1

1 2

Ablation of eNOS Does Not Promote Adipose Tissue Inflammation 3 4 5 6 Thomas J. Jurrissen1, Ryan D. Sheldon1,2, Michelle L. Gastecki1, Makenzie L. 7 Woodford1, Terese M. Zidon1, R. Scott Rector1,2,3, Victoria J. Vieira-Potter1, Jaume 8 Padilla1,4,5 9 10 1Nutrition and Exercise Physiology, University of Missouri, Columbia, MO 11 2Research Service-Harry S Truman Memorial Veterans Affairs Medical Center, 12 Columbia, MO 13 3Medicine-Division of Gastroenterology and Hepatology, University of Missouri, 14 Columbia, MO 15 4Dalton Cardiovascular Research Center, University of Missouri, Columbia, MO 16 5Child Health, University of Missouri, Columbia, MO 17 18 19 20 21 22 Running Title: eNOS deficiency and adipose tissue inflammation 23 24 Key Words: Obesity, western diet, adipose tissue, endothelial nitric oxide synthase 25 26 27 28 29 30 31 32 33 Correspondence: 34 Jaume Padilla, Ph.D. 35 Department of Nutrition & Exercise Physiology 36 204 Gwynn Hall 37 University of Missouri 38 Columbia, MO 65211 39 padillaja@missouri.edu 40

41

2

ABSTRACT 42

Adipose tissue (AT) inflammation is a hallmark characteristic of obesity and an 43

important determinant of insulin resistance and cardiovascular disease; therefore, a 44

better understanding of factors regulating AT inflammation is critical. It is well 45

established that reduced vascular endothelial nitric oxide (NO) bioavailability promotes 46

arterial inflammation; however, the role of NO in modulating inflammation in AT remains 47

disputed. In the present study, ten-week old C57BL6 wild-type and eNOS knockout 48

male mice were randomized to either a control diet (10% kcal from fat) or Western diet 49

(44.9% kcal from fat, 17% sucrose, and 1% cholesterol) for 18 weeks (n=7-8/group). In 50

wild-type mice, Western diet-induced obesity led to increased visceral white AT 51

expression of inflammatory genes (e.g., MCP1, TNFα, CCL5 mRNAs) and markers of 52

macrophage infiltration (e.g., CD68, ITGAM, EMR1, CD11C mRNAs and Mac-2 53

protein), as well as reduced markers of mitochondrial content (e.g., OXPHOS complex I 54

and IV protein). Unexpectedly, these effects of Western diet on visceral white AT were 55

not accompanied by decreases in eNOS phosphorylation at Ser1177 or increases in 56

eNOS phosphorylation at Thr495. Also counter to expectations, eNOS knockout mice, 57

independent of the diet, were leaner and did not exhibit greater white or brown AT 58

inflammation compared to wild-type mice. Collectively, these findings do not support the 59

hypothesis that reduced NO production from eNOS contributes to obesity-related AT 60

inflammation. 61

62

63

3

INTRODUCTION 64

With obesity, expansion of visceral white adipose tissue (AT) is accompanied by 65

infiltration of immune cells and increased expression of inflammatory cytokines (7). A 66

better understanding of the factors regulating AT inflammation is important given the 67

cumulative evidence implicating AT inflammation as a determinant of obesity-related 68

insulin resistance and cardiovascular disease (3, 14). While it is well-known that 69

reduced vascular endothelial nitric oxide (NO) bioavailability, a hallmark characteristic of 70

obesity and type 2 diabetes, contributes to arterial inflammation (4, 6, 9, 10, 15), 71

conflicting evidence exists regarding the role of NO in modulating inflammation in AT. 72

One study revealed that male eNOS knockout mice exhibit increased expression of 73

inflammatory genes and markers of immune cell infiltration in visceral white (epididymal) 74

AT (5); however, that finding does not appear to be reinforced by data from others (12, 75

