nedd4 most likely participates in the regulation of the channel by SGKs. However, the kinases modify the gating kinetics of SCN5A channels, an effect which cannot be explained by delayed retrieval from the cell membrane (42). At this stage it is not clear whether SGK1 phosphorylates SCN5A. Interestingly, replacement by site-directed mutagenesis of the serine at the putative SGK consensus sequence by alanine leads to an opposite shift of the gating properties of the channel (42).
8. Cardiac and epithelial K+ channels KCNE1/KCNQ1
The K+ channel complex KCNE1/KCNQ1 is expressed in several tissues including heart and kidney (311). It activates slowly upon depolarization (311). As has become apparent from heterologous expression studies in Xenopus oocytes, the current is markedly upregulated by coexpression with SGK1, SGK2, and/or SGK3 (100). Again, part of the effect is due to phosphorylation of Nedd4–2 (100). The effects of SGK1 on KCNE1/KCNQ1 are expected to shorten the cardiac action potential and may influence transport in KCNE1-/KCNQ1-expressing epithelia (see below).
9. Inner ear K+ channels KCNQ4
KCNQ4 is expressed in inner and outer hair cells of the inner ear where it determines electrical excitability. The critical role of KCNQ4 in hearing function is underlined by the fact that loss-of-function mutations of the KCNQ4 gene cause hearing loss (148, 176, 211). SGK1 significantly increases peak KCNQ4 activity and also modulates the influence of previous channel activation (284). These effects were sensitive to mutations in the SGK1 consensus sequence of the channel. In theory, through its effect on KCNQ4, SGK1 could contribute to the known beneficial effect of glucocorticoids in hearing loss or vertigo (15, 144, 292).
10. Voltage-gated K+ channels Kv1.3, Kv1.5, and Kv4.3
As is apparent from Xenopus oocyte expression system studies, SGK1, SGK2, and SGK3 enhance the K+ current induced by the voltage-gated K+ channels Kv1.3 (124, 140, 359), Kv1.5 (322), and Kv4.3 (21). Similarly, overexpression of PDK1, SGK1, SGK2, or SGK3 in human embryonic kidney (HEK) cells leads to marked upregulation of K+ channels with gating properties identical to Kv1.3 (123, 124, 359). The same channels are activated by treatment of the HEK cells with IGF-I, an effect reversed by inhibition of PI 3-kinase (124). Kv channel activity is seemingly normal in fibroblasts from gene-targeted mice lacking functional SGK1 (sgk1–/–), but cannot be regulated by serum or glucocorticoids (293). Accordingly, in contrast to fibroblasts from wild-type animals, fibroblasts from sgk1–/– mice do not significantly decrease Kv channel activity after serum depletion and do not significantly increase Kv channel activity after treatment with dexamethasone, IGF-I, or both (293).
11. Cation channels created by 4F2/LAT
The expression of the heterodimeric amino acid transporter 4F2/LAT in Xenopus oocytes not only induces amino acid transport, but also a cation conductance (344). This cation conductance, but not the amino acid transport, is upregulated by coexpression with SGK1 (344).
SGK1 increases the insertion of the kainate receptor GluR6 into the cell membrane (308), whereas SGK3 stimulates the -amino-3-hydroxy-5-methyl-4-isoxazolic acid (AMPA) receptor GluR1 (307). Thus both kinases could participate in the regulation of neuroexcitability.
SGK1 increases NHE3 activity, an effect requiring the scaffold protein NHERF2 (379). The kinase is effective through phosphorylation of the carrier protein at Ser-663 (353). The increased NHE3 activity is associated with an increased amount of carrier proteins in the surface membrane (353). NHE3 participates in Na+ reabsorption and H+ secretion in a variety of epithelia and contributes to the regulation of cytosolic pH in a number of epithelial and nonepithelial cells (346).
SGK1 has been shown to stimulate the activity of the Na+-glucose cotransporter SGLT1 (93) and the facilitative glucose transporter GLUT1 (241). On the other hand, overexpression of SGK1 failed to enhance glucose uptake into 3T3-L1 cells (276). Regulation of SGLT1 (93), but not GLUT1 (241), by SGK1 is at least partially mediated by phosphorylation of Nedd4–2. The mechanisms underlying the SGK1-dependent regulation of GLUT1 remain ill-defined (241).
