Strong Association of Serum- and Glucocorticoid-Regulated Kinase 1 with Peripheral and Adipose Tissue Inflammation in Obesity

ORIGINAL ARTICLE Strong association of serum- and glucocorticoid-regulated kinase 1 with peripheral and adipose tissue inﬂammation in obesity

MH Schernthaner-Reiter1, F Kiefer1, M Zeyda1,2, TM Stulnig1,2, A Luger1 and G Vila1

OBJECTIVE: The serum- and glucocorticoid-regulated kinase 1 (SGK1) is an early transcriptional target of glucocorticoids and is activated via insulin. Here we investigate the regulation of SGK1 expression in human obesity, diet-induced murine obesity and human monocytic cell line THP-1 monocytes. SUBJECTS AND METHODS: SGK1 expression was studied in subcutaneous and omental adipose tissue (AT) of 20 morbidly obese and 20 age- and gender-matched non-obese controls in murine diet-induced obesity and the THP-1 cell line. The regulation of SGK1 by inflammatory signals was tested in THP-1 cells. RESULTS: Murine diet-induced obesity is associated with a significant upregulation of Sgk1 in gonadal AT. Sgk1 expression is highest in the macrophage-rich stromal vascular fraction and lower in adipocytes. In humans, AT SGK1 is predominantly expressed in CD14+ macrophages and significantly upregulated in omental and subcutaneous AT of obese subjects. SGK1 mRNA expression in both omental and subcutaneous AT correlates with body mass index, circulating leptin and C-reactive protein, and the local expression of inflammatory markers including monocyte chemotactic protein-1 and macrophage inflammatory protein-1α. The expression of SGK1 in THP-1 cells is upregulated by inflammatory signals, such as lipopolysaccharide and tumour necrosis factor-α, as well as during the induction of monocyte-to-macrophage maturation. CONCLUSIONS: Our data present the first link between SGK1 and obesity-associated inflammation. SGK1 expression is stimulated in response to inflammatory signals and increased in AT macrophages. The characterisation of SGK1 functions in obesity and immunity may help identify potential therapeutic targets in the treatment of obesity. International Journal of Obesity (2015) 39, 1143–1150; doi:10.1038/ijo.2015.41

INTRODUCTION upregulated in response to glucocorticoids and promotes adipocyte 10,11 Serum- and glucocorticoid-regulated kinase 1 (SGK1) is a serine/ (AC) differentiation. threonine kinase and an early transcriptional target of glucocorti- White AT of obese mice and men is characterised by macrophage 12 coids, which is post-translationally activated by insulin through infiltration in proportion to weight and BMI, respectively. This AT phosphorylation via the phosphatidylinositol-3 kinase-dependent inflammation is linked to the low-grade systemic inflammation 13 pathway.1 Sgk1-deficient (Sgk1− / −) mice have a mild phenotype, accompanying obesity, andisstimulatedbyhypoxia,metabolic demonstrating reduced responses to glucocorticoids, insulin and endotoxaemia and lipotoxicity/saturated fatty acids released from 14,15 inflammation.2,3 SGK1 upregulates the cell surface expression of hypertrophic ACs. AT macrophages (ATMs) display a specific the insulin-dependent glucose transporters GLUT1 and GLUT4,3,4 phenotype and secrete a number of inflammatory molecules 16 17 and could thus have a role in regulating insulin sensitivity. In including tumour necrosis factor-α (TNF-α) and interleukin-6. addition, SGK1 activates sodium glucose cotransporter 1 (SGLT1) Here we investigate the association of SGK1 with AT inflamma- and accelerates intestinal glucose absorption,5 thereby contribut- tion in murine and human obesity, as well as the regulation of ing to hyperglycaemia, hyperinsulinaemia and eventually obesity.6 SGK1 expression in response to inflammatory stimuli in human It was recently shown that SGK1 inhibition leads to reduced THP-1 monocytes. We show that SGK1 expression is significantly T-helper type 17-mediated inflammation.7 increased in both human and murine obese AT, specifically in the The upregulation of SGK1 via glucocorticoids and insulin murine macrophage-rich stromal vascular cell (SVC) fraction and in + suggests a link between SGK1 and obesity, a condition human CD14 macrophages. In THP-1 monocytes, SGK1 expres- associated with mild hypercortisolaemia and insulin resistance.8 sion is strongly induced during monocyte-to-macrophage matura- In humans, an SGK1 variant containing a combination of two tion, as well as by stimuli of monocyte/macrophage activation. specific single-nucleotide polymorphisms was associated with significantly higher body mass index (BMI).5 Moreover, SGK1 upregulation has also been linked to comorbidities of obesity SUBJECTS AND METHODS including hypertension, diabetes, stroke and cancer.9 SGK1 expres- Subjects sion is increased in human adipose tissue (AT) in obesity and The studies were approved by the Ethics committee of the Medical diabetes at both mRNA and protein levels.10 Furthermore, SGK1 is University of Vienna. Written informed consent was obtained from all

