Received: 20 March 2021 Revised: 22 October 2021 Accepted: 15 November 2021 DOI: 10.1111/bph.15776 RESEARCH ARTICLE Microsomal prostaglandin E synthase-1 is involved in the metabolic and cardiovascular alterations associated with obesity Constanza Ballesteros-Martínez1 | Raquel Rodrigues-Díez1,2 | Luis M. Beltrán3,4 | Rosa Moreno-Carriles5 | Ernesto Martínez-Martínez6 | María González-Amor1,2 | Jose Martínez-González2,7 | Cristina Rodríguez2,8 | Victoria Cachofeiro2,6 | Mercedes Salaices1,2 | Ana M. Briones1,2 1 noma de Madrid, Instituto de Investigacio Departamento de Farmacología, Facultad de Medicina, Universidad Auto n Hospital Universitario La Paz (IdiPaz), Madrid, Spain 2 CIBER de Enfermedades Cardiovasculares, ISCIII, Madrid, Spain 3 Servicio de Medicina Interna, Hospital Universitario La Paz, IdiPaz, Madrid, Spain 4 Servicio de Medicina Interna, Hospital Virgen del Rocío - IBiS Sevilla, Departamento de Medicina, Universidad de Sevilla, Sevilla, Spain 5 Servicio de Angiología y Cirugía vascular, Hospital Universitario La Princesa, Madrid, Spain 6 n Sanitaria Gregorio Maraño Departamento de Fisiología, Facultad de Medicina, Universidad Complutense de Madrid, Instituto de Investigacio n (IiSGM), Madrid, Spain 7 n Biomédica (IIB) Sant Pau, Barcelona, Spain Instituto de Investigaciones Biomédicas de Barcelona (IIBB-CSIC), Instituto de Investigacio 8 Institut de Recerca Hospital de la Santa Creu i Sant Pau (IRHSCSP), IIB-Sant Pau, Barcelona, Spain Correspondence Ana M. Briones, Departamento de Background and Purpose: Microsomal prostaglandin E synthase-1 (mPGES-1) is an Farmacología, Facultad de Medicina, noma de Madrid, inducible isomerase responsible for prostaglandin E2 production in inflammatory con- Universidad Auto C/Arzobispo Morcillo 4, 28029 Madrid, Spain. ditions. We evaluated the role of mPGES-1 in the development and the metabolic Email: ana.briones@uam.es and cardiovascular alterations of obesity. Funding information Experimental Approach: mPGES-1+/+ and mPGES-1/ mice were fed with normal n. Programa Ministerio Ciencia e Innovacio or high fat diet (HFD, 60% fat). The glycaemic and lipid profile was evaluated by glu- Juan de la Cierva, Grant/Award Number: IJCI- 2017-31399; Comunidad de Madrid, Grant/ cose and insulin tolerance tests and colorimetric assays. Vascular function, structure Award Number: B2017/BMD-3676 and mechanics were assessed by myography. Histological studies, q-RT-PCR, and AORTASANA, FEDER-a way to build Europe; Fondo Europeo de Desarrollo Regional, Grant/ western blot analyses were performed in adipose tissue depots and cardiovascular Award Number: SAF2016-80305P; Ministerio n, Grant/Award tissues. Gene expression in abdominal fat and perivascular adipose tissue (PVAT) de Ciencia e Innovacio Number: FIS PI18/0257 from patients was correlated with vascular damage. Abbreviations: AA, arachidonic acid; AAA, abdominal aortic aneurysm; BMI, body mass index; BP, blood pressure; CLS, crown-like structures; cPGES, cytosolic prostaglandin E synthase; DEA- NO, diethylamine NONOate; GTT, glucose tolerance test; HFD, high fat diet; IMT, intima media thickness; ITT, insulin tolerance test; KHS, Krebs–Henseleit solution; mPGES-1, microsomal prostaglandin E synthase-1; mPGES-2, microsomal prostaglandin E synthase-2; MRAs, mesenteric resistance arteries; PFA, paraformaldehyde; PLA2α, phospholipase A2α; PVAT, perivascular adipose tissue; PWV, pulse wave velocity; WAT, white adipose tissue; WGA, wheat germ agglutinin. Constanza Ballesteros-Martínez and Raquel Rodrigues-Díez contributed equally to this study. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. © 2021 The Authors. British Journal of Pharmacology published by John Wiley & Sons Ltd on behalf of British Pharmacological Society. Br J Pharmacol. 2022;179:2733–2753. wileyonlinelibrary.com/journal/bph 2733 14765381, 2022, 11, Downloaded from https://bpspubs.onlinelibrary.wiley.com/doi/10.1111/bph.15776 by Readcube (Labtiva Inc.), Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 2734 BALLESTEROS-MARTÍNEZ ET AL. Key Results: Male mPGES-1/ mice fed with HFD were protected against body weight gain and showed reduced adiposity, better glucose tolerance and insulin sensi- tivity, lipid levels and less white adipose tissue and PVAT inflammation and fibrosis, compared with mPGES-1+/+ mice. mPGES-1 knockdown prevented cardiomyocyte hypertrophy, cardiac fibrosis, endothelial dysfunction, aortic insulin resistance, and vascular inflammation and remodelling, induced by HFD. Obesity-induced weight gain and endothelial dysfunction of resistance arteries were ameliorated in female mPGES-1/ mice. In humans, we found a positive correlation between mPGES-1 expression in abdominal fat and vascular remodelling, vessel stiffness, and systolic blood pressure. In human PVAT, there was a positive correlation between mPGES-1 expression and inflammatory markers. Conclusions and Implications: mPGES-1 inhibition might be a novel therapeutic approach to the management of obesity and the associated cardiovascular and meta- bolic alterations. KEYWORDS adipose tissue alterations, inflammation, mPGES-1, obesity, vascular function and remodelling 1 | I N T RO DU CT I O N What is already known Obesity, particularly visceral obesity, is one of the major risk factors in the development of metabolic disorders such as Type 2 diabetes and • mPGES-1 deficiency prevents obesity-induced weight insulin resistance (Vecchié et al., 2018). In normal conditions, adipose gain. tissue participates in the maintenance of body homeostasis, including normal vascular structure and tone. However, dysfunctional obese adipose tissue is characterized by a proinflammatory status and secre- What does this study add tion of proinflammatory mediators associated with the development of cardiovascular disease (Fuster et al., 2016; Vecchié et al., 2018). • mPGES-1 deficiency prevents obesity-induced metabolic In physiological conditions, prostaglandin E2 (PGE2) is synthesized and cardiovascular alterations. from tissue phospholipids and arachidonic acid (AA) by the sequential • In humans, mPGES-1 positively correlates with inflamma- effects of phospholipase A2α (PLA2α), cyclooxygenase (COX)-1, and tion and vascular damage. two constitutive PGE2 synthases (PGES): microsomal PGES (mPGES)-2 and cytosolic PGES (cPGES). However, in inflammatory conditions, inducible COX-2 and mPGES-1 isomerase are the main What is the clinical significance enzymes involved in the biosynthesis of PGE2 (Koeberle & Werz, 2015; Samuelsson et al., 2007). PGE2 is one of the most abun- • mPGES-1 inhibition could prevent obesity-induced meta- dant prostanoids released by adipose tissue (Fain et al., 2002) and bolic and vascular alterations. there is increasing evidence that PGE2 affects adipogenesis and has • mPGES-1 inhibitors might be a safe therapeutic tool in anti-lipolytic effects (García-Alonso & Clària, 2014), thus promoting obese patients. the enlargement of fat depots. Moreover, preadipocytes release proinflammatory mediators in a PLA2/PGE2 dependent manner (Leiguez et al., 2020). However, studies evaluating changes in PGE2 levels or in enzymes involved in PGE2 synthesis in adipose tissues in obesity have reported conflicting results. Thus, Hétu and (Jaworski et al., 2009), COX-2 (Fain et al., 2001; Ghoshal et al., 2011), Riendeau (2007) found significantly decreased levels of PGE2 in sub- or mPGES-1 (Pierre et al., 2018) are protected against body weight cutaneous and inguinal white adipose tissue (WAT) in murine obesity, gain, adipose tissue enlargement, and/or inflammation induced by a whereas other authors found augmented PGE2 levels in isolated adi- high fat diet (HFD). On the other hand, no differences in gain weight pocytes from obese rats (Chan et al., 2016) or in human obese WAT following mPGES-1 deletion have been described in animals fed a (García-Alonso et al., 2016). Importantly, mice lacking PLA2α modified HFD containing a lower amount of fat for shorter times 14765381, 2022, 11, Downloaded from https://bpspubs.onlinelibrary.wiley.com/doi/10.1111/bph.15776 by Readcube (Labtiva Inc.), Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License BALLESTEROS-MARTÍNEZ ET AL. 2735 (Sasaki et al., 2021). Of note, the contribution of mPGES-1 to glucose Angiology and Vascular Surgery Unit of the Hospital La Princesa. and lipid metabolism in obesity has been less explored. Inclusion criteria include patients with an AAA ≥ 5.5 cm (≥5 cm in In addition to metabolic alterations, obesity is associated with car- women) in diameter, proven by imaging technique, symptomatic or diovascular complications such as endothelial dysfunction, thickened asymptomatic, or those who during the follow-up would have experi- vascular wall, increased vascular stiffness, and cardiac hypertrophy enced rapid growth (more than 0.5 cm in a 6-month interval) and that and fibrosis (Briones et al., 2014; Martínez-Martínez et al., 2015; met the criteria for open surgery. Exclusion criteria include inflamma- Prieto et al., 2014) that clearly contribute to enhance cardiovascular tory aneurysm, active neoplastic conditions, HIV positive serology and risk. Different factors are involved in the cardiovascular alterations pregnancy. Characteristics of the studied population are included in associated with obesity. Specifically, the participation of prostanoids Table S2. Both abdominal and aortic adipose tissue were frozen at derived from COX-2 in the functional vascular alterations induced by 80 C for qRT-PCR analysis. obesity has been demonstrated (Briones et al., 2014; Even et al., 2014; Muñoz et al., 2015; Prieto et al., 2014; Vessières et al., 2010). However, the contribution of mPGES-1 to the cardiovas- 2.2 | Animal models cular alterations associated to obesity is currently unknown. We hypothesized that in obesity, there is an inflammatory pro- All experimental procedures and animal care were approved by the cess mediated by mPGES-1-derived prostanoids that contributes to noma de Ethical Committee of Research of the Universidad Auto the structural and functional cardiovascular alterations and to the  n General de Medio Ambiente, Comunidad Madrid (UAM) and Direccio metabolic changes associated with this pathology. We performed de Madrid, Spain (PROEX 345/14, PROEX 183.2/20), and conducted studies in a model of obesity in mPGES-1 knockout mice and in in accordance with the current Spanish and European Policy for Ani- humans. mal Protection RD53/2013. The