SGK1 has been shown to upregulate a variety of amino acid transporters including ASCT2 (SLC1A5) (247), the glutamine transporter SN1 (40), as well as the glutamate transporters EAAT1 (39), EAAT2 (37), EAAT3 (283), EAAT4 (43), and EAAT5 (41). Upregulation of amino acid transporters has also been shown by SGK2 and SGK3 (37, 39–41, 246, 283).
4. Na+-dicarboxylate cotransporter NaDC-1
As shown in Xenopus oocytes, NaDC-1 is stimulated by SGK1 and SGK3, an effect requiring the participation of NHERF2 (38).
Both SGK1 and SGK3 stimulate the activity of CreaT (SLC6A8), and this effect is partially mediated by interference with Nedd4–2 activity (291).
Coexpression of SGK1 enhances the activity of the intestinal Na+-phosphate cotransporter NaPiIIb, an effect at least partially accounted for by phosphorylation of Nedd4–2 (243). The proximal renal tubular phosphate transporter NaPiIIa was, however, shown to be insensitive to SGK1 coexpression (381).
In the kidney, SGK1 colocalizes with the Na+-K+-ATPase (12). In Xenopus oocytes, expression of SGK1 increases the activity of both endogenous (285) and exogenous (332, 381) Na+-K+-ATPase. The effect is at least partially due to enhanced Na+-K+-ATPase abundance in the cell membrane (381). The ability to modulate Na+-K+-ATPase activity is shared by the SGK2 and SGK3 isoforms (141).
As discussed above, SGK1 phosphorylates Nedd4–2, thus interfering with the binding of the ubiquitin ligase to its targets (17, 86, 298, 299, 386).
SGK1 further phosphorylates, and thus inhibits, glycogen synthase kinase 3 (GSK3) (276). Accordingly, SGK1 may abrogate the effect of this kinase on glycogen synthase. Moreover, it may modify the many cellular functions under control of this important signaling molecule (74), including inhibition of matrix protein formation (see below). However, studies with a PDK1 mutant mouse with a disrupted PIF-binding docking site, which activates PKB/Akt but not SGK (75), suggest that GSK3 is usually regulated by PKB/Akt rather than SGK.
SGK1 phosphorylates and thus inhibits mitogen-activated protein kinase/ERK kinase kinase 3 (MEKK3), which is activated following exposure to growth factors, oxidative stress, and hyperosmotic stress (69). Thus SGK1 may interfere with MEKK3-dependent cellular functions, which signal to the transcription factors NFAT and NFB (1, 36, 62).
SGK1 phosphorylates and inactivates the B-Raf kinase (382) and may thus interfere with B-Raf kinase-dependent cellular functions, which are critical for stimulation of cell proliferation by growth factors (19, 301).
SGK1 phosphorylates and thereby inhibits the phosphomannose mutase 2 (218). SGK1 may thus interact with the glycosylation of glycoproteins (218).
SGK1 associates with and phosphorylates the cAMP responsive element binding protein (CREB) and thus interferes with CREB-dependent gene transcription (83). Moreover, SGK1 enhances the activity of NFB (384) by association with and activation of IB kinase beta (IKK) (384). SGK1 enhances the ability of IKK to phosphorylate IB, leading to degradation of IB and thus NFB activation (384) (Fig. 2).
|
At least in keratinocytes, SGK3 interferes with the Lef-1/-catenin transcription pathway, as nuclear accumulation of -catenin is deranged in keratinocytes from gene-targeted mice lacking functional SGK3 (214).
Pancreatic -cells express negligible amounts of SGK1 under normal conditions (170, 322). However, treatment with glucocorticoids markedly upregulates SGK1 expression (322). SGK1 in turn upregulates the voltage-sensitive Kv1.5 channels. These channels lead to more rapid repolarization of depolarized -cells, and thus blunt the Ca2+ increase during stimulation (322). As a result, SGK1 contributes to, or even accounts for, the acute inhibitory effect of glucocorticoids on insulin release.
Plasma aldosterone concentration is enhanced in SGK1 knockout mice (368). The hyperaldosteronism is presumably the adequate response to extracellular volume depletion (see below) rather than due to a more direct role of SGK1 in the regulation of aldosterone release.
