Persistent organic pollutants (POPs) increase rage signaling to promote downstream cardiovascular remodeling

Abstract Exposure to environmental contaminants and consumption of a high, saturated fatty diet has been demonstrated to promote precursors for metabolic syndrome (hyperglycemia, hyperinsulinemia, and hypertriglyceridemia). The purpose of this study was to determine if exposure to the most prevalent environmental persistent organic pollutants (POPs) would act as causative agents to promote metabolic syndrome independent of dietary intake. We hypothesized that POPs will activate the advanced glycated end‐product (AGE)‐and receptor for AGE (RAGE) signaling cascade to promote downstream signaling modulators of cardiovascular remodeling and oxidative stress in the heart. At 5‐weeks of age nondiabetic (WT) and diabetic (ob/ob) mice were exposed POPs mixtures by oral gavage twice a week for 6‐weeks. At the end of 6‐weeks, animals were sacrificed and the hearts were taken for biochemical analysis. Increased activation of the AGE‐RAGE signaling cascade via POPs exposure resulted in elevated levels of fibroblast differentiation (α‐smooth muscle actin) and RAGE expression indicated maladaptive cardiac remodeling. Conversely, the observed decreased superoxide dismutase‐1 and ‐2 (SOD‐1 and SOD‐2) expression may exacerbate the adverse changes occurring as a result of POPs treatment to reduce innate cardioprotective mechanisms. In comparison, ventricular collagen levels were decreased in mice exposed to POPs. In conclusion, exposure to organic environmental pollutants may intensify oxidative and inflammatory stressors to overwhelm protective mechanisms allowing for adverse cardiac remodeling.

POPs upwards in the food chain. 2 Exposure to POPs occurs primarily through the consumption of food, especially meat, fish, and dairy products, containing environmental pollutants. Interestingly, POPs also have the capability to travel long-range distances via their natural tendency to enter the gas phase under environmental temperatures. 3 Therefore, POPs found in food do not always come from the pesticides used in the vicinity of industries located near food-producing farms. 3 Epidemiological studies indicate exposure to certain organochlorine (OC) pesticides is positively associated with the development of insulin resistance, diabetes, and the metabolic syndrome. 4,5 Animal studies have revealed that exposure to POPs in the presence of fatty fish oil led to the development of insulin resistance, abdominal obesity, and hepatosteatosis. 6,7 The analysis of data from the National Health and Nutrition Examination Survey (NHANES) from 1999 to 2002 indicated that diabetic prevalence in individuals was strongly positively associated with concentrations of POPs serum concentrations, with OC pesticides being most strongly and consistently associated with the metabolic syndrome. 8,9 Though the underlying mechanism(s) of action governing the effects of POPs exposure remains poorly understood, these studies have shown that POPs, especially the OC pesticides or their metabolites as well as polychlorinated biphenyls (PCBs), are positively correlated with the susceptibility to and onset of T2DM. [7][8][9][10] Additionally, recent epidemiological studies have indicated that elevated levels of POPs are positively associated with alterations in left ventricle geometry and function in the elderly (PMID 23743808 and PMID 23562393). 11,12 Thus, the possibility arises that elevated levels of POPs may alter cardiac remodeling and function.
Increased circulating levels of glucose, as observed in diabetic patients, will react nonenzymatically with amino groups at the carbonyl ends of proteins to form reversible Schiff bases and then Amadori compounds through a process known as the Maillard reaction. 13 The resulting product is an irreversibly crosslinked derivative termed advanced glycation end products (AGEs). 14 Other pathways for AGE production include autoxidation of glucose from an increase in oxidative stress, and the conversion of glucose to sorbitol and then to fructose, known as the polyol pathway. [15][16][17] Evidence also suggests that AGEs form during natural aging because of exposure of long-lived proteins to euglycemic levels. 14,[18][19][20][21] Due to the diverse pathways leading to AGE genesis, AGE chemical structures tend to vary; however, accumulation of assorted AGEs will alter cellular and intracellular structures as well as function through activation of the multi-ligand receptor for AGEs or RAGEs. RAGEs are transmembrane cell surface receptors whose cytoplasmic domain is key for downstream signaling. 14 The AGE-RAGE signaling pathway can trigger multiple cellular signaling pathways in which various downstream signaling pathways, such as Ras-extracellular signal regulated protein kinase (ERK1/2), Jak/Stat, and the transcription regulator nuclear factor-κB (NF-κB) are involved in the generation of reactive oxygen species (ROS). [22][23][24] In a previous a study by Mulligan et