1Clinical Division of Endocrinology and Metabolism, Department of Internal Medicine III, Medical University of Vienna, Vienna, Austria and 2Christian Doppler Laboratory for Cardiometabolic Immunotherapy, Department of Internal Medicine III, Medical University of Vienna, Vienna, Austria. Correspondence: Dr G Vila, Clinical Division of Endocrinology and Metabolism, Department of Internal Medicine III, Medical University of Vienna, Waehringer Guertel 18-20, Vienna A-1090, Austria. E-mail: [email protected] Received 30 October 2014; revised 18 January 2015; accepted 1 February 2015; accepted article preview online 26 March 2015; advance online publication, 19 May 2015 SGK1 in adipose tissue inflammation MH Schernthaner-Reiter et al 1144 subjects before the study. Collection of human AT samples was described insulin resistance was calculated by glucose (mg dl − 1) × insulin (mg dl − 1)/ previously.18 Briefly, paired samples of omental or subcutaneous AT 405. BMI was calculated by mass (kg)/height (m)2. were obtained from two groups of either morbidly obese patients (BMI = 53.1 ± 2.5 kg m− 2, n = 20), undergoing laparoscopic surgery for gastric banding, or age- and gender-matched non-obese control subjects Animals (BMI = 25.2 ± 0.7 kg m −2, n = 20) undergoing laparoscopic cholecystectomy, Male C57BL/6 J mice (Charles River Laboratories, Sulzfeld, Germany) aged fundoplication or resection of liver cysts. Exclusion criteria were the 7 weeks were either placed on a high-fat (HF) diet (60% kcal fat, D12492; presence of infection, inflammation, neoplastic or systemic disease, Research Diets Inc., New Brunswick, NJ, USA) to induce obesity or on a low- diabetes or other uncontrolled endocrine diseases, or the current use of fat diet (10% kcal fat, D12450B; Research Diets Inc.) to serve as lean antibiotic, anti-inflammatory or antiobesity medication. Blood samples controls. Mice were given ad libitum access to food and water, and were were obtained by cubital venepuncture after an overnight fast and stored maintained on a 12 h light/dark cycle. After 20 weeks, they were killed at − 80 °C. AT specimens obtained during excision were washed in saline by cervical dislocation. Gonadal white AT (GWAT) was removed and buffer. After removal of blood and visible blood vessels, samples were immediately snap frozen in liquid nitrogen. Study protocols were approved immediately snap frozen in liquid nitrogen. Subcutaneous AT for AT by the Ethics committee of the Medical University of Vienna and followed fractionation was obtained from patients undergoing reductional plastic the guidelines on accommodation and care of animals formulated by the − 2 surgery (BMI = 29.8 ± 2.5 kg m , n = 5). European Convention for the Protection of Vertebrate Animals Used for Experimental and other Scientific Purposes. Laboratory parameters Cholesterol, high- and low-density lipoprotein cholesterol, triglycerides, Cell culture glucose, glycated haemoglobin (HbA1c) and C-reactive protein levels were THP-1 monocytes and 3T3-L1 pre-ACs were obtained from the measured by routine laboratory methods. Plasma free fatty acids were American Type Culture Collection (Manassas, VA, USA) and kept under measured using a microfluorometric technique (Wako Chemicals USA, standard cell culture conditions (37 °C, 5% CO2). THP-1 were maintained in Richmond, VA, USA). Leptin, adiponectin (Linco Research, St Charles, MO, RPMI 1640, supplemented with 10% fetal bovine serum, 2 mML-glutamine USA) and insulin (Serono Diagnostics, Freiburg, Germany) levels were (all obtained from Gibco, Carlsbad, CA, USA), 50 000 IU l − 1 of penicillin determined by radioimmunoassay. The homoeostatic model assessment of and streptomycin (Lonza, Basel, Switzerland), 0.05 mM 2-mercaptoethanol

Figure 1. Sgk1 expression in 3T3-L1 pre-ACs and in murine AT in diet-induced obesity. (a)Sgk1mRNAexpressioninGWATofobese(HF,n = 12) or lean (low-fat, n = 8) mice, results are shown relative to LF. (b) Sgk1 mRNA expression in AC and SVC fractions in GWAT of obese (n = 9) or lean (n = 3) mice, results are shown relative to LF AC. (c) Sgk1 mRNA expression in 3T3-L1 after stimulation with dexamethasone (Dex; 100 nM) for 24 h versus unstimulated control, results are shown relative to control. (a–c) Error bars show the s.e.m. **Po0.01, ***Po0.001, as determined by unpaired Student’s t-tests. (d)ImmunoblotofSgk1andβ-actin in 3T3-L1 pre-ACs after stimulation with Dex (200 nM) for 24 h versus unstimulated control. (e) Sgk1 (green), Mac-2 (red) and nuclei 4ʹ,6-diamidino-2-phenylindole (DAPI; blue) visualised by immunofluorescence on paraffin-embedded sections of low-fat diet (LFD) and HF diet (HFD) murine AT under × 40 magnification. The merged figure shows an overlay of the green and red channels. Images shown are representative of the LFD and HFD group, respectively. UBC, ubiquitin C.

