Activation of AT₂ receptors prevents diabetic complications in female db/db mice by NO‐mediated mechanisms

Abstract

Background and Purpose

The AT₂ receptor plays a role in metabolism by opposing the actions triggered by the AT₁ receptors. Activation of AT₂ receptors has been shown to enhance insulin sensitivity in both normal and insulin resistance animal models. In this study, we investigated the mechanism by which AT₂ receptors activation improves metabolism in diabetic mice.

Experimental Approach

Female diabetic (db/db) and non‐diabetic (db/+) mice were treated for 1 month with the selective AT₂ agonist, compound 21 (C21, 0.3 mg·kg⁻¹·day⁻¹, s.c.). To evaluate whether the effects of C21 depend on NO production, a subgroup of mice was treated with C21 plus a sub‐pressor dose of the NOS inhibitor l‐NAME (0.1 mg·ml⁻¹, drinking water).

Key Results

C21‐treated db/db mice displayed improved glucose and pyruvate tolerance compared with saline‐treated db/db mice. Also, C21‐treated db/db mice showed reduced liver weight and decreased hepatic lipid accumulation compared with saline‐treated db/db mice. Insulin signalling analysis showed increased phosphorylation of the insulin receptor, Akt and FOXO1 in the livers of C21‐treated db/db mice compared with saline‐treated counterparts. These findings were associated with increased adiponectin levels in plasma and adipose tissue and reduced adipocyte size in inguinal fat. The beneficial effects of AT₂ receptors activation were associated with increased eNOS phosphorylation and higher levels of NO metabolites and were abolished by l‐NAME.

Conclusion and Implications

Chronic C21 infusion exerts beneficial metabolic effects in female diabetic db/db mice, alleviating type 2 diabetes complications, through a mechanism that involves NO production.

Abbreviations

C21

compound 21

FOXO1

transcription factor forkhead box O1

InsR

insulin receptor

l‐NAME

N ^ω‐nitro‐l‐arginine methyl ester hydrochloride

RAS

renin–angiotensin system

UCP1

uncoupling protein 1

What is already known

• Stimulation of the AT₂ receptor has beneficial metabolic effects.

What this study adds

• The AT₂ receptor agonist C21 ameliorates diabetes‐induced complications in female db/db mice through NO

• C21 increases plasma and adipose levels of adiponectin and improves insulin signaling in the liver.

What is the clinical significance

• Drugs targeting the AT₂ receptor may represent a therapeutic option to prevent diabetes

• This action may be useful in alleviating diabetes and its complications in females.

1. INTRODUCTION

Major features of type 2 diabetes associated with obesity include dysregulated glycaemic profile, dyslipidaemia, increased adiposity with redistribution of fat in central areas, hyperinsulinaemia, increased glucose production, hyperglycaemia and hepatic steatosis (Kahn, Hull, & Utzschneider, 2006). Given that the consequences of type 2 diabetes are serious and include long‐term severe organ and tissue damage (Papatheodorou, Banach, Bekiari, Rizzo, & Edmonds, 2018), finding new therapeutic strategies to counteract this disease is essential.

The renin–angiotensin system (RAS) plays a key role in the development of metabolic syndrome and type 2 diabetes (Henriksen, 2007). Activation of the AT₁ receptor increases the generation of inflammatory cytokines and augments oxidative stress, which ultimately diminish the tissue sensitivity to insulin and lead to the development of insulin resistance (Forrester et al., 2018; Iimura et al., 1995). Contrarily, blockade of the AT₁ receptor consistently improves insulin sensitivity and glucose homeostasis in various animal models of insulin resistance and/or type 2 diabetes (Henriksen, 2007; Iimura et al., 1995; Munoz, Giani, Dominici, Turyn, & Toblli, 2009; Rodriguez et al., 2018; Shiuchi et al., 2004). The role of the AT₂ receptors in metabolic control is less well known, but recent evidence suggests that pharmacological activation of the AT₂ receptor blocks the production of inflammatory cytokines and lowers oxidative stress, thereby improving insulin sensitivity and secretion (Carey, 2017; Santos et al., 2019). Although AT₁ antagonists and ACE inhibitors are administered for the purpose of alleviating diabetes complications such as diabetic nephropathy and vascular disease in humans (Umanath & Lewis, 2018), novel formulations that modulate the AT₂ receptor may represent a therapeutic breakthrough (Paulis et al., 2016). In this sense, the discovery and characterization of the selective small‐molecule agonist, compound 21 (C21), has been a major advance in the field (Paulis et al., 2016; Santos et al., 2019; Steckelings et al., 2011). Administration of C21 exerts a beneficial effect on insulin sensitivity in both diabetic and insulin‐resistant animal models (Ohshima et al., 2012; Shao, Yu, & Gao, 2014; Shao, Zucker, & Gao, 2013; Than et al., 2017; Wang, Wang, Li, & Leung, 2017). In addition, administration of C21 enhances insulin delivery and metabolic action in rat skeletal muscle (Yan et al., 2018). Together with our recent observation that prolonged C21 administration enhances insulin sensitivity in mice (Quiroga et al., 2018), these findings support the potential role of AT₂ receptor as a modulator of insulin action and glucose homeostasis.

To further characterize the therapeutic effect of AT₂ receptor activation in metabolic‐related diseases, we studied the effect of C21 on db/db mice, a mouse model of type 2 diabetes and obesity‐induced hyperglycaemia. To gain insight into the potential participation of NO in the beneficial effects of C21, a group of diabetic mice was treated concomitantly with a sub‐pressor dose of the NOS inhibitor Nω‐nitro‐l‐arginine methyl ester hydrochloride (l‐NAME). In general, studies on mouse models of type 2 diabetes have historically employed male mice (Mauvais‐Jarvis, 2015). However, preclinical studies have demonstrated that the beneficial AT₂ receptor‐mediated effects on cardiovascular and renal function are enhanced in females, primarily in rodent models (Hilliard et al., 2012; Hilliard et al., 2014; Silva‐Antonialli et al., 2004). The main hypothesis behind the current work is that C21 treatment ameliorates type 2 diabetes‐associated complications through a NO‐mediated mechanism. To that end, we analysed glucose and pyruvate tolerance, circulating and tissue TNF‐α, changes in hepatic steatosis and fibrosis, and circulating and adipose tissue content of both adiponectin and resistin in db/db female mice that were treated for a month with C21 in absence or presence of l‐NAME. To gain further insight into the mechanism behind the beneficial effects of C21, we analysed the status of the insulin signalling system (insulin receptor [IsnR], Akt and transcription factor forkhead box O1 [FOXO1]) as well as the phosphorylation and protein levels of eNOS in both liver and adipose tissue of the experimental animals.

