Obesity induces white adipose tissue (WAT) dysfunction characterized by unremitting inflammation and fibrosis, impaired adaptive thermogenesis and increased lipolysis. Prostaglandins (PGs) are powerful lipid mediators that influence the homeostasis of several organs and tissues. The aim of the current study was to explore the regulatory actions of PGs in human omental WAT collected from obese patients undergoing laparoscopic bariatric surgery. In addition to adipocyte hypertrophy, obese WAT showed remarkable inflammation and total and pericellular fibrosis. In this tissue, a unique molecular signature characterized by altered expression of genes involved in inflammation, fibrosis and WAT browning was identified by microarray analysis. Targeted LC-MS/MS lipidomic analysis identified increased PGE2 levels in obese fat in the context of a remarkable COX-2 induction and in the absence of changes in the expression of terminal prostaglandin E synthases (i.e. mPGES-1, mPGES-2 and cPGES). IPA analysis established PGE2 as a common top regulator of the fibrogenic/inflammatory process present in this tissue. Exogenous addition of PGE2 significantly reduced the expression of fibrogenic genes in human WAT explants and significantly down-regulated Col1α1, Col1α2 and αSMA in differentiated 3T3 adipocytes exposed to TGF-β. In addition, PGE2 inhibited the expression of inflammatory genes (i.e. IL-6 and MCP-1) in WAT explants as well as in adipocytes challenged with LPS. PGE2 anti-inflammatory actions were confirmed by microarray analysis of human pre-adipocytes incubated with this prostanoid. Moreover, PGE2 induced expression of brown markers (UCP1 and PRDM16) in WAT and adipocytes, but not in pre-adipocytes, suggesting that PGE2 might induce the trans-differentiation of adipocytes towards beige/brite cells. Finally, PGE2 inhibited isoproterenol-induced adipocyte lipolysis. Taken together, these findings identify PGE2 as a regulator of the complex network of interactions driving uncontrolled inflammation and fibrosis and impaired adaptive thermogenesis and lipolysis in human obese visceral WAT.
Citation: García-Alonso V, Titos E, Alcaraz-Quiles J, Rius B, Lopategi A, López-Vicario C, et al. (2016) Prostaglandin E2 Exerts Multiple Regulatory Actions on Human Obese Adipose Tissue Remodeling, Inflammation, Adaptive Thermogenesis and Lipolysis. PLoS ONE 11(4): e0153751. https://doi.org/10.1371/journal.pone.0153751
Editor: Patricia T. Bozza, Fundação Oswaldo Cruz, BRAZIL
Received: November 18, 2015; Accepted: April 4, 2016; Published: April 28, 2016
Copyright: © 2016 García-Alonso et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The microarray data he been deposited in Gene Expression Omnibus (GEO): GSE7141.
Funding: Supported by Spanish Ministerio de Economía y Competitividad (MEC) (SAF12/32789, PIE14/00045 and SAF15/63674-R) under European Regional Development Funds (ERDF). CIBERehd is funded by the Instituto de Salud Carlos III. This study was carried out at the Center Esther Koplowitz. V.G.-A. and B.R. had fellowships from MEC. C.L.-V. was supported by IDIBAPS/Fundació Clínic. A.L. was funded by a Marie Curie Action and J.A.-Q. is a recipient of an Agaur/BFU fellowship (Generalitat de Catalunya).
Competing interests: The authors he declared that no competing interests exist.
IntroductionProstaglandin (PG) E2 is one of the most abundant lipid mediators in the human body. It is constitutively produced in nearly all tissues by the coordinate enzymatic activities of cyclooxygenases (COX) and terminal PGE synthases [1–4]. PGE2 is a powerful molecule that exerts multiple biological effects depending on the tissue environment and the cell type [1–4]. In this regard, in addition to being recognized as an important mediator of inflammation, pain and fever, PGE2 also plays an important role in the regulation of vascular tone and cell proliferation and differentiation [5–7].
White adipose tissue (WAT) is a complex and highly active endocrine organ that plays a key role in the regulation of energy metabolism. In obese individuals, WAT expands its energy-buffering capacity by fat cell hypertrophy and/or by hyperplasia from committed progenitors [8]. This adipose tissue expansion leads to a plethora of functional derangements including hypoxia, lack of nutrients and tissue remodeling, which are major contributors to the chronic “low-grade” state of mild inflammation present in WAT of obese individuals [9–11]. This persistent and unresolved inflammatory state in WAT is, in turn, responsible for the excessive synthesis of extracellular matrix components and the subsequent interstitial deposition of fibrotic material [12,13]. Increased interstitial WAT fibrosis decreases extracellular matrix flexibility and reduces the tissue plasticity, which ultimately leads to adipocyte dysfunction [12]. The ultimate consequence of these derangements is the development of a number of comorbidities associated with obesity including insulin resistance and type 2 diabetes, non-alcoholic fatty liver disease, atherosclerosis and cardiovascular disease [14,15].