Figure 2. SGK1 expression in human obesity. (a) SGK1 expression was analysed in human omentous AT (omAT) and subcutaneous AT (scAT) biopsies from obese patients (n = 20) and age- and gender- matched non-obese control subjects (n = 20). (b) Human scAT was fractionated into ACs, matrix, macrophage-depleted SVCs and CD14+ macrophages, and SGK1 mRNA expression was determined by real-time quantitative PCR. Error bars show the s.e.m. NS, non-significant; *Po0.05, ***Po0.001, as determined by analysis of variance followed by Dunn’s test. (c) SGK1 (green), CD68 (red) and nuclei 4ʹ,6-diamidino-2-phenylindole (DAPI; blue) were visualised by immunofluorescence on frozen sections of lean and obese scAT using × 40 magnification. The merged figure shows an overlay of the green and red channel. Images shown are representative of the lean and obese group, respectively.

(Sigma, St Louis, MO, USA), and 1 mM each of sodium pyruvate and non- material retained on the 200-μm mesh filters is referred to as the matrix essential amino acids solution (both obtained from Gibco). fraction. The flow-through containing the floating AC fraction was For stimulations, 2 × 106 THP-1 cells were seeded into six-well plates centrifuged again and immediately lysed in Trizol. The pellet contained and left to recover overnight. Macrophage maturation alone was carried the stromal vascular fraction, which was treated with haemolysis buffer, out by addition of PMA (phorbol 12-myristate 13-acetate; Sigma) for 24 h. passed through 70 μm mesh filters and either lysed in Trizol reagent or Stimulations with 100 ng ml − 1 lipopolysaccharide (LPS; Sigma) or further fractionated into ATMs and a macrophage-depleted SVC fraction. 50 ng ml − 1 TNF-α (R&D Systems; Minneapolis, MN, USA) were performed The isolation of ATMs was described previously.20 Briefly, CD14 magnetic for 24 h with or without previous macrophage maturation with 25 ng ml − 1 beads (Miltenyi Biotec, Bergisch Gladbach, Germany) were used to PMA for 24 h. Cells were then centrifuged at 300 g for 10 min and pellets separate ATMs, and their purity was determined by flow cytometry of were lysed in Trizol reagent (Invitrogen, Carlsbad, CA, USA) for RNA CD14-positive cells, and by determination of CD68 and adiponectin mRNA extraction. expression in all fractions by real-time quantitative PCR.19 3T3-L1 were maintained in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum, 2 mML-glutamine and 50 000 IU l − 1 of penicillin and streptomycin. For stimulations, 3T3-L1 pre- RNA isolation, reverse transcription and real-time quantitative PCR ACs were seeded into six-well plates, grown to 80% confluency and Levels of mRNA expression were determined by RNA isolation followed stimulated with 100 or 200 nM dexamethasone (Sigma), as stated, for 24 h by reverse transcription and real-time quantitative PCR, as described 21 fl before lysis in Trizol (Invitrogen) or Weinberg buffer (50 mM HEPES, pH 7.0, previously. Brie y, RNA was isolated from cell lysates in Trizol by phase 0.5% Nonidet P-40, 250 mM NaCl, 5 mM EDTA, 5 mM NaF, 0.2 mM Na VO , separation with chloroform and subsequent isopropanol precipitation 3 4 ’ 50 mM β-glycerophosphate, 1 mM phenylmethylsulfonyl fluoride, 2 mM according to manufacturer s instructions. Precipitated RNA was washed − 1 − 1 μ benzamidine, 10 μgml aprotinin and 10 μgml leupeptin). in 70% ethanol and resuspended in 20 l sterile diethylpyrocarbonate- treated water. An amount of 2 μg of total RNA was pretreated with DNAse (Invitrogen) AT fractionation for 10 min at room temperature. Samples were then heated to 65 °C for As described previously,19 visible blood vessels and connective tissues 10 min and placed on ice for 2 min. The RNA template was incubated with were removed from murine GWAT or human subcutaneous AT. The tissue dNTPs (Promega, Fitchburg, WI, USA), random primers, RNase OUT and was cut into small pieces, washed in phosphate-buffered saline and dithiothreitol in 5 × first strand buffer (all obtained from Invitrogen) at digested with 35 μgml− 1 liberase 3 (Roche, Mannheim, Germany) and 25 °C for 2 min, and reverse transcription was started by addition of 50 U ml − 1 DNAse I (Sigma) for 1 h at 37 °C. The tissue was then passed Superscript II reverse transcriptase (Invitrogen) in a final volume of 20 μl. through 200 μm mesh filters and centrifuged at 1000 g for 10 min. The Samples were incubated at 42 °C for 50 min; then the reaction was stopped