experiments were conducted in accordance with the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals and the ARRIVE (Percie du Sert 2 | METHODS et al., 2020) and BJP guidelines (Lilley et al., 2020). mPGES-1/ mice and their wild-type littermates (mPGES-1+/ + 2.1 | Human studies ) on C57BL/6J background were generated by crossing heterozy- gous knockout males in DBA genetic background kindly donated These studies were conducted according to the principles of the Dec- by Pfizer with wild-type females on C57BL/6J background, bred at laration of Helsinki and approved by the Institutional Research Ethics the Animal Care Facility of the Faculty of Medicine, UAM, for Committees of Hospital Universitario La Paz (PI-1747) and Hospital 11 generations. Universitario La Princesa (PI-825) (Madrid, Spain). All subjects were Two-month-old male and female mPGES-1+/+ and mPGES-1/ aware of the research nature of the study and gave informed consent. mice were randomly divided in two groups: Control, fed with a stan- Abdominal adipose tissue was obtained from patients during abdomi- dard diet (5LF2* EURodent Diet, 6.6% kcal from fat, LabDiet, USA); nal surgery in the Hepatobiliary Surgery Unit of the Hospital La Paz and obese, fed with high fat diet (HFD) (TD.06414, 60.3% kcal from and the vascular study was carried out in the Cardiovascular Unit in fat, ENVIGO, USA) for 13 weeks. This HFD model has been selected the same hospital. Patients (18- to 75-years-old) scheduled for elec- based on previous studies as it reproduces some of the features of tive abdominal surgery were offered to participate in the study. Exclu- human obesity (Buettner et al., 2007; Wang & Liao, 2012). sion criteria included inflammatory and neoplastic conditions, active Animals were randomly distributed in the different experimental smoking, or use of recreational drugs, established cardiovascular dis- groups with each group having the same number of animals by design. ease and pregnancy. Recruited patients were subjected to complete We did not utilize statistical methods to predetermine sample or anamnesis, physical examination, and blood pressure (BP), pulse wave group sizes, and the number of animals used was estimated from our velocity (PWV), and intima media thickness (IMT) measurements. BP previous experience with this animal model. Investigators for outcome was determined using an automatic oscillometric device OMRON-M3 assessments were not blinded to group allocation unless specified. All (HEM-7200-E-Omron Healthcare, Kyoto, Japan) according to interna- mice were bred at the conventional Animal Care Facility of the Faculty tional guidelines recommendations (Mancia et al., 2013). PWV was  noma de Madrid (UAM) under con- of Medicine, Universidad Auto determined by applanation tonometry using SphygmoCor device trolled conditions at 25 C in a 12-h light/dark cycle, with ad libitum (AtCor Medical, Australia). IMT was measured according to interna- access to water and food. Animals were housed in groups of two or tional consensus recommendations (Touboul et al., 2012) using a four in standard polypropylene cages containing rich bedding made of Mylab Seven system (Esaote, Genova, Italy) with a 4- to 13-MHz dried wood chips. probe. The system employed a dedicated software radiofrequency- Mice were weighed weekly. Blood pressure was measured by tracking technology to obtain IMT (QIMT®) (Dan et al., 2011). The tail-cuff plethysmography at the end of the procedure after previous characteristics of the patients studied are summarized in Table S1. training; eight individual observations per animal were performed and Perivascular adipose tissue (PVAT) was obtained from patients averaged. All the measurements were made at the same time of the undergoing open surgery for abdominal aortic aneurysm (AAA) in the day. The animal study complied with the 3Rs. 14765381, 2022, 11, Downloaded from https://bpspubs.onlinelibrary.wiley.com/doi/10.1111/bph.15776 by Readcube (Labtiva Inc.), Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 2736 BALLESTEROS-MARTÍNEZ ET AL. 2.3 | Tissue preparation incubated with haematoxylin for 4 min, washed and stained with eosin for 5 min. For each mouse, 3 images of adipose tissue were Mice were killed by CO2 narcosis. Aorta and first-order branches taken and analysed using ImageJ software (ImageJ, NIH, Bethesda, of the mesenteric artery were cleaned and placed in cold (4 C) MD, USA, RRID:SCR_001935). Images were duplicated and calibrated. Krebs–Henseleit solution (KHS) (in mmolL1) (115 NaCl, Background was subtracted, and the excess of noise was removed. A 25 NaHCO3, 4.7 KCl, 1.2 MgSO4.7H2O, 2.5 CaCl2, 1.2 KH2PO4, threshold was applied to the images which were then converted to 11.1 glucose, and 0.01 Na2EDTA) bubbled with a 95% O2–5% CO2 binary format. Borders of adipocytes were enhanced and adipocytes mixture (pH = 7.4). Segments of aorta and mesenteric arteries were with broken membranes were separated manually with the “cut draw- immediately used for functional, structural, or mechanical studies ing” tool. Finally, adipocytes area was measured as described (Parlee using wire or pressure myography or immediately frozen and et al., 2014). Data analysis was carried out with the average of the  stored at 80 C. images of each mouse. Aortic PVAT was also removed and frozen at 80 C for qPCR Aortic rings were stained with haematoxylin and eosin as analysis. WAT and perirenal adipose tissue were cleaned, weighed described above to determine aortic structure. Vessel internal and and frozen at 80 C or fixed in 4% paraformaldehyde (PFA, pH = 7.4) external perimeter (Pi, Pe) were measured using ImageJ Software  and kept at 4 C for histological studies. Adiposity index (%) was calcu- (ImageJ, NIH). Cross-sectional area (CSA) was calculated as CSA = ( lated: [(abdominal + perirenal fat pads)/(final body weight  sum of Pe2-Pi2)/4π. Results were quantified without knowledge of the sample abdominal and perirenal fat pads)  100]. treatments. Left cardiac ventricles were isolated, weighed, and frozen at Cardiac, WAT and aorta sections were stained with Picrosirius   80 C or fixed in 4% PFA and kept at 4 C for histological studies. red to detect collagen fibres. Sections were incubated 2 min with Cardiac hypertrophy was calculated as the left ventricle weight nor- picric acid and washed. Then, they were stained with Sirius red solu- malized by the tibia length. tion for 1 h and dehydrated. The area of cardiac interstitial fibrosis was identified as the ratio of interstitial fibrosis to the total tissue area after excluding vessel areas from the region of interest. For each 2.4 | Insulin and glucose tolerance tests mouse, 5 fields of two different sections were analysed and averaged. In adipose tissue samples, the area of pericellular fibrosis was identi- The glycaemic profile was studied 2 weeks before the end of the fied as the ratio of collagen deposition to the total tissue area after experiment. Mice were fasted 4 h for the insulin tolerance test (ITT) excluding vessels area from the region of interest. This value was nor- and 16 h for the glucose tolerance test (GTT). Animals were weighed malized by the number of adipocytes. For each sample, 10–15 fields and basal glucose levels were measured. Then, insulin (Insulin Actrapid were analysed using a 40  objective (Leica DM 2000; Leica Camera 100 UI/ml, NOVO NORDISK, 0.8 UI/Kg of mouse) or hypertonic glu- AG, Wetzlar, Germany) and quantified (Leica Q550 IWB; Leica Cam- 1 cose solution (10%, Fresenius Kabi, 20 μlkg of mouse) were era AG) and averaged. In aorta, images were captured using an Olym- injected i.p.. Glycaemia was measured from tail blood at different time pus BX50 microscope at 10 magnification. The percentage of points (15, 30, 60, 90, and 120 min) using a standard glucometer fibrosis was determined as the ratio between collagen area in the (FreeStyle Optium). media and the total medial area. Results were quantified without knowledge of the sample treatments. Mean cardiomyocyte area was evaluated in Wheat Germ Aggluti- 2.5 | Measurement of lipid levels nin (WGA)-stained cardiac sections. Tissue sections were incubated in a wet chamber with WGA-CY3 (5 μgml1, Biotium) and Hoechst Total cholesterol (Ref. 1001091, SpinReact), triglycerides (Ref. 33342 (0.01 mgml1, Thermofisher) in PBS-Tween (PBS-T, 0.05%), 1001313, SpinReact), LDL-Cholesterol-D (Ref. 41023, SpinReact) and during 1 h at room temperature. Samples were washed with PBS-T, oxLDL (Ref. ab242301, Abcam) levels were evaluated in plasma sam- covered-slipped using Prolong and kept at 4 C. Three stacks per ani- ples following manufacturers´ instructions. mal were captured with a confocal microscope (Leica TCS SP5, Leica Microsystems) and individual projections were reconstructed. At least 30 cardiomyocytes in each field were measured using Metamorph 2.6 | Histological studies Image Analysis Software (Molecular Devices Corp. Downingtown, PA, USA. RRID:SCR_002368) and averaged. PFA fixed tissue samples were dehydrated, embedded in paraffin, and cut in 5 μm-thick sections. Then, sections were deparaffinized and hydrated (5 min Xilol, 5 min EtOH 100%, 5 min EtOH 70%, 5 min 2.7 | Immunohistochemistry studies H2O) for further analysis. All stains were coverslipped with DPX mounting medium. Immunohistochemistry was performed in 5 μm thick paraffined WAT WAT sections were stained with haematoxylin and eosin to mea- sections to detect macrophage infiltration. After de-waxing, antigen sure adipocyte size and number. After rehydration, samples were retrieval was performed using Proteinase-K method by incubation 14765381, 2022, 11, Downloaded from https://bpspubs.onlinelibrary.wiley.com/doi/10.1111/bph.15776 by Readcube (Labtiva Inc.), Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License BALLESTEROS-MARTÍNEZ ET AL. 2737 with Proteinase-K (Roche, 20 mgml1 in Tris-acetato-EDTA buffer) Wall thickness ðWTÞ ¼ ðDe0Ca −Di0Ca Þ=2; for 3 min and washed with PBS-Tween (PBS-T, 0.5%). Endogenous wall : lumen ¼ ðDe0Ca −Di0Ca Þ=2Di0Ca : peroxidase activity was blocked by 10% H2O2 (Panreac, 10 min). Free protein binding sites were blocked with 2% fetal bovine serum (FBS, Circumferential wall strain (ε) = (Di0Ca  D00Ca)/D00Ca, where Sigma, 1 h). Then, samples were incubated overnight with a rat pri- D00Ca is the internal diameter at 3 mmHg and Di0Ca is the observed mary antibody against mouse F4/80 (clone Cl:A3-1. Bio-Rad Cat# internal diameter for a given