The plasma concentrations of parathyroid hormone and of 1,25(OH)2D3 have been shown to be similar in SGK1 knockout mice and their wild-type littermates (280).
SGK1 is strongly upregulated by osmotic and isotonic cell shrinkage (349). In contrast, SGK1 expression is stimulated by swelling of A6 cells (272). It is tempting to speculate that SGK1-dependent regulation of cation channels contributes to the regulation of cell volume, which involves cation channels in a variety of cells (61, 97, 125, 153, 173, 187, 188, 337, 363). The entry of Na+ through nonselective cation channels depolarizes the cells, thus favoring parallel entry of Cl–. The entry of NaCl and the osmotically obliged water then allows regulatory cell volume increase (181). Interestingly, dexamethasone, which stimulates the transcription of SGK1 (361, 362), enhances the stimulating effect of osmotic cell shrinkage on cation channels in macrophages (125). However, the molecular identity of these channels and the mechanisms of their regulation by glucocorticoids and osmotic cell shrinkage have remained elusive. Thus the functional significance of SGK1 in the stimulation of cation channels and in cell volume regulation remains to be established.
SGK1 has been shown to activate the volume-sensitive osmolyte and anion channel (VSOAC) (354). In Xenopus oocytes, SGK1 enhances the activity of the cell volume-regulated Cl– channel ClC2 (242). Activation of those channels leads to the exit of Cl– and cell membrane depolarization, which in turn favors exit of K+. The cellular loss of KCl leads to regulatory cell volume decrease. Those observations are in seeming contrast to the strong genomic upregulation of SGK1 by cell shrinkage (349) and would suggest a role of SGK1 in regulatory cell volume decrease rather than in regulatory cell volume increase. Possibly, the genomic upregulation of SGK1 during cell shrinkage serves to increase the ability of the cell to cope with alterations of cell volume in either direction.
G. Cell Proliferation and Apoptosis
Both SGK1 (53, 287, 384) and SGK3 (203, 372) have been shown to confer cell survival. The antiapoptotic effect of SGK1 and SGK3 has been attributed in part to phosphorylation of forkhead transcription factors (53, 203, 372), such as FKRHL1 (287). Moreover, SGK3 has been shown to phosphorylate and thus inactivate Bad (203, 372). Phosphorylated Bad binds to the chaperone 14–3-3 and is thus prevented from traveling to the mitochondria, where it triggers apoptosis (204).
Activation of SGK1 in Madin-Darby canine kidney (MDCK) cells by adhesion to fibronectin provided an antiapoptotic signal (287). Following detachment, the survival of MDCK cells was compromised but could be maintained by activation of SGK1 with HGF (287).
SGK1 has further been shown to inhibit apoptosis of breast cancer cells (384). In those cells SGK1 exerted its effects by modulating NFB (384). Accordingly, SGK1-dependent cell survival was abolished by a dominant negative form of IKK or knockout of IKK (384). On the other hand, SGK1 apparently does not share the capability of PKB/Akt to phosphorylate Bax, and thus to prevent the targeting of this important proapoptotic factor to the mitochondria (320).
Proliferative signals appear to shuttle SGK1 into the nucleus (55, 112). The effect of SGK1 and SGK3 on cell proliferation may further involve their ability to regulate Kv1.3 (Fig. 3). The upregulation of Kv1.3 channel activity may be important for the proliferative effect of growth factors, as IGF-I-induced cell proliferation is disrupted by several blockers of Kv channels (124). Stimulation of K+ channels by mitogenic factors and correlation of K+ channel activity with proliferation has been demonstrated in a variety of cells (201, 239, 266, 300). K+ channels are thought to be important for the maintenance of the cell membrane potential (Fig. 3), which is in turn required for proper function of the Ca2+ release-activated Ca2+ channel (ICRAC) (248). The latter channel mediates Ca2+ entry upon stimulation of cells with a wide variety of mitogenic factors, a prerequisite for triggering cell proliferation (27). Both Kv1.3 channels (134, 310) and ICRAC (200) are inhibited by stimulation of CD95, a receptor, which among other functions, disrupts activation and triggers apoptosis of lymphocytes (181, 185).
|
Impaired proliferation and enhanced apoptosis of keratinocytes were observed in gene-targeted mice lacking functional SGK3, leading to delayed hair growth (9, 214).
