SGK1-Dependent Upregulation of Connective Tissue Growth Factor by Angiotensin II

Angiotensin II has previously been shown to trigger fibrosis, an effect involving connective tissue growth factor (CTGF). The signaling pathways linking angiotensin II to CTGF formation are, however, incompletely understood. A gene highly expressed in fibrosing tissue is the serum- and glucocorticoid-inducible kinase SGK1. The present study explored whether SGK1 is transcriptionally regulated by angiotensin II and participates in the angiotensin II-dependent regulation of CTGF expression. To this end, experiments have been performed in human kidney fibroblasts and mouse lung fibroblasts from gene-targeted mice lacking SGK1 (sgk1–/–) and their wild-type littermates (sgk1+/+). In human renal fibroblasts, SGK1 and CTGF protein expression were enhanced by angiotensin II (10 nM) within 4 h. In sgk1+/+ mouse fibroblasts, SGK1 transcript levels were significantly increased after 4 h of angiotensin II treatment. Angiotensin II stimulated both transcript and protein abundance of CTGF in fibroblasts from sgk1+/+ mice, effects significantly blunted in fibroblasts of sgk1–/– mice. In conclusion, angiotensin II stimulates the expression of SGK1, which is in turn required for the stimulating effect of angiotensin II on the expression of CTGF. Thus, SGK1 presumably contributes to the profibrotic effect of angiotensin II.

The present study aimed to elucidate the possible participation of SGK1 in the stimulation of CTGF expression by angiotensin II.

Preparation of Fibroblasts
Human fibroblast cell lines derived from normal kidney (TK 173) were cultured as described previously [44,45] . Briefly, the cells were cultivated in culture flasks at 37 ° C in 5% CO 2 atmosphere in DMEM, containing 5.6 m M glucose supplemented with 10% fetal calf serum, 20 m M L -glutamine and 100 U/ml penicillin/100 mg/ml streptomycin and passaged every week.
To determine the role of SGK1 in angiotensin II-induced CTGF expression, fibroblasts were collected from SGK1 knockout mice ( sgk1 -/-) and their wild-type littermates ( sgk1 +/+ ) [46] . Several fibroblast preparations have been used. The yield required for Western blotting was achieved with lung fibroblasts. To harvest primary lung fibroblasts from sgk1 -/and sgk1 +/+ mice, whole lungs were removed under anesthesia from 8-to 14-week-old animals and transferred to 90-mm cell culture dishes containing 2 ml of DMEM supplemented with 10% fetal calf serum, 100 U/ml penicillin, 100 mg/ml streptomycin and 2 m M L -glutamine (Gibco-Invitrogen, Karlsruhe, Germany). The tissue was cut into small pieces and cultured under standard cell culture conditions (37 ° C, 5% CO 2 ). Cell growth was observed 2-4 days after initial plating. Fibroblasts were identified by positive staining for fibronectin and used in experiments between passages 2 and 6. Animal experiments were conducted according to the guidelines of the German Animal Welfare Law and were approved by local authorities.

Western Blotting
For Western blotting, human renal fibroblasts or sgk1 +/+ and sgk1 -/mouse lung fibroblasts were grown in 60-mm culture dishes for 32 h. Serum was removed for 18 h and cells were thereafter treated with angiotensin II, 10 n M , for 4 h. After treatment, cells were lysed and whole cell lysates (50 g) were subjected to 10% SDS-PAGE in 10% Tris-glycine buffer. The proteins were transferred to nitrocellulose membranes and the membrane incubated for 1 h in PBS containing 5% fat-free milk and 0.1% Tween (blocking buffer) to block unspecific binding sites. The membranes were incubated overnight at 4 ° C with a goat polyclonal CTGF primary antibody (from Santa Cruz, Heidelberg, Germany, diluted 1: 400 in blocking buffer, or from R&D Systems, Wiesbaden, Germany, at a concentration of 0.1 g/ml) or with a rabbit polyclonal SGK1 primary antibody (from Upstate Biotechnology, Dundee, UK, diluted 1: 1,000, or from Pineda, Berlin, Germany, 1: 100). The CTGF antibody from R&D Systems recognizes two isoforms of about 36 and 40 kDa. The 36-kDa band was chosen for analysis. After incubation with HRP-conjugated anti-goat or anti-rabbit secondary antibodies (Santa Cruz) for CTGF and SGK1, respectively, proteins binding the antibodies were visualized with enhanced chemiluminescence according to the manufacture's instructions (Amersham, Freiburg, Germany). Membranes were probed with GAPDH antibody (Santa Cruz) to control loaded and transferred amounts of protein. Densitometric analysis of CTGF protein bands was performed using Scion Image (Scion, Frederick, Md., USA). The specific bands were referred to the staining of GAPDH or of non-specific bands. Due to differences of the exposure times of different Western blots, the densities of the bands were normalized to the values of the respective non-treated controls or wildtypes.

