Animals

All animals were treated in accordance with the protocol approved by the Ethics Committee of Animal Research at Beijing Tongren Hospital, Capital Medical University, Beijing, China. Animals were housed in controlled-temperature rooms (22 °C) with a 12:12-h dark–light cycle.

ACE2 cleaves Ang II to produce Ang(1–7), which acts mainly through the Mas receptor. Thus, the deletion of ACE2 or Mas would affect the expression levels or the action of Ang(1–7), and ACE2 knockout (KO) and Mas KO mice were chosen to explore the role of the lack of Ang(1–7) on metabolism. Male ACE2 KO mice(8 weeks of age) and their age and sex-matched C57BL/6 J wild-type (WT) littermates, and male Mas KO mice and their age and sex-matched C57BL/6 J WT littermates were obtained from the Nanjing Biological Medicine Research Institute (n = 5–7/group), and were fed with high-fat diet (HFD)(60 kcal% fat) (Research Diets, New Brunswick, NJ, USA) for 8 weeks, and body weight was measured every week.

Male db/db mice at 8 weeks of age were purchased from the Nanjing Biological Medicine Research Institute, and were fed with HFD(60 kcal% fat) (Research Diets, New Brunswick, NJ, USA) (n = 5/group). Db/db mice were concomitantly treated with either normal saline (NS), Ang(1–7) (576 µg/kg/day; H-1715, Bachem, USA), or Ang(1–7) combined with Mas receptor inhibitor A779 (1152 µg/kg/day; H-2888, Bachem, USA) for 4 weeks using mini osmotic pumps (Alzet-Durect, Cupertino, CA, USA Model #1004) placed subcutaneously [22], and body weight was measured every week.

Fat analysis by magnetic resonance imaging

Body fat distribution of ACE2 KO, Mas KO and WT mice was assayed by magnetic resonance imaging (MRI) (Varian 7 T/160 mm animal MRI scanner) equipped with a gradient coil system producing a gradient of up to 400 mT/m in each of the three dimensions. The calculation of fat volume is based on the fat segmentation method [23].

Tissue collection

All mice were sacrificed after anesthesia. Subcutaneous and visceral fat tissues were rapidly removed and weighed, immediately frozen in liquid nitrogen, and stored at − 80 °C for posterior analysis.

Determination of lipid profile, adipokines and Ang(1–7) in epididymal adipose tissue

Frozen epididymal adipose tissue from different mice models was allowed to thaw to room temperature and homogenized by macerating with a disposable glass pipet. Adipose total cholesterol (TC) and TG were measured by enzymatic colorimetric method using commercial kits (Beijing Labo Biotech, CO, LTD). Adipose leptin and adiponectin were measured by ELISA kits (eBioscience, USA). In addition, Ang(1–7) levels in epididymal adipose tissue of WT and ACE2 KO mice were also detected by ELISA kits (Cloud-Clone Corp., China).

Histology

Epididymal adipose tissue from different mice models was fixed in buffered formalin, embedded in paraffin, sectioned, mounted on slides, deparaffinized and stained with hematoxylin/eosin to evaluate the adipocyte morphology according to a standard protocol [24]. Images were captured on a slide scanner (3DHISTECH, Pannoramic MIDI), and the representative visions were chosen by the Pannoramic viewer imaging system. For adipocyte size measurements, two images per mice were analyzed using the Pannoramic viewer imaging system.

Cell culture and treatments

3T3-L1 pre-adipocytes were purchased from Cell Resource Center, IBMS, CAMS/PUMC, China. Cells were maintained in Dulbecco’s Modified Eagle’s Medium (DMEM, GIBCO, Carlsbad, CA, USA) supplemented with 10% calf serum and penicillin–streptomycin (100 U/ml, GIBCO) at 37 °C with 5% CO2 under humidified conditions. Upon confluence, the growth medium was changed the following day and replaced with differentiation medium consisting of DMEM, 0.5 mM 3-isobutly-1-methylxanthine (IBMX, cat. No. 2547, Sigma Aldrich, St. Louis, MO, USA), 1 uM dexamethasone (cat. No. D1756, Sigma Aldrich), 10% fetal bovine serum (FBS, GIBCO), and 10ug/ml insulin (I5500-100 mg, Sigma Aldrich) for 2 days. After additional 2 days in DMEM containing 10% FBS and 10 μg/ml insulin, the growth medium was replaced with DMEM supplemented with 10% FBS for another 2–4 days. Oil Red O staining was performed to confirm the differentiation of adipocytes.

The differentiated cells were treated with NS as control or pre-loaded with 400 µM of palmitic acid (PA, Sigma-Aldrich, St. Louis, MA, USA) for 24 h to induce ER stress [25], then treated with NS, 10–9 mmol/L Ang(1–7) or both Ang(1–7) and 10–6 mmol/L A779 for 24 h [26]. Oil Red O staining was performed to assess lipid accumulation.

Oil Red O staining

Differentiated 3T3-L1 cells were washed three times with PBS and then fixed with 4.0% formaldehyde for 25 min. Subsequently, cells were stained with freshly diluted Oil Red O solution for 20 min at room temperature. Finally, cells were washed with double distilled water, and visualized by light microscopy and photographed.

Western blot

Epididymal adipose tissue and 3T3-L1 cells were homogenized in RIPA buffer, and protein concentration in lysates was assessed by the BCA method using a commercially available kit (Beyotime, China). Equal amounts of protein were resolved by 10% SDS-PAGE gel, transferred electrically onto PVDF membranes (Millipore, Billerica, MA, USA). After blocking with 5% nonfat milk, primary antibodies were added overnight at 4 °C. The antibodies targeted the following proteins: β-Actin (4970, CST, USA), ACE2 (Sc-20998, Santa Cruz, USA), Mas (AAR-013, Alomone labs, Israel), Fatty Acid Synthase (FAS) (3180, CST), acetyl-CoA carboxylase α (ACCα) (Sc-30212, Santa Cruz), sterol regulatory element-binding protein-1c (SREBP-1c) (Sc-366, Santa Cruz), C/EBP homologous protein (CHOP) (2895, CST), Glucose regulated protein 78 (GRP78) (ab21685, Abcam, England), activating transcription factor 4 (ATF4) (11,815, CST). Then the membranes were incubated with horseradish peroxidase-conjugated secondary antibodies at room temperature for 1.5 h. The bands were finally detected with ECL Detection Regents (Bio-Rad, USA).

Statistical analysis

Data were expressed by mean ± SEM. Different groups were compared by Student t test or one-way ANOVA (with Bonferroni post-hoc tests to compare replicate means). Prism 5 (GraphPad Software, San Diego, CA) was used for all statistical analyses. P < 0.05 (two-side) indicated a statistically significant difference.