© 2015 Macmillan Publishers Limited International Journal of Obesity (2015) 1143 – 1150 SGK1 in adipose tissue inflammation MH Schernthaner-Reiter et al 1146 by incubation at 70 °C for 10 min. cDNA was diluted with 100 μl Table 1. Clinical and biochemical correlates of SGK1 expression in diethylpyrocarbonate water before further use. omental and subcutaneous adipose tissue One microlitre of the original cDNA transcript was used as a template in TaqMan gene expression assays with 1.25 μl of the gene-specific SGK1 omental SGK1 subcutaneous 6-carboxyfluorescein–carboxytetramethyl-rhodamine-labelled SGK1, TNF-α, CD68, monocyte chemotactic protein-1 and macrophage inflammatory a b Weight 0.52 0.64 protein-1α, regulated on activation, normal T-cell expressed and secreted Waist circumference 0.46a 0.54a a b (RANTES or chemokine (C-C motif) ligand 5, CCL5), monocyte chemotactic BMI 0.53 0.62 protein-2 (or CCL8), CCL7, CCL11 or C-C chemokine receptor types 1, 2 and Fasting glucose − 0.00 0.19 μ c 5 probes (Applied Biosystems, Carlsbad, CA, USA) and 12.5 lof2× Fasting insulin 0.28 0.41 Universal PCR Master Mix (Applied Biosystems) in a total volume of 25 μl. HOMA-IR 0.23 0.40c PCR was performed in duplicate on an ABI Prism 7000 Cycler (Applied HbA1c 0.40c 0.46a − − Biosystems). Values of gene expression for each sample were normalised Fasting triglycerides 0.21 0.08 fl Free fatty acids 0.28 0.21 to 6-carboxy uorescein-labelled 18 S (mouse and human data), Total cholesterol − 0.08 − 0.17 glyceraldehyde-3 phosphate dehydrogenase or ubiquitin C (cell culture Serum leptin 0.46a 0.65b experiments, all obtained from Applied Biosystems) values, as stated. Serum adiponectin − 0.23 −0.22 Serum CRP 0.49a 0.56b Immunoblotting Abbreviations: BMI, body mass index; CRP, C-reactive protein; HbA1c, glycated Cells were washed with phosphate-buffered saline and lysed in ice-cold haemoglobin; HOMA-IR, homeostatic model assessment of insulin resistance; Weinberg buffer containing 1:100 protease inhibitor and phosphatase SGK1, serum- and glucocorticoid-regulated kinase 1. Spearman’s correlation inhibitor cocktails (both obtained from Sigma). Samples were centrifuged coefficients of SGK1/18 S mRNA expression in omental and subcutaneous at 2500 g for 10 min, and supernatants were subjected to SDS- adipose tissue with patients’ clinical and biochemical parameters were polyacrylamide gel electrophoresis followed by western blotting. Mem- determined after performing Shapiro–Wilk test. aPo0.01. bPo0.001. cPo0.05. branes were probed with rabbit anti-SGK (Sigma) and monoclonal mouse anti-β-actin (Novus Biologicals, Littleton, CO, USA) as described.22

Figure 3. Scatterplots representing the relationship of log-transformed subcutaneous AT (scAT) SGK1/18 S mRNA expression with BMI (a) and log-transformed C-reactive protein (CRP; b), and of log-transformed omental AT (omAT) SGK1/18 S mRNA expression with BMI (c) and log-transformed CRP (d) with Pearson’s correlation coefﬁcients (r).

International Journal of Obesity (2015) 1143 – 1150 © 2015 Macmillan Publishers Limited SGK1 in adipose tissue inflammation MH Schernthaner-Reiter et al 1147 Immunofluorescence SGK1 expression in relation to clinical and biochemical Paraffin-embedded sections of murine GWAT were deparaffinised and parameters, and to local inflammatory markers rehydrated. Frozen sections of human subcutaneous AT were fixed in ice- Correlation analyses were performed between the mRNA expres- cold acetone for 5 min and then rehydrated in phosphate-buffered saline sion levels of SGK1 in human omental AT and subcutaneous AT, for 5 min. The sections were incubated with primary antibodies anti-SGK and the patients’ clinical and biochemical parameters. SGK1 mRNA (Sigma), anti-Mac-2 (Serotec, Kidlington, UK) or anti-CD68 (Dako, Glostrup, expression correlates with weight, waist circumference, BMI, Denmark) at 4 °C overnight, alongside negative controls without primary HbA1c, leptin and with the systemic inflammatory marker antibodies. The sections were further incubated with secondary antibodies C-reactive protein in both omental and subcutaneous AT Alexa 488-conjugated anti-rabbit IgG, Alexa 594-conjugated anti-rat IgG or Alexa 594-conjugated anti-mouse IgG (all obtained from Invitrogen) for 1 h (Table 1 and Figure 3). Multiple regression analysis reveals that at room temperature, and 4ʹ,6-diamidino-2-phenylindole was applied for BMI is an independent predictor of both omental and subcuta- 10 min before the slides were mounted using Vectashield mounting neous SGK1 expression (Po0.05 for both). The expression of medium (Vector Laboratories, Servion, Switzerland) and covered. Fluores- multiple local AT inflammatory mediators, including TNF-α and cence was visualised on a fluorescence microscope (Leitz, Solms, Germany CD68, in omental as well as subcutaneous AT highly correlate with and Nikon, Tokyo, Japan). SGK1 expression (Table 2).