2. METHODS

2.1. Mice and study design

All animal procedures were approved by the Cedars‐Sinai Institutional Animal Care and Use Committee and conducted in accordance with the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals. The animals were treated humanely and all efforts were made to minimize the animals' suffering and the number of mice used in the study. Animal studies are reported in compliance with the ARRIVE guidelines (Percie du Sert et al., 2020) and with the recommendations made by the British Journal of Pharmacology (Lilley et al., 2020). Animals were maintained under controlled light and temperature conditions and had free access to water and standard chow diet. Experiments were performed on 16‐week‐old db/db female mice (B6.BKS(D)‐Lepr^db/J, stock #697) (IMSR_JAX:000697) and their respective db/+ (heterozygous) controls purchased from The Jackson Laboratory (Bar Harbor, ME, USA).

The group size of all animal experiment was designed to at least n = 5. Taking into consideration the expected attrition owing to long‐term treatment, we increased the group size of mice to n = 7. All group size was estimated by power analysis (Schmidt, Lo, & Hollestein, 2018). To evaluate the effects of chronic AT₂ receptor agonism, mice (n = 7 per group) received a 4‐week subcutaneous treatment with either sterile saline (vehicle controls) or C21 (0.3 mg·kg⁻¹·day⁻¹, Vicore Pharma AB, Gothenburg, Sweden) using osmotic minipumps (Alzet, Catalogue #1004, Cupertino, CA, USA). We generated groups of equal size using a completely randomized design in the experiment. The experiments were blind to the operators and analysers. The dose of C21 was selected based on previous published protocols (Nag, Khan, Samuel, Ali, & Hussain, 2015). To test the role of NO, a subgroup of mice (n = 7) received C21 plus a sub‐pressor dose of l‐NAME (0.1 mg·ml⁻¹, Bachem, Torrance, CA, USA) in drinking water. Systolic BP (SBP) was measured weekly in conscious mice using the tail‐cuff method (Visitech BP2000 System, Visitech Systems Inc., Apex, NC, USA) after proper training. SBP was measured 25 times for each time point. At Day 24, mice were fasted for 6 h and a glucose tolerance test (GTT) was conducted as described below. After a 2‐day resting period (Day 26), mice were fasted again for 6 h to perform a pyruvate tolerance test (PTT) as detailed below. At Day 28, mice were killed via conscious decapitation after a 6‐h fast. Liver, heart, and inguinal and perirenal adipose tissues were harvested and used for histological analysis, cytokine assessment and the study of key insulin signalling‐related proteins by Western blotting as described in sections below.

2.2. Urine collection and albuminuria

For urine collection, mice were placed in metabolic cages (Tecniplast, Varese, VA, USA) for 6 h with free access to water and standard chow diet (Bio Serv, Flemington, NJ, USA). Urine was collected in refrigerated tubes, centrifuged at 2,000 g for 10 min at 4°C and aliquoted for further analysis. Urine albumin levels were measured using an elisa commercial kit (Albuwell, Exocell, Philadelphia, PA, USA). Systemic NOS inhibition by l‐NAME was confirmed by measuring urinary excretion of NO metabolites (NO_x) using the Griess assay (Promega Corp., Fitchburg, WI, USA). Creatinine levels were assessed using the Creatinine LiquiColor Test (EKF Diagnostics Inc, Boerne, TX, USA) as previously described (Eriguchi et al., 2018). Both urine albumin and NO_x excretion values were corrected by urine creatinine levels.

2.3. Glucose tolerance test (GTT) and pyruvate tolerance test (PTT)

Mice were fasted for 6 h prior to commencement of a GTT. Baseline glucose levels were sampled from tail blood using a Contour Blood Glucose Monitoring System (Bayer HealthCare LLC, Tarrytown, NY, USA). Then, mice were injected intraperitoneally with 2 g of glucose per kilogram of body weight using a 25% w/v solution prepared in sterile PBS. Blood glucose level was measured at 15, 30, 60 and 120 min after injection. After 2 days, a PTT was performed. For this, mice were fasted for 6 h and then injected intraperitoneally with 1 g of pyruvate per kilogram of body weight using a 10% w/v solution prepared in sterile PBS. Blood glucose levels were then measured at 15, 30, 60 and 120 min after injection via tail‐end bleed.

2.4. Insulin, adiponectin and resistin assessment in plasma

Mice were killed by conscious decapitation, blood was collected in EDTA‐coated tubes and centrifuged at 2,000 g for 10 min at 4°C and plasma was stored at −80°C until analysis. Plasma insulin was measured using the Ultra‐Sensitive Mouse Insulin elisa Kit (Crystal Chem, Elk Grove Village, IL, USA), plasma adiponectin was measured using the Mouse Adiponectin elisa Kit (Crystal Chem) and plasma resistin was measured using a Mouse Resistin elisa Kit (RayBiotech, Peachtree Corners, GA, USA) following the manufacturer's instructions.

2.5. Tissue homogenization and elisa

Perirenal and inguinal adipose tissue (200 mg) was homogenized in 500 μl of Mammalian Protein Extraction Reagent (M‐PER buffer, Thermo Scientific, Waltham, MA, USA) containing 0.5‐mM disodium EDTA, 0.2‐mM PMSF, 9 μg·ml⁻¹ of aprotinin and 5 μg·ml⁻¹ of phosphatase inhibitor cocktail 3 (P2850, Sigma‐Aldrich, St. Louis, MO, USA). Each sample was homogenized for 5 min at a medium‐speed setting with an Ultra‐Turrax T25 (IKA Labortechnik, Wilmington, NC, USA) and placed on ice for 30 min. Samples were then centrifuged at 20,000 g for 10 min at 4°C, the lipid layer was removed and protein concentrations were determined using a bicinchoninic acid protein assay (BCA assay, Pierce Thermo Scientific, Rockford, IL, USA). For the liver, 100 mg of tissue was homogenized in 500 μl of radioimmunoprecipitation assay buffer (RIPA Pierce, Thermo Scientific) containing 0.5‐mM disodium EDTA, 0.2‐mM PMSF, 9 μg·ml⁻¹ of aprotinin and 5 μg·ml⁻¹ of phosphatase inhibitor cocktail (P2850, Sigma‐Aldrich). After centrifugation at 20,000 g for 10 min at 4°C, the supernatant was collected and the protein concentration was determined using the BCA assay. The levels of IL‐1β and TNF‐α were measured in whole‐tissue homogenates using commercially available elisa kits (eBioscience, San Diego, CA, USA).