intravascular pressure both measured in MCA497RT, RRID:AB_1102558, 1/100). Samples were washed with 0Ca2+ medium. Circumferential wall stress (σ) = (P  Di0Ca)/(2WT), PBS-T and incubated with a secondary HRP-conjugated anti-rat anti- where P is the intraluminal pressure (1 mmHg = body (Millipore Cat# AP136P, RRID:AB_11214444) for 1 h. HRP 1.334  103 dynescm2) and WT is the wall thickness at each activity was visualized using diaminobenzidine (VectorLab) as sub- intraluminal pressure in 0Ca2+-KHS. Incremental distensibility repre- strate. Sections were counterstained with haematoxylin, dehydrated sents the percentage of change in the arterial internal diameter for and mounted in DPX mounting medium. Crown-like structures (CLS) each interval change in intraluminal pressure, and was calculated as were defined in the sections as single and entire dead adipocytes, follows: partly or totally encircled by macrophages. Incremental distensibility ¼ ΔDi0Ca =ðDi0Ca  ΔPÞ  100: 2.8 | Vascular function The arterial stiffness independent of the geometry was deter- mined by Young's elastic modulus (E = stress/strain). The stress– Reactivity of mouse aorta and mesenteric resistance arteries was strain relationship is nonlinear; therefore, it is more appropriate to studied in a wire myograph (Danish Myo Tech, Aarhus, Denmark). obtain a tangential or incremental elastic modulus (Einc) by determin- After a 30-min equilibration period in oxygenated KHS, arterial ing the slope of the stress–strain curve (Einc = δσ/δε). Einc was segments were stretched to their optimal lumen diameter for obtained by fitting the stress–strain data from each animal to an expo- active tension development. Contractility of segments was tested nential curve using the equation: σ = σ origeβε, where σ orig is the stress + + by an initial exposure to a high-K solution (K -KHS, at the original diameter (diameter at 3 mmHg). Taking derivatives of 120 mmolL1). Arteries were discarded if K+-KHS response was the equation, we determined that Einc = βσ. For a given σ-value, Einc is lower than 0.5 mNmm1. Concentration–response curves to directly proportional to β. An increase in β implies an increase in Einc, acetylcholine (1 nmolL1 to 30 μmolL1), insulin (10 nmolL1 to which indicates an increase in stiffness. 30 μmolL1), or the NO donor diethylamine NONOate (DEA-NO, 10 nmolL1 to 30 μmolL1) were obtained after precontraction with phenylephrine. To minimize variation, vasodilator responses 2.11 | Organization of internal elastic lamina were expressed as the percentage of relaxation of the tone induced by phenylephrine. The 3D structure of the internal elastic lamina of mesenteric arteries pressure-fixed at 45 mmHg was studied using fluorescence confocal microscopy (Leica TCS SP5, Leica Microsystems) based on the auto- 2.9 | Pressure myography fluorescent properties of elastin (Ex 488 nm and Em 500–560 nm). Serial optical sections were captured from the adventitia to the lumen The mechanical and structural properties of mesenteric arteries were (z step = 0.5 μm) using a X63 oil objective (Zoom 4). At least four sta- studied with a pressure myograph (Danish Myo Tech, Aarhus, cks of images from different regions per artery were taken. From each Denmark). PVAT-free vessels were mounted on two glass stack of images, individual projections of the internal elastic lamina microcannulae, set to a pressure of 45 mmHg and equilibrated for were reconstructed and mean fenestrae area and number were calcu- 30 min at 37 C in calcium-free KHS (0 mM Ca2+; omitting calcium lated taking into account the luminal surface area occupied by the and adding 1 mmolL1 EGTA). A pressure-diameter curve was internal elastic lamina in a 1-mm length segment. obtained by increasing intraluminal pressure in 20 mmHg steps from 3 to 120 mmHg. Every intraluminal pressure was maintained for 2 min. Internal and external diameters were measured under passive 2.12 | RT-PCR assay conditions (Di0Ca, De0Ca) at each intraluminal pressure using ImageJ software. Total RNA was obtained with TRI Reagent® (Sigma-Aldrich); 1 μg was reversed transcribed to cDNA with NZY First-Strand cDNA Synthesis Kit (Nzytech). Quantitative PCR (qPCR) was performed in a 7500 Fast 2.10 | Calculation of passive mechanical and ABI System (Invitrogen Life Technologies) in duplicate to ensure the structural parameters reliability of qRT-PCR values, using specific q-PCRs probes and assays (Table S3). Data were analysed using the average of the duplicate Structural and mechanical parameters were calculated as follows: values. β2-microgobulin, ribosomal protein large P0, or 18 s were used 14765381, 2022, 11, Downloaded from https://bpspubs.onlinelibrary.wiley.com/doi/10.1111/bph.15776 by Readcube (Labtiva Inc.), Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 2738 BALLESTEROS-MARTÍNEZ ET AL. as calibrators. PCR cycle programmes were as follows: initial denatur- declared group size is the number of these biological replicates, and ation (30 s, 95 C), followed by 40 cycles (95 C, 5 s and 60 C, 30 s). statistical analysis was performed using these independent values. Melting curve was performed in SYBR green reactions to show PCR Statistical analysis was undertaken only for studies in which each product specificity. To calculate the relative index of gene expression, group size was at least n = 5. In some experiments, group sizes we employed the 2ΔΔCt method. mRNA expression in mice was nor- became unequal during the study owing to biological loss (i.e., death malized against mPGES-1+/+ control samples to minimize variation. of the mouse), technical failure, undetectable expression or presence ΔΔCt For human samples, mRNA expression is expressed as Log (2 ) of of outliers that were excluded using predefined criteria. The ROUT the corresponding gene after normalization with its calibrator. method was used to exclude data from the analysis. After Shapiro– Wilk normality test to determine data distribution, differences between groups were determined by one-way or two-way ANOVA or 2.13 | Lipid peroxidation its corresponding non-parametric test (Kruskal–Wallis) attending to the normal distribution of the samples, followed by Bonferroni's or Lipid peroxidation in frozen segments of aorta and heart was deter- Dunn's post hoc test when appropriate. In human studies, gene mined using a lipid peroxidation (malondialdehyde) assay kit expression was log-transformed to obtain a normal distribution. Cor- (ab118970, Abcam) following the manufacturer’s instructions. relations were assessed using Spearman's correlation tests. Data were analysed with the GRAPHPAD PRISM 7 software (Graphpad Soft- ware, Inc., San Diego, CA, USA. RRID:SCR_002798). P < 0.05 was 2.14 | PGE2 levels considered significant. PGE2 levels were determined in frozen segments of aorta and heart using the PGE2 ELISA Kit – Monoclonal (No. 514010, Cayman). Sam- 2.17 | Materials ples were homogenized (0.1-molL1 phosphate, pH 7.4, 1mmolL1 EDTA) with a Dounce homogenizer. The assay was performed follow- All drugs were dissolved in distilled water and purchased from Sigma- ing the manufacturer’s instructions. PGE2 concentrations were calcu- Aldrich (Madrid, Spain), unless otherwise specified. Primary antibodies lated as the amount of PGE2 per microgram of protein. used were supplied as follows: TGF-β, Abcam, Ref. ab190503, 1/1000), α-SMA, GeneTex, Ref. GTX100034, 1/250), GAPDH (Cell Signaling Technology Cat# 8884, Ref: D16H11, 1/5000) and F4/80 2.15 | Western blot analysis (Bio-Rad Cat# MCA497, clone Cl:A3-1. AbD Serotec, 1/100). Total protein from the left ventricle were obtained with TRI Reagent® (Sigma-Aldrich) following the manufacturer’s instructions. Cardiac 2.18 | Nomenclature of targets and ligands proteins were separated by SDS-PAGE and transferred to Hybond-c Extra nitrocellulose membranes (Hybond-P; Amersham Biosciences, Key protein targets and ligands in this article are hyperlinked to Piscataway, NJ) with the Trans-Blot Turbo Transfer System (Bio.Rad). corresponding entries in the IUPHAR/BPS Guide to PHARMACOL- Membranes were probed overnight with primary antibodies for TGF-β OGY (http://www.guidetopharmacology.org) and are permanently (Abcam, Ref. ab19053, RRID:AB_778340, 1/1000), α-smooth muscle archived in the Concise Guide to PHARMACOLOGY 2021/22 actin (α-SMA, GeneTex, Ref. GTX100034, RRID:AB_1240408, 1/250), (Alexander, Christopoulos et al., 2021; Alexander, Cidlowski and GAPDH (Cell Signaling Technology Cat# 8884, RRID: et al., 2021; Alexander, Fabbro et al., 2021; Alexander, Kelly AB_11129865, 1:5000) that was used as a loading control. Secondary et al., 2021). antibodies were incubated for 1 h. Signals were detected using the ECL system (Amersham Pharmacia Biotech). Results are expressed as an n-fold increase over the values of the control group in densitomet- 3 | RE SU LT S ric arbitrary units, to minimize variation. 3.1 | mPGES-1 deficiency protects against body weight gain and improves the adverse glycaemic and 2.16 | Data and statistical analysis lipid profile induced by a HFD The data and statistical analysis in this study complied with the rec- In males given the HFD, Ptges mRNA levels did not change in WAT, ommendations on experimental design and analysis in pharmacology PVAT, aorta, or heart, although a tendency to decrease was observed (Curtis et al., 2018). Data are expressed as mean ± SEM of the number in WAT (Figure S1A). PGE2 levels were not changed, either by diet or of animals or patients studied. Data are reported as dot plots that rep- genotype in aorta or heart (Figure S1B,C). resent different biological replicates. When this is not possible, the mPGES-1/ mice fed on normal diet gained similar weight over number of animals (n) is reported in figures or figure legends. The the course of the experiment and had similar basal glycemia, glucose 14765381, 2022, 11, Downloaded from https://bpspubs.onlinelibrary.wiley.com/doi/10.1111/bph.15776 by Readcube (Labtiva Inc.), Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License BALLESTEROS-MARTÍNEZ ET AL. 2739 F I G U R E 1 mPGES-1 deficiency protects against high fat diet (HFD)-induced body weight gain and improves the glycaemic and lipid profiles. (a,b) Body weight gain, (c) basal glucose levels, (d) glucose tolerance test (GTT), (e) area under the curve (AUC) of GTT, and (f,g) insulin tolerance test (ITT), in mPGES-1+/+ and mPGES-1/ mice fed with normal diet (control) and HFD. (h–j) Triglycerides, total cholesterol, and