Pharmacologic blockade of the natriuretic peptide clearance receptor promotes weight loss and enhances insulin sensitivity in type 2 diabetes

Abstract

While natriuretic peptides (NPs) are primarily known for their renal and cardiovascular actions, NPs stimulate lipolysis in adipocytes and induce a thermogenic program in white adipose tissue (WAT) that resembles brown fat. The biologic effects of NPs are negatively regulated by the NP clearance receptor (NPRC), which binds and degrades NPs. Knockout (KO) of NPRC protects against diet induced obesity and improves insulin sensitivity in obese mice. To determine if pharmacologic blockade of NPRC enhanced the beneficial metabolic actions of NPs in type 2 diabetes, we blocked NP clearance in a mouse model of type 2 diabetes using the specific NPRC ligand ANP(4-23). We found that treatment with ANP(4-23) caused a significant decrease in body weight by increasing energy expenditure and reducing fat mass without a change in lean body mass. The decrease in fat mass was associated with a significant improvement in insulin sensitivity and reduced serum insulin levels. These beneficial effects were accompanied by a decrease in infiltrating macrophages in adipose tissue, and reduced expression of inflammatory markers in both serum and WAT. These data suggest that inhibiting NP clearance may be an effective pharmacologic approach to promote weight loss and enhance insulin sensitivity in type 2 diabetes. Optimizing the therapeutic approach may lead to useful therapies for obesity and type 2 diabetes.

INTRODUCTION

Obesity is a public health problem that has reached epidemic proportions.¹ The worldwide prevalence of overweight individuals was 2.3 million in 2015, and the prevalence has continued to increase in many countries over the last decade.¹ Overweight and obese individuals are at a substantial risk for the development of comorbidities such as type 2 diabetes.^1,2 Moreover, it is estimated that the lifetime risk of developing diabetes is 1 in 3 for United States (US) citizens born after 2000.² The cost of caring for this patient population was over 3.2 billion dollars in the US in 2017.³ Obesity and its complications are, therefore, a significant and costly medical problem. As a result, new and effective treatment approaches are needed to reduce the prevalence of overweight individuals.

While NPs are primarily known for their cardiovascular and renal actions, NPs also have important effects in other tissues and cell types.⁴ NPs potently stimulate lipolysis in adipocytes and induce a thermogenic program in WAT that resembles brown fat.^5–10 Epidemiologic data suggest that NPs play a role in reducing the risk of developing obesity and its complications such as type 2 diabetes. For example, low NP levels are associated with insulin resistance in both lean and obese individuals.¹¹ Conversely, higher NP levels appear to be protective against obesity.¹²

The effects of NPs are mediated by binding to the cell surface NP receptors (NPRs), and stimulating cGMP generation.⁴ Atrial NP (ANP) and brain NP (BNP) circulate to act on target organs and tissues by binding to NP receptor-A (NPRA).⁴ The C-type natriuretic peptide (CNP) is produced and acts locally by binding to NPRB.⁴ In contrast, the NP clearance receptor (NPRC) binds and degrades all 3 NPs.^4,13 As a result, NPRC negatively regulates the actions of NPs.^4,13 The biologic importance of NPRC in adipocytes was investigated in mice lacking NPRC specifically in adipocytes.¹⁴ The KO mice were resistant to obesity induced by a high fat diet, and exhibited both increased energy expenditure and improved insulin sensitivity.¹⁴ These data suggest that pharmacologic inhibition of NP clearance by NPRC might be a useful strategy for treating obesity and its complications such as type 2 diabetes.

In addition to NPRC, the biological effects of NPs are negatively regulated by cleavage of NPs by endoproteases including the neutral endopeptidase neprilysin,^13,15 and by intracellular hydrolysis of cGMP by a family of phosphodiesterases (PDEs).¹⁵ The relative role of NPRC and neprilysin in clearing NPs from the circulation has been investigated using pharmacologic inhibitors.^13,16–19 In these experiments, infusion of NPRC antagonists enhanced both ANP concentrations and the physiologic effects of natriuretic peptides to a greater extent than pharmacologic neprilysin inhibition.^13,16–19 Moreover, NPRC is highly expressed in adipocytes and previous studies suggested that inhibition of either neprilysin or PDE5 did not play a significant role in regulating NP signaling in adipocytes.^7,20 In contrast, more recent studies found an important role for PDE9 in diet induced obesity.^21,22 Taken together, these studies suggest that the effects of NPs in adipocytes is regulated by both increased clearance of NPs from the circulation by NPRC, and postreceptor mechanisms including hydrolysis of cGMP by PDE9.^7,20–23 This negative regulation of NP signaling by NPRC may be particularly important in type 2 diabetes because hyperinsulinemia reciprocally upregulates NPRC and downregulates NPRA in adipocytes, which further limits the local effects of NPs.^6,7,14,24

To test the therapeutic potential of pharmacologic blockade of NPRC in type 2 diabetes, we treated BTBR-ob/ob with the NPRC ligand ANP (4-23) that specifically binds to NPRC without binding NPRA or NPRB.^25–27 BTBR-ob/ob mice are leptin-deficient and develop severe obesity and type 2 diabetes.^28,29 These mice are also one of the few mouse models of type 2 diabetes that develops robust diabetic kidney disease (DKD).^28,30 The initial focus of our studies was DKD, but review of the literature suggested that NPRC blockade might have important effects on obesity and insulin sensitivity. As a result, we modified our studies to focus on the metabolic effects of ANP(4-23). We found that ANP(4-23) reduced body weight by decreasing fat mass without a change in lean body mass, and improved insulin sensitivity in BTBR-ob/ob mice. These data suggest that pharmacologic blockade of the NPRC may be a useful therapeutic approach for reducing both obesity and insulin resistance in type 2 diabetes mellitus.