2.6. Real‐time PCR

To quantify the gene expression of the AT₂ receptor in the liver, total RNA was isolated from liver samples using RNeasy Mini Kit (Qiagen, Valencia, CA, USA) following the manufacturer's instructions. Synthesis of cDNA was performed using 1 μg of total RNA and Maxima™ H Minus cDNA Synthesis Master Mix (Thermo Fisher, Rockford, IL, USA). Real‐time PCR was performed in a 10‐μl reaction mixture consisting of 5‐μl iTaq Universal SYBR Green Supermix (Bio‐Rad, Hercules, CA, USA), 1‐μl cDNA and 0.3 μM of primers for each specific target. Amplification was performed at 95°C for 15 min, followed by 45 cycles at 95°C for 20 s and 60°C for 60 s. The relative amount of the target mRNA was normalized with the housekeeping gene GAPDH. The primers were as follows:‐ AT₂ receptor 5′‐GGGTAAACAGACCCAGCAAA‐3′ and 5′‐CTGGAACTGTGCCCAGAAAT‐3′ and GAPDH 5′‐AACTTTGGCATTGTGGAAGG‐3′ and 5′‐GGATGCAGGGATGATGTTCT‐3′.

2.7. Western blotting

The Immuno‐related procedures used comply with the recommendations made by the British Journal of Pharmacology (Alexander et al., 2018). Liver and adipose tissue extracts were denatured, resolved by SDS‐PAGE, transferred into PVDF membranes (Millipore Immobilon‐FL, EMD Millipore, Billerica, MA, USA) and finally probed with specific antibodies for adiponectin/Acrp30 (AB_2221787) (R&D Systems, Minneapolis, MN, USA), uncoupling protein 1 (AB_2241459) (UCP‐1, Thermo Fisher), InsR β subunit (AB_1950593) (GeneTex, Irvine, CA, USA) and phospho‐Tyr 1158/1162/1163 IR β subunit (AB_568831) (Millipore, Burlington, MA, USA), Akt (AB_915783) and phospho‐Ser 473 Akt (AB_2315049) (Cell Signaling, Danvers, MA, USA), and FOXO1 (AB_823503) and phospho‐Ser 256 FOXO1 (AB_329831) (Cell Signaling). Signals were detected with the Odyssey Infrared Imaging System (LI‐COR, Lincoln, NE, USA) and quantified by accompanying software. Test values were normalized to control group values for unwanted sources of variation. To assess the error of the control group, each individual control value was divided by average intensity obtained for the control group (db/+ mice). The intensity value of each individual band was divided by the average intensity obtained for the control group (db/+ mice). The units shown in bar graphs were obtained by considering the average value of intensity of each specific band in the control group as 100% and transforming them to 1 arbitrary unit ± SD.

For protein loading control and normalization, membranes were re‐probed with antibodies against β‐actin (AB_476697) for the liver or GAPDH (AB_796208) (Sigma‐Aldrich) for adipose tissue. The level of each protein evaluated was normalized to the level of β‐actin or GAPDH from control samples to avoid sources of variation. The MW of proteins was estimated using pre‐stained protein markers (Bio‐Rad). Table S1 catalogues the amounts assayed, antibodies, vendors and dilutions used.

2.8. Masson's trichrome and Oil Red O staining

For the assessment of interstitial hepatic fibrosis, liver samples were fixed with 10% buffered formalin and embedded in paraffin. Four‐micrometre‐thick sections were deparaffinized, rehydrated and stained with Masson's trichrome. The analysis of lipid droplets in the liver was performed using the Oil Red O staining (Mehlem, Hagberg, Muhl, Eriksson, & Falkevall, 2013). For this, immediately after collection, liver samples were snap frozen in liquid nitrogen and stored at −80°C. Later, tissues were embedded in Tissue‐Tek and 8‐μm‐thick sections were cut and air dried. Then sections were fixed in formalin for 15 min, washed for 2 min with tap water and rinsed with 60% isopropanol for 2 min. To make the Oil Red O working solution, 30 ml of Oil Red O stock solution (Sigma‐Aldrich, Catalogue #O1391) was diluted with 20 ml of distilled water and protected from light. Fixed sections were then stained with the working solution horizontally for 15 min and rinsed three times with 60% isopropanol (1 min each). Samples were counterstained with haematoxylin for 1 min and immediately evaluated. Images were acquired using an RVL‐100‐G Microscope (Echo Laboratories, San Diego, CA, USA) at 20× magnification.

2.9. Morphological analysis of adipose tissue

Perirenal and inguinal adipose tissue samples were preserved in 10% formalin solution and embedded in paraffin. Three‐micrometre‐thick paraffin‐embedded sections were then deparaffinized in xylene and rehydrated to water through ethanol at 100°C, 96°C and 80°C and stained with haematoxylin and eosin (H&E) (Quiroga et al., 2018). For each individual sample, adipocyte size was measured in five microscopic fields in single‐blinded condition and expressed in μm². The size of 60 adipocytes was determined for each mouse. Images were acquired as described above.

2.10. Data and analysis

The data and statistical analysis comply with the recommendations of the British Journal of Pharmacology on experimental design and analysis in pharmacology (Curtis et al., 2018). Animals were randomly distributed between groups. All statistical analyses were performed using GraphPad Prism 8 (GraphPad Software, San Diego, CA, USA) (SCR_002798). The threshold for statistical significance was set at the level of P being 0.05. Therefore, in all cases, P < 0.05 was considered statistically significant. Statistical analysis was undertaken only for studies where each group size was at least n = 5. No additional data, involving group sizes of n < 5, were used or analysed. Every individual value obtained was included in the analysis and no outliers were detected. Data are presented as individual dot plots and the mean ± SD. Differences among experimental groups were compared by two‐way ANOVA. Tukey's post hoc tests were performed only if the F value in ANOVA achieved the required level of statistical significance (P < 0.05) and there was no significant variance in homogeneity. The declared group size is the number of independent values. Statistical analysis was done using these independent values.

2.11. Nomenclature of targets and ligands

Key protein targets and ligands in this article are hyperlinked to corresponding entries in the IUPHAR/BPS Guide http://www.guidetopharmacology.org (Harding et al., 2018), and are permanently archived in the Concise Guide to PHARMACOLOGY 2019/20 (Alexander et al., 2019).