SGK1 has been shown to participate in the osmotic regulation of the type A natriuretic peptide receptor (NPR-A). The receptor is upregulated by osmotic shrinkage through a signaling pathway involving p38 kinase and SGK1 (64). SGK1 has further been shown to phosphorylate the Ca2+-regulated heat-stable protein of apparent molecular mass 24 kDa (CRHSP24) (18).
As mentioned before, target proteins specifically phosphorylated by SGK1 (and not by PKB/Akt) are NDRG1 and NDRG2 (227, 229). The phosphorylation by SGK1 primes the protein for phosphorylation by GSK3 (227). The functional impact of SGK1-dependent phosphorylation of NDRG1 and NDRG2, however, remains elusive. NDRG2 is a transcriptional target of mineralocorticoids, pointing to some role in the regulation of salt balance (46).
SGK1 has further been shown to phosphorylate filamin C (228) and the microtubule-associated protein tau (71). SGK1 forms together with tau and the protein 14–3-3 a ternary protein complex leading to phosphorylation of tau (71). As described below, hyperphosphorylation of tau is a typical feature of Alzheimer's disease.
|
IV. ROLE OF SERUM- AND GLUCOCORTICOID-INDUCIBLE PROTEIN KINASES IN THE REGULATION OF INTEGRATED FUNCTIONS |
|---|
|
TopPreviousNextReferences |
|---|
When expressed in Xenopus oocytes, SGK1 and SGK3 have been shown to stimulate the activity of the Na+-glucose cotransporter SGLT1 (93), which serves to absorb intestinal glucose. Notably, sgk1–/– mice show reduced glucocorticoid-induced intestinal glucose uptake (132), whereas sgk3–/– mice actually exhibit decreased basal intestinal glucose transport (279). At least partially due to its stimulating effect on glucose transporter GLUT1 (241), SGK1 also favors cellular glucose uptake from the circulation into several tissues including brain, fat, and skeletal muscle (45). Accordingly, after an intraperitoneal glucose load, the plasma glucose concentration increases to higher peak values and declines slower in SGK1 knockout mice (45). Likewise, the plasma glucose-lowering effect of an intraperitoneal insulin application is attenuated in SGK1 knockout mice (45).
Salt excess, which suppresses the release of the mineralocorticoid aldosterone, impairs the peripheral glucose uptake and thus leads to delayed decrease of plasma glucose concentrations following a glucose load (66, 121, 333). The mechanism linking salt excess to peripheral glucose uptake remained elusive until recently. Experiments in mice revealed that the mineralocorticoid deoxycorticosterone acetate (DOCA) restores glucose tolerance and insulin sensitivity in wild-type mice on a diet with excess salt (45). In contrast to the observations in wild-type animals, the glucose tolerance was not significantly modified by salt excess in SGK1 knockout mice (45). Moreover, salt excess dissipated the differences in glucose uptake between wild-type and SGK1 knockout mice (45). Thus the decreased activity of SGK1 contributes to, or even accounts for, the decreased glucose tolerance during salt excess. These observations uncover a critical role of SGK1 in the stimulation of cellular glucose uptake by insulin. Accordingly, SGK1 does not only integrate the effects of mineralocorticoids and insulin on renal tubular Na+ transport (see below) but similarly affects glucose transport.
SGK1 further modifies carbohydrate metabolism by phosphorylation of GSK3 and enhancement of glycogen synthase activity in hepatocytes (276). Thus SGK1 and SGK3 enhance intestinal glucose uptake and SGK1 helps to transport glucose into target organs like brain, fat, skeletal muscle, and liver. The effect of SGK1 on glucose transport is additive to the well-known function of PKB/Akt, which shares, along with the SGK isoforms, the consensus sequence, the ability to stimulate glucose transporters, and the role in the regulation of GSK3 (74).
B. Kidney Function and Salt Appetite
The ability of SGK1 to stimulate ENaC, ROMK1, ClC-Ka, TRPV5, and presumably further renal epithelial ion channels is expected to affect renal electrolyte excretion (for reviews, see Refs. 306, 327, 328).