LDL- cholesterol levels in plasma samples of mice from all experimental groups. In bar graphs, each dot represents one mouse. Data shown are means ± SEM, with individual values in (b,c,e,h–j). *P < 0.05, significantly different from control of each genotype. +P < 0.05, significantly different from mPGES-1+/+ under the same diet; one-way (b,c,e,h–j) or two-way ANOVA (a,d,f,g) or non-parametric Kruskal–Wallis test. The number of animals in each group is shown in parentheses or below each bar 14765381, 2022, 11, Downloaded from https://bpspubs.onlinelibrary.wiley.com/doi/10.1111/bph.15776 by Readcube (Labtiva Inc.), Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 2740 BALLESTEROS-MARTÍNEZ ET AL. tolerance and insulin sensitivity and lipid levels that were similar to 3.4 | mPGES-1 deficiency protects against HFD- those in mPGES-1+/+ mice (Figure 1). mPGES-1 gene deletion partly induced vascular and PVAT inflammation prevented the increase in body weight gain in HFD-fed animals, com- pared with wild-type littermates (Figure 1a,b). These effects were HFD increased aortic mRNA levels of several inflammatory genes accompanied by an improvement in basal glycaemia, glucose toler- such Ccl2, Il1b, Tnfa, interferon stimulated gene 15 (Isg15), and ance, insulin sensitivity, and lipid levels in obese animals (Figure 1c–j), those for NADPH oxidase 1, 2, and 4 (Nox1, Cybb, Nox4) in without significant changes in plasma oxLDL (mPGES-1+/+ Control: mPGES-1+/+ but not in mPGES-1/ mice (Figure 4a,b). Unexpect- +/+ / 0.015 ± 0.003; mPGES-1 HFD: 0.013 ± 0.004; mPGES-1 Con- edly, under basal conditions, mPGES-1/ knockdown enhanced / 1 trol: 0.020 ± 0.002; mPGES-1 HFD: 0.019 ± 0.003 μgml , n=6 the expression of Ccl2, Il1b, and Nox1 (Figure 4a,b). No differences all groups). in MDA levels were observed in the four experimental groups (Figure 4c). Given the importance of PVAT as a key location for accumulation 3.2 | mPGES-1 deficiency protects against HFD- of immune cells in obesity, we addressed the consequences of induced adiposity, fibrosis, and macrophage infiltration mPGES-1 deletion on the inflammatory profile of aortic PVAT. In agreement with results observed in WAT, mRNA levels of Adgre1 and mPGES-1 knockdown did not affect the weight of adipose tissues in Ccl2 were increased in response to HFD in mPGES-1+/+ but not in male mice fed on normal diet (Table S4). However, HFD did increase mPGES-1/ mice (Figure 4d,e). No changes in Tnfa or the lympho- the perirenal and the abdominal adipose tissue weight, and adiposity cyte marker Cd3 expression were observed between groups +/+ / index, more in mPGES-1 than in mPGES-1 mice (Table S4). Adi- (Figure 4f,g). The expression of Adipoq (Figure 4h) did not change with pocyte size and collagen deposition in WAT was similar in mPGES- HFD but it was diminished in HFD-fed mPGES-1/ mice. We did not 1/ and mPGES-1+/+ mice fed on normal diet (Figure 2a,b). HFD observe differences in inflammatory markers or adiponectin in PVAT increased adipocyte size and adipose tissue fibrosis in mPGES-1+/+ from mice given normal diet (Figure 4d–h). / mice, but not in mPGES-1 mice (Figure 2a,b). No differences in the expression of Lep, Adipoq, Pparg, or macro- phage markers Ccl2 and Adgre1 expression or CLS presence) were 3.5 | mPGES-1 deficiency partially protects against observed between genotypes in mice fed on normal diet (Figure 2c–e, HFD-induced structural and mechanical alterations g–i). Strikingly, mPGES-1 knockdown increased basal Tnfa expression (Figure 2f). HFD increased Lep mRNA in WAT from mPGES-1+/+ but HDF increased aorta cross-sectional area in mPGES-1+/+ but not in not in knockout mice (Figure 2c). Conversely, the HFD-induced mPGES-1/ mice (Figure 5a). Structural and mechanical properties adiponectin expression was greater in mPGES-1/ than in mPGES- were similar in mesenteric resistance arteries from control male +/+ 1 mice (Figure 2d). Pparg expression was similarly augmented by mPGES-1+/+ and mPGES-1/ mice although mPGES-1/ mice the HFD in both groups (Figure 2e). HFD tended to increase Tnfa showed greater wall stress likely due to slight, but not significant, dif- expression in mPGES-1+/+ animals whereas the opposite was ferences in lumen diameter and wall thickness (Figure 5b–g). HFD observed in knockout mice (Figure 2f). Finally, HFD increased expres- decreased lumen diameter in arteries from both genotypes although sion of Ccl2 and the macrophage marker Adgre1 and the presence of statistical significance was only reached in knockout mice (Figure 5b). CLS in mPGES-1+/+ mice, but not in mPGES-1/ mice (Figure 2g–i). Moreover, wall thickness was higher in mPGES-1+/+ than mPGES-1/  mice fed with HFD (Figure 5c). This geometric adaptation in mPGES-1 knockout mice normalized circumferential wall stress that 3.3 | mPGES-1 deficiency protects against HFD- was otherwise decreased in obese wild type animals, probably to pro- induced endothelial dysfunction and vascular insulin tect the vessel wall from possible changes in intraluminal pressure resistance (Figure 5d). Strain was decreased by HFD in both genotypes (Figure 5e). Moreover, incremental distensibility was decreased and Systolic blood pressure (SBP) was similar in all the experimental vascular stiffness was increased by HFD in mPGES-1+/+ but not in groups (Table S5). In male mice, HFD impaired endothelium- knockout mice (Figure 5f,g). dependent vasorelaxation to acetylcholine and to insulin in aorta We then looked at vascular and PVAT fibrosis markers. We found from mPGES-1+/+ but not in knockout mice (Figures 3a,b and S2). no differences between groups in collagen and Tgfb expression in Similar results were observed in mesenteric resistance arteries aorta (Figure S3A,B). HFD increased Fn1 expression in aortic PVAT (Figure 3e,f). No differences in vasodilator responses to DEA-NO from mPGES-1+/+ but not in mPGES-1/ mice (Figure S3C). How- were observed between groups, either in aorta (Figure 3c,d) or in ever, Tgfb and Ctgf gene expression remained unchanged by diet or mesenteric arteries (Figure 3g,h), thus excluding changes in vascular genotype in aortic PVAT (Figure S3D,E). No differences were found in smooth muscle cells sensitivity to NO as responsible for the 3D elastin structure in the internal elastic lamina of mesenteric arter- observed effects. ies between groups (Figure S3F). 14765381, 2022, 11, Downloaded from https://bpspubs.onlinelibrary.wiley.com/doi/10.1111/bph.15776 by Readcube (Labtiva Inc.), Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License BALLESTEROS-MARTÍNEZ ET AL. 2741 F I G U R E 2 mPGES-1 deficiency prevents adiposity, fibrosis, and inflammation in abdominal adipose tissue. (a) Adipocyte size in haematoxylin–eosin-stained sections of abdominal WAT from mPGES-1+/+ and mPGES-1/ mice fed with normal diet (control) and high fat diet (HFD). Representative images are shown. (b) Quantification and representative images of collagen volume fraction (CVF) of Sirius red stained- WAT. mRNA levels of Lep (c), Adipoq (d), Pparg (e), Tnfa (f), Ccl2 (g), and Adgre1 (h) in adipose tissue from each experimental group. (i) Quantification and representative images of crown-like structures in sections of abdominal WAT. Arrows show crown-like structures. In all bar graphs, each dot represents one mouse. Data shown are individual values with means ± SEM. *P < 0.05, significantly different from control of each genotype. +P < 0.05, significantly different from mPGES-1+/+ under the same diet; one-way ANOVA or non-parametric Kruskal–Wallis test. The number of animals in each group is shown below each bar 14765381, 2022, 11, Downloaded from https://bpspubs.onlinelibrary.wiley.com/doi/10.1111/bph.15776 by Readcube (Labtiva Inc.), Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 2742 BALLESTEROS-MARTÍNEZ ET AL. F I G U R E 3 mPGES-1 deficiency protects against endothelial dysfunction induced by high fat diet (HFD) in aortic and mesenteric resistance arteries. Concentration-response curves to acetylcholine (ACh) and diethylamine NONOate (DEA-NO) of aorta (a–d) and mesenteric resistance arteries (MRA) (e-h) from mPGES-1+/+ and mPGES-1/ mice fed with normal diet (control) and HFD. Data represent mean ± SEM. *P < 0.05. significantly different from control; two- way ANOVA. The number of animals in each group is shown in parentheses 3.6 | mPGES-1 deficiency partially protects against mPGES-1+/+ but not in knockout mice (Figure 6a,b). We found no sig- HFD-induced cardiac hypertrophy and fibrosis nificant differences in the expression of hypertrophy markers Acta2, Nppa, or α-SMA (Figure 6c–e). Heart structure or extracellular matrix proteins were unaltered in male No changes in gene expression of the profibrotic markers Tgfb, mice from both genotypes fed on normal diet (Figures 6 and 7). HFD ColIa2, or Col3a1 were observed between groups (Figure 7a–c). induced cardiac hypertrophy and cardiomyocyte hypertrophy in However, HFD increased TGF-β protein and collagen deposition at 14765381, 2022, 11, Downloaded from https://bpspubs.onlinelibrary.wiley.com/doi/10.1111/bph.15776 by Readcube (Labtiva Inc.), Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License BALLESTEROS-MARTÍNEZ ET AL. 2743 F I G U R E 4 mPGES-1 deficiency protects against vascular and perivascular adipose tissue (PVAT) inflammation, induced by a high fat diet (HFD). mRNA levels of Ccl2, Il1b, Tnfa, and Isg15 (a), and the NADPH oxidase subunits Nox1, Cybb (Nox2), and Nox4 (b) in aorta homogenates from mPGES-1+/+ and mPGES-1/ mice fed with normal diet (control) and HFD. (c) Malondialdehyde (MDA) levels in aorta homogenates of each experimental group. mRNA levels of Adgre1 (d), Ccl2 (e), Tnfa (f), Cd3 (g), and Adipoq (h), in PVAT homogenates from mPGES-1+/+ and mPGES-1/ mice fed with normal diet (control) and HFD. In all bar graphs, each dot represents one mouse. Data shown are individual values with means ± SEM. *P < 0.05, significantly different from control of each genotype. +P < 0.05, significantly different from mPGES-1+/+ under the same diet; one-way ANOVA or non-parametric Kruskal–Wallis test. The number of animals in each group is shown below each bar 14765381, 2022, 11, Downloaded from https://bpspubs.onlinelibrary.wiley.com/doi/10.1111/bph.15776 by Readcube (Labtiva Inc.), Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 2744 BALLESTEROS-MARTÍNEZ ET AL. F I G U R E 5 mPGES-1 deficiency partly protects against structural alterations in aorta and mesenteric resistance