MATERIALS AND METHODS

Materials

Primary antibodies used for the study included: (1) A mouse monoclonal antibody to actin (clone C4, catalog #: MA1501, Sigma-Aldrich, St. Louis, MO), (2) A mouse monoclonal antibody to NPRC (clone OTI4H1, catalog #: TA501044, Origene Technologies, Rockville, MD), (3) A rabbit polyclonal antibody to UCP1 (catalog #: ab23841, Abcam, Cambridge, UK), (4) A rat monoclonal antibody to F4/80 (clone C1:A3-1, catalog #: MCA497R. Bio-Rad, Hercules, CA), and (5) The following monoclonal or polyclonal rabbit antibodies from (Cell Signaling Technology): Total Akt (catalog #: 4691), phospho-Akt (T308) (catalog #: 13038), adiponectin (catalog #: 2789), F4/80 (catalog #: 70076), Glut4 (catalog #: 2213) hormone sensitive lipase (HSL) (catalog #: 18381), and phospho-HSL (S660) (catalog #: 45804). Secondary antibodies from Cell Signaling Technology included: (1) An antimouse HRP-linked polyclonal antibody (catalog #: 7076), (2) An antirabbit HRP-linked polyclonal antibody (catalog #: 7074), and (3) An antirat Alexa Fluor 555 Conjugate (catalog #: 4417). The following additional materials were obtained from Sigma-Aldrich: ANP(4-23) (catalog #: SCP0022), a protease inhibitor cocktail (catalog #: P8340) and a phosphatase inhibitor 3 cocktail (catalog #: P0044).

Animal experiments

Experiments were performed with male BTBR T+tf/J mice (BTBR) mice using the animal facility in the Medical Sciences Research Building 1 (MSRB1) at Duke University, Durham, NC. Age matched male WT BTBR and BTBR-ob/ob mice were obtained from Jackson Laboratories (Bar Harbor, ME). Mice were housed in groups of 2–4 mice per cage at 72°F (22.2°C) in a 12-hour light-dark cycle. Sample size for the experiments was estimated based on the primary outcome (weight loss) from a previous adipocyte specific NPRC KO study in mice.¹⁴ For the experiments, 6 WT BTBR mice and 14 BTBR-ob/ob mice were assigned to treatment with 10 nmol/kg ANP(4-23). An identical number of WT BTBR mice and BTBR-ob/ob mice were treated with vehicle. The drug and vehicle were given by intraperitoneal (IP) injection 5 days per week (Monday to Friday) starting at 4 weeks of age. During the study, the order of treatment was random. The 10 nmol/kg dosage of ANP(4-23) was chosen based on a previous publication,³¹ and we confirmed that this dosage of ANP(4-23) significantly enhanced urinary cGMP excretion in a previous study.³² Mice were weighed every Monday and the dosage of ANP(4-23) adjusted for changes in body weight. The rodent diet (Purina Rodent Chow #5053) was obtained from Nestlé Purina Petcare, St. Louis, MO. Food was weighed prior to feeding every Monday, Wednesday, and Friday. Before refeeding, the remaining food was weighed to monitor food consumption. Fasting blood glucose levels were obtained in the morning after a 12-hour fast at 2, 4, and 6 weeks of age using the Contour Next testing system (Bayer, Parsippany, NJ). Twenty-four-hour urine samples were collected after 4 weeks of treatment and prior to harvest using metabolic cages specifically designed for collection of mouse urine (Hatteras Instruments, Cary, NC). Indirect calorimetry and magnetic resonance imaging (MRI) were performed as described below during week 7. At the end of the study, insulin tolerance tests (ITTs) were performed as described below and then mice were euthanized after the last urine collection by injecting pentobarbital, IP, followed by harvesting of internal organs. Immediately following euthanasia, blood was obtained by cardiac puncture and kidneys, heart, liver, epididymal WAT, interscapular BAT, and retroperitoneal WAT were removed and weighed. Tissues samples were removed and immediately snap frozen in liquid nitrogen for extraction of mRNA and protein for quantitative RT-PCR and immunoblotting, respectively. Additional tissue samples were saved in 10% buffered formalin (Azer Scientific, Morgantown, PA), and snap frozen in liquid nitrogen using optimal cutting temperature (OCT) compound. During the study, there were no deaths in the WT BTBR groups but 6 deaths in the severely diabetic BTBR-ob/ob mice during the last 2 weeks of the study [4 vehicle treated mice and 2 mice treated with ANP(4-23)]. All experiments conformed to the Guide for the Care and Use of Laboratory Animals and were approved by the Duke Medical Center’s Institutional Animal Care and Use Committee. A schematic of the experimental protocol is provided in Supplemental Fig 1.

Indirect calorimeter

Indirect calorimetry was performed using the Comprehensive LAb Monitoring System (CLAMS) from Columbus Instruments, Columbus, OH. Before starting the experiment, system was calibrated with the standard gas mixture. Animals were housed in a single cage chamber for the measurement of oxygen consumption, carbon dioxide release, heat production, and respiratory exchange ratio (RER). The measurements were made according to the directions of the manufacturer over 24 hours, and the stable data were used to evaluate the metabolic response of the animals.