3. RESULTS

3.1. C21 treatment and metabolic parameters

Female diabetic (db/db) mice and their corresponding non‐diabetic (db/+) littermates received subcutaneous C21 or saline for 1 month via osmotic minipumps. Another subgroup of mice was simultaneously treated with C21 and a sub‐pressor dose of l‐NAME in drinking water. Initially, we confirmed that l‐NAME inhibited systemic NOS without affecting BP. C21 induced an increase in urinary NO metabolites (NO_x: nitrites and nitrates) in db/db mice compared with db/db mice receiving saline (~2‐fold, Figure S1A). Although this increase was completely prevented with l‐NAME, no significant changes in BP were observed (Figure S1A,B). In both db/db and db/+ mice treated with C21 + l‐NAME, systolic BP remained unchanged during the whole experiment (Figure S1B). The absence of significant albuminuria (Figure S1C) and cardiac hypertrophy (Table 1) in l‐NAME‐treated mice adds further evidence that no changes in BP occurred during the treatment.

TABLE 1.

Female diabetic (db/db) mice and their respective control (db/+) received a subcutaneous treatment with either saline or C21 for 4 weeks

Tissue	db/+	db/db	db/+	db/db	db/+	db/db
	Saline	Saline	C21	C21	C21 + l‐NAME	C21 + l‐NAME
Heart (g)	0.14 ± 0.02	0.17 ± 0.03	0.15 ± 0.02	0.17 ± 0.02	0.14 ± 0.02	0.16 ± 0.02
Inguinal fat (g)	0.4 ± 0.1	1.8 ± 0.5 *	0.4 ± 0.1	1.2 ± 0.5 * , # , †	0.34 ± 0.06	1.7 ± 0.2 *
Perirenal fat (g)	0.5 ± 0.2	2.6 ± 0.6 *	0.6 ± 0.1	1.8 ± 0.4 * , # , †	0.5 ± 0.1	2.8 ± 0.6 *
Liver (g)	0.9 ± 0.1	3.7 ± 0.7 *	0.9 ± 0.2	2.7 ± 0.8 * , #	0.9 ± 0.1	3.2 ± 0.5 *

Note: A subgroup of C21‐treated mice also received l‐NAME in the drinking water. Tissue weight for the heart, liver, and inguinal and perirenal adipose tissues were measured at the end of the experiment. Data are expressed as mean ± SD.

Abbreviations: C21, compound 21; l‐NAME, N ^ω‐nitro‐l‐arginine methyl ester hydrochloride.

^* P < 0.05 versus the corresponding db/+ group.

^# P < 0.05 versus db/db saline group.

^† P < 0.05 versus db/db C21 + l‐NAME group.

As expected, body weight and blood glucose were significantly increased in db/db compared with non‐diabetic db/+ mice at the beginning of the experiment (Figure S2). Treatment with C21 or C21 + l‐NAME did not modify these parameters (Figure 1a,b). Plasma insulin and the Homeostatic Model Assessment of Insulin Resistance (HOMA‐IR) index were also significantly increased in db/db compared with non‐diabetic controls (Figure 1c,d). Interestingly, treatment of db/db mice with C21 induced a significant reduction (39%) of the HOMA‐IR that was reverted by the concomitant administration of l‐NAME (Figure 1c,d). Interestingly, C21 induced a significant increase of plasma adiponectin in db/db mice (~2.3‐fold, Figure 1e) that was blunted in the presence of l‐NAME (Figure 1e). Plasma resistin levels were similar among all experimental groups (Figure 1f).

FIGURE 1.

Chronic compound 21 (C21) treatment improves insulin sensitivity and increases plasma levels of adiponectin in diabetic mice. (a) Body weight, (b) blood glucose, (c) plasma insulin, (d) Homeostatic Model Assessment of Insulin Resistance (HOMA‐IR), (e) plasma adiponectin, and (f) plasma resistin were assessed at the end of the experiment. Data are expressed as mean ± SD. ^* P < 0.05 (n = 7 per group). l‐NAME, N ^ω‐nitro‐l‐arginine methyl ester hydrochloride

To gain further insight into the metabolic profile of db/db mice and the effect of C21, a GTT and a PTT were performed at the end of the experiment. Diabetic db/db mice displayed a significant intolerance to an acute glucose load compared with non‐diabetic db/+ littermates. This observation was true for non‐treated db/db and db/db mice receiving either C21 or C21 + l‐NAME (Figure 2a). However, when comparing the AUC, C21‐treated db/db mice showed a significant AUC reduction (from 4,178 ± 216 to 3,544 ± 304 mg·dl⁻¹·min⁻¹, Figure 2b) compared with non‐treated db/db mice. Similarly, a PTT to evaluate liver gluconeogenesis revealed that, compared with db/+ mice, db/db mice have an exaggerated blood glucose elevation in response to pyruvate (Figure 2c). Again, this response was also improved after C21 treatment (from 4,059 ± 202 to 3,296 ± 217 mg·dl⁻¹·min⁻¹, Figure 2d). The concomitant administration of C21 plus l‐NAME blunted the beneficial effect of C21 on both glucose (Figure 2b) and pyruvate handling (Figure 2d).

FIGURE 2.

Chronic administration of compound 21 (C21) improves glucose and pyruvate tolerance in female db/db mice. (a) Glucose tolerance test (GTT) and (c) pyruvate tolerance test (PTT) were performed as described in Section 2. For this, basal glucose was measured after a 6‐h fast and reanalysed at 15, 30, 60, and 120 min after injection of either glucose or pyruvate. (b, d) The AUC was calculated for each experimental group. Data are expressed as mean ± SD. ^* P < 0.05 (n = 7 per group). l‐NAME, N ^ω‐nitro‐l‐arginine methyl ester hydrochloride

3.2. Effects of C21 on inguinal and perirenal adipose tissue

Both inguinal and perirenal adipose tissue weights were increased in db/db mice compared with db/+ mice. Inguinal fat showed a ~4.5‐fold increase, while perirenal fat displayed a ~5.2‐fold increase in untreated db/db mice compared with db/+ mice (Table 1). In C21‐treated db/db mice, inguinal adipose tissue weight was significantly reduced compared with non‐treated db/db mice (from 1.8 ± 0.5 to 1.2 ± 0.5 mg, Table 1). Perirenal fat weight of db/db mice was also lower in the C21‐treated animals (2.6 ± 0.6 mg in untreated db/db mice vs. 1.8 ± 0.4 mg in C21‐treated animals, Table 1). In db/db mice treated with C21 + l‐NAME, both the inguinal and perirenal adipose tissue weights were indistinguishable from untreated db/db mice, suggesting that the effect of C21 was counteracted by l‐NAME (Table 1).