arteries, induced by a high fat diet (HFD). (a) Wall cross sectional area (CSA) and representative images of haematoxylin and eosin-stained aortas from mPGES-1+/+ and mPGES-1/ mice fed with normal diet (control) and HFD. Structural (b,c) and mechanical (d–g) parameters in mesenteric resistance arteries from each experimental group. In bar graphs, each dot represents one mouse. Data shown are means ± SEM, with individual values in (a, g inset). * P < 0.05, significantly different from control of each genotype. +P < 0.05, significantly different from mPGES-1+/+ under the same diet; one-way (a, g inset) or two-way ANOVA or non-parametric Kruskal–Wallis test (b–f). The number of animals in each group is shown in parentheses or below each bar 14765381, 2022, 11, Downloaded from https://bpspubs.onlinelibrary.wiley.com/doi/10.1111/bph.15776 by Readcube (Labtiva Inc.), Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License BALLESTEROS-MARTÍNEZ ET AL. 2745 F I G U R E 6 mPGES-1 deficiency protects against cardiac hypertrophy, induced by a high fat diet (HFD). (a) Left ventricle (LV) weight normalized to tibia length (TL) and (b) mean cardiomyocyte size and representative images of WGA-stained cardiomyocytes from mPGES-1+/+ and mPGES-1/ mice fed with normal (control) and HFD. (c) Acta-2 and (d) Nppa mRNA expression in cardiac lysates. (e) Protein expression of α-SMA in cardiac homogenates from each experimental group. Representative blots are also shown. In all bar graphs, data shown are individual values with means ± SEM. *P < 0.05, significantly different from control; one-way ANOVA or non-parametric Kruskal–Wallis test. The number of animals in each group is shown below each bar the heart in wild-type mice but not in mPGES-1 deficient mice 3.7 | mPGES-1 deficiency also protects against (Figure 7d,e). Moreover, there was a positive correlation between some of the metabolic and cardiovascular alterations cardiac fibrosis and hypertrophy (Figure 7f). Also, ColIa2 but not associated with obesity in female mice Col3a1 positively correlated with Tgfb expression (Figure 7g,h). We then looked at cardiac oxidative stress and inflammatory markers No differences in body weight gain or glucose levels were observed in and found no significant differences between groups neither in female mPGES-1+/+ and mPGES-1/ fed on normal diet (Figure S5). gene expression nor in MDA levels (Figure S4). However, similarly to male mice, HFD-induced weight gain was 14765381, 2022, 11, Downloaded from https://bpspubs.onlinelibrary.wiley.com/doi/10.1111/bph.15776 by Readcube (Labtiva Inc.), Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 2746 BALLESTEROS-MARTÍNEZ ET AL. F I G U R E 7 mPGES-1 deficiency partly protects against cardiac fibrosis induced by a high fat diet (HFD). mRNA levels of Tgfb (a), ColIa2 (b), and Col3a1 (c) in hearts from mPGES-1+/+ and mPGES-1/ mice fed with normal (control) and HFD. (d) Protein expression of TGF-β in cardiac homogenates from each experimental group. Representative blots are also shown. (e) Representative images and quantification of collagen volume fraction (CVF) of Sirius red stained hearts sections from all groups. (f) Correlation by Spearman t-test of the percentage of cardiac hypertrophy, measured as the ratio of left ventricular (LV) weight to tibia length (TL), with fibrosis, measured as CVF; correlations of the expression of ColIa2 (g) and Col3a1 (h) with expression of Tgfb. The data shown in (f) and (g) were significantly (P < 0.05) correlated. Data shown in bar graphs are individual values with means ± SEM. *P < 0.05, significantly different from control; one-way ANOVA or non-parametric Kruskal– Wallis test. The number of animals in each group is shown below each bar 14765381, 2022, 11, Downloaded from https://bpspubs.onlinelibrary.wiley.com/doi/10.1111/bph.15776 by Readcube (Labtiva Inc.), Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License BALLESTEROS-MARTÍNEZ ET AL. 2747 significantly higher in female mPGES-1+/+ mice than in mPGES-1/ by COX inhibitors while producing strong anti-inflammatory effects animals (Figure S5A) although this difference seems to be less marked (Bergqvist et al., 2020). Thus, there is an urgent need to understand than in males (Figure 1a). Perirenal fat weight was significantly the role of mPGES-1 in the cardiovascular system in the context of increased by HFD only in wild type mice (Table S4). Moreover, WAT highly prevalent clinical conditions such as obesity. Our study demon- weight and adiposity index were smaller in HFD-fed mPGES-1/ strates that mPGES-1 deletion protects from the metabolic and car- mice, compared with mPGES-1+/+ mice (Table S4). diovascular alterations associated with obesity by mechanisms likely Basal glucose levels and GTT were comparably increased by HFD to be related to a decrease in inflammatory mediators. in both genotypes (Figure S5B–D). No differences in insulin sensitivity We observed that mPGES-1/ mice developed less gain weight were observed between genotypes (Figure S5E,F). and had smaller adipocytes in response to HFD than wild type ani- Regarding cardiovascular alterations, we found no differences mals, which agrees with a recent study performed in mPGES-1 knock- between groups in SBP (Table S5). In aorta from females, HFD did not out mice on DBA genetic background (Pierre et al., 2018). Reasons for modify endothelium-dependent relaxation to acetylcholine or insulin this adipose tissue phenotype in HFD-mPGES-1/ mice are likely to in any of the genotypes (Figures S6A,B and S7A,B), indicating some include an increase in energy expenditure without changes in brow- vascular protection by gender. However, HFD induced endothelial ning process (Pierre et al., 2018). In addition, PGE2 seems to have +/+ dysfunction in mesenteric resistance arteries from mPGES-1 but anti-lipolytic functions whereas PGI2 stimulates lipolysis (Chatzipanteli not in mPGES-1/ mice (Figure S6E,F). No differences in vasodilator et al., 1992), and mPGES-1 deficient mice have increased urinary, responses to DEA-NO were observed between the experimental macrophage, adipose tissue and vascular PGI2 levels in different set- groups neither in aorta (Figure S6C,D) nor in mesenteric arteries tings including obesity (Avendaño et al., 2018; Cheng et al., 2006; (Figure S6G,H). HFD-induced heart hypertrophy was similar in Sasaki et al., 2021; Wang et al., 2006, 2008). Of note, the smaller mPGES-1+/+ and mPGES-1/ mice (Figure S7C). HFD-induced weight gain observed in mPGES-1/ mice was associ- ated with a better metabolic profile, as shown by the improved glu- cose tolerance and insulin sensitivity, and lipid levels. Accordingly, 3.8 | mPGES-1 expression in human abdominal fat inhibition of the COX-2 enzyme in humans (González-Ortiz correlates with vascular remodelling, vascular et al., 2005) and in animal models (Hsieh et al., 2010), in the context stiffness, and SPB of obesity, was associated with a better metabolic status. However, Pierre et al. (2018) found similar basal levels of glucose in mPGES-1+/ + Given the beneficial effects of mPGES-1 deletion in the animal model, and mPGES-1/, although a complete characterization was not we evaluated a possible relationship between expression of mRNA for performed. Mechanisms responsible for insulin resistance in obesity mPGES-1 in abdominal adipose tissue and vascular damage in include adipose tissue augmented inflammatory status characterized humans. PTGES expression did not correlate with BMI (Figure 8a) but by macrophage infiltration and enhanced TNF-α production, and dys- it positively correlated with SBP (Figure 8b), IMT (Figure 8c), and vas- regulation of adiponectin and leptin levels (Petersen & Shulman, 2018; cular stiffness measured as PWV (Figure 8d). Stern et al., 2016). Indeed, insulin resistance in obese individuals is associated with adipose tissue dysfunction, inflammation, and an imbalance of circulating adipokines independently of total body fat 3.9 | mPGES-1 expression in PVAT of aortic mass (Stern et al., 2016). Additionally, a rigid extracellular matrix may abdominal aneurysm patients correlates with exert shear stress on expanding adipocytes, leading to adipocyte cell inflammation death and consequent inflammation (Datta et al., 2018). Accordingly, we observed that HFD mPGES-1/ mice showed increased levels of Because mPGES-1 deletion reduced obesity-induced inflammatory Adipoq and decreased Lep levels and collagen deposition and showed cells infiltration in PVAT (Figure 4), we used human aortic PVAT to protection against inflammatory features that might have contributed analyse a potential correlation between PTGES expression and that of to a better metabolic profile. immune cell markers in this specific adipose tissue depot. As shown in In obesity, adipose tissue dysfunction leads to an inflammatory Figure 8e, expression of PTGES in PVAT did not correlate with body state that could trigger cardiovascular alterations (Fuster et al., 2016). mass index. However, PTGES expression positively correlated with the Thus, insulin resistance, PVAT inflammation and NADPH oxidase macrophage marker CD68 (Figure 8f), the T lymphocytes markers overexpression with the subsequent reduction in NO availability asso- (CD3G, CD4, and CD8A) (Figure 8g–i) and with LEP (Figure 8j). ciated with obesity, trigger endothelial dysfunction (Engin, 2017; Kobayasi et al., 2010). Production of prostanoids including TXA2, 6-keto-PGF1α and PGE2 was increased in aorta from a rat model of 4 | DISCUSSION obesity, induced by monosodium glutamate administration (Majewski et al., 2018). We did not observe changes in cardiac, vascular, or PVAT Selective mPGES-1 inhibitors that specifically constrain PGE2 synthe- mPGES-1 mRNA expression in response to obesity and PGE2 levels sis without affecting synthesis of the protective PGI2 are currently were not modified either by diet or genotype in aorta or heart, under development to circumvent the cardiovascular hazards induced suggesting that most of the detected PGE2 derives from the 14765381, 2022, 11, Downloaded from https://bpspubs.onlinelibrary.wiley.com/doi/10.1111/bph.15776 by Readcube (Labtiva Inc.), Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 2748 BALLESTEROS-MARTÍNEZ ET AL. F I G U R E 8 mPGES-1 expression in human abdominal adipose tissue or perivascular adipose tissue (PVAT) correlates with parameters of vascular damage and inflammation. Correlation between mRNA levels of PTGES in