Body composition analysis

MRI studies were performed using the Bruker LF90 II “Minispec” body composition analyzer (Bruker Optics, Billerica, MA) TD-NMR system. This device measures the mass of fat, lean tissue, and body fluid. Before starting the measurements, the system was calibrated with a device of equivalent size to the animals being studied. Mice were placed in an acrylic cylinder customized to fit the animal’s size and were loosely restrained within the cylinder by pushing a plunger to maintain the animal within a length of ~20 cm inside the cylinder. The cylinder was then positioned inside the bore of the magnet. The measurements of fat mass, lean body mass, and fluid were recorded automatically, and animals returned to their home cage in about 1 minute.

Measurement of adipocyte surface area

Retroperitoneal WAT and interscapular BAT were fixed in 10% formalin and then switched to 70% ethanol. Fixed adipose was sectioned and stained with hematoxylin and eosin (H&E) by the Duke Research Animal Pathology Service. Five digital images were taken of each section at 20× magnification using a Nikon Eclipse TB-2000 microscope with a Roper Scientific Photometrics digital camera. Adipocyte cross-sectional area was determined by evaluating digital images using Adobe Photoshop CS6 Extended software after a calibrating the measurement tool to convert pixels to a known surface area. The adipocyte surface area in the digital images was then outlined using the “lasso” tool and the number of pixels in the adipocyte area converted to μm2. Forty measurements by 3 different observers blinded to treatment group were made for each mouse for a total of 120 measurements per animal. Measurements were made by 3 different observers without knowledge of either genotype or experimental groups. Concurrence was greater than 90% for the average values between the individuals making the measurements.

Immunofluorescence studies

Retroperitoneal WAT was fixed in 10% formalin and then switched to 70% ethanol. Fixed adipose tissue was sectioned by the Duke Research Animal Pathology Service. Tissue sections were then deparaffinized by treating sequentially for 5 minutes with xylene, xylene/ethanol, 95% ethanol, 70% ethanol, and 50% ethanol, and then washing twice with deionized water. Slides were then washed briefly with phosphate buffered saline (PBS), placed in 10 mM sodium citrate buffer (pH 6.0), and heated to 100°C for 20 minutes. After washing with PBS, tissues were permeabilized by treatment with 0.1% Triton X-100 (Sigma-Aldrich, St. Louis, MO) and nonspecific binding was blocked by incubation in 2% goat serum (Sigma-Aldrich) in PBS for 20 minutes. Sections were then incubated with a 1:100 dilution of the rat monoclonal antibody to F4/80 (clone C1:A3-1) at 4°C overnight. After washing 3 times with PBS, slides were incubated with the antirat antibody (Alexa Fluor 555 Conjugate) for 2 hours at room temperature. Slides were then washed 3 times in PBS and incubated with a 1:100 dilution of 40 μM Alexa Fluor 488 phalloidin (Thermo Fisher Scientific, Waltham, MA) for 20 minutes in the dark at temperature. After 3 washes of 5 minutes duration in PBS, glass coverslips were mounted using Vectashield Vibrance antifade mounting medium (Vector Laboratories (Burlingame, CA) containing 0.9 μg/mL DAPI (4′,6-diamidino-2-phenylindole). Slides were examined using a Nikon Eclipse TE-2000S inverted fluorescent microscope and images taken with a Roper Scientific Photometrics digital camera.

Note: Additional methodologic details are provided in the Supplemental Materials.

RESULTS

Effect of ANP(4-23) on food intake, body and organ weights, and body composition

For the experiments, we treated 4-week-old BTBR mice for 8 weeks with ANP(4-23) as described in the Methods section (see schematic in Supplemental Fig 1). As shown in Fig 1, A, food intake was increased in BTBR-ob/ob mice compared to similarly treated WT BTBR controls at all time points. Treatment with ANP(4-23) reduced daily food intake per week in BTBR-ob/ob compared to vehicle treated BTBR-ob/ob mice, but the decrease in food intake was not statistically significant at any of the weekly time points. Fig 1, B shows cumulative daily food intake over the 8-week study. Treatment with ANP(4-23) reduced cumulative food intake in BTBR-ob/ob mice but the reduction in food intake was not statistically significant.

Fig 1.

Food intake, body weights, and body composition. (A) Food intake was increased at all weekly time points in BTBR-ob/ob mice compared to similarly treated WT BTBR controls. (B) Cumulative weekly food intake was decreased in the BTBR-ob/ob mice treated with ANP(4-23) compared to BTBR-ob/ob controls, but this difference was not statistically significant. (C) Body weight was significantly increased in BTBR-ob/ob mice compared to similarly treated WT BTBR mice at all weekly time points. There was also a significant decrease in body weight in BTBR-ob/ob mice treated with ANP(4-23) compared to vehicle treated BTBR-ob/ob mice at the 7- and 8-week time points. (D–F) These data show the results for MRI studies at the 7-week time point. Body weight, fat mass, and lean body mass were significantly different in BTBR-ob/ob mice compared to similarly treated WT BTBR mice. Treatment with ANP(4-23) caused a significant decrease in body weight in BTBR-ob/ob mice compared to vehicle-treated BTBR-ob/ob mice that was caused by a significant decrease in fat mass with little change in lean body mass. *P < 0.0001 vs similarly treated WT BTBR mice by an ANOVA followed by a repeated measures test, **P < 0.05 vs vehicle-treated BTBR-ob/ob by an ANOVA followed by a repeated measures test, Panel B was analyzed by a t-test. Panels D–F were analyzed by an ANOVA followed by Sidak’s multiple comparisons test.

