Human adipocyte function is impacted by mechanical cues
Fibrosis is a hallmark of human white adipose tissue (WAT) during obesity‐induced chronic inflammation. The functional impact of increased interstitial fibrosis (peri‐adipocyte fibrosis) on adjacent adipocytes remains unknown. Here we developed a novel in vitro 3D culture system in which human adipo...
Saved in:
| Published in | The Journal of pathology Vol. 233; no. 2; pp. 183 - 195 |
|---|---|
| Main Authors | , , , , , , , , , |
| Format | Journal Article |
| Language | English |
| Published |
Chichester, UK
John Wiley & Sons, Ltd
01.06.2014
Wiley Subscription Services, Inc |
| Subjects | |
| Online Access | Get full text |
Cover
Loading…
| Summary: | Fibrosis is a hallmark of human white adipose tissue (WAT) during obesity‐induced chronic inflammation. The functional impact of increased interstitial fibrosis (peri‐adipocyte fibrosis) on adjacent adipocytes remains unknown. Here we developed a novel in vitro 3D culture system in which human adipocytes and decellularized material of adipose tissue (dMAT) from obese subjects are embedded in a peptide hydrogel. When cultured with dMAT, adipocytes showed decreased lipolysis and adipokine secretion and increased expression/production of cytokines (IL‐6, G‐CSF) and fibrotic mediators (LOXL2 and the matricellular proteins THSB2 and CTGF). Moreover, some alterations including lipolytic activity and fibro‐inflammation also occurred when the adipocyte/hydrogel culture was mechanically compressed. Notably, CTGF expression levels correlated with the amount of peri‐adipocyte fibrosis in WAT from obese individuals. Moreover, dMAT‐dependent CTGF promoter activity, which depends on β1‐integrin/cytoskeleton pathways, was enhanced in the presence of YAP, a mechanosensitive co‐activator of TEAD transcription factors. Mutation of TEAD binding sites abolished the dMAT‐induced promoter activity. In conclusion, fibrosis may negatively affect human adipocyte function via mechanosensitive molecules, in part stimulated by cell deformation. Published by John Wiley & Sons, Ltd |
|---|---|
| Bibliography: | ark:/67375/WNG-FGZ0LFL9-D istex:71322AEA4B9537AB986FE4418EDE5B44399A83E3 ArticleID:PATH4347 Appendix S1. Supplementary materialFrom WAT extracellular matrix to dMAT. (A) Photomicrographs from OCT imaging of non-digested material of adipose tissue (MAT) obtained from subcutaneous adipose tissue (SAT) sample of obese subject before (left photomicrograph) and after (right photomicrographs) the decellularization process. During this process, adipocytes (*) surrounded by collagen fibres, and other cells composing WAT are eliminated without alterations of the collagen fibre structure. Scale bars = 500 µm and 100 µm. (B) Observation of dMAT by immunofluorescence analysis using antibodies directed against type I, III collagen (red, Cy3-conjugated anti-rabbit IgG) and type VI collagen (green, Cy2-conjugated anti-mouse IgG) (scale bar = 100 µm) and second harmonic generation, SHG (scale bar = 50 µm). (C) Observation of obese SAT by immunofluorescence analysis. Diversity of collagens were determined using antibodies directed against type I, III, IV collagen (green, Cy2-conjugated anti-rabbit IgG), type V, type VI, XVIII collagen (red, Cy3-conjugated anti-mouse IgG) and FN (green, Cy2-conjugated anti-mouse IgG). Nuclei were stained with DAPI. Scale bar = 100 µm. (D) Observation of fibrotic depots extracted from obese SAT by OCT imaging and immunofluorescence microscopy using antibodies directed against type I (red, Cy3-conjugated anti-rabbit IgG) and type VI collagen (green, Cy2-conjugated anti-mouse IgG). The BMI of the obese subject was 48; age 46 years3D culture of human unilocular adipocytes, tissue microstructure and maintenance of viable mature adipocytes. (A) Experimental procedure of human mature adipocyte culture in the 3D hydrogel. Mature adipocytes were isolated from SAT of lean subjects and embedded in the peptidic hydrogel. The photograph shows floating hydrogel, containing unilocular adipocytes in culture media. (B) Comparison of insulin response