The regulation of ENaC by SGK1 (Fig. 1) is considered to participate in the regulation of renal Na+ excretion by aldosterone, insulin, and IGF-I (33, 34, 355). While the effect of aldosterone is only partially dependent on the presence of SGK1, and effects of aldosterone and SGK1 are additive, antidiuretic hormone (ADH) or insulin does not further stimulate ENaC in cells expressing active SGK1 (16). The latter findings indicate that activation of ENaC by ADH or insulin depends on SGK1 and/or reflects independent pathways induced by ADH/insulin and SGK1 that converge on the same target structures, e.g., ENaC or Nedd4–2 phosphorylation (298).
Experiments in SGK1 knockout (sgk1–/–) mice indeed revealed subtle but relevant impairment of renal salt retention (368). Under normal salt supply arterial blood pressure as well as salt intake and excretion are virtually identical in the sgk1–/– mice and their wild-type littermates (sgk1+/+). However, the plasma aldosterone concentration is significantly higher in sgk1–/– mice fed a normal Na+ and K+ diet. This may point to impaired regulation of renal Na+ reabsorption (368) or reflect an impairment of renal K+ secretion, which also depends on ENaC-mediated Na+ reabsorption (see below). Moreover, a NaCl-deficient diet discloses the impaired ability of the sgk1–/– mice to adequately lower renal Na+ output. Even though NaCl excretion decreases substantially under NaCl depletion in both sgk1–/– and sgk1+/+ mice, the renal NaCl loss remains significantly larger in sgk1–/– mice than in sgk1+/+ mice, despite an increase of plasma aldosterone concentration, decrease of arterial blood pressure, decrease of glomerular filtration rate (GFR), and enhanced proximal tubular Na+ reabsorption in the sgk1–/– mice (368). The phenotype of the sgk1–/– mice is less severe than the phenotype of ENaC knockout mice (155) and mineralocorticoid receptor knockout mice (25). Thus renal ENaC function and renal mineralocorticoid action are partially but not completely dependent on the presence of SGK1 (Fig. 1).
Acute intravenous application of insulin at a superphysiological dose (under euglycemic conditions) significantly reduced fractional urinary Na+ excretion without affecting blood pressure and GFR in sgk1+/+ mice, an effect significantly blunted in sgk1–/– mice (149). Those experiments demonstrate a critical role for SGK1 in insulin-induced renal Na+ retention.
SGK2 and SGK3 are similarly able to stimulate Na+ channel activity in the Xenopus oocyte expression system (117). Renal Na+ retention is, however, not impaired in the SGK3 knockout mouse (214). Moreover, renal salt loss does not appear to be more pronounced in mice lacking both SGK1 and SGK3 (133). Thus the in vitro observations may not necessarily reflect the in vivo function of SGK2 and SGK3.
Lack of SGK1 further affects the ability to excrete a K+ load. Despite increased basal plasma aldosterone levels, the ability of sgk1–/– mice to excrete an acute K+ load is reduced. Moreover, sgk1–/– mice excrete a chronic K+ load only after marked increases of plasma K+ and aldosterone concentrations (151), which are two major determinants of renal K+ excretion. The impaired ability of sgk1–/– mice to excrete a chronic K+ load is not due to decreased ROMK1 abundance in the luminal membrane of the distal nephron (151), but appears to be due to a decreased electrical driving force secondary to decreased ENaC and/or Na+-K+-ATPase activity. It may also be the result of impaired ROMK1 channel activity or due to decreased expression, trafficking, or activity of another K+ channel.
The SGK1-sensitive voltage-gated K+ channel Kv1.3 (see above) is expressed in the basolateral membrane of the inner medulla (102, 373) and presumably contributes to K+ reabsorption under low K+ intake (373). The ability of SGK1-deficient mice to lower urinary K+ excretion in response to a low K+ intake appeared normal and does not apparently require SGK1-dependent regulation of Kv1.3 (151).