al, the experimental design and
POPs mixture exposure were reported to successfully promote nonalcoholic fatty liver disease (NAFLD) hepatic steatosis in ob/ob mice. 24,25 The purpose of this study was to determine if the POPs exposure in the same animals would act as causative agents to promote metabolic syndrome outcomes in the heart independent of dietary intake. We hypothesized that the AGE-RAGE signaling cascade will be activated by POPs exposure to stimulate downstream signaling modulators to promote cardiovascular remodeling. Given the critical role of activation of the AGE-RAGE axis in the pathogenesis of T2DM and cardiac remodeling, these findings demonstrate that exposure to POPs promotes cardiometabolic risk and cardiac remodeling, due as least in part to POPs-induced activation of the AGE-RAGE axis. In states of metabolic dysfunction such as obesity and diabetes, ROS levels become exceedingly high, and when compounded with POPs exposure, innate antioxidant defense mechanisms become overwhelmed with increased levels of ROS from the activation of RAGE/NF-κB signaling. 24,26 The following results demonstrate a potential role for AGE-RAGE signaling in the POPs-mediated remodeling.

| Animal model and POPs mixture
In a previous a study by Mulligan et al, the experimental design and POPs mixture were reported to successfully promote NAFLD hepatic steatosis in ob/ob mice. 24 Briefly, four-week old male WT and ob/ob mice were randomly divided into treatment groups with 10 animals per group.
Treatment groups are as follows: (a) WT, (b) ob/ob, (c) WT POPs, and (d) ob/ob POPs. The POPs mixture contained the five most prevalent OC pesticide or their metabolites as well as the five most prevalent PCBs present in contaminated salmon. 7,24 The following chemicals were contained within the POPs mixture: pp-dichlorodiphenyldichloroethylene (DDE), pp-dichlorodiphenyldichloroethane (DDD), hexachlorobenzene, dieldrin, transnonachlor, PCB-153, PCB-138, PCB-118, PCB-77, and PCB-126, Table 1. A twice weekly oral gavage method was used to administer POPs treatments for 7 weeks at which time the animals were sacrificed. It should be noted that POPs were delivered in a corn oil vehicle (1 mL/kg) to mimic oil dissolved POPs, as would occur in contaminated food. The weekly intake of POPs yielded equivalent concentrations of POPs as that in previously published studies based on contaminated salmon levels and therefore reflects an environmentally relevant exposure via dietary intake. 7,10,24 Serum and adipose concentrations of POPs were not measured, as they were used as experimental endpoints for alternative studies previously reported. 24 Prior to sacrificing, mice were weighed and fasting glucose levels were recorded by glucometer. Hearts were removed, weighed, and sectioned for paraformaldehyde fixation and freezing at −80 C for biochemical assays. Serum was collected to measure fasting insulin levels which were previously published by Mulligan et al. 24