Statistical analysis SGK1 regulation in THP-1 monocytes Human data of SGK1 expression, clinical and biochemical parameters, and The regulation of SGK1 by different inflammatory signals and local AT inflammatory parameters were tested for normality using Shapiro– pathways was further studied in THP-1 monocytes. Induction of Wilk test. Spearman rank correlations were computed to assess the monocyte-to-macrophage maturation by PMA is paralleled by a relationship between local SGK1 expression and other variables, as SGK1 concentration-dependent upregulation of SGK1 mRNA expression was not normally distributed. Multiple regression analysis was used for α fi (Figure 4a), and simultaneously an upregulation of TNF- identi cation of independent relationships. Differences between means expression (Figure 4b), confirming the transition of the monocyte were tested by two-sided independent samples Student’s t-test, one- or to a pro-inflammatory state. LPS similarly causes a strong two-way analysis of variance, or equivalent non-parametric tests followed by post hoc analysis, as stated. Diagrams show mean ± s.e.m. Correlations induction of SGK1 expression; however, this effect is blunted in and multiple regression analyses were performed using IBM SPSS Statistics PMA-matured THP-1 where LPS does not have an additional effect 20 (IBM Corp., Armonk, NY, USA). Other analyses were performed using (Figure 4c). Conversely, SGK1 upregulation in response to TNF-α GraphPad Prism (GraphPad Software, La Jolla, CA, USA). P-values o0.05 stimulation in THP-1 monocytes does not reach significance, but were considered statistically significant. TNF-α exerts synergistic effects with PMA on SGK1 expression in the matured macrophage state (Figure 4d). LPS conversely also induces TNF-α expression in THP-1 monocytes, as well as in the RESULTS PMA-matured macrophages (Figure 4e). These results suggest that Sgk1 expression in murine AT SGK1 may be involved in monocyte-to-macrophage maturation as To investigate Sgk1 expression in diet-induced obesity, mice were well as the inflammatory activation of the macrophage. maintained on a HF diet or low-fat control diet for 20 weeks, resulting in induction of obesity in the HF mice (weight DISCUSSION 50.6 ± 1.1 g versus 32.9 ± 1.2 g). Sgk1 expression in GWAT was fi This study investigated the potential association of SGK1 with AT signi cantly increased in HF mice compared with their lean fl fi controls by more than fourfold (Figure 1a). Given that AT consists in ammation in murine and human obesity. The main ndings of of several different cell populations, we investigated Sgk1 expression in the AC fraction and the macrophage-rich SVC fl fraction. Sgk1 expression was significantly higher in the SVCs Table 2. Relationship of SGK1 expression to local in ammatory compared with ACs in both lean and obese GWAT (Figure 1b). parameters To further specify the cell population expressing high levels of SGK1 omental SGK1 subcutaneous Sgk1 within the SVC fraction, Sgk1 expression was tested in 3T3-L1 pre-ACs in which Sgk1 was basally expressed and strongly TNF-α 0.20 0.59a stimulated by dexamethasone on the mRNA and protein level CD68 0.65a 0.76a a a (Figures 1c and d). In immunofluorescence microscopy of murine MCP-1 (CCL2) 0.66 0.68 α a a GWAT, Sgk1 is mainly co-expressed in cells positive for the MIP-1 (CCL3) 0.61 0.70 RANTES (CCL5) 0.39b 0.78a macrophage marker Mac-2 (galectin-3; Figure 1e), suggesting a MCP-2 (CCL8) 0.66a 0.73a role of Sgk1 in obesity, specifically in ATMs. CCL7 0.61c 0.78a CCL11 0.45c 0.27 a a SGK1 expression in human AT CCR1 0.53 0.75 CCR2 0.45c 0.61a The expression of SGK1 in omental and subcutaneous AT biopsies CCR5 0.49c 0.67c of 20 morbidly obese subjects and 20 age- and gender-matched control subjects were compared. SGK1 expression in obese Abbreviations: CCL7 and CCL11, chemokine (C-C motif) ligands 7 and 11; omental AT and subcutaneous AT was significantly higher CCR1, 2 and 5, C-C chemokine receptor types 1, 2 and 5; MCP-1, monocyte chemotactic protein-1 (chemokine (C-C motif) ligand 2); MCP-2, monocyte compared with the respective lean controls (Figure 2a). To further chemotactic protein-2 (chemokine (C-C motif) ligand 8); MIP-1α, macro- differentiate the localisation of SGK1 expression within the phage inflammatory protein-1α (chemokine (C-C motif) ligand 3); RANTES, different cell populations in human AT, AT fractionation was regulated on activation, normal T-cell expressed and secreted (chemokine performed in subcutaneous AT. AT SGK1 is predominantly (C-C motif) ligand 5); SGK1, serum- and glucocorticoid-regulated kinase 1; expressed in CD14+ macrophages, with significantly lower TNF-α, tumour necrosis factor-α. Spearman’s correlation coefficients of expression in ACs (Figure 2b). In immunofluorescence experi- SGK1/18 S mRNA expression in omental and subcutaneous adipose tissue with local expression of other inflammatory parameters were determined ments, SGK1 AT expression co-localises to cells expressing CD68 after performing Shapiro–Wilk test, which showed non-normal distribution (Figure 2c). CD68 is a transmembrane and lysosomal glycoprotein of subcutaneous SGK1 expression. aPo0.001. bPo0.05. cPo0.01. that is highly expressed in monocytes and macrophages.23