We next investigated the effect of ANP(4-23) on body weight and body composition. Fig 1, C depicts weekly body weight in BTBR-ob/ob mice and WT BTBR controls. As shown in Fig 1, C, body weight was significantly increased in BTBR-ob/ob mice compared to similarly treated WT BTBR mice. In addition, there was a significant decrease in body weight in BTBR-ob/ob mice treated with ANP(4-23) compared to vehicle treated BTBR-ob/ob mice at the 7- and 8-week time points. In contrast, there was no significant effect of ANP(4-23) on body weight in the WT BTBR groups. Fig 1, D–F shows the effects of ANP(4-23) on body composition measured by MRI at the 7-week time point. The decrease in body weight in BTBR-ob/ob mice was caused by a decrease in fat mass with little change in lean body mass. There were no significant differences in body weight, fat mass or lean body mass between the WT BTBR groups.

Fig 2, A shows the effect of ANP(4-23) on the weights of heart, liver, kidneys, and selected adipose depots. There was a significant increase in the liver weight, and the weights of retroperitoneal and epididymal fat in both groups of BTBR-ob/ob mice compared to similarly treated WT BTBR controls. Treatment with ANP(4-23) also reduced the weight of epididymal adipose tissue. Fig 2, B shows the effects of ANP(4-23) on adipocyte surface area in retroperitoneal fat of BTBR mice. Adipocyte surface area was similar in both groups of WT BTBR mice and these data were combined for the data analysis. There was a significant increase in adipocyte surface area in BTBR-ob/ob mice compared to WT BTBR mice. Adipocyte size was decreased in BTBR-ob/ob mice treated with ANP(4-23), but this difference was not statistically significant. Fig 2, C shows the frequency distribution of adipocyte size by deciles, and Fig 2, D shows representative pictures of adipocyte size in the 4 treatment groups.

Fig 2.

Body and organ weights. (A) There was a significant increase in the liver weight, and the weights of retroperitoneal and epididymal fat in both groups of BTBR-ob/ob mice compared to similarly treated WT BTBR controls. Treatment with ANP(4-23) significantly reduced epididymal fat mass. (B) There was a significant increase in adipose surface area in retroperitoneal fat of vehicle treated BTBR-ob/ob mice compared to WT BTBR mice. Treatment with ANP(4-23) decreased adipocyte surface area in BTBR-ob/ob mice compared to vehicle treated BTBR-ob/ob mice, but this difference was not statistically significant. (C) Panel C shows the frequency distribution of adipocyte size by decile. (D) Panel D shows representative pictures of adipocyte size in the 4 groups. ^†P < 0.001 vs similarly treated WT BTBR mice by an ANOVA followed by Sidak’s multiple comparisons test, *P < 0.01 vs vehicle treated BTBR-ob/ob mice by an ANOVA followed by Sidak’s multiple comparisons test. Panels B and C were analyzed by an ANOVA followed by Sidak’s multiple comparisons test.

Effect of ANP(4-23) on energy expenditure

Given the modest effects of ANP(4-23) on food intake, we assessed energy expenditure by indirect calorimetry. Since there was no difference in lean body mass between BTBR-ob/ob treatment groups, data were normalized to lean body mass. Fig 3 shows results for the indirect calorimetry experiments. As shown in Fig 3, A and B, oxygen consumption (VO₂) and carbon dioxide release (VCO₂) were significantly increased in BTBR-ob/ob mice compared with WT BTBR mice. In contrast, respiratory exchange ratio (RER) was significantly decreased in in BTBR-ob/ob mice compared to WT BTBR mice. In addition, there was a significant decrease in RER in WT BTBR mice treated with ANP(4-23) compared to WT BTBR mice treated with vehicle. Fig 3, D shows energy generation in BTBR mice. Energy expenditure is strongly correlated with body size and body composition.³³ These relationships are shown in Fig 3, E and F for body weight and lean body mass. To account for the effects of body composition on energy expenditure in BTBR-ob/ob mice, we performed an analysis of covariance (ANCOVA)³⁴ with body weight and lean body mass as covariates. In this analysis, treatment with ANP(4-23) increased energy expenditure after controlling for either body weight or lean body mass as covariates (P < 0.025 for both covariates).

Fig 3.

Indirect calorimetry studies. (A and B) There was a significant increase in oxygen consumption (VO₂) and carbon dioxide release (VCO₂) as indicated. (C) Respiratory exchange ratio (RER) was significantly decreased in BTBR-ob/ob mice compared with WT BTBR mice. In addition, treatment with ANP(4-23) significantly reduced RER in WT BTBR mice treated with ANP(4-23) compared to WT BTBR mice treated with vehicle. (D–F) These panels show energy expenditure and the relationship between energy expenditure and either body weight or body mass in the treatment groups. Energy expenditure was increased in BTBR-ob/ob mice treated with ANP(4-23) by ANCOVA using either body weight or lean body mass as a covariate (panel D). Panels A–C were analyzed by an ANOVA followed by Sidak’s multiple comparisons test. Panels D–F were analyzed by an analysis of covariance (ANCOVA).

Effect of ANP(4-23) on hyperglycemia, insulin sensitivity, and hyperlipidemia

We next assessed the effect of ANP(4-23) on glycemia and circulating insulin levels. Fig 4, A shows fasting blood glucose levels in BTBR-ob/ob mice. Measurements were made in the morning after a 12-hour fast using a glucometer. At the 2-week time point, glucose levels were lower in BTBR-ob/ob mice treated with ANP(4-23) compared to vehicle treated BTBR-ob/ob mice, but this difference was not statistically significant. Later in the study, glucose levels were consistently above the linear range of the glucometer. At the end of the study, we measured glucose, insulin and hemoglobin A1c (HbA1c) levels in serum collected at the time of harvest (Fig 4, B–D). Both glucose and HbA1c levels were similarly increased in BTBR-ob/ob mice compared to WT BTBR controls. In contrast, insulin levels were significantly reduced in BTBR-ob/ob mice treated with ANP(4-23) compared to vehicle treated BTBR-ob/ob mice (Fig 4, D). Cholesterol and triglyceride levels were significantly increased in BTBR-ob/ob mice compared to similarly treated WT BTBR mice and were not significantly affected by treatment with ANP(4-23) (Fig 4, E and F).