| Protein isolation
After hearts were harvested, proteins were isolated as follows. 500 μL of modified Hunter's Buffer, containing 1% Triton X-100, 75 mM NaCl, 5 mM Tris (pH 7.4), 0.5 mM orthovanadate, 0.5 mM EDTA, 0.5 mM EGTA, 0.25% NP-40 and protease inhibitors, was added to 50 mg of heart tissue. On ice, tissue was minced and then sonicated for 5 second bursts until it was disrupted. Lysates were centrifuged for 15 minutes at 20 000g at 4 C. The supernatant was transferred to new 1.5 mL centrifuge tubes and stored at −80 C. Protein concentration for each sample was determined using the bicinchoninic acid (BCA) assay (Pierce Biotechnology) according to manufacturer's instructions. Blots were developed using Pierce enhanced chemiluminescence (ECL) solution (Pierce Biotechnology), exposed to X-ray film, and immunoreactive bands were quantified using NIH Image J. Measured densitometric values were normalized to their respective loading controls (ie, GAPDH, total NF-κB) and then normalized to WT-Sham.

| Western blot protein analysis
Changes in protein expression are described as being a fold increase or fold decrease of Sham. Representative blots shown in Results section may include membranes that have been probed, stripped, and reprobed multiple times. For these images, it will be denoted in the figure legends.

| Histological analysis
Morphological evaluation of hematoxylin and eosin (H&E) stained slides was performed on hearts from WT, ob/ob, WT + POPs and ob/ ob + POPs mice for neutrophil infiltration into the myocardium.
Briefly, hearts were fixed in 10% buffered formalin and embedded in paraffin blocks. Blocks were sectioned at 5-μm thickness and stained with H&E, and estimation of neutrophils was obtained by using bright field microscopy.

| Collagen content
Myocardial collagen concentration and solubility was determined as described. 27,28 In brief, 50 mg of cardiac tissue was mixed with 3 mL water and frozen at −80 C until Sircol analysis. Sircol collagen assay (Biocolor, Ltd., Ireland) was performed according to manufacturer's instructions. The conditions for pepsin digestion were chosen, so that significant solubilization of the unmodified or soluble collagen was identified at 2 hours and 24 hours, which maximum solubilization (>98%) was observed. 27,28 Insoluble collagen content was determined by measuring collagen concentration after 18 hours acid hydrolysis using the procedure of Stegemann and Stalder. 27 Collagen content was expressed as μg of collagen per mg of wet weight of heart tissue.
Morphological evaluation of picric acid-sirius red stained collagen was performed on hearts from WT, ob/ob, WT + POPs and ob/ob + POPs mice as previously described. 29,30 Briefly, hearts were fixed in 10% buffered formalin and embedded in paraffin blocks. Blocks were sectioned at 5-μm thickness and stained with Picric acid-sirius red F3BA. Estimates of the fractions of collagen fibrils were obtained by using polarized light. Due to the birefringent quality of the stain, collagen refracted a distinct color based upon the size of the collagen fibrils: red and yellow (thick filaments) and green (thin filaments).
Quantitative analysis is accomplished by light microscopy with a video-based image-analyzer system. Color thresholds were set for biphasic analysis to capture and generate a percent collagen content per 20× field within the specified RGB wavelength ranges separate T A B L E 1 Administered persistent organic pollutants (POPs). Persistent organic pollutants (POPs) and their administered concentrations were recorded in the

| Statistical analysis
For morphometric data analysis, western blot protein analysis, and collagen content analysis, a one-way ANOVA performed with Tukey's post hoc analysis to examine differences between WT, WT-POPs, ob/ ob, and ob/ob-POPs mice. Statistical analysis was performed using GraphPad Prism 5 software. Statistical differences are defined as data having P < .05. Error bars represent ± SE of the mean (SEM).