Figure 4. Regulation of SGK1 and TNF-α in THP-1. (a, b) SGK1 (a) and TNF-α (b) mRNA expression in THP-1 after macrophage maturation with PMA at different concentrations. (c) SGK1 expression in THP-1 monocytes after stimulation with either TNF-α (50 ng ml − 1) or LPS (100 ng ml − 1) for 24 h. (d) SGK1 expression in PMA-matured THP-1 macrophages that were stimulated with TNF-α or LPS for 24 h. (e) TNF-α mRNA expression after LPS stimulation of THP-1 monocytes and PMA-matured THP-1 macrophages. Signiﬁcances for multiple comparisons are given below. Experiments were performed in triplicate, and results shown are representative of three separate experiments. Results are shown relative to control. Error bars show the s.e.m. NS, non-signiﬁcant; *Po0.05, **Po0.01,***Po0.001, as determined by one-way analysis of variance (ANOVA) followed by Dunnett’s test (a, b), two-way ANOVA followed by Newman–Keuls test (c, d) or two-way ANOVA followed by unpaired Student’s t-tests (e).

our study are as follows: (1) the predominant expression of SGK1 small amounts of AT SGK1 are actually expressed in ACs. AC in ATMs within the AT, (2) the upregulation of SGK1 in both hypertrophy is associated with increased oxidative and hypoxic subcutaneous and omental AT in human obesity, (3) the strong stress, both of which were previously shown to induce SGK1 correlation of SGK1 expression in AT with clinical parameters of expression in other tissues thereby preventing apoptosis.24–26 obesity, as well as systemic and local inflammatory parameters, Through its action on glucose transporters, SGK1 could and (4) the stimulatory effect of TNF-α and LPS on SGK1 contribute to the altered glucose metabolism and insulin expression in THP-1 monocytes. response of ACs in obesity.27 This hypothesis is also supported We demonstrate for the first time a relationship between by our present findings showing an association between SGK1 and obesity-associated inflammatory response; our subcutaneous AT SGK1 expression and circulating insulin, findings could also have important implications for the homoeostatic model assessment of insulin resistance and development of metabolic and cardiovascular complications, HbA1c. The link between SGK1 and obesity has been shown in which are strongly associated with obesity-driven inflammation. previous publications, and is underpinned also by the fact that Interestingly, AT SGK1 expression correlates with parameters of SGK1 is induced and activated by glucocorticoids and insulin, obesity and systemic inflammation, but not with secondary respectively, while obesity is associated with hyperinsulinism characteristics of obesity such as circulating triglycerides or and glucocorticoid overproduction.9 Expression studies showed cholesterol. BMI is the strongest predictor of SGK1 expression increased SGK1 expression in obese AT,10 while functional inAT.Inaddition,wedemonstrateforthefirsttimethatonly studies reveal that SGK1 influences adipogenesis.11 Our data