Further histological analysis of inguinal adipose tissue revealed that db/db mice receiving saline showed a significant increase in adipocyte size compared with non‐diabetic controls. Although C21‐treated db/db mice still displayed larger adipocytes compared with non‐diabetic mice, their size was significantly reduced compared with non‐treated db/db mice (8,346 ± 1,519 vs. 5,617 ± 1,186 μm², Figure 3a,b). Even more important, adiponectin levels were significantly increased by C21 in inguinal adipose tissue from db/db mice (~2.4‐fold increase, Figure 3c). Uncoupling protein 1 (UCP1; SLC25A7), a marker of adipose tissue browning, was significantly reduced in db/db mice compared with db/+ controls. Although C21 induced a slight increase of UCP1 adipose tissue content in db/db mice, this change did not reach statistical significance (Figure 3d). As observed for other parameters, blockade of NO production with l‐NAME blunted the effects of C21 regarding adipocyte size and tissue adiponectin content (Figure 3a–c). Contrarily, perirenal adipose tissue from db/db mice appeared less responsive to C21 as none of the parameters evaluated were modified in this fat depot after a 4‐week administration of C21 or C21 + l‐NAME (Figure S3).

FIGURE 3.

Compound 21 (C21) decreases adipocyte cell size and increases adiponectin levels in inguinal adipose tissue. These effects are prevented with N ^ω‐nitro‐l‐arginine methyl ester hydrochloride (l‐NAME). Paraffin sections (3 μm) of inguinal adipose tissue were stained with haematoxylin and eosin, and images were acquired as described in Section 2. (a) Representative histological images were obtained in adipose tissue sections at 20× magnification. (b) To evaluate the adipocyte size distribution, the average area of 60 adipocytes was measured in five different microscopic fields for each sample. (c) Adiponectin and (d) uncoupling protein 1 (UCP1) were measured in homogenates of inguinal adipose tissue by Western blot. GAPDH confirms equal protein amount in each sample. Data are expressed as mean ± SD. ^* P < 0.05 (n = 7 for all groups)

3.3. Effects of C21 on hepatic steatosis and fibrosis

To evaluate whether the beneficial effects of C21 were also observed in the liver, we evaluated hepatic lipid deposition and fibrosis at the end of the experiment. Diabetic db/db mice displayed an increased liver weight compared with their db/+ counterparts. However, C21‐treated db/db mice showed a significant reduction in liver weight compared with non‐treated db/db mice (2.7 ± 0.8 vs. 3.7 ± 0.7 g, Table 1). C21 also decreased hepatic lipid deposition in db/db mice. Histological analysis of liver sections stained with the Oil Red O showed that C21‐treated db/db mice had less lipid deposition (~48% reduction, Figure 4a,b) and smaller lipid droplets (~60% reduction, Figure 4c) compared with non‐treated db/db mice. Finally, histological evaluation of liver fibrosis by Masson's trichrome staining revealed a significant increase in extracellular matrix deposition in the liver of untreated db/db mice, with a significant reduction in the C21‐treated db/db mice (~50%, Figure 4d,e). Although NO inhibition with l‐NAME did not modify the effect of C21 on liver weight (Table 1), it abolished the protective effect of C21 on hepatic lipid deposition. Diabetic mice receiving both C21 and l‐NAME showed increased lipid deposition (Figure 4b) and larger lipid droplets (Figure 4c) in the liver compared with C21‐treated db/db mice.

FIGURE 4.

Chronic compound 21 (C21) administration decreases lipid accumulation in the liver of db/db mice. These effects were partially blunted with N ^ω‐nitro‐l‐arginine methyl ester hydrochloride (l‐NAME). (a) Frozen liver sections (8 μm) were stained with Oil Red O as described in Section 2. (b) The area of positive staining per field and (c) the average size of lipid droplets were calculated using ImageJ. (d) For fibrosis assessment, paraffin sections (3 μm) of liver tissue were stained with Masson's trichrome staining. (e) Percentage of fibrosis (indicated with yellow arrows) per total area was calculated using ImageJ. Analysis was performed in 10 representative histological images obtained at 20× magnification. Bars represent 50 μm. Data are expressed as mean ± SD. ^* P < 0.05 (n = 7 for all groups)

3.4. Role of C21 on hepatic insulin signalling of female db/db mice

To get a deeper insight into the molecular mechanism that mediates the beneficial effects of C21 on metabolism, we evaluated the phosphorylation and total abundance of InsR, Akt and FOXO1, key intracellular molecules associated with the insulin signalling pathway in the liver. In non‐treated db/db mice, despite very high plasma insulin levels, the phosphorylation of InsR was negligible and not different from db/+ mice (Figure 5a). However, in C21‐treated db/db mice, basal IR phosphorylation was significantly higher compared with non‐treated db/db mice (3.5 ± 1.5 vs. 1.5 ± 0.4 arbitrary units). The greater hepatic InsR activation displayed by C21‐treated db/db mice was blunted by concomitant administration of l‐NAME (Figure 5a). Akt phosphorylation showed no changes between non‐treated db/db and db/+ mice. However, db/db mice treated with C21 displayed an increased Akt phosphorylation compared with non‐treated db/db mice (2.6 ± 0.9 vs. 1.6 ± 0.4 arbitrary units, Figure 5b). In accordance to InsR phosphorylation, l‐NAME reduced Akt phosphorylation in db/db mice to levels that were similar to those found at baseline (Figure 5b). No significant changes were observed in total protein expression of either InsR or Akt (Figure 5a,b). Hepatic FOXO1 phosphorylation, a major regulator of hepatic gluconeogenesis, was increased in db/db mice compared with db/+ littermates. However, C21‐treated db/db mice displayed significantly higher FOXO1 phosphorylation compared with non‐treated db/db (Figure 5c). As shown for the other signalling molecules analysed, the effect of C21 was blunted in the presence of a sub‐pressor dose of l‐NAME (Figure 5c). When compared with their respective db/+ group, db/db mice displayed a decreased expression of total FOXO1 in the liver that was not modified by C21 treatment (Figure 5c). Finally, the FOXO1 phosphorylation‐to‐total protein ratio was higher in db/db mice treated with C21 compared with non‐treated db/db mice (Figure 5c).