human abdominal adipose tissue and body mass index (BMI) (a), systolic blood pressure (SBP) (b), intima-media thickness (IMT) (c), and pulse wave velocity (PWV) (d). Correlation between mRNA levels of PTGES in PVAT of patients with aortic aneurysms and body mass index (BMI) (e), CD68 (f), CD3 (g), CD4 (h), CD8 (i), and LEP (j). Each dot represents one patient. Correlations were calculated with Spearman t-test and were significant (P < 0.05) for all data sets shown except for those with BMI, shown in (a) and ( e). The number of patients is shown in each graph 14765381, 2022, 11, Downloaded from https://bpspubs.onlinelibrary.wiley.com/doi/10.1111/bph.15776 by Readcube (Labtiva Inc.), Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License BALLESTEROS-MARTÍNEZ ET AL. 2749 constitutive PGE2 synthases. However, we demonstrate for the first mPGES-1/ mice had less infiltration of macrophages that would time, that mPGES-1 has a key role in the vascular alterations observed eventually result in lower production of inflammatory cytokines. in obesity. Thus, the HFD-induced endothelial dysfunction of conduc- Importantly, this might have clinical implications as we observed posi- tance and resistance arteries and the vascular insulin resistance tive correlations between mPGES-1 expression and macrophages and observed in wild-type mice, was completely prevented by mPGES-1 T cells infiltration in aortic PVAT from patients with AAA. Increased knockdown. Moreover, mPGES-1 deletion partly prevented vascular synthesis of the vasculoprotective PGI2 in mPGES-1 deficient mice remodelling and mechanical alterations induced by HFD, although the might also contribute to the beneficial effects of mPGES-1 deletion in specific proteins involved need to be further studied. The role of obesity, as shown previously in hypertension and other vascular mPGES-1 in vascular damage has already been explored in the con- pathologies (Avendaño et al., 2018; Wang et al. 2006 and 2008), and text of other pathologies, with divergent results. Specifically, in this warrants further investigation. In this sense, in healthy humans, mPGES-1 deficient mice, we recently observed vascular protection the pharmacological inhibitor of mPGES-1, LY3023703 increased the against hypertension-induced endothelial dysfunction, hyper- levels of urinary PGI2 compared with placebo (Jin et al., 2016) and the contractility and structural alterations (Avendaño et al., 2018), and mPGES-1 inhibitor C3 compound increased PGI2 release in internal other authors found less aneurysm formation (Wang et al., 2008), ath- mammary artery and saphenous vein, associated with decreased vas- erosclerosis development (Wang et al., 2006) and neointima thicken- cular tone (Ozen et al., 2017). Finally, a contribution of the better met- ing in response to injury (Wang et al., 2011). However, other studies abolic profile and greater weight loss observed in mPGES-1 knockout have suggested a protective role of mPGES-1 against myocardial mice cannot be fully excluded. ischaemia reperfusion injury by vascular mechanisms (Zhu We found that cardiac hypertrophy positively correlated with the et al., 2019). More importantly, we found positive correlations percentage of fibrosis, which is known to impair cardiac function between adipose tissue mPGES-1 and clinical parameters of vascular (Martínez-Martínez et al., 2015). We also provided novel evidence damage, that is, IMT and PWV in the absence of correlation with BMI, demonstrating that mPGES-1 deletion prevents obesity-induced car- suggesting that the relationship between adipose tissue mPGES-1 and diac and cardiomyocytes hypertrophy and fibrosis and expression of vascular damage in humans is not merely a consequence of altered some profibrotic markers suggesting some degree of cardiac protec- body weight. Indeed, HFD tended to decrease the expression of Ptges tion. A limitation of our study is that we have not performed a full mRNA in mice WAT and Sasaki et al. (2021) found no changes in characterization of the cardiac damage at the functional and molecular PGE2 levels in WAT between wild-type and mPGES-1/ mice fed on level. Indeed, to our surprise, we did not detect significant changes in a HFD. Our data extend previous observations showing a positive cor- inflammatory mediators in our cardiac samples. The effects of relation between peripheral blood mononuclear cells (PBMCs), mPGES-1 deletion in cardiac damage have provided controversial mPGES-1 and BMI in humans (Avendaño et al., 2018). Furthermore, results depending on the experimental model. In mice, the lack of we reported that mPGES-1 expressed in PBMCs was strongly associ- mPGES-1 prevents the acute post-myocardial infarction death (Wu ated with blood pressure and vascular remodelling (Avendaño et al., 2009) or fibrosis induced by agonists of β- adrenoceptors (Ji et al., 2018). In agreement, mice lacking mPGES-1 conditionally in et al., 2020). However, mPGES-1 deletion impairs left ventricular con- myeloid cells, when crossed into hyperlipidaemic LDLR-deficient ani- tractile function after acute myocardial infarction (Degousee mals, showed reduced atherogenesis (Chen et al., 2014). Altogether, et al., 2008) or after chronic angiotensin II infusion (Harding these results suggest a role for mPGES-1, likely from a range of differ- et al., 2011) and impairs cardiac microvascular perfusion in a myocar- ent cell types including vascular cells, immune cells, and adipocytes, in dial ischaemia reperfusion model (Zhu et al., 2019). Although further the vascular damage associated with obesity. Moreover, although our dedicated studies are needed, the different contribution of preclinical model of obesity did not show increased blood pressure, we haemodynamic, metabolic, or inflammatory mechanisms in models of observed a positive correlation of adipose tissue mPGES-1 and SBP in hypertension, myocardial infarction, and obesity could underlie these humans, a clinical condition frequently observed in human obesity. discrepancies. Underlying mechanisms responsible for the beneficial effects of Another important result is that some of the beneficial effects mPGES-1 deletion in obesity-induced vascular damage are likely to induced by mPGES-1 deletion in males were also reproduced in include both vascular mechanisms and improved crosstalk with sur- females. There is a historical paucity of studies performed in females rounding PVAT. In this sense, it is accepted that PVAT might have a probably due to the metabolic and cardiovascular protection elicited role in vascular damage in obesity and other pathologies such as AAA by oestrogens on insulin sensitivity, glucose tolerance, body weight because of the infiltration of inflammatory macrophages and T cells gain, and adiposity in models of HFD-induced obesity (Handgraaf / (Fernández-Alfonso et al., 2017). In mPGES-1 mice, HFD did not et al., 2013; Riant et al., 2009). In fact, we found that, given the same increase aortic gene expression of inflammatory cytokines involved in HFD, female mice gained similar weight to males but showed no endothelial dysfunction or vascular remodelling including Ccl2, Il1b, endothelial dysfunction or vascular insulin resistance of the aorta, Tnfa, or Isg15, and NADPH oxidase isoforms (Nox1/2/4/), as opposed although in resistance arteries, endothelial function was similarly to wild type mice. In agreement, mPGES-1 deletion reduced vascular impaired. As in males, the absence of mPGES-1 in females partly oxidative stress in both hypertensive and aneurysmal models protected against HFD-induced body weight gain, adiposity, and (Avendaño et al., 2018; Wang et al., 2008). Moreover, PVAT from improved resistance arteries endothelial dysfunction induced by HFD. 14765381, 2022, 11, Downloaded from https://bpspubs.onlinelibrary.wiley.com/doi/10.1111/bph.15776 by Readcube (Labtiva Inc.), Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 2750 BALLESTEROS-MARTÍNEZ ET AL. However, a reduced weight gain seems to be more pronounced in analysis. The data supporting the findings of this study are available male mPGES-1/ than in female knockout mice and mPGES-1 dele- from the corresponding authors upon reasonable request. Some data tion did not improve the insulin sensitivity, glucose tolerance or gross may not be made available because of privacy or ethical restrictions. cardiac hypertrophy induced by HFD. In conclusion, we demonstrate that in male mice, mPGES-1 dele- OR CID tion confers protection against body weight gain, adipose tissue Raquel Rodrigues-Díez https://orcid.org/0000-0002-6348-1505 enlargement and the glycaemic and lipid alterations induced by HFD. Ana M. Briones https://orcid.org/0000-0001-8218-5579 Moreover, we demonstrate for the first time that the absence of mPGES-1 protects against adipose tissue and cardiac hypertrophy and RE FE RE NCE S fibrosis, and endothelial dysfunction and vascular remodelling induced Alexander, S. P., Christopoulos, A., Davenport, A. P., Kelly, E., Mathie, A., by obesity. Underlying mechanisms are likely to include decreased Peters, J. A., Veale, E. L., Armstrong, J. F., Faccenda, E., inflammation in adipose tissue depots and the vasculature. Moreover, Harding, S. D., Pawson, A. J., Southan, C., Davies, J. A., Abbracchio, M. P., Alexander, W., Al-hosaini, K., Bäck, M., some of the beneficial effects of mPGES-1 deletion were also Barnes, N. M., Bathgate, R., … Ye, R. D. (2021). The Concise Guide reproduced in female mice. In human tissue samples, mPGES-1 to PHARMACOLOGY 2021/22: G protein-coupled receptors. British expression in abdominal or perivascular adipose tissue correlated with Journal of Pharmacology, 178(S1), S27–S156. https://doi.org/10. parameters of vascular remodelling, stiffness, and inflammation. Thus, 1111/bph.15538 Alexander, S. P., Cidlowski, J. A., Kelly, E., Mathie, A., Peters, J. A., selective inhibitors of mPGES-1 might be a useful therapeutic tool Veale, E. L., Armstrong, J. F., Faccenda, E., Harding, S. D., against the metabolic and cardiovascular alterations induced by Pawson, A. J., Southan, C., Davies, J. A., Coons, L., Fuller, P. J., obesity. Korach, K. S., & Young, M. J. (2021). The Concise Guide to PHARMACOLOGY 2021/22: Nuclear hormone receptors. British Journal of Pharmacology, 178(S1), S246–S263. https://doi.org/10. ACKNOWLEDGEMEN TS 1111/bph.15540 n This work was supported by the Ministerio de Ciencia e Innovacio Alexander, S. P., Fabbro, D., Kelly, E., Mathie, A., Peters, J. A., Veale, E. L., and