Fig 4.

Effect of ANP(4-23) on glucose and insulin levels and insulin sensitivity. (A) Fasting blood glucose levels in BTBR-ob/ob mice were not significantly different at any of the time points studied but, glucose levels were consistently above the linear range of the glucometer after the 2-week time point. (B and C) Serum glucose and hemoglobin A1c (HbA1c) levels were similarly elevated in vehicle and ANP(4-23) treated BTBR-ob/ob mice compared to WT controls. (D) Insulin levels were significantly elevated in BTBR-ob/ob mice treated with vehicle compared to vehicle treated WT BTBR mice, and this increase with significantly reduced by treatment with ANP(4-23). (E and F) Triglyceride and cholesterol levels were increased in BTBR-ob/ob mice compared to similarly treated WT BTBR mice. Panel A was analyzed by an ANOVA followed by a repeated measures test, Panels B–D were analyzed by an ANOVA followed by Sidak’s multiple comparisons test.

We also performed ITTs in BTBR-ob/ob mice (Fig 5, A) and in WT BTBR mice (Fig 5, B). Treatment with ANP(4-23) caused a significant decrease in glucose levels in both BTBR-ob/ob mice and WT BTBR mice at 60 and 90 minutes after the insulin injection. Treatment with ANP(4-23) also significantly increased the area under the curves in both BTBR-ob/ob mice and WT BTBR animals (Fig 5, C and D). Consistent with the ITTs, basal phospho-Akt levels were increased in retroperitoneal adipose tissue from both WT BTBR mice and BTBR-ob/ob mice treated with ANP (4-23) compared to WT BTBR and BTBR-ob/ob mice treated with vehicle (Fig 5, E and F). A similar increase in basal phospho-Akt levels has been observed in other models that potentiate the effects of endogenous NPs,^14,22 which may be caused, in part, by the stimulatory effects of NPs on glucose induced insulin secretion.^6,35–37 In addition, Glut4 protein levels were decreased in both groups of BTBR-ob/ob mice compared to similarly treated WT BTBR mice (Fig 5, G). There was also a significant decrease in Glut4 mRNA levels in vehicle treated BTBR-ob/ob mice compared to WT BTBR mice (Fig 5, H). Treatment with ANP(4-23) significantly increased Glut4 mRNA expression in BTBR-ob/ob mice compared to BTBR-ob/ob mice treated with vehicle. Lastly, the adipokine adiponectin protects against insulin resistance; however, adiponectin was decreased at both the protein and mRNA levels in BTBR-ob/ob mice compared to WT BTBR mice and did not increase with ANP(4-23) treatment (Supplemental Fig 2).

Fig 5.

Effect of ANP(4-23) on insulin sensitivity. (A and B) Insulin tolerance tests (ITTs) found that treatment with ANP(4-23) caused a significant decrease in glucose levels at the 60- and 90-minute time points in both BTBR-ob/ob mice and WT BTBR mice treated with ANP(4-23) compared to the vehicle treated groups. (C and D) There was a significant increase in the area under the curves in both BTBR-ob/ob mice and WT BTBR mice treated with ANP(4-23) compared to vehicle treated mice. (E and F) Treatment with ANP(4-23) caused a significant increase in phospho-Akt levels in WT BTBR mice and BTBR-ob/ob mice compared to the vehicle treated groups. (E and G) Glut4 protein levels were decreased in BTBR-ob/ob mice compared to WT BTBR mice in the vehicle-treated groups. (H) Glut4 mRNA levels were decreased in vehicle-treated BTBR-ob/ob mice compared to WT BTBR mice, and treatment with ANP(4-23) significantly increased Glut4 expression in BTBR mice-ob/ob mice compared to vehicle treated BTBR-ob/ob mice. Panels A and B were analyzed by an ANOVA followed by a repeated measures test, Panels C and D were analyzed by a t-test. Panels F–H were analyzed by an ANOVA followed by Sidak’s multiple comparisons test, *P < 0.05 or **P < 0.01 vs the vehicle treated groups by an ANOVA followed by Sidak’s multiple comparisons test.

We next performed ANCOVA analyses to determine if treatment with ANP(4-23) affected indices of glycemia and insulin sensitivity after controlling for differences in body weight. We found a significant effect of pharmacologic NPRC blockade on insulin levels (P = 0.034) and insulin sensitivity (area under the curve) (P < 0.0001) using body weight as a covariate. We were unable to demonstrate a significant effect of ANP(4-23) on glycemia (P = 0.053) or HbA1c levels (P = 0.39). These data suggest that enhancing endogenous NP levels reduces insulin levels and increases insulin sensitivity after accounting for differences in body weight.