between floating adipocytes (2D) and adipocytes cultured for 24 h into the hydrogel (3D). Insulin response was evaluated by insulin (10 nm) stimulation of serine473 phosphorylation of Akt (pS473Akt) for 10 min. A representative western blot is presented among three separate experiments. (C) Comparison of LDH activity (cytotoxicity), lipolytic activity (glycerol release) and leptin secretion between 24 h floating adipocytes (2D, white bars) and adipocytes cultured for 24 h into the hydrogel (3D, black bars). Data are mean ± SEM of five separate experiments from lean SAT. **p < 0.01, ***p < 0.001, 2D adipocytes versus 3D adipocytes. (D) Metabolic and secretory functions of human adipocytes from lean SAT were followed for 7 days (d) in the 3D setting. Lipolytic activity was evaluated by glycerol release after isoproterenol stimulation (1 μm) for 4 h. Leptin secretion was measured by ELISA assay. Data are mean ± SEM of six separate experiments. AU, arbitrary unitsAnalysis of adipocyte flattening in culture with obese dMAT. (A) Observation of a 100 µm section of the hydrogel by OCT imaging containing unilocular adipocytes from lean SAT with dMAT. The transverse view (higher panel) shows organization of collagen fibres in the 3D setting. Scale bar = 50 µm. Adipocytes were reconstructed in 3D to appreciate dMAT-induced changes in the adipocyte morphology (lower panel). A representative 3D adipocyte reconstruction is presented among 26 analyses. (B) Histogram of the flattening measured in two dimensions for adipocytes in 3D culture without dMAT (blue bars) and with dMAT (red bars) (n > 500 adipocytes)Lipolytic activity of adipocytes cultured with obese dMAT in the 3D hydrogel. (A) Representative experiment of dose-dependent effects of dMAT on glycerol release in basal (white line) and stimulated conditions, isoproterenol (1 μm) (black line, Iso) from lean adipocytes cultured for 3 days in the 3D setting, corresponding to a representative experiment of three independent experiments (R2= 0.9267, basal conditions, and R2 = 0.9064, β-adrenergic stimulated conditions). *p < 0.05. (B) Dose-dependent effects of dMAT on adipocyte cytotoxicity corresponding to a representative experiment of three independent experiments. Adipocyte cytotoxycity was evaluated by LDH activity after 3 days in the 3D culture model. The SAT of lean subjects (mean BMI = 23.05 ± 0.38 kg/m2) was used to prepare adipocytes and SAT of obese subjects (mean BMI = 45.2 ± 1.5 kg/m2) to prepare dMATMetabolic and secretory functions of adipocytes submitted to mechanical deformation and cultured with obese dMAT in the 3D hydrogel. (A) Scheme of the set-up used to apply controlled deformation in a multi-well plate. The set-up allows the application of increasing deformation in the one direction of the plate, whereas the other direction is used to test reproducibility. (B-D) Effects of the mechanical constraints on adipocyte cytotoxicity (B) and secretion of leptin (C) and adiponectin (D), which were evaluated for each deformation from 0% to 50%. Adipocyte cytotoxycity was evaluated by LDH activity and the leptin and adiponectin secretions were measured using an ELISA after 3 days in the 3D setting. Data are presented as mean ± SEM of five independent experiments. NS, not significant. (E) The heat map with graded shades from green to red represents the secretion level (significant levels > 10 pg/ml) of 14 cytokines and chemokines screened by Multiplex analysis. Brace identifies the seven showing the highest secretion level. (F) Significant changes in the secretions of inflammatory molecules (IL-6 and G-CSF) by adipocytes in the 3D setting with dMAT (AD + dMAT, black bars) submitted to a mechanical deformation (50%, AD + DEF, grey bars) or cultured with dMAT under mechanical deformation (AD + dMAT + DEF, dark grey bars), compared to adipocytes cultured alone (control AD, white bars). The results are presented as the fold differences over the control. Data are presented as mean ± SEM of three independent experiments. *p < 0.05, AD versus AD + dMAT or AD + DEF or AD + dMAT + DEF. The