We recently described that PGE2 participates in the differentiation of WAT pre-adipocytes into beige/brite cells [16]. Since beige/brite cells are able to dissipate large amounts of chemical energy as heat by uncoupling protein 1 (UCP1), which uncouples the synthesis of ATP from the respiratory chain [17], this finding was interpreted as forable within the context of metabolic homeostasis of obese WAT. The aim of the current study was to translate and expand this finding to human obesity by investigating the potential metabolic benefits of PGE2 in WAT remodelling, inflammation, adaptive thermogenesis and lipolysis in omental adipose tissue obtained from obese individuals undergoing bariatric surgery. Our data provide evidence that PGE2 exerts pleiotropic regulatory effects in the complex homeostasis of WAT in human obesity.
Materials and Methods ReagentsPGE2 was obtained from Cayman Chemicals (Ann Arbor, MI). Krebs-Ringer bicarbonate buffer, Dulbecco's Modified Eagle's Medium (DMEM), fatty acid-free (FAF)-BSA and liberase were from Roche (Basel, Switzerland). Nylon mesh filters (100 μm) were obtained from BD Biosciences (San Jose, CA). TRIzol was from Invitrogen (Carlsbad, CA) and L-Glutamine from Biological Industries (Kibbutz Beit Haemek, Israel). FBS and Dulbecco’s PBS with and without calcium and magnesium (DPBS+/+ and DPBS-/-, respectively) were obtained from Lonza (Vervieres, Belgium). The free glycerol assay kit was from Sigma (St. Louis, MO). The High-Capacity Archive Kit and TaqMan Gene Expression Assays were provided by Applied Biosystems (Foster City, CA). Adipocyte maintenance medium was from Zen Bio (Research Triangle Park, NC). The Pierce BCA Protein Assay kit was from Thermo Scientific (Carlsbad, CA).
Study participants and sample collectionTwelve obese individuals undergoing laparoscopic bariatric surgery and 10 patients without a history of obesity undergoing laparoscopic cholecystectomy were recruited at the Gastroenterology Surgery Unit of the Hospital Clínic of Barcelona. The participants were selected according to their body mass index (BMI) calculated as mass/height2 and categorized as non-obese control (BMI29.9 kg/m2) groups. Individuals with the presence of inflammatory bowel disease or cancer, obese patients with previous bariatric surgical procedures or patients with active prescription for any non-steroidal anti-inflammatory drug (NSAID) or any other medication that could affect PG synthesis were excluded from the study. Demographic and clinical data were collected from the electronic medical records of the patients at the time of surgery. Omental adipose tissue samples were harvested and washed twice with DPBS and minced into approximately 100 mg pieces. Tissue samples were either fixed in 10% formalin or snap-frozen in liquid nitrogen and stored at -80°C for further analysis. Some adipose tissue samples were used ex vivo or for the isolation of primary adipocytes and pre-adipocytes. This study was approved by the Investigation and Ethics Committee of the Hospital Clínic and written informed consent was obtained from all the participants (protocol # 12–7239).
Assessment of adipose tissue fibrosis by Sirius Red and Masson’s trichrome stainingFibrosis in adipose tissue was assessed in paraffin sections by Sirius Red staining. Briefly, sections were incubated for 10 minutes in 0.5% thiosemicarbazide and stained in 0.1% Sirius Red F3B in saturated picric acid for 1 hour, and subsequently washed with an acetic acid solution (0.5%). Fibrosis was also assessed by Masson’s trichrome staining at the Pathology Department of the Hospital Clínic.
Assessment of adipose tissue fibrosis by bright field and polarization microscopySections of visceral adipose tissue stained with picrosirius red were visualized in a DMRB-Leica microscope (Wetzlard, Germany), equipped with a wide-field Leica DFC 450 camera and a λ analyzer and polarizer. Acquisition of the images was performed with the LAS 4.0 Leica software at 20xPL FLUOTAR Dry Numerical Aperture 0.50 and 40X PL FLUOTAR 1–0.5 oil. The same region of the sample was imaged using bright field to visualize picrosirius stained tissue and polarized light to observe collagen fibbers. Image analysis was performed using FIJI-Image J (Wayne Rasband, NIH). Briefly, images of bright field and polarized light were background subtracted, filtered and automatically thresholded to segment and quantify areas of cellular membrane in bright field images and areas of fibrotic regions visualized under polarized light. Total fibrosis quantification (sum of collagen I and III fibers) was expressed as the ratio of fibrous regions/cellular membranes as described in reference 13. The same procedure was used for quantification of pericellular fibrosis.