Quantification of mRNA by Real-Time RT/PCR
For real-time PCR total RNA was isolated from cultured fibroblasts of sgk1 +/+ and sgk1 -/mice using the Qiagen RNeasy Fibrous Tissue Midi Kit (Qiagen, Hilden, Germany). SGK1 or CTGF mRNA were transcribed using Taq polymerase (Roche Diagnostics GmbH, Mannheim, Germany) and quantified by PCR using a light cycler system (Roche Diagnostics GmbH, Roche Applied Science, Mannheim, Germany). For the detection of mouse CTGF mRNA the specific primers used were: sense: 5 -ACG ACG CCA AGG ACC GCA-3 ; antisense: 5 -TTG TAA TGG CAG GCA CAG-3 . For detection of SGK1 transcript levels the specific primers used were: sense: 5 -TGTCTTGGGGCTGTCCTGTATG-3 ; antisense: 5 -GCTTCTGCTGCTTCCTTCACAC-3 . The transcript levels of the housekeeping gene GAPDH were also determined for each sample using a commercial primer kit (Search LC, Heidelberg, Germany). Amplification of the target DNA was performed during 35 cycles, each 10 s at 95 ° C, 10 s at 68 ° C and 16 s at 72 ° C. Melting curve analysis confirmed the specificity of amplified products, which were then separated on 1.5% agarose gels to verify the size of the PCR product. Results were calculated as a ratio of the target vs. housekeeping gene transcripts.

Statistical Analysis
Data are provided as means 8 SEM, n represents the number of independent experiments. All data were tested for significance using ANOVA or t-test, as appropriate, and only results with p ! 0.05 were considered statistically significant.

SGK1 and CTGF Expression Is Stimulated by Angiotensin II in a Fibroblast Cell Line
To explore whether the serum-and glucocorticoid-inducible kinase SGK1 is a transcriptional target of angiotensin II in human tissues, experiments were performed with fibroblast cell lines derived from normal kidneys. As illustrated in figure 1 , after exposure to angiotensin II (10 n M ) for 4 h the SGK1 protein abundance was significantly increased in the fibroblasts. Next, the expression of CTGF was analyzed ( fig. 2 ). Incubation of the cells with angiotensin II for 4 h significantly increased the expression of CTGF. These observations indicate that angiotensin II upregulates the expression of SGK1 and CTGF.

Angiotensin II Increased the Expression of SGK1 in Primary Mouse Fibroblasts
To test whether SGK1 is functionally relevant for the stimulation of CTGF expression by angiotensin II, experiments have been performed in lung fibroblasts from SGK1 knockout mice ( sgk1 -/-) and their wild-type littermates ( sgk1 +/+ ). First, the effect of angiotensin II on SGK1 expression was analyzed in lung fibroblasts from wildtype mice. After exposure of primary lung fibroblasts to angiotensin II (10 n M ), SGK1 mRNA was significantly increased within a 2-hour treatment ( fig. 3 ).

CTGF Expression Is Not Stimulated by Angiotensin II in the Absence of SGK1
CTGF transcript levels were then analyzed in primary lung fibroblasts from both wild-type and sgk1 -/mice. As Angiotensin II (AII) stimulates SGK1 expression in human fibroblasts. a Representative Western blot demonstrating SGK1 protein abundance after a 4-hour incubation in the absence or presence of 10 n M AII. A non-specific band served as loading control. b Arithmetic means 8 SEM (n = 6) of SGK1 over GAPDH protein abundance or normalized to a non-specific band in fibroblasts from normal human kidneys after a 4-hour incubation in the absence (left bar) or presence (right bar) of 10 n M AII. * Significant (p ! 0.05) difference from control.  figure 4 , basal CTGF transcript levels were not different between sgk1 -/and sgk1 +/+ fibroblasts. Angiotensin II, 10 n M , however, increased CTGF transcript levels in wild-type but not in sgk1 -/mice fibroblasts.
These observations indicate that transcriptional upregulation of CTGF requires the participation of SGK1. When CTGF protein was quantified by Western blotting the upregulation in wild-type cells by angiotensin II was indeed mirrored by an increased amount of CTGF protein in these cells. In sgk1 -/mouse fibroblasts, in contrast, CTGF protein levels were not significantly increased by angiotensin II ( fig. 5 ). These results strongly indicate that the stimulation of CTGF protein expression by angiotensin II depends on the presence of SGK1.

Discussion
The present observations demonstrate that the serumand glucocorticoid-inducible kinase SGK1 is a transcriptional target of angiotensin II. Thus, angiotensin II upregulates SGK1 expression not only by increasing aldosterone release and subsequent mineralocorticoid stimulation of SGK1 transcription [33,34] , but as well by a more direct stimulation of SGK1 transcription.
To the extent that SGK1 is upregulated by mineralocorticoids, it could similarly participate in the stimulation of fibrosis by mineralocorticoid excess, which, for instance, has been shown to induce cardiac fibrosis [55,56] in a pressure-independent manner via cardiac mineralocorticoid receptors [57,58] . Inhibition of those receptors would abrogate the aldosterone-induced fibrosis but favor increase of angiotensin II release and thus promote angiotensin II-induced fibrosis. SGK1 has indeed been shown to participate in the stimulation of cardiac CTGF formation during mineralocorticoid excess [39] . Moreover, SGK1 has been shown to potentiate the stimulating effect of hyperglycemia on matrix protein formation [35] .
Mechanisms linking SGK1 with CTGF expression could at least in theory involve nuclear factor NF B [31] . SGK1 associates with and activates I B kinase ␤ (IKK ␤ ), which in turn phosphorylates I B ␣ , leading to degradation of I B ␣ and thus activation of NF B [59] . The stimulating effect of the mineralocorticoid DOCA on cardiac CTGF indeed requires both, SGK1 and NF B [39] . Angiotensin II similarly signals through NF B [60][61][62] , but may not require NF B for stimulation of fibrosis [63] . SGK1 further phosphorylates glycogen synthase kinase 3 ␤ , an effect, however, apparently not critical for cardiac fibrosis [64] . Moreover, SGK1 may be effective through its well-known role in the regulation of channels and transporters [31,[65][66][67] . In any case, additional experimentation will be necessary to elucidate whether the SGK1-dependent pathway stimulated by angiotensin II involves more than NF B.