FIGURE 5.

Compound 21 (C21) increases the phosphorylation of insulin signalling molecules in the liver of db/db mice. The phosphorylation level and total abundance of (a) the insulin receptor (InsR), (b) Akt, and (c) transcription factor forkhead box O1 (FOXO1) were evaluated in liver homogenates by Western blot. Western blot membranes were re‐probed with an anti‐β‐actin antibody for loading control, and each sample was corrected by its corresponding β‐actin value. The phosphorylation‐to‐protein ratio was calculated for each sample. Positive controls for IR and Akt (rightmost bands) are liver homogenates from db/+ mice injected with 5 μg of insulin. Data are expressed as mean ± SD. ^* P < 0.05 (n = 7 for all groups). l‐NAME, N ^ω‐nitro‐l‐arginine methyl ester hydrochloride

3.5. C21 decreases TNF‐α abundance in the liver

To determine whether the beneficial effects of C21 were associated with the inhibition of inflammation, we evaluated the hepatic and adipose tissue content of TNF‐α and IL‐1β, two pro‐inflammatory cytokines widely associated with metabolic abnormalities during obesity and diabetes. Hepatic TNF‐α abundance was significantly increased in db/db mice compared with non‐diabetic controls (161 ± 17 vs. 55 ± 17 pg·mg⁻¹ of protein, Figure 6a). Interestingly, db/db mice receiving C21 for 1 month showed a significant reduction of TNF‐α compared with non‐treated db/db mice (83 ± 37 vs. 161 ± 28 pg·mg⁻¹ of protein, Figure 6a). In line with previous findings, co‐administration of C21 with l‐NAME blunted this protective response indicating a role of NO in the C21‐mediated anti‐inflammatory effect (Figure 6a). IL‐1β was also increased in the liver of db/db mice compared with db/+ mice (54 ± 9 vs. 15 ± 5 pg·mg⁻¹ of protein, Figure 6b). Notably, this inflammatory cytokine was significantly reduced by C21 treatment (from 54 ± 9 to 35 ± 14 pg·mg⁻¹ of liver). Again, C21 + l‐NAME treatment blunted this reduction (Figure 6b). Similar studies were performed in inguinal and perirenal adipose tissue. In those adipose tissue depots, both TNF‐α and IL‐1β were significantly elevated in all db/db experimental groups compared with non‐diabetic db/+ mice (Figure 6c–f) and remained unchanged after C21 treatment. These data indicate that, at least in this experimental model of type 2 diabetes, the anti‐inflammatory effect of C21 is primarily observed in the liver. The analysis of hepatic AT₂ receptor expression by RT‐PCR adds further evidence supporting a role of C21 in the liver of female diabetic mice. As shown in Figure S4, db/db mice have, although not significant, higher expression of AT₂ receptor in the liver compared with db/+ control littermates. This increase was similar in all db/db experimental groups (Figure S4).

FIGURE 6.

Diabetic db/db mice treated with compound 21 (C21) display lower levels of pro‐inflammatory cytokines in the liver. (a) TNF‐α and (b) IL‐1β were measured in total liver (a, b; n = 7 for all groups) and inguinal (c, d; n = 6 for db/+ saline and n = 7 for all other groups) and perirenal (e, f; n = 6 for db/+ saline and n = 7 for all other groups) adipose tissue homogenates using a commercial elisa as described in Section 2. Data are expressed as mean ± SD. l‐NAME, N ^ω‐nitro‐l‐arginine methyl ester hydrochloride. ^* P < 0.05

3.6. C21 increases the phosphorylation of eNOS in the liver and inguinal adipose tissue of female db/db mice

Further evidence supporting the role of NO in C21 beneficial effects was obtained from the assessment of total abundance and phosphorylation of eNOS at Ser1177 in liver and inguinal adipose tissue by Western blot. In the liver of non‐treated mice, the level of eNOS phosphorylation was similar between db/+ and db/db mice. However, in inguinal adipose tissue, db/db displayed significantly lower phosphorylated eNOS compared with the non‐treated db/+ mice (1.0 ± 0.3 vs. 0.4 ± 0.2 arbitrary units). In db/+ mice, C21 increased phosphorylated eNOS levels in the liver (from 1.0 ± 0.2 to 2.4 ± 0.8 arbitrary units, Figure 7a) but not in inguinal fat (Figure 7b). However, in db/db mice, C21 significantly increased the phosphorylation of eNOS in both liver (from 1.2 ± 0.3 to 4.0 ± 1.1 arbitrary units, Figure 7a) and inguinal fat (0.4 ± 0.2 to 1.3 ± 0.3 arbitrary units, Figure 7b). In both db/+ and db/db mice, l‐NAME blunted C21‐induced eNOS phosphorylation. Mice treated with C21 plus l‐NAME displayed eNOS phosphorylation levels that were indistinguishable from non‐treated mice (Figure 7a,b). In the liver, both db/+ and db/db mice displayed similar levels of total eNOS (Figure 7a). In inguinal adipose tissue, the levels of eNOS were significantly decreased in db/db compared with db/+ (Figure 7b). Neither C21 nor l‐NAME modified the total abundance of eNOS in these tissues (Figure 7b). Further analysis revealed that C21 increased the phosphorylation‐to‐protein ratio of eNOS in both liver and inguinal adipose tissue of db/db mice. This increase was prevented by l‐NAME chronic administration (Figure 7a,b).

FIGURE 7.

Compound 21 (C21) increased the phosphorylation at Ser1177 of endothelial NOS (eNOS) in liver and inguinal adipose tissue of female db/db mice that was prevented with a chronic administration of a subpressor dose of N ^ω‐nitro‐l‐arginine methyl ester hydrochloride (l‐NAME) (0.1 mg·ml⁻¹ in the drinking water). The phosphorylation at Ser1177 and the total abundance of eNOS were evaluated in (a) liver and (b) inguinal fat homogenates by Western blot. Western blot membranes were re‐probed with an anti‐β‐actin (liver) or anti‐GAPDH (inguinal fat) antibody for loading control, and each sample was corrected by its corresponding β‐actin or GAPDH value. The phosphorylation‐to‐protein ratio was calculated for each sample. Data are expressed as mean ± SD. ^* P < 0.05 (n = 7 for all groups)

4. DISCUSSION

The main findings of our current study were as follows:‐ (i) Treatment with C21 improved glucose and pyruvate tolerance in db/db mice; (ii) C21‐treated db/db mice showed reduced liver weight, decreased hepatic levels of TNF‐α, IL‐1β, and lipid accumulation, and lower liver fibrosis compared with db/db mice receiving vehicle; (iii) these changes were associated with a significant improvement of insulin signalling (increased InsR, Akt and FOXO1 phosphorylation) in the livers of diabetic animals; (iv) chronic AT₂ receptor agonism led to increased levels of adiponectin both in the circulation and in inguinal adipose tissue as well as reduced adipocyte size in inguinal fat; (v) C21 increased the phosphorylation of eNOS at Ser1177 in liver and inguinal adipose tissue and (vi) most of the beneficial effects resulting from AT₂ receptor stimulation through C21 administration were significantly prevented by the concomitant administration of l‐NAME and indicating a major participation for NO in these positive effects.