International Journal of Obesity (2015) 1143 – 1150 © 2015 Macmillan Publishers Limited SGK1 in adipose tissue inflammation MH Schernthaner-Reiter et al 1149 complement these observations finding a hitherto unknown 4 Palmada M, Boehmer C, Akel A, Rajamanickam J, Jeyaraj S, Keller K et al. SGK1 relationship between SGK1 and obesity-driven inflammation. kinase upregulates GLUT1 activity and plasma membrane expression. Diabetes SGK1 expression in AT is predominantly localised in ATMs not 2006; 55:421–427. only in HF diet-induced obesity in mice but also in humans, linking 5 Dieter M, Palmada M, Rajamanickam J, Aydin A, Busjahn A, Boehmer C et al. fl Regulation of glucose transporter SGLT1 by ubiquitin ligase Nedd4-2 and kinases SGK1 with AT-associated in ammation. The relationship of SGK1 12 – with AT inflammation is further supported by the strong SGK1, SGK3, and PKB. Obes Res 2004; : 862 870. 6 Fujita Y, Kojima H, Hidaka H, Fujimiya M, Kashiwagi A, Kikkawa R. Increased correlations with local inflammatory markers, such as monocyte fl α intestinal glucose absorption and postprandial hyperglycaemia at the early step chemotactic protein-1 and macrophage in ammatory protein-1 of glucose intolerance in Otsuka Long-Evans Tokushima Fatty rats. Diabetologia (Table 2), and is similarly strong in both omental and subcuta- 1998; 41: 1459–1466. neous AT. Furthermore, SGK1 expression also correlates with the 7 Heikamp EB, Patel CH, Collins S, Waickman A, Oh MH, Sun IH et al. The AGC kinase systemic inflammatory marker C-reactive protein. SGK1 regulates TH1 and TH2 differentiation downstream of the mTORC2 com- Several previous studies have established a link between SGK1 plex. Nat Immunol 2014; 15:457–464. and inflammation.3,7 SGK1 was suggested to have a role in pro- 8 Anagnostis P, Athyros VG, Tziomalos K, Karagiannis A, Mikhailidis DP. Clinical inflammatory cytokine signalling in granulocytes,28 but the review: The pathogenetic role of cortisol in the metabolic syndrome: a hypothesis. 94 – expression and function of SGK1 in monocytes and macrophages J Clin Endocrinol Metab 2009; : 2692 2701. 9 Lang F, Voelkl J. Therapeutic potential of serum and glucocorticoid inducible have not been studied previously. The strong link between SGK1 22 – fl kinase inhibition. Expert Opin Investig Drugs 2013; : 701 714. and in ammatory parameters prompted us to investigate the 10 Li P, Pan F, Hao Y, Feng W, Song H, Zhu D. SGK1 is regulated by metabolic-related profile of SGK1 expression following stimuli of monocyte/ factors in 3T3-L1 adipocytes and overexpressed in the adipose tissue of subjects macrophage activation. SGK1 expression increases parallel to with obesity and diabetes. Diabetes Res Clin Pract 2013; 102:35–42. TNF-α expression during monocyte-to-macrophage maturation, 11 Di Pietro N, Panel V, Hayes S, Bagattin A, Meruvu S, Pandolfi A et al. Serum- and corroborating a direct regulation of SGK1 during inflammation. glucocorticoid-inducible kinase 1 (SGK1) regulates adipocyte differentiation via Indeed, stimulation studies confirm that both LPS and TNF-α forkhead box O1. Mol Endocrinol 2010; 24:370–380. induce SGK1 expression. These inflammatory stimuli extend the 12 Weisberg SP, McCann D, Desai M, Rosenbaum M, Leibel RL, Ferrante AW Jr. list of factors contributing to excessive SGK1 expression in AT Obesity is associated with macrophage accumulation in adipose tissue. J Clin Invest 2003; 112: 1796–1808. beyond the well-known hormones glucocorticoids, insulin and 13 Maachi M, Pieroni L, Bruckert E, Jardel C, Fellahi S, Hainque B et al. Systemic aldosterone. As macrophages are considered as the effector cells low-grade inflammation is related to both circulating and adipose tissue TNFal- fl 29 of AT in ammation, the increased expression of SGK1 in mature pha, leptin and IL-6 levels in obese women. Int J Obes Relat Metab Disord 2004; 28: macrophages could be of pathophysiological significance. The 993–997. correlation of SGK1 with markers of local AT inflammation and 14 Suganami T, Mieda T, Itoh M, Shimoda Y, Kamei Y, Ogawa Y. Attenuation of macrophage infiltration, such as CD68 and monocyte chemotactic obesity-induced adipose tissue inflammation in C3H/HeJ mice carrying a Toll-like protein-1, suggests that SGK1 may have a role in systemic receptor 4 mutation. Biochem Biophys Res Commun 2007; 354:45–49. inflammation and macrophage recruitment into dysfunctional 15 Ye J. Emerging role of adipose tissue hypoxia in obesity and insulin resistance. Int 33 – obese AT. In this context, SGK1 could be involved monocyte-to- J Obes (Lond) 2009; :54 66. macrophage maturation as supported by our observation that 16 Fried SK, Bunkin DA, Greenberg AS. Omental and subcutaneous adipose tissues of obese subjects release interleukin-6: depot difference and regulation by gluco- SGK1 expression increases in matured THP-1 cells. corticoid. J Clin Endocrinol Metab 1998; 83:847–850. Obesity is associated with the dysregulation of a multitude of 17 Fain JN, Madan AK, Hiler ML, Cheema P, Bahouth SW. Comparison of the release 29 proteins and signalling pathways, including a number of of adipokines by adipose tissue, adipose tissue matrix, and adipocytes from 30 proteins overexpressed in ATMs. Hence, further functional visceral and subcutaneous abdominal adipose tissues of obese humans. studies are warranted to establish a causal relationship for each Endocrinology 2004; 145: 2273–2282. obesity-associated alteration including SGK1. 