The expression of SGK1 mRNA in the kidney is not restricted to the distal nephron but includes glomeruli, the thick ascending limb, and possibly the early distal convoluted and proximal tubules (for review, see Ref. 328). Recent coexpression experiments in Xenopus oocytes showed that SGK1 can activate a K+ channel that is formed by KCNQ1/KCNE1 (100). In the S2 and S3 segments of the proximal tubule, this K+ channel mediates K+ fluxes into the lumen to prevent membrane depolarization during electrogenic reabsorption of Na+, e.g., Na+-glucose cotransport (324, 325). In the S2 and S3 segments of the proximal tubule, glucose reabsorption is accomplished by the high-affinity Na+-glucose cotransporter SGLT1, which is activated by SGK1 (93). Whether insulin-activated SGK1 regulates and coordinates KCNQ1/KCNE1 and SGLT1 in vivo, e.g., in response to postprandial hyperglycemia to prevent renal glucose loss, remains to be determined.
TRPV5 Ca2+ channel expression is markedly decreased in mice lacking functional SGK1 (280), a finding consistent with a stimulating effect of SGK1 on TRPV5 (101, 246). Renal Ca2+ excretion is, however, decreased in sgk1–/– mice (280), the opposite of what was expected. Presumably, Ca2+ reabsorption in the proximal tubule and in the thick ascending limb of the loop of Henle is enhanced due to stimulation of renal tubular Na+ reabsorption in these nephron segments. In SGK1 knockout animals, the stimulation of proximal tubular and loop of Henle Na+ reabsorption may serve to compensate for the defective Na+ reabsorption in the further distal nephron (368).
SGK1 mRNA is abundant in kidney medulla, and SGK1 was shown to mediate osmotic induction of type A natriuretic peptide receptor (NPR-A) gene promoter in cultured cells from rat inner medullary collecting duct (64). Thus SGK1 may contribute to cell volume regulation in the inner medulla during osmotic challenges. Basal or maximal urinary concentrating ability, however, is not affected in sgk1–/– mice (368), and thus the role of SGK1 in the inner medulla remains to be determined.
SGK1 not only participates in electrolyte balance by its influence on renal elimination, but mediates the effect of mineralocorticoids on salt appetite (326) . As illustrated in Figure 4, treatment of animals with the mineralocorticoid DOCA leads to excessive salt intake in wild-type mice but not in SGK1-deficient mice. Thus SGK1 plays at least a dual role in mineralocorticoid-regulated NaCl homeostasis (Fig. 5). SGK1 dependence of both NaCl intake and renal NaCl reabsorption suggests that excessive SGK1 activity leads to arterial hypertension by simultaneous stimulation of oral NaCl intake and renal NaCl retention (see below).
|
|
SGK1 is heavily expressed in the entire gastrointestinal tract (79) with particularly high expression in enterocytes (352). SGK1 colocalizes with Nedd4–2 in villus enterocytes (243). The localization suggests the involvement of SGK1 and Nedd4–2 in the regulation of intestinal transport systems. Intestinal SGK1 transcript levels are decreased by adrenalectomy, but not increased by a single dose of aldosterone (79). Instead, intestinal SGK1 expression may be primarily stimulated by glucocorticoids.
Intestinal function is seemingly normal in mice lacking functional SGK1 (133). However, in those mice the stimulating effect of dexamethasone on electrogenic glucose transport and the Na+/H+ exchanger NHE3 are blunted or even abolished (133). Those in vivo observations parallel the in vitro stimulating effect of SGK1 on the electrogenic Na+-glucose cotransporter SGLT1 (93) and on NHE3 (353, 379).
The regulation of ENaC by SGK1 in the colon is less clear. The abundance of SGK1 in the distal colon did not change significantly after Na+ depletion or after a single dose of aldosterone (79). Other studies, however, reported that SGK1 mRNA levels (29, 49, 290) and protein levels (29) were significantly elevated in the distal colon in response to application of aldosterone. The transepithelial potential and amiloride-sensitive equivalent short-circuit current in the distal colon are higher in SGK1 knockout animals under both control and low-salt conditions (unpublished observations). Thus SGK1 is apparently not required for stimulation of ENaC in the distal colon.
In contrast to SGK1-deficient mice, gene-targeted mice lacking functional SGK3 show decreased glucose transport even in the absence of glucocorticoid treatment (279). Similarly, intestinal transport is impaired in the SGK1/SGK3 double knockout mouse (132). Again, the obse