Effect of ANP(4-23) on inflammation in WAT

Obesity-induced insulin resistant is caused, in part, by inflammation in adipose tissue, which is characterized by accumulation of macrophages that secrete inflammatory cytokines such as monocyte chemotactic protein-1 (MCP1; also termed C-C Motif Chemokine Ligand 2 or CCL2). This cytokine promotes further accumulation of macrophages in adipose tissues and contributes to insulin resistance.³⁸ As shown in Fig 6, A, MCP1 mRNA levels were significantly increased in retroperitoneal WAT from vehicle treated BTBR-ob/ob mice, and was reduced by treatment with ANP(4-23). Similarly, MCP1 protein levels were significantly increased in serum from vehicle treated BTBR-ob/ob mice, and this increase was significantly reduced by ANP(4-23) (Fig 6, B). We next determined if ANP(4-23) affected accumulation of macrophages in WAT using antibodies to the heavily glycosylated macrophage marker F4/80.^39,40 Fig 6, C and D shows immunofluorescence studies in BTBR-ob/ob mice. We detected frequent collections of macrophages in WAT from BTBR-ob/ob mice treated with vehicle. In contrast, collections of macrophages were difficult to detect in WAT from ANP(4-23) treated BTBR-ob/ob mice and in WT controls. We quantitated macrophage infiltration in WAT by immunoblotting for F4/80. As shown in Fig 6, E and F, F4/80 protein levels were significantly increased in WAT in vehicle treated BTBR-ob/ob mice compared to WT controls, and treatment with ANP(4-23) significantly reduced this increase in MCP1 protein levels in BTBR-ob/ob mice. A similar pattern of mRNA expression was observed for the cytokines IL1-β, IL6, and TNF-α (Supplemental Fig 3).

Fig 6.

Effect of ANP(4-23) on MCP1 levels and macrophages in WAT. (A) MCP1 mRNA was significantly increased in WAT from vehicle treated BTBR-ob/ob mice, and this increase was reduced by ANP(4-23). (B) MCP1 protein levels were significantly increased in serum from vehicle treated BTBR-ob/ob mice and this increase was significantly reduced by ANP(4-23). (C and D) Immunofluorescence studies detected frequent collections of macrophages in WAT from BTBR-ob/ob mice treated with vehicle but, collections of macrophages were only rarely seen in BTBR-ob/ob mice treated with ANP(4-23). (E and F) F4/80 protein levels were significantly increased in WAT in vehicle-treated BTBR-ob/ob mice compared to WT controls, and treatment with ANP(4-23) significantly reduced this increase in MCP1 protein. Data were analyzed by an ANOVA followed by Sidak’s multiple comparisons test. Macrophages in WAT were stained sequentially with the rat F4/80 antibody and the antirat Alexa Fluor 555 antibody, and then tissue was counter-stained with Alexa Fluor 488 phalloidin and DAPI (4′,6-diamidino-2-phenylindole) as described in the Methods section.

Effect of ANP(4-23) on lipolysis and expression of UCP1 and PGC1-α

Previous studies have suggested that NPs promote weight loss by increasing lipolysis,^5–7,41 and inducing a thermogenic program in adipose tissue,^5–7 which is characterized by induction of both UCP1 (uncoupling protein 1) and a key regulator of mitochondrial biogenesis PGC1-α (peroxisome proliferator-activated receptor gamma coactivator 1-α).^5–7,14 As shown in Fig 7, A and B, there was increased phosphorylation of hormone sensitive lipase (HSL) on an activating site in interscapular BAT. In addition, protein levels of PGC1-α were significantly increased in BAT by treatment with ANP(4-23) without a change in PGC1-α mRNA levels (Fig 7, C–E). UCP1 protein and mRNA levels were not significantly affected by treatment with ANP(4-23) (Fig 7, F–H). In addition, there was no significant effect of ANP(4-23) on phospho-HSL levels or expression of UCP1 at either the protein and/or mRNA level in retroperitoneal WAT (Supplementary Fig 4). (Note: PGC1-α protein could not be detected in WAT.) Additional gene expression results are presented in Supplementary Figs 5 and 6.

Fig 7.

Expression of phospho-HSL, PGC1-α, and UCP1 in BAT. (A and B) ANP(4-23) significantly increased phosphorylation of HSL on an activating site in BAT. (C–E) ANP(4-23) significantly increased expression of UCP1 protein in BAT without a change in BAT mRNA levels. (F-H) ANP(4-23) did not significantly affect expression of UCP1 protein or mRNA in BAT. Data were analyzed by an ANOVA followed by Sidak’s multiple comparisons test.

Effect of ANP(4-23) on urinary cGMP excretion and expression of NPRC and neprilysin in WAT

We measured urinary cGMP excretion in the final urine collection to assess the effectiveness of NPRC blockade. As shown in Fig 8, A, there was a significant increase in urinary cGMP excretion in WT BTBR mice treated with ANP(4-23) compared to vehicle treated WT BTBR group. A similar pattern was seen in BTBR-ob/ob mice, but this difference was modest and not statistically significant. There was also a significant decrease in urinary cGMP excretion in both groups of BTBR-ob/ob mice compared to similarly treated WT BRBR mice. Given that previous studies found that obesity increases NPRC expression in both animals and humans,^6,7,14 we next examined NPRC levels in retroperitoneal WAT. As shown in Fig 8, B and C, NPRC expression was increased in both groups of BTBR-ob/ob mice and this difference was statistically significant for vehicle treated BTBR-ob/ob mice. In contrast, we did not find an effect of obesity or ANP(4-23) on expression of neprilysin protein levels (Fig 8, D and E).

Fig 8.

Urinary cGMP excretion and expression of NPRC and neprilysin in WAT. (A) ANP(4-23) significantly increased urinary cGMP excretion in WT BTBR mice. A similar pattern of cGMP excretion was observed in BTBR-ob/ob mice but the amount of cGMP excreted was significantly lower in both groups of BTBR-ob/ob mice compared to WT BTBR mice. (B and C) NPRC expression was increased in both groups of BTBR-ob/ob mice compared to WT BTBR mice, and the difference was statistically significant for the vehicle-treated BTBR-ob/ob mice. (D and E) Neprilysin protein levels were similar in WT and BTBR-ob/ob mice, and were not changed by treatment with ANP(4-23). Data were analyzed by an ANOVA followed by Sidak’s multiple comparisons test. Note: the y-axis of Panel A is logarithmic.