Previous studies in animal models of metabolic disorders indicated beneficial effects of AT₂ receptor stimulation using C21 in vivo, namely, reduction of body weight and fat mass and improvement of insulin sensitivity in type 2 diabetic KKAy mice (Ohshima et al., 2012) and in high‐fructose/high‐fat‐fed rats (Shum et al., 2013). Interestingly, C21 also improved glucose tolerance and insulin sensitivity in mice fed with a regular diet (Quiroga et al., 2018). The db/db mouse is extreme in having a marked increase in body weight, fat mass, hyperglycaemia and pronounced hyperinsulinaemia. Thus, the improvement in glucose tolerance and pyruvate tolerance (an indication of hepatic gluconeogenesis) after C21 treatment is both novel and marked. Aside from the InsR, Akt signalling is central to hepatocellular insulin action and coordinates all insulin effects in the liver including the gluconeogenic capacity (Petersen & Shulman, 2018). The transcription factor FOXO1 stimulates a gluconeogenic transcriptional programme and is repressed after phosphorylation by Akt (Petersen & Shulman, 2018). Interestingly, although protein levels of FOXO1 were not modified by C21 treatment, this therapy led to an increase in FOXO1 phosphorylation, indicating further repression of gluconeogenesis. Thus, we postulate that the beneficial metabolic effects of C21 treatment in db/db mice could be ascribed to the concomitant increased phosphorylation of major hepatic insulin signalling components. This observation is novel and adds to the knowledge of the mechanisms by which AT₂ receptor stimulation improves insulin action.

One of the major findings of the current study is the identification of favourable effects of C21 on liver steatosis and fibrosis. Many lines of evidence have indicated that obesity is closely linked to a chronic inflammatory state, which contributes to metabolic disorders (Shoelson, Lee, & Goldfine, 2006). Among these complications, non‐alcoholic steatohepatitis (NASH), a progressive liver disease characterized by hepatic steatosis that leads to inflammation, fibrosis, and cirrhosis, stands out (Tilg & Moschen, 2010). Currently, there is no treatment for this disease, which is thought to begin with excessive fat accumulation in the liver, followed by aggravating factors such as oxidative stress, inflammatory cytokines, and endotoxins (Tilg & Moschen, 2010). In preclinical studies, angiotensin II has been associated with the development of NASH since blockade of AT₁ receptors has been reported to ameliorate this condition (Ran, Hirano, & Adachi, 2004). The AT₁ receptor, which is localized in hepatocytes, bile duct cells, hepatic stellate cells, myofibroblasts, Kupffer cells, and vascular endothelial cells, mediates most of the actions of angiotensin II in the liver (Warner, Lubel, McCaughan, & Angus, 2007). However, some studies also reported AT₂ receptor gene (Agtr2) expression in liver tissue (Bataller et al., 2003; Nabeshima et al., 2006). Interestingly, we observed that AT₂ receptor is slightly increased in db/db mice compared with non‐diabetic littermates. This suggests that our current results showing decreased fat accumulation and anti‐fibrogenic effects associated with AT₂ receptor activation are likely due to direct activation of liver AT₂ receptors in diabetic mice. TNF‐α is a key player involved in the establishment of liver fibrosis. This cytokine, mainly released by hepatocytes or Kupffer cells, activates hepatic stellate cells in a paracrine manner and mediates chronic liver injury and inflammation (Yang & Seki, 2015). Our current results showing decreased hepatic fibrosis after treatment with C21 correlate well with the decreased hepatic TNF‐α and IL‐1β levels observed in C21‐treated diabetic mice. Previous studies showed that AT₁ receptor blockade suppresses the recruitment of macrophages to the liver and decreases TNF‐α expression (Toblli et al., 2008). In view of our current data, the amelioration of NASH observed after AT₁ receptor antagonism could be ascribed, at least in part, to the interaction of angiotensin II with liver AT₂ receptors.

In humans, adipose cell enlargement in association with decreased expression of lipogenic genes and expansion of visceral adipose tissue are mediators of obesity‐related insulin resistance (McLaughlin et al., 2016; Salans, Knittle, & Hirsch, 1968). Chronic treatment with C21 has been shown to reduce the size of hypertrophic adipocytes and adipose tissue mass in high‐fat diet‐fed rats (Nag et al., 2015; Shum et al., 2013). Also, we have previously reported a decreased number of large adipocytes and increased number of new, small adipocytes after treatment of C57BL/6 mice with C21 (Quiroga et al., 2018). Our current results in db/db mice showing decreased inguinal fat and reduced adipocyte size after treatment with the AT₂ agonist highlight the potential therapeutic activity of AT₂ receptor activation in the treatment of obesity‐associated disorders.

Adiponectin is a hormone secreted by white adipose tissue with insulin‐sensitizing properties. Thus, reduction in the levels of this adipokine is associated with the development of metabolic syndrome and type 2 diabetes (Fang & Sweeney, 2006). In mice, administration of adiponectin lowers glycaemia and improves insulin action (Berg, Combs, Du, Brownlee, & Scherer, 2001; Kubota et al., 2002). Thus, the increase in adiponectin levels detected after treatment with C21 might contribute to the improvement of glucose and pyruvate tolerance. In line with current findings, treatment with C21 has been reported to increase serum adiponectin levels both in diet‐induced insulin‐resistant rats (Shum et al., 2013) and in normal mice (Quiroga et al., 2018; Than et al., 2017). Collectively, previous reports and current findings suggest that increased tissue and circulating adiponectin levels could be considered a read‐out of in vivo AT₂ receptor activation. We and others have previously analysed the effects of AT₂ receptor activation on the levels of UCP1 in adipose tissue and found an increase in this marker of browning as indicative of improved thermogenesis (Quiroga et al., 2018; Than et al., 2017). Similar results were recently reported in mice fed with a high‐fat diet (Nag, Patel, Mani, & Hussain, 2019). In contrast to these findings, UCP1 levels in adipose tissue were not altered by C21 treatment neither in db/+ nor in db/db mice. Differences in the animal model, dose, time, and route of administration could explain this result. Also, a limitation of the current study is that males were not employed to draw a broader conclusion on the beneficial effects of C21 in db/db mice.