Fondo Europeo de Desarrollo Regional (FEDER)/FSE Armstrong, J. F., Faccenda, E., Harding, S. D., Pawson, A. J., (SAF2016-80305P), Instituto de Salud Carlos III (ISCIII; FIS Southan, C., Davies, J. A., Boison, D., Burns, K. E., Dessauer, C., Gertsch, J., Helsby, N. A., Izzo, A. A., Koesling, D., … Wong, S. S. PI18/0257); Comunidad de Madrid (CM) (B2017/BMD-3676 (2021). The Concise Guide to PHARMACOLOGY 2021/22: Enzymes. AORTASANA) FEDER-a way to build Europe. MGA was supported by British Journal of Pharmacology, 178(S1), S313–S411. https://doi.org/ a FPI-UAM fellowship and RRD by a Juan de la Cierva contract (IJCI- 10.1111/bph.15542 2017-31399). The authors thank Victor Gutierrez his help with some Alexander, S. P., Kelly, E., Mathie, A., Peters, J. A., Veale, E. L., experiments. Armstrong, J. F., Faccenda, E., Harding, S. D., Pawson, A. J., Southan, C., Buneman, O. P., Cidlowski, J. A., Christopoulos, A., Davenport, A. P., Fabbro, D., Spedding, M., Striessnig, J., Davies, J. A., AUTHOR CONTRIBUTIONS Ahlers-Dannen, K. E., … Zolghadri, Y. (2021). The Concise Guide CB-M and RR-D wrote the manuscript and researched data. LMB, to PHARMACOLOGY 2021/22: Other Protein Targets. British Journal RM-C, EM-M, MG-A, and CR researched data. AM-B reviewed/edited of Pharmacology, 178(S1), S1–S26. https://doi.org/10.1111/bph. 15537 and wrote the manuscript. VC, MS, and JM-G contributed to discus- Avendaño, M. S., García-Redondo, A. B., Zalba, G., González-Amor, M., sion and reviewed/edited manuscript. All authors approved the Aguado, A., Martínez-Revelles, S., Beltrán, L. M., Camacho, M., manuscript. Cachofeiro, V., Alonso, M. J., Salaices, M., & Briones, A. M. (2018). mPGES-1 (microsomal prostaglandin E synthase-1) mediates vascular dysfunction in hypertension through oxidative stress. Hypertension CONF LICT OF IN TE RE ST (Dallas, Tex: 1979), 72(2), 492–502. https://doi.org/10.1161/ The authors have declared that no conflict of interest exists. HYPERTENSIONAHA.118.10833 Bergqvist, F., Morgenstern, R., & Jakobsson, P. J. (2020). A review on DECLARATION OF TRAN SPARENCY AND S CIENTI FIC mPGES-1 inhibitors: From preclinical studies to clinical applications. Prostaglandins & Other Lipid Mediators, 147, 106383. https://doi.org/ RI GOUR 10.1016/j.prostaglandins.2019.106383 This Declaration acknowledges that this paper adheres to the princi- Briones, A. M., Aras-Lo  pez, R., Alonso, M. J., & Salaices, M. (2014). Small ples for transparent reporting and scientific rigour of preclinical artery remodeling in obesity and insulin resistance. Current Vascular research as stated in the BJP guidelines for Design and Analysis, Pharmacology, 12(3), 427–437. https://doi.org/10.2174/ 1570161112666140423221319 Immunoblotting and Immunochemistry, and Animal Experimentation, Buettner, R., Schölmerich, J., & Bollheimer, L. C. (2007). High-fat diets: and as recommended by funding agencies, publishers and other orga- Modeling the metabolic disorders of human obesity in rodents. Obesity nizations engaged with supporting research. (Silver Spring, Md.), 15(4), 798–808. https://doi.org/10.1038/oby. 2007.608 DATA AVAI LAB ILITY S TATEMENT Chan, P. C., Hsiao, F. C., Chang, H. M., Wabitsch, M., & Hsieh, P. S. (2016). Importance of adipocyte cyclooxygenase-2 and prostaglandin Constanza Ballesteros-Martínez, Raquel Rodrigues-Diez, and Ana E2-prostaglandin E receptor 3 signaling in the development of M. Briones had full access to all the data in the study and take respon- obesity-induced adipose tissue inflammation and insulin resistance. sibility for the integrity of the data and the accuracy of the data FASEB Journal: Official Publication of the Federation of American 14765381, 2022, 11, Downloaded from https://bpspubs.onlinelibrary.wiley.com/doi/10.1111/bph.15776 by Readcube (Labtiva Inc.), Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License BALLESTEROS-MARTÍNEZ ET AL. 2751 Societies for Experimental Biology, 30(6), 2282–2297. https://doi.org/ García-Alonso, V., Titos, E., Alcaraz-Quiles, J., Rius, B., Lopategi, A., Lo pez- 10.1096/fj.201500127 Vicario, C., Jakobsson, P.-J., Delgado, S., Lozano, J., & Clària, J. (2016). Chatzipanteli, K., Rudolph, S., & Axelrod, L. (1992). Coordinate control of Prostaglandin E2 exerts multiple regulatory actions on human obese lipolysis by prostaglandin E2 and prostacyclin in rat adipose tissue. adipose tissue remodeling, inflammation, adaptive thermogenesis and Diabetes, 41(8), 927–935. https://doi.org/10.2337/diab.41.8.927 lipolysis. PLoS ONE, 11(4), e0153751. https://doi.org/10.1371/journal. Chen, L., Yang, G., Monslow, J., Todd, L., Cormode, D. P., Tang, J., pone.0153751 Grant, G. R., DeLong, J. H., Tang, S. Y., Lawson, J. A., Pure, E., & Ghoshal, S., Trivedi, D. B., Graf, G. A., & Loftin, C. D. (2011). Fitzgerald, G. A. (2014). Myeloid cell microsomal prostaglandin E Cyclooxygenase-2 deficiency attenuates adipose tissue differentiation synthase-1 fosters atherogenesis in mice. Proceedings of the National and inflammation in mice. The Journal of Biological Chemistry, 286(1), Academy of Sciences of the United States of America, 111(18), 6828– 889–898. https://doi.org/10.1074/jbc.M110.139139 6833. https://doi.org/10.1073/pnas.1401797111 González-Ortiz, M., Pascoe-González, S., Kam-Ramos, A. M., & Cheng, Y., Wang, M., Yu, Y., Lawson, J., Funk, C. D., & Fitzgerald, G. A. Hernández-Salazar, E. (2005). Effect of celecoxib, a cyclooxygenase- (2006). Cyclooxygenases, microsomal prostaglandin E synthase-1, and 2-specific inhibitor, on insulin sensitivity, C-reactive protein, homocys- cardiovascular function. The Journal of Clinical Investigation, 116(5), teine, and metabolic profile in overweight or obese subjects. Metabolic 1391–1399. https://doi.org/10.1172/JCI27540 Syndrome and Related Disorders, 3(2), 95–101. https://doi.org/10. Curtis, M. J., Alexander, S., Cirino, G., Docherty, J. R., George, C. H., 1089/met.2005.3.95 Giembycz, M. A., Hoyer, D., Insel, P. A., Izzo, A. A., Ji, Y., Handgraaf, S., Riant, E., Fabre, A., Waget, A., Burcelin, R., Lière, P., MacEwan, D. J., Sobey, C. G., Stanford, S. C., Teixeira, M. M., Krust, A., Chambon, P., Arnal, J.-F., & Gourdy, P. (2013). Wonnacott, S., & Ahluwalia, A. (2018). Experimental design and Prevention of obesity and insulin resistance by estrogens requires analysis and their reporting II: Updated and simplified guidance for ERα activation function-2 (ERαAF-2), whereas ERαAF-1 is dispens- authors and peer reviewers. British Journal of Pharmacology, 175(7), able. Diabetes, 62(12), 4098–4108. https://doi.org/10.2337/db13- 987–993. 0282 Dan, H. J., Wang, Y., Zeng, M. X., Luan, Y. Y., & Hu, B. (2011). Evaluation Harding, P., Yang, X. P., He, Q., & Lapointe, M. C. (2011). Lack of micro- of intima-media thickness and vascular elasticity of the common somal prostaglandin E synthase-1 reduces cardiac function following carotid artery in patients with isolated systolic hypertension using angiotensin II infusion. American Journal of Physiology. Heart and Circu- ultrasound radiofrequency-data technology. Clinical Physiology and latory Physiology, 300(3), H1053–H1061. https://doi.org/10.1152/ Functional Imaging, 31(4), 315–319. ajpheart.00772.2010 Datta, R., Podolsky, M. J., & Atabai, K. (2018). Fat fibrosis: Friend or foe? Hétu, P. O., & Riendeau, D. (2007). Down-regulation of microsomal prosta- The Journal of Clinical Investigation Insight, 3(19), e122289. https://doi. glandin E2 synthase-1 in adipose tissue by high-fat feeding. Obesity org/10.1172/jci.insight.122289 (Silver Spring, Md.), 15(1), 60–68. https://doi.org/10.1038/oby. Degousee, N., Fazel, S., Angoulvant, D., Stefanski, E., Pawelzik, S. C., 2007.514 Korotkova, M., Arab, S., Liu, P., Lindsay, T. F., Zhuo, S., Butany, J., Hsieh, P. S., Lu, K. C., Chiang, C. F., & Chen, C. H. (2010). Suppressive Li, R.-K., Audoly, L., Schmidt, R., Angioni, C., Geisslinger, G., effect of COX2 inhibitor on the progression of adipose inflammation Jakobsson, P.-J., & Rubin, B. B. (2008). Microsomal prostaglandin E2 in high-fat-induced obese rats. European Journal of Clinical Investiga- synthase-1 deletion leads to adverse left ventricular remodeling after tion, 40(2), 164–171. https://doi.org/10.1111/j.1365-2362.2009. myocardial infarction. Circulation, 117(13), 1701–1710. https://doi. 02239.x org/10.1161/CIRCULATIONAHA.107.749739 Jaworski, K., Ahmadian, M., Duncan, R. E., Sarkadi-Nagy, E., Varady, K. A., Engin, A. (2017). Endothelial dysfunction in obesity. Advances in Experi- Hellerstein, M. K., Lee, H.-Y., Samuel, V. T., Shulman, G. I., Kim, K.-H., mental Medicine and Biology, 960, 345–379. https://doi.org/10.1007/ de Val, S., Kang, C., & Sul, H. S. (2009). AdPLA ablation increases lipol- 978-3-319-48382-5_15 ysis and prevents obesity induced by high-fat feeding or leptin defi- Even, S. E., Dulak-Lis, M. G., Touyz, R. M., & Cat, A. N. D. (2014). Crosstalk ciency. Nature Medicine, 15(2), 159–168. https://doi.org/10.1038/nm. between adipose tissue and blood vessels in cardiometabolic syn- 1904 drome: Implication of steroid hormone receptors (MR/GR). Hormone Ji, S., Guo, R., Wang, J., Qian, L., Liu, M., Xu, H., Zhang, J., Guan, Y., Molecular Biology and Clinical Investigation, 19(2), 89–101. https://doi. Yang, G., & Chen, L. (2020). Microsomal prostaglandin E2 synthase-1 org/10.1515/hmbci-2014-0013 deletion attenuates isoproterenol-induced myocardial fibrosis in mice. Fain, J. N., Ballou, L. R., & Bahouth, S. W. (2001). Obesity is induced in The Journal of Pharmacology and Experimental Therapeutics, 375(1), 40– mice heterozygous for cyclooxygenase-2. Prostaglandins & Other Lipid 48. https://doi.org/10.1124/jpet.120.000023 Mediators, 65(4), 199–209. https://doi.org/10.1016/S0090-6980(01) Jin, Y., Smith, C. L., Hu, L., Campanale, K. M., Stoltz, R., Huffman, L. G., 00136-8 McNearney, T. A., Yang, X. Y., Ackermann, B. L., Dean, R., Regev, A., & Fain, J. N., Kanu, A., Bahouth, S. W., Cowan, G. S. Jr., Hiler, M. L., & Landschulz, W. (2016). Pharmacodynamic comparison of LY3023703, Leffler, C. W. (2002). Comparison of PGE2, prostacyclin and leptin a novel microsomal prostaglandin e synthase 1 inhibitor, with release by human adipocytes versus explants of adipose tissue in pri- celecoxib. Clinical Pharmacology and Therapeutics, 99(3), 274–284. mary culture. Prostaglandins, Leukotrienes, and Essential Fatty Acids, https://doi.org/10.1002/cpt.260 67(6), 467–473. https://doi.org/10.1054/plef.2002.0430 