DISCUSSION

We tested a pharmacologic approach to enhance cGMP generation in adipose tissue by inhibiting NP clearance. We found that pharmacologic blockade of NPRC reduced both body weight and hyperinsulinemia in BTBR-ob/ob mice. The decrease in body weight was mediated by reducing fat mass without altering lean body mass. Indirect calorimetry suggested that the decrease in fat mass was caused, at least in part, by enhanced energy expenditure. These findings were associated with increased phosphorylation of HSL on an activating site in BAT and increased PGC1-α protein levels, a reduction in MCP1 protein in serum, and a decrease in expression of both MCP1 mRNA and the macrophage marker F4/80 in WAT. Taken together, these data suggest that pharmacologic blockade of NPRC may be a useful therapeutic approach for promoting weight loss and enhancing insulin sensitivity in type 2 diabetes.

NPs promote weight loss by enhancing cGMP generation and activating protein kinase G (PKG) which, in turn, stimulates lipolysis and induces a thermogenic program in WAT that resembles brown fat.^5–7 The effects of PKG are mediated by: (1) Phosphorylating HSL on an activating site, which increases lipolysis, and (2) Inducing expression of both UCP1 and PGC1-α,^5–7 which increases uncoupled oxidative phosphorylation and enhances oxidative capacity.^42,43 This “browning” of WAT increases fatty acid oxidation and decreases adiposity.⁵ In the present study, we found that treatment with ANP(4-23) enhanced energy expenditure in BTBR mice and promoted phosphorylation of HSL in BAT. These effects were associated with enhanced expression of PGC1-α protein levels. In contrast, we were unable to demonstrate an effect ANP (4-23) on expression of UCP1 in BAT. Moreover, we did not detect changes in phospho-HSL levels or expression of either PGC1-α or UCP1 in retroperitoneal WAT. One explanation for these findings is that the browning capacity of retroperitoneal WAT is reduced compared to other adipose depots.^44–46 It is also possible the browning effects of ANP(4-23) had dissipated by the time of harvest due to the rapid degradation of ANP(4-23) in vivo.²⁷ In addition, expression of NPRC was increased in BTBR-ob/ob mice (Fig 8), which may have enhanced clearance of NPs from the circulation and, in turn, reduced the effectiveness of the treatment approach. In support of the latter possibility: (1) Urinary cGMP excretion was decreased in BTBR-ob/ob mice compared WT BTBR mice by an order of magnitude, and (2) Treatment with ANP(4-23) did not significantly increase urinary cGMP excretion in BTBR-ob/ob mice by the end of the study (see Fig 8). We speculate that optimizing the treatment approach may further enhance the benefits of this therapeutic strategy in future studies.

A recent manuscript by Hobbs and coworkers⁴⁷ provides an alternative explanation for the modest effects of ANP(4-23) on phospho-HSL, UCP1, and PGC1-α expression in WAT. In this regard, a large literature suggests that NPRC acts both as a clearance and signaling receptor.^47,48 In this model, NPRC clears ANP, BNP, and CNP from the circulation, but binding of CNP to NPRC uniquely promotes intracellular signaling.^47,48 These investigators suggest that CNP has dual effects on adipogenesis and thermogenesis that includes: (1) Inducing a thermogenic program in adipose tissue by stimulating NPRB and promoting cGMP generation, and (2) Stimulating Gi signaling by NPRC, which inhibits cAMP generation and limits the thermogenic actions of the β-adrenergic system in adipose tissue. These observations are relevant to the current study because ANP(4-23) is an NPRC agonist.^25,47 As a result, treatment with ANP(4-23): (1) Blocks NP clearance, enhances circulating NP levels, and stimulates cGMP signaling in adipocytes, and (2) Stimulates Gi signaling by binding to NPRC and inhibits cAMP generation by the β-adrenergic system generation. In this scenario, the beneficial effects of ANP(4-23) on obesity may be tempered by inhibiting the thermogenic actions of the β-adrenergic system.

In addition to affecting energy metabolism, NPs may have other actions that promote weight loss. NPs suppress appetite and promote satiety in humans, perhaps by increasing expression of the gut-derived hormone ghrelin.⁴⁹ In the present study, cumulative daily food intake was decreased in BTBR-ob/ob mice treated with ANP(4-23) compared to vehicle treated BTBR-ob/ob mice (Fig 1, A and B), but this difference was not statistically significant. Another possibility is that treatment with ANP(4-23) increased the level of physical activity in BTBR-ob/ob mice and contributed to enhanced energy generation in this treatment group. Unfortunately, these data were not collected by our indirect calorimetry system. However, the relationship between energy expenditure and activity is complex.⁵⁰ At thermoneutral temperatures (approximately 30°C), mice expend a significant amount of energy on activity.⁵⁰ In contrast, energy expenditure from activity does not make a significant contribution to total daily energy expenditure at temperatures below thermoneutrality.⁵⁰ Consequently, we think it is unlikely that increased activity significantly affected the results, given that the studies were performed at 72°F (22.2°C). Lastly, both ANP and CNP have effects on pancreatic and bile secretions as well as decrease salt and water absorption in the intestine,⁵¹ which might affect the availability of nutrients for energy generation.