One of the most important findings of the present study is that NO mediates most of the beneficial actions of C21 in type 2 diabetes‐related complications. Although the knowledge of the physiological roles of the AT₂ receptor has increased significantly, the information regarding the signalling pathways triggered by AT₂ receptor activation is scarce. Activation of AT₂ receptor is known to be associated with stimulation of NO production (Siragy & Carey, 1997; Steckelings, Kaschina, & Unger, 2005; Takata et al., 2015). In this area, the spectrum of AT₂ receptor signalling has recently been expanded with the report showing the participation of the Ser/Thr kinase Akt in AT₂ receptor‐mediated NO production (Peluso et al., 2018). Also, a recently published study shows that AT₂ receptor activation inhibits fatty acid uptake in adipocytes through a NO‐mediated mechanism (Nag et al., 2019). However, the exact contribution of NO to the beneficial effects of C21 in vivo was not yet addressed. Several studies have underscored NO to have both anti‐obesogenic and insulin‐sensitizing properties (Sansbury & Hill, 2014). NO increases fat oxidation in peripheral tissues and decreases lipid synthesis in the liver. It also increases transport of insulin and glucose to key peripheral tissues and regulates gluconeogenesis (Sansbury & Hill, 2014). Our studies demonstrate that chronic C21 administration increased the phosphorylation of eNOS at Ser1177 in liver and inguinal adipose tissue of db/db mice. This was associated with increased urinary NO_x. Both eNOS phosphorylation and urinary NO_x were reduced in db/db mice receiving a sub‐pressor dose of l‐NAME. Thus, preventing C21‐induced NO synthesis blunted many of the beneficial effects of C21 during diabetes. In non‐diabetic mice, C21 did not modify NO_x, and it only increased eNOS phosphorylation in liver but not in inguinal adipose tissue, suggesting that non‐diabetic mice are less responsive to AT₂ receptor activation than diabetic counterparts.

Overall, our findings demonstrate that activation of AT₂ receptors through the administration of the selective AT₂ agonist C21 improved glucose and pyruvate tolerance in female db/db mice. These changes were associated with decreased liver weight, fibrosis, lipid deposition, and TNF‐α abundance as well as with improved insulin signalling in this tissue. Inguinal fat weight and adipocyte size were significantly reduced after C21 administration, while both tissue and circulating levels of adiponectin were increased, indicative of improved insulin sensitivity. Further mechanistic analysis revealed that NO might mediate many of the C21 beneficial effects during diabetes. In conclusion, drugs targeting the AT₂ receptor may represent a therapeutic option to prevent diabetes and alleviate its complications.

AUTHOR CONTRIBUTIONS

F.P.D, U.M.S., and J.F.G. conceived and designed the research; F.P.D., L.C.V., J.Z.Y.S., D.T.Q., E.A.B., and J.F.G. performed the experiments; F.P.D., L.C.V., J.Z.Y.S., E.A.B., and J.F.G. interpreted the results of the experiments; F.P.D., L.C.V., K.E.B., and J.F.G. analysed the data; F.P.D. and J.F.G. prepared the figures; F.P.D., L.C.V., J.Z.Y.S., and J.F.G. drafted the manuscript; and all the authors revised, edited, and approved the final version of the manuscript.

CONFLICT OF INTEREST

The authors declare no conflicts of interest.

DECLARATION OF TRANSPARENCY AND SCIENTIFIC RIGOUR

This Declaration acknowledges that this paper adheres to the principles for transparent reporting and scientific rigour of preclinical research as stated in the BJP guidelines for Design & Analysis, Immunoblotting and Immunochemistry and Animal Experimentation and as recommended by funding agencies, publishers and other organizations engaged with supporting research.

Supporting information

Figure S1 . C21 induced an increase of urinary nitrites/nitrates (NOx) that was prevented with a chronic administration of a sub‐pressor dose of L‐NAME (0.1 mg/ml in drinking water). Six‐hour urine collection was performed at the end of the C21 treatment. Urinary NOx were assessed using the Griess assay and corrected by urinary creatinine (A; n = 6 for db/+ C21 and n = 7 for all other groups). Blood pressure was evaluated every week by the tail‐cuff method after proper mouse training (B; n = 7 for all groups). Urine albumin was measured by a commercial ELISA and corrected by urinary creatinine (C; n = 6 for db/+ C21 and n = 7 for all other groups). Data are expressed as mean ± SD. *P < 0.05.

Figure S2. Basal body weight (A) and blood glucose (B) before starting the C21 treatment. Data are expressed as mean ± SD. *P < 0.05 (n = 7 for all groups).

Figure S3. C21 did not induce significant changes in adipocyte cell size and adiponectin levels in perirenal adipose, tissue. Paraffin sections (3 μm) of perirenal adipose tissue were stained with hematoxylin and eosin and images acquired as described in Methods. Representative histology images were obtained in adipose tissue sections at x20 magnification (A). To evaluate the adipocyte size distribution, the average area of 60 adipocytes was measured in five different microscopic fields for each sample (B). Adiponectin (C) was measured in homogenate of perirenal adipose tissue by Western blot. Glyceraldehyde 3‐phosphate dehydrogenase (GAPDH) confirms equal protein amount for each sample. Data are expressed as mean ± SD. *P < 0.05 (n = 7 for all groups).

Figure S4. AT₂R mRNA expression in the liver. Data are expressed as mean ± SD (n = 6 for db/+ Saline, db/db saline, db/+ C21 and n = 7 for all other groups).

Table S1. Immunoblot protocol and antibody details. kDa refers to apparent molecular weight determined by SDS‐PAGE molecular weight markers. GAPDH: Glyceraldehyde 3‐phosphate dehydrogenase, IR/IusR: insulin receptor. UCP1: Uncoupling Protein 1, O/N: overnight.

Dominici FP, Veiras LC, Shen JZY, et al. Activation of AT₂ receptors prevents diabetic complications in female db/db mice by NO‐mediated mechanisms. Br J Pharmacol. 2020;177:4766–4781. 10.1111/bph.15241