Ex vivo experiments in adipose tissue explantsHuman WAT explants collected under sterile conditions were placed in P60 plates in pre-warmed (37°C) DPBS+/+ containing antibiotics (100 U/ml penicillin and 100 mg/ml streptomycin). Connective tissue and blood vessels were removed by dissection before cutting the tissue into small pieces. The explants were washed with DPBS at 37°C by centrifugation (400g, 1 min) to remove blood cells and pieces of tissue containing insufficient adipocytes to float. Thereafter, the explants were cultured in DMEM with L-glutamine (2 mM), antibiotics and 10% FAF-BSA. Tissue explants of approximately 40 mg were subsequently cultured in 12-well plates with vehicle (0.4% ethanol) or PGE2 (1 μM) in 1% FAF-BSA for 24 h.
Isolation of adipocytes and pre-adipocytesWAT was excised, weighed and rinsed twice in cold carbogen-gassed Krebs-Ringer bicarbonate buffer supplemented with 1.5% FAF-BSA and 2 mM EDTA and centrifuged at 800g for 1 min at 4°C to remove free erythrocytes and leukocytes. Tissue suspensions (300–600 mg) were placed in 5 ml of digestion buffer containing Krebs-Ringer supplemented with 1.5% FAF-BSA and 1 mg/ml of liberase and incubated at 37°C for 30 min as described [18]. Floating cells corresponding to adipocytes were collected and cultured in DMEM with FBS (10%), L-glutamine (2 mM), antibiotics and HEPES (20 mM). The remaining supernatant was centrifuged at 800g for 5 min and pelleted cells corresponding to the stromal vascular cell (SVC) fraction were incubated with erythrocyte lysis buffer (0.15 M NH4Cl, 10 mM KHCO3, and 0.1 mM EDTA) for 5 min and centrifuged at 500g for 5 min, as described (16). Thereafter, SVC were incubated in a culture flask containing pre-adipocyte priming DMEM/F-12/Ham’s mediun with L-glutamine, HEPES, 10% FBS and antibiotics for 24 hours.
Pre-adipocyte differentiation into beige/brite cellsFreshly isolated WAT pre-adipocytes were cultured in DMEM containing 3% FBS, 100 nM insulin and antibiotics and incubated with vehicle (0.01% ethanol) or PGE2 (0.1 μM) for 3 hours. The induction medium was replaced with fresh adipocyte maintenance medium (33 μM biotin, 17 μM pantothenate, 1 μM dexamethasone) in the presence of 1 nM triiodothyronine (T3) for 2 additional days. Thereafter, the cells were incubated in adipocyte medium until day 10.
Differentiation of 3T3-L1 adipocytesMouse 3T3-L1 cells were seeded onto 24-well plates (75.000 cells/well) with DMEM supplemented with 10% FBS and maintained in a 5% CO2 atmosphere. Cells were allowed to grow to confluence for 2 days and then exposed to DMEM containing a differentiation cocktail (5 μg/ml insulin, 0.25 μM dexamethasone and 0.5 mM IBMX) supplemented with antibiotics and 2 mM L-glutamine in the presence of vehicle (0.01% DMSO), PGE2 (0.1, 1 and 5 μM) or compound III (1, 5 and 8 μM). After 48 h, the cells were grown in fresh DMEM with 10% FBS and 5 μg/ml insulin. Finally, the medium was replaced by fresh DMEM after 48h. Six days later, the cells were incubated with vehicle or LPS (100 ng/ml) for 6 h. In some experiments cells were allowed to grow to confluence for 2 days and then incubated in DMEM containing a differentiation cocktail in the presence of vehicle (0.01% DMSO), TGF-β (10 ng/ml) and PGE2 (0.1, 1 and 5 μM). After 48 h, the cells were grown in fresh DMEM with 10% FBS and 5 μg/ml insulin. Finally, the medium was replaced by fresh DMEM after 48 h.
High-throughput transcriptomic analysisTotal RNA was obtained from omental WAT collected from obese patients (n = 4) and non-obese individuals (n = 4). RNA was also obtained from primary cultures of human pre-adipocytes (n = 8) incubated in the absence or presence of PGE2 (1 μM) for 3 h. Affymetrix Human Genome U219 expression arrays containing more than 36,000 transcripts and variants (Affymetrix, Inc., Santa Clara, CA) were used to process all the samples. The preparation of cRNA probes, hybridization, and scanning of arrays were performed according to the manufacturer’s protocol and carried out at the Functional Genomics Unit of IDIBAPS. Affymetrix gene expression data were normalized with the robust multiarray algorithm using a custom probe set definition that mapped probes directly to Entrez Gene Ids (HGU219_Hs_ENTREZG). A filtering step excluding probes not reaching a coefficient of variation of 0.02 was employed, and the final number of probes reached a total of 13,928 probes. For the detection of differentially expressed genes, a linear model was fitted to the data and empirical Bayes moderated statistics were calculated using the Limma package from Bioconductor (Seattle, WA). Adjustment of p-values was made by the determination of false discovery rates (FDR) using the Benjamini-Hochberg procedure [19]. All computations were made using R statistical software. Genes representing a fold change of 1.5 or greater and a moderate p-value