NPRC blockade also had beneficial effects on the inflammatory response in WAT of obese mice. Adiposity stimulates accumulation of macrophages in WAT that secrete inflammatory cytokines and contribute to insulin resistance.³⁸ MCP1 is important cytokine implicated in macrophage accumulation in adipose tissues.³⁸ This cytokine is secreted by both adipocytes and macrophages, and its expression is enhanced in obesity.³⁸ We found that treatment with ANP(4-23) reduced the high levels of MCP1 mRNA in WAT of BTBR-ob/ob mice. Similarly, MCP1 protein levels were enhanced in serum of BTBR-ob/ob mice and treatment with ANP(4-23) significantly attenuated the circulating levels of this cytokine (Fig 6). The reduction in MCP1 levels was associated with a significant decrease in the macrophage marker F4/80 in WAT. These data suggest that enhancing cGMP in adipose tissue has anti-inflammatory effects. In support of this hypothesis, cGMP signaling inhibits multiple pathways implicated in inflammatory responses including TGF-β, Rho GTPases, and calcium sensitive signaling.¹⁵ These pathways play important roles in inflammation by modulating the immune cell migration, fibrosis, and the severity of immune response.^52–55 Taken together, these data suggest that the anti-inflammatory actions of the NPRC blockade may play an important role in the beneficial effects of this treatment approach.

In addition to effects on BTBR-ob/ob mice, pharmacologic NPRC blockade had beneficial actions on insulin sensitivity in WT BTBR mice (Fig 5, B and D). These animals are insulin resistant,⁵⁶ and pharmacologic blockade of NPRC improved insulin sensitivity without a change in body weight, fat mass, or lean body mass (Fig 1). Consistent with these findings, ANCOVA analyses found that treatment with ANP(4-23) significantly improved insulin levels and insulin sensitivity after controlling for body weight as a covariant. In support of these observations, higher NP levels are associated with improved insulin sensitivity in both animal and human studies.^{6,10,11,57,58} For example, genetic variants that promote enhanced BNP levels are associated with a lower risk of type 2 diabetes.^6,57,58 Similarly, examination of data from the Framingham Heart Study and the Malmö Diet and Cancer Study found that lower levels of NPs were associated with increased susceptibility to insulin resistance in both obese and nonobese individuals.^6,11 Moreover, genetically modified mice expressing high levels of BNP are protected from insulin resistance when fed a high fat diet.^6,10 While we can only speculate about the mechanism(s), the findings in the present studies suggest that enhancing NP levels may improve insulin sensitivity independent of obesity.

While the treatment approach improved insulin sensitivity, we did not detect an effect of ANP(4-23) on either fasting glucose levels or HbA1c levels (Fig 4) despite an improvement in insulin sensitivity. We suspect there were several confounding factors, which may have contributed to these observations. First, fasting glucose levels were above the upper limits of detection for our glucometer in BTBR-ob/ob mice, which reduced our ability to detect a reduction in blood glucose levels, even if a difference was present. Second, we did not detect an effect on HbA1c levels at the end of the study. Given that HbA1c levels reflect average glucose levels over several weeks, it is possible that the lack of an effect on HbA1c levels was due to an insufficient time for improved insulin sensitivity to improve HbA1c levels.

Lastly, NPs are primarily known for their renal and cardiovascular actions. In this regard, NPs promote vasorelaxation and potently stimulate natriuresis.⁶ In addition, a recent landmark study suggested that augmenting the effects of NPs protects patients with heart failure from death and hospitalization.⁵⁹ NPs also have direct renal protective actions in animal models of kidney disease.^60,61 Moreover, potentiating the effects of NPs reduces blood pressure⁶² and proteinuria⁶³ in humans with chronic kidney diseases (CKD). Thus, pharmacologic blockade of NPRC may favorably impact multiple systemic disease processes by potentiating the effects of NPs.

In conclusion, we found that pharmacologic blockade of NPRC: (1) Reduced body weight by decreasing fat mass without a change in lean body mass, and (2) Improved insulin sensitivity in type 2 diabetes. The goal of these experiments was to establish proof-of-concept and provide the rationale for further developing effective, degradation resistant inhibitors of NP clearance that might further enhance the therapeutic benefits of pharmacologic NPRC blockade. Based on our findings, we posit that further development of the therapeutic approach may lead to useful therapies for obesity and type 2 diabetes.

At A Glance Commentary.

Wang L, et al.

Background

Natriuretic peptides (NPs) stimulate lipolysis in adipocytes and induce a thermogenic program in white adipose tissue that resembles brown fat. The biologic effects of NPs are negatively regulated by the NP clearance receptor (NPRC).

Translational Significance

Pharmacologic blockade of NPRC was used to inhibit NP clearance. Inhibition of NP clearance decreased body weight by increasing energy expenditure and reducing fat mass without a change in lean body mass. These beneficial effects improved insulin sensitivity and reduced serum insulin levels. Pharmacologic blockade of NPRC may be an effective approach to promote weight loss and enhance insulin sensitivity.

Abbreviations:

ANCOVA

analysis of covariance

ANP

atrial NP

BAT

brown adipose tissue

BNP

brain NP

CKD

chronic kidney diseases

CLAMS

Comprehensive LAb Monitoring System

CNP

C-type natriuretic peptide

HbA1c

hemoglobin A1c

HSL

hormone sensitive lipase

ITT

insulin tolerance test

KO

knockout

MCP1

monocyte chemotactic protein-1

MRI

magnetic resonance imaging

NPRC

NP clearance receptor

NPRs

NP receptors

NPs

natriuretic peptides

PBS

phosphate buffered saline

PDEs

phosphodiesterases

RER

respiratory exchange ratio

UCP1

uncoupling protein 1

US

United States

VCO₂

carbon dioxide release

VO₂

oxygen consumption

WAT

white adipose tissue

1-α

peroxisome proliferator-activated receptor gamma coactivator