| Physiological and biochemical markers of obesity and T2DM resulting from POPs exposure and genotype
The data presented herein were gathered using heart tissue samples from identical animals as those used by Mulligan et al. 24 Briefly, ob/ob POPs and ob/ob mice body weights were significantly heavier than that of their WT littermates indicating obesity. 24 Additionally, ob/ob POPs and ob/ob mice had higher fasting blood glucose and insulin levels. 24 When hearts were examined for changes in morphology, such as hypertrophy or dilatation, there were no differences in heart weight, Figure 1A. Significant differences were observed in the ventricular to body weight ratios, Figure 1B. Diabetic ob/ob mice had significantly smaller ventricular to body weight ratios than their lean WT littermates.
These findings indicate a larger body weight or obese phenotype observed in the ob/ob mice. Most notably Mulligan et al found significant increases in fasting triglyceride and total cholesterol in ob/ob mice compared to WT; however, no differences were observed between untreated and POPs treated animals. 24 There was also marked triglyceride accumulation in the liver indicating increased hepatic steatosis. 24 Yet, these changes were significantly higher in ob/ob POPs than ob/ob groups. From this prior study, it was found that POPs treatment signifi- F I G U R E 1 A, Ventricular weight. Hearts were examined for changes in morphology, such as hypertrophy, there was no differences in heart weight between wildtype (WT) and diabetic (ob/ob) hearts. POPs treatment did not alter heart weight. n = 10-11 hearts per group. B, Ventricular weight to body weight ratio. Significant differences were observed in the ventricular weight to body weight ratios in diabetic ob/ob mice. There was a significant decrease in ratios in diabetic as compared to their lean wildtype (WT) littermates. This finding indicates a larger body weight or obese phenotype observed in the ob/ob mice that was unaffected by persistent organic pollutant (POPs) exposure. n = 10-11 hearts per group. A one-way ANOVA with Tukey's post hoc analysis was performed to examine differences between WT, WT-POPs, ob/ob, and ob/ob-POPs mice. Statistical differences are denoted on the graph. Error bars represent ± SE of the mean (SEM)

| α-Smooth muscle actin (SMA)
Previously published studies by our laboratory demonstrated a role for AGE-RAGE signaling in fibroblast stress responses. 30 Of note, increased α-smooth muscle actin (SMA) protein expression has been shown to be present in diabetic fibroblast undergoing fibroblast-to-myofibroblast differentiation in response to elevated AGE-RAGE signaling. 30 Western blot analysis was used to determine changes in fibroblast differentiation in the ventricular samples. α-SMA protein expression was found to be elevated in ob/ob and in POPs treated WT and ob/ob animals compared to hearts of WT mice, Figure 3. Interestingly, WT-POPs and ob/ob α-SMA levels were not significantly different. Thus, indicating a potential rise in the stressed fibroblast phenotype within the hearts of WT-POPs mice independent of circulating glucose levels. These changes may be attributed to increases in AGE-RAGE signaling.

| Phospho-NF-κB
In addition to stimulating fibroblast differentiation, AGE-RAGE cascade activation will also alter a number of proinflammatory modulators, such as NF-κB. 32 Western blot analysis showed slightly elevated levels of phospho-NF-κB, the active form of NF-κB, demonstrated to increase levels of oxidative stress, in ob/ob and WT-POPs animals as compared to WT mice; however, these changes were not significant, Figure 4. Furthermore, POPs exposure in ob/ob hearts significantly decreased NF-κB phosphorylation levels below those measured in ob/ ob and WT-POPs mouse ventricular samples.

| SOD-1 and SOD-2 levels
Superoxide dismutase-1 and -2 (SOD-1 and SOD-2) are generally ubiquitously expressed, innate antioxidants used to neutralize superoxide radicals in the body. 19 In this study SOD-1 and -2 protein expression levels were measured to determine how efficiently POPs treated animals were able to compensate for increased oxidative stressors. POPs exposure in ob/ob mice significantly decreased both SOD-1 and SOD-2, Figure 5A,B, respectively, as compared to WT, ob/ob, WT-POPs groups. We predict that decreased SOD-1 and -2 levels in ob/ob-POPs mice correlated with their inability to compensate with rising levels of oxidative stressors in the hearts. These data may suggest the ob/ob-POPs mice have lost the ability to neutralize superoxide radicals.
F I G U R E 2 RAGE protein expression. RAGE protein expression was significantly increased in WT-POPs, ob/ob, and ob/ob-POPs treated hearts as compared to WT. n = 10-11 hearts per group. Representative blots shown may include membranes that have been probed, stripped, and reprobed multiple times. A one-way ANOVA with Tukey's post hoc analysis was performed to examine differences between WT, WT-POPs, ob/ob, and ob/ob-POPs mice. Statistical differences are denoted on the graph. Error bars represent ± SE of the mean (SEM) Representative blots shown may include membranes that have been probed, stripped, and reprobed multiple times. A one-way ANOVA with Tukey's post hoc analysis was performed to examine differences between WT, WT-POPs, ob/ob, and ob/ob-POPs mice. Statistical differences are denoted on the graph. Error bars represent ± SE of the mean (SEM)