SAT of lean subjects (mean BMI = 23.05 ± 0.38 kg/m2) was used to prepare adipocytes and SAT of obese subjects (mean BMI = 45.2 ± 1.5 kg/m2) to prepare dMATMetabolic and secretory functions of adipocytes cultured with dMAT from SAT of lean subjects (BMI 22.6 ± 0.7, n = 12) in the 3D setting. (A) ELISA analysis of the secretion of leptin, adiponectin and IL-6 by adipocytes cultured alone (AD, white bars) or with dMAT prepared from SAT of lean subjects (0.05 mg/ml, AD + dMAT, black bars) in the 3D setting. (B) Lipolytic activity was evaluated by the glycerol release from either adipocytes cultured alone (AD) or with dMAT prepared from SAT of lean subjects (0.05 mg/ml, AD + dMAT) in the 3D setting. Glycerol release was measured in the basal (white bars) or stimulated conditions (isoproterenol, 1 μm; black bars, Iso) for 4 h. Data are presented as mean ± SEM of four independent experiments. ns, not significant. The SAT of lean subjects was used to prepare adipocytes and dMAT (mean BMI = 23.05 ± 0.38 kg/m2 and mean BMI = 22.6 ± 0.7 kg/m2, respectively)Role of β1-integrin in mediating alterations of adipocytes induced by obese dMAT. ELISA analysis of leptin and adiponectin secretion by adipocytes in the 3D setting with dMAT (AD + dMAT, black bars) compared to either adipocytes cultured alone (control AD, white bars). Adipocytes were treated with IgG1 or β1-integrin neutralizing antibody (ab β1-int). Data are presented as mean ± SEM of five independent experiments. *p < 0.05, AD versus AD + dMAT, ns, not significant. The SAT of lean subjects (mean BMI = 23.05 ± 0.38 kg/m2) was used to prepare adipocytes and SAT of obese subjects (mean BMI = 45.2 ± 1.5 kg/m2) to prepare dMATEffect of obese dMAT on kinase phosphorylation profile, actin cytoskeleton remodelling and gene expression. (A) Cell lysates prepared from four independent experiments of adipocytes cultured for 30 min in the 3D hydrogel alone (control, AD) or exposed to dMAT (AD + dMAT) or 50% mechanical deformation (AD + DEF) were pooled and analysed for the phosphorylation of different kinases using the Human Phospho-Kinase Antibody Array Kit, as described in Materials and methods: (1) TOR; (2) Src; (3) Hck; (4) EGFR; (5) CREB; (6) Lyn; (7) Yes; (8) Chk-2; (9) PRAS-40; (10) ERK 1/2; (11) MSK 1/2; (12) HSP27; (13) FAK; (14) AMPKα2; (15) STAT2; (16) STAT6; (17) GSK-3α/β; (18) AKT (S473); (19) STAT5a; (20) STAT5b; (21) STAT5a/b; (22) AKT (T308); (23) P70 S6K (T389); (24) p70 S6K (T421/S424); (25) STAT3; (26) HSP60; (27) P53 (S46); (28) p53 (S15); (29) RSK 1/2/3; (30) c-jun. PC, positive control; NC, negative control. (B) Human preadipocytes isolated from SAT of lean subjects (BMI: 24 kg/m2) were differentiated in the 3D hydrogel for 7 days with (dMAT) or without dMAT (control) and examined by confocal microscopy for actin staining (red, phalloidin-AlexaFluor 546); scale bar = 50 µm. (C) A heat map representation of the mRNA expression of several genes involved in metabolism, ER stress, inflammation and ECM remodelling, as quantified using real-time PCR and normalized to 18S in adipocytes cultured alone (control AD) or with dMAT in the 3D culture model (AD + dMAT). Graded shades from green to red represent the fold differences between the control adipocytes and AD + dMAT. Data are presented as mean ± SEM of six independent experiments. #, mechanosensitive genes. (D) LOX gene expressions was quantified by real-time PCR and normalized to 18S in adipocytes isolated from SAT of lean subjects. Results are expressed as fold differences between adipocytes cultured in the 3D hydrogel for 3 days alone (AD, white bar), with dMAT (AD + dMAT, black bar) or submitted to compression (AD + DEF, grey bar). Data are represented as mean ± SEM of five separate experiments. ns, non significant. The SAT of lean subjects (mean BMI = 23.05 ± 0.38 kg/m2) was used to prepare adipocytes and SAT of obese subjects (mean BMI = 45 No conflicts of interest were declared. ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 content type line 23 |
| ISSN: | 0022-3417 1096-9896 1096-9896 |
| DOI: | 10.1002/path.4347 |