Kobayasi, R., Akamine, E. H., Davel, A. P., Rodrigues, M. A., Carvalho, C. R., Fernández-Alfonso, M. S., Somoza, B., Tsvetkov, D., Kuczmanski, A., & Rossoni, L. V. (2010). Oxidative stress and inflammatory mediators Dashwood, M., & Gil-Ortega, M. (2017). Role of perivascular adipose contribute to endothelial dysfunction in high-fat diet-induced obesity tissue in health and disease. Comprehensive Physiology, 8(1), 23–59. in mice. Journal of Hypertension, 28(10), 2111–2119. https://doi.org/ https://doi.org/10.1002/cphy.c170004 10.1097/HJH.0b013e32833ca68c Fuster, J. J., Ouchi, N., Gokce, N., & Walsh, K. (2016). Obesity-induced Koeberle, A., & Werz, O. (2015). Perspective of microsomal prostaglandin changes in adipose tissue microenvironment and their impact on car- E2 synthase-1 as drug target in inflammation-related disorders. Bio- diovascular disease. Circulation Research, 118(11), 1786–1807. chemical Pharmacology, 98(1), 1–15. https://doi.org/10.1016/j.bcp. https://doi.org/10.1161/CIRCRESAHA.115.306885 2015.06.022 García-Alonso, V., & Clària, J. (2014). Prostaglandin E2 signals white-to- Leiguez, E., Motta, P., Maia Marques, R., Lomonte, B., Sampaio, S. V., & brown adipogenic differentiation. Adipocytes, 3(4), 290–296. https:// Teixeira, C. (2020). A representative GIIA phospholipase A2 activates doi.org/10.4161/adip.29993 preadipocytes to produce inflammatory mediators implicated in 14765381, 2022, 11, Downloaded from https://bpspubs.onlinelibrary.wiley.com/doi/10.1111/bph.15776 by Readcube (Labtiva Inc.), Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 2752 BALLESTEROS-MARTÍNEZ ET AL. obesity development. Biomolecules, 10(12), 1593. https://doi.org/10. Riant, E., Waget, A., Cogo, H., Arnal, J. F., Burcelin, R., & Gourdy, P. (2009). 3390/biom10121593 Estrogens protect against high-fat diet-induced insulin resistance and Lilley, E., Stanford, S. C., Kendall, D. E., Alexander, S., Cirino, G., glucose intolerance in mice. Endocrinology, 150(5), 2109–2117. Docherty, J. R., George, C. H., Insel, P. A., Izzo, A. A., Ji, Y., https://doi.org/10.1210/en.2008-0971 Panettieri, R. A., Sobey, C. G., Stefanska, B., Stephens, G., Samuelsson, B., Morgenstern, R., & Jakobsson, P. J. (2007). Teixeira, M., & Ahluwalia, A. (2020). ARRIVE 2.0 and the British Membrane prostaglandin E synthase-1: A novel therapeutic target. Journal of Pharmacology: Updated guidance for 2020. British Journal of Pharmacological Reviews, 59(3), 207–224. https://doi.org/10.1124/pr. Pharmacology, 177(16), 3611–3616. https://doi.org/10.1111/bph. 59.3.1 15178 Sasaki, Y., Kuwata, H., Akatsu, M., Yamakawa, Y., Ochiai, T., Yoda, E., Majewski, M., Jurgon ski, A., Fotschki, B., & Juskiewicz, J. (2018). The toxic Nakatani, Y., Yokoyama, C., & Hara, S. (2021). Involvement of prosta- effects of monosodium glutamate (MSG)—The involvement of nitric cyclin synthase in high-fat-diet-induced obesity. Prostaglandins and oxide, prostanoids and potassium channels in the reactivity of thoracic Other Lipid Mediators, 153, 106523. https://doi.org/10.1016/j. arteries in MSG-obese rats. Toxicology and Applied Pharmacology, 359, prostaglandins.2020.106523 62–69. https://doi.org/10.1016/j.taap.2018.09.016 Stern, J. H., Rutkowski, J. M., & Scherer, P. E. (2016). Adiponectin, leptin, Mancia, G., Fagard, R., Narkiewicz, K., Redo  n, J., Zanchetti, A., Böhm, M., and fatty acids in the maintenance of metabolic homeostasis through Christiaens, T., Cifkova, R., De Backer, G., Dominiczak, A., adipose tissue crosstalk. Cell Metabolism, 23(5), 770–784. https://doi. Galderisi, M., Grobbee, D. E., Jaarsma, T., Kirchhof, P., Kjeldsen, S. E., org/10.1016/j.cmet.2016.04.011 Laurent, S., Manolis, A. J., Nilsson, P. M., Ruilope, L. M., … Task Force Touboul, P. J., Hennerici, M. G., Meairs, S., Adams, H., Amarenco, P., Members. (2013). 2013 ESH/ESC guidelines for the management of Bornstein, N., Csiba, L., Desvarieux, M., Ebrahim, S., Hernandez, R. H., arterial hypertension: The task force for the management of arterial Jaff, M., Kownator, S., Naqvi, T., Prati, P., Rundek, T., Sitzer, M., hypertension of the European Society of Hypertension (ESH) and of Schminke, U., Tardif, J.-C., Taylor, A., … Woo, K. S. (2012). the European Society of Cardiology (ESC). Journal of Hypertension, Mannheim carotid intima-media thickness and plaque consensus 31(7), 1281–1357. https://doi.org/10.1097/01.hjh.0000431740. (2004–2006-2011). An update on behalf of the advisory board of the 32696.cc 3rd, 4th and 5th watching the risk symposia, at the 13th, 15th and Martínez-Martínez, E., Lo   pez-Andres, N., Jurado-Lo  pez, R., Rousseau, E., 20th European Stroke Conferences, Mannheim, Germany, 2004, Brus- Bartolomé, M. V., Fernández-Celis, A., Rossignol, P., Islas, F., sels, Belgium, 2006, and Hamburg, Germany, 2011. Cerebrovascular Antequera, A., Prieto, S., Luaces, M., & Cachofeiro, V. (2015). Diseases (Basel, Switzerland), 34(4), 290–296. https://doi.org/10.1159/ Galectin-3 participates in cardiovascular remodeling associated with 000343145 obesity. Hypertension (Dallas, Tex.: 1979), 66(5), 961–969. Vecchié, A., Dallegri, F., Carbone, F., Bonaventura, A., Liberale, L., Muñoz, M., Sánchez, A., Pilar Martínez, M., Benedito, S., Lo  pez- Portincasa, P., Frühbeck, G., & Montecucco, F. (2018). Obesity pheno- Oliva, M. E., García-Sacristán, A., Hernández, M., & Prieto, D. (2015). types and their paradoxical association with cardiovascular diseases. COX-2 is involved in vascular oxidative stress and endothelial dysfunc- European Journal of Internal Medicine, 48, 6–17. https://doi.org/10. tion of renal interlobar arteries from obese Zucker rats. Free Radical 1016/j.ejim.2017.10.020 Biology & Medicine, 84, 77–90. https://doi.org/10.1016/j. Vessières, E., Belin de Chantemèle, E. J., Toutain, B., Guihot, A. L., freeradbiomed.2015.03.024 Jardel, A., Loufrani, L., & Henrion, D. (2010). Cyclooxygenase-2 inhibi- Ozen, G., Gomez, I., Daci, A., Deschildre, C., Boubaya, L., Teskin, O., tion restored endothelium-mediated relaxation in old obese Zucker rat Uydeş-Dog an, B. S., Jakobsson, P.-J., Longrois, D., Topal, G., & mesenteric arteries. Frontiers in Physiology, 1, 145. https://doi.org/10. Norel, X. (2017). Inhibition of microsomal PGE synthase-1 reduces 3389/fphys.2010.00145 human vascular tone by increasing PGI2: A safer alternative to COX-2 Wang, C. Y., & Liao, J. K. (2012). A mouse model of diet-induced obesity inhibition. British Journal of Pharmacology, 174(22), 4087–4098. and insulin resistance. Methods in Molecular Biology (Clifton, N.J.), 821, https://doi.org/10.1111/bph.13939 421–433. https://doi.org/10.1007/978-1-61779-430-8_27 Parlee, S. D., Lentz, S. I., Mori, H., & MacDougald, O. A. (2014). Quantifying Wang, M., Ihida-Stansbury, K., Kothapalli, D., Tamby, M. C., Yu, Z., size and number of adipocytes in adipose tissue. Methods in Enzymol- Chen, L., Grant, G., Cheng, Y., Lawson, J. A., Assoian, R. K., ogy, 537, 93–122. https://doi.org/10.1016/B978-0-12-411619-1. Jones, P. L., & Fitzgerald, G. A. (2011). Microsomal prostaglandin e2 00006-9 synthase-1 modulates the response to vascular injury. Circulation, Percie du Sert, N., Hurst, V., Ahluwalia, A., Alam, S., Avey, M. T., Baker, M., 123(6), 631–639. https://doi.org/10.1161/CIRCULATIONAHA.110. Browne, W. J., Clark, A., Cuthill, I. C., Dirnagl, U., Emerson, M., 973685 Garner, P., Holgate, S. T., Howells, D. W., Karp, N. A., Lazic, S. E., Wang, M., Lee, E., Song, W., Ricciotti, E., Rader, D. J., Lawson, J. A., Lidster, K., MacCallum, C. J., Macleod, M., … Würbel, H. (2020). The Puré, E., & FitzGerald, G. A. (2008). Microsomal prostaglandin E ARRIVE guidelines 2.0: Updated guidelines for reporting animal synthase-1 deletion suppresses oxidative stress and angiotensin II- research. PLoS Biology, 18(7), e3000410. https://doi.org/10.1371/ induced abdominal aortic aneurysm formation. Circulation, 117(10), journal.pbio.3000410 1302–1309. https://doi.org/10.1161/CIRCULATIONAHA.107. Petersen, M. C., & Shulman, G. I. (2018). Mechanisms of insulin action and 731398 insulin resistance. Physiological Reviews, 98(4), 2133–2223. https://doi. Wang, M., Zukas, A. M., Hui, Y., Ricciotti, E., Puré, E., & org/10.1152/physrev.00063.2017 FitzGerald, G. A. (2006). Deletion of microsomal prostaglandin E Pierre, C., Guillebaud, F., Airault, C., Baril, N., Barbouche, R., synthase-1 augments prostacyclin and retards atherogenesis. Save, E., Gaigé, S., Bariohay, B., Dallaporta, M., & Troadec, J. D. Proceedings of the National Academy of Sciences of the United States (2018). Invalidation of microsomal prostaglandin E synthase-1 of America, 103(39), 14507–14512. https://doi.org/10.1073/pnas. (mPGES-1) reduces diet-induced low-grade inflammation and adipos- 0606586103 ity. Frontiers in Physiology, 9, 1358. https://doi.org/10.3389/fphys. Wu, D., Mennerich, D., Arndt, K., Sugiyama, K., Ozaki, N., Schwarz, K., 2018.01358 Wei, J., Wu, H., Bishopric, N. H., & Doods, H. (2009). Comparison of Prieto, D., Contreras, C., & Sánchez, A. (2014). Endothelial microsomal prostaglandin E synthase-1 deletion and COX-2 inhibition dysfunction, obesity and insulin resistance. Current Vascular in acute cardiac ischemia in mice. Prostaglandins & Other Lipid Media- Pharmacology, 12(3), 412–426. https://doi.org/10.2174/ tors, 90(1–2), 21–25. https://doi.org/10.1016/j.prostaglandins.2009. 1570161112666140423221008 06.006 14765381, 2022, 11, Downloaded from https://bpspubs.onlinelibrary.wiley.com/doi/10.1111/bph.15776 by Readcube (Labtiva Inc.), Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License BALLESTEROS-MARTÍNEZ ET AL. 2753 Zhu, L., Xu, C., Huo, X., Hao, H., Wan, Q., Chen, H., Zhang, X., Breyer, R. M., Huang, Y., Cao, X., Liu, D. P., FitzGerald, G. A., & How to cite this article: Ballesteros-Martínez, C., Wang, M. (2019). The cyclooxygenase-1/mPGES-1/endothelial pros- Rodrigues-Díez, R., Beltrán, L. M., Moreno-Carriles, R., taglandin EP4 receptor pathway constrains myocardial ischemia- reperfusion injury. Nature Communications, 10(1), 1888. https://doi. Martínez-Martínez, E., González-Amor, M., org/10.1038/s41467-019-09492-4 Martínez-González, J., Rodríguez, C., Cachofeiro, V., Salaices, M., & Briones, A. M. (2022). Microsomal prostaglandin E synthase-1 is involved in the metabolic and cardiovascular SUPPORTING INFORMATION alterations associated with obesity. British Journal of Additional supporting information may be found in the online version Pharmacology, 179(11), 2733–2753. https://doi.org/10.1111/ of the article at the publisher's website. bph.15776