| Histological analysis
Morphological evaluation of H&E was performed on hearts from WT, ob/ob, WT + POPs and ob/ob + POPs mice. Neutrophil infiltration into the myocardium was estimated. There were no significant differences in neutrophil population across the study groups ( Figure 6).

| Collagen content
Collagen content was measured to determine downstream effectors of the AGE-RAGE signaling cascade. Ventricular fibrosis has been demonstrated to be a significant outcome of increased RAGE activation. 30 POPs exposure in both WT and ob/ob mouse hearts had significantly decreased soluble and insoluble collagen levels, Figure 7A,B. As a result, solubleinsoluble ratios were also decreased, Figure 7C. Collectively, these results can indicate low collagen production, increased collagen degradation by matrix metalloproteases (MMPs), as well as decreased collagen crosslinking, which can impact cardiac performance. These findings were also observed in picric acid-sirius red stained ventricular cross-sections; however, the percent of stained collagen was significantly decreased in ob/ob-POPs hearts ( Figure 8).

| DISCUSSION
From previous studies, increased POPs exposure exacerbated hepatic steatosis and adipocyte dysfunction resulting in elevated production of proinflammatory and oxidative stressors. 24 These changes in the inflammatory profile have been demonstrated to exacerbate type 2 diabetes mellitus (T2DM); however, the mechanism regulating this process is unknown. 10  The AGE-RAGE signaling cascade activation has been shown to increase proinflammatory and oxidative stress modulators, such as NF-κB, SOD-1, and SOD-2, to adapt to changes in inflammation and ROS levels. 19,26 NF-κB protein expression was shown to be significantly diminished in ob/ob-POPs mice. Decreases in NF-κB levels oftentimes coincide with loss of protective mechanisms against apoptosis. 37 This finding may indicate an inability of the ob/ob-POPs animals to sustain a sufficient inflammatory response to POPs exposure.
F I G U R E 4 Phospho-NF-κB/total-NF-κB protein expression. Phospho-NF-κB (active NF-κB) as shown by western blot analysis was found to be significantly decreased in ob/ob-POPs hearts as compared to ob/ob and WT-POPs hearts. n = 10-11 hearts per group. Representative blots shown may include membranes that have been probed, stripped, and reprobed multiple times. A one-way ANOVA with Tukey's post hoc analysis was performed to examine differences between WT, WT-POPs, ob/ob, and ob/ob-POPs mice. Statistical differences are denoted on the graph. Error bars represent ± SE of the mean (SEM) Further studies would need to be performed in order to confirm significance of these findings. In the current study, protein expression levels of two key neutralizers of ROS, SOD-1 and SOD-2, were measured to determine the compensatory ability of animals to adapt to POPs as potential increased oxidative stressors. 24  While prior studies have used neutrophil population as a predictor for changes in SOD levels, our findings suggest changes in neutrophil numbers may have either occurred early in POPs exposure timeframe or depleted over time with chronic exposure culminating in decreased SOD levels. 38,39 Future studies will be performed using known positive control agents, such as doxorubicin. Furthermore, additional studies will need to be performed in order to understand the exact mechanism.