18 Huber J, Kiefer FW, Zeyda M, Ludvik B, Silberhumer GR, Prager G et al. Taken together, SGK1 is strongly upregulated following CC chemokine and CC chemokine receptor profiles in visceral and subcutaneous inflammatory stimuli during monocyte-to-macrophage matura- adipose tissue are altered in human obesity. J Clin Endocrinol Metab 2008; 93: – tion, while SGK1 expression in ATMs is increased in human obesity 3215 3221. 19 Kiefer FW, Zeyda M, Todoric J, Huber J, Geyeregger R, Weichhart T et al. and significantly correlates with local and systemic markers of fl Osteopontin expression in human and murine obesity: extensive local up-regulation obesity and in ammation. In this context, SGK1 could be a in adipose tissue but minimal systemic alterations. Endocrinology 2008; 149: mediator of peripheral and ATM function, and should be further 1350–1357. investigated as a therapeutic target in efforts to disrupt the vicious 20 Zeyda M, Farmer D, Todoric J, Aszmann O, Speiser M, Gyori G et al. Human cycle that causes AT dysfunction and systemic inflammation in adipose tissue macrophages are of an anti-inflammatory phenotype but capable obesity. of excessive pro-inflammatory mediator production. Int J Obes (Lond) 2007; 31: 1420–1428. 21 Nabokikh A, Ilhan A, Bilban M, Gartner W, Vila G, Niederle B et al. Reduced TGF- CONFLICT OF INTEREST beta1 expression and its target genes in human insulinomas. Exp Clin Endocrinol Diabetes 2007; 115: 674–682. The authors declare no conflict of interest. 22 Reiter MH, Vila G, Knosp E, Baumgartner-Parzer SM, Wagner L, Stalla GK et al. Opposite effects of serum- and glucocorticoid-regulated kinase-1 and gluco- ACKNOWLEDGEMENTS corticoids on POMC transcription and ACTH release. Am J Physiol Endocrinol Metab 2011; 301: E336–E341. We thank L-I Ionasz for expert technical assistance. 23 Holness CL, Simmons DL. Molecular cloning of CD68, a human macrophage marker related to lysosomal glycoproteins. Blood 1993; 81:1607–1613. 24 Nishida Y, Nagata T, Takahashi Y, Sugahara-Kobayashi M, Murata A, Asai S. REFERENCES Alteration of serum/glucocorticoid regulated kinase-1 (sgk-1) gene expression in 1 Park J, Leong ML, Buse P, Maiyar AC, Firestone GL, Hemmings A. Serum and rat hippocampus after transient global ischemia. Brain Res Mol Brain Res. 2004; glucocorticoid-inducible kinase (SGK) is a target of the PI 3-kinase-stimulated 123:121–125. signaling pathway. EMBO J 1999; 18:3024–3033. 25 Rusai K, Wagner B, Roos M, Schmaderer C, Strobl M, Boini KM et al. The serum and 2 Artunc F, Amann K, Nasir O, Friedrich B, Sandulache D, Jahovic N et al. glucocorticoid-regulated kinase 1 in hypoxic renal injury. Cell Physiol Biochem Blunted DOCA/high salt induced albuminuria and renal tubulointerstitial damage 2009; 24:577–584. in gene-targeted mice lacking SGK1. J Mol Med (Berl) 2006; 84:737–746. 26 Kobayashi T, Cohen P. Activation of serum- and glucocorticoid-regulated protein 3 Lang F, Stournaras C, Alesutan I. Regulation of transport across cell membranes by kinase by agonists that activate phosphatidylinositide 3-kinase is mediated by the serum- and glucocorticoid-inducible kinase SGK1. Mol Membr Biol 2014; 31: 3-phosphoinositide-dependent protein kinase-1 (PDK1) and PDK2. Biochem J 29–36. 1999; 339: 319–328.

© 2015 Macmillan Publishers Limited International Journal of Obesity (2015) 1143 – 1150 SGK1 in adipose tissue inflammation MH Schernthaner-Reiter et al 1150 27 Lumeng CN, Deyoung SM, Saltiel AR. Macrophages block insulin action in 29 McNelis JC, Olefsky JM. Macrophages, immunity, and metabolic disease. Immunity adipocytes by altering expression of signaling and glucose transport proteins. 2014; 41:36–48. Am J Physiol Endocrinol Metab 2007; 292:E166–E174. 30 Cancello R, Henegar C, Viguerie N, Taleb S, Poitou C, Rouault C et al. Reduction of 28 Cowling RT, Birnboim HC. Expression of serum- and glucocorticoid-regulated macrophage infiltration and chemoattractant gene expression changes in white kinase (sgk) mRNA is up-regulated by GM-CSF and other proinflammatory adipose tissue of morbidly obese subjects after surgery-induced weight loss. mediators in human granulocytes. J Leukoc Biol 2000; 67: 240–248. Diabetes 2005; 54: 2277–2286.