In states of metabolic dysfunction such as obesity and diabetes, ROS levels become exceedingly high, and when compounded with POPs Ventricular fibrosis has been demonstrated to be a significant outcome of increased RAGE activation and myofibroblast differentiation. 30 We found both WT-POPs and ob/ob-POPs mouse hearts had F I G U R E S 7 A, Soluble collagen. Soluble collagen was significantly decreased in WT-POPs and ob/ob-POPs compared to WT and ob/ob mice hearts. n = 10-11 hearts per group. B, Insoluble collagen. Insoluble collagen was significantly decreased in WT-POPs and ob/ob-POPs compared to WT and ob/ob mice hearts. n = 10-11 hearts per group. C, Soluble:Insoluble ratio. Soluble:Insoluble collagen was significantly decreased in WT-POPs and ob/ob-POPs compared to WT and ob/ob mice hearts. n = 10-11 hearts per group. A one-way ANOVA with Tukey's post hoc analysis was performed to examine differences between WT, WT-POPs, ob/ob, and ob/ob-POPs mice. Statistical differences are denoted on the graph. Error bars represent ± SE of the mean (SEM) significantly decreased soluble and insoluble collagen levels as well as decreased soluble and insoluble collagen levels, and consequently soluble-insoluble ratios were also significantly decreased in the aforementioned groups. Changes in soluble levels are indicative of changes in collagen production and/or collagen degradation by MMPs. 29,30 While changes in insoluble collagen represent alterations in collagen maturation and crosslinking leaving collagen strands susceptible to degradative enzymes. 40 Both, of which, can directly impact cardiac performance. 29,30,40 Similar findings were also observed in picric acidsirius red stained ventricular cross-sections ( Figure 7); however, the percent of stained collagen was only significantly decreased in ob/ob-POPs hearts. Without cardiac functional analysis, it is difficult to speculate on significance of findings. One implication of these findings may be attributed to the loss of NF-κB transcriptional regulation affects collagen synthesis. [41][42][43] Alternatively, loss of SOD-1 and -2 in ob/ob-POPs could result in a potential elevation in ROS levels resulting in elevated MMP activity, and consequently, increased collagen degradation. [44][45][46] Of note, collagen levels were not significantly elevated in ob/ob hearts, which leptin has been noted to play a role in fibrosis. [47][48][49] In summary, this study implicated organic pollutants (POPs) as potential promoters of RAGE protein expression. Our findings demonstrate that exposure to POPs may increase cardiometabolic risk and cardiac remodeling, due in part to POPs-induced activation of the AGE-RAGE signaling cascade. Results also showed that the POPsinduced decreases in NF-κB, SOD-1, and SOD-2 to potentially dampen cardioprotective mechanisms needed to offset POPsmediated oxidative and inflammatory stressors in the ob/ob-POPs mice. The mechanisms by which this occurs is currently not F I G U R E 8 A-E, Picric acid-sirius red stained ventricles. Stained collagen was visualized using polarized light (left panel) as well as brightfield (right panel) microscopy. A, WT representative images. B, ob/ob representative images. C, WT-POPs representative images. D, ob/ob-POPs representative images. E, Stained collagen was significantly decreased in ob/ob-POPs compared to WT hearts. n = 5-6 hearts per group. A oneway ANOVA with Tukey's post hoc analysis was performed to examine differences between WT, WT-POPs, ob/ob, and ob/ob-POPs mice. Statistical differences are denoted on the graph. Error bars represent ± SE of the mean (SEM). Scale bars equals 100 μm understood. Therefore, the combination of these results with previous studies highlight the critical role of AGE-RAGE cascade activation in the pathogenesis of T2DM and implicate this pathway in environmental perturbation of the cardiac ramifications of T2DM. The positive correlation between POPs exposure and onset of T2DM give rise to the need for future studies analyzing the potential mechanisms connecting POPs exposure to increased RAGE expression and subsequent adverse cardiac outcomes.