13). In a microarray analysis of visceral white AT from female eNOS knockout versus 76

wild-type mice, inflammatory genes were not reported amongst the genes with 77

significantly altered expression (13); however, since data on inflammatory genes were 78

not reported, conclusions cannot be drawn regarding whether or not those eNOS 79

knockout mice had greater or equivalent AT inflammation compared to controls. We 80

recently showed in lean and obese male rats that chronic administration of L-NAME, a 81

NOS inhibitor, did not increase markers of inflammation in AT (12) but did in the liver 82

(17); nevertheless, these data should be interpreted with the understanding that L-83

NAME inhibits production of NO from all three NOS enzymes (eNOS, iNOS, nNOS). 84

Based on these contrasting findings, we set out the present study to more 85

comprehensively interrogate the role of NO in modulating AT inflammation. We 86

4

assessed several markers of inflammation in visceral white AT of eNOS knockout and 87

wild-type male mice fed a control versus Western diet. Given that brown AT is more 88

vascularized than white AT and that eNOS is predominantly expressed in endothelial 89

cells, the impact of eNOS deficiency on inflammatory markers was also examined in the 90

interscapular brown AT depot. If, indeed, obesity-induced AT inflammation is partially 91

mediated by reduced NO signaling, then we would expect the AT of lean eNOS 92

deficient mice to phenotypically resemble that of obese mice. That is, we would 93

hypothesize that the pro-inflammatory effect of eNOS deletion would normalize 94

differences in AT inflammation between lean and obese mice. Further, we hypothesized 95

that eNOS ablation would have greater inflammatory effects in brown AT compared to 96

white AT. 97

METHODS 98

Experimental Design 99

Male C57BL6 wild-type and eNOS knockout mice (B6.129P2-Nos3tm1Unc/J) on 100

a C57BL6 background were purchased from Jackson Laboratory (Bar Harbor, ME) at 101

10 weeks of age. Mice were randomized to either normal chow (i.e., control diet) or 102

Western diet (n=7-8/group) ad libitum for 18 weeks. Control diet (3.85 kcal/g of food) 103

contained 10% kcal fat, 70% kcal carbohydrate, and 20% kcal protein, with 3.5% kcal 104

sucrose (D12110704, custom formulated; Research Diets Inc., New Brunswick, NJ). 105

Western diet (4.68 kcal/g of food) contained 44.9% kcal fat, 35.1% kcal carbohydrate, 106

and 20% kcal protein, with 1% cholesterol and 17% kcal sucrose (D09071604, custom 107

formulated; Research Diets Inc.). All mice were pair-housed under standard 108

temperature conditions (~22°C) and humidity with a light cycle from 0700 to 1900 and a 109

5

dark cycle from 1900 to 0700. At 28 weeks of age, mice were euthanized following a 5-110

hour fast and tissues were harvested for analysis. All procedures were approved in 111

advance by the University of Missouri Institutional Animal Care and Use Committee. 112

Body composition 113

The percent body fat was measured by a nuclear magnetic resonance imaging 114

whole-body composition analyzer (EchoMRI 4in1/1100; Echo Medical Systems, 115

Houston, TX). This noninvasive measure was performed on conscious mice. 116

Fasting Blood Parameters 117

Plasma glucose, cholesterol, triglycerides, alanine aminotransferase, aspartate 118

aminotransferase, and non-esterified fatty acids (NEFA) assays were performed by a 119

commercial laboratory (Comparative Clinical Pathology Services, Columbia, MO) on an 120

Olympus AU680 automated chemistry analyzer (Beckman-Coulter, Brea, CA) using 121

assays according to manufacturer’s guidelines. Plasma insulin concentrations were 122

determined using a commercially available, mouse-specific ELISA (Alpco Diagnostics, 123

Salem, NH). Plasma nitrate + nitrite (NOx) levels were measured using the 124

Nitrate/Nitrite Fluorometric Assay Kit (Cayman Chemical, #780051, Ann Arbor, MI) 125

according to manufacturer’s instructions. 126

Histological Assessments 127

Formalin-fixed epididymal white AT samples were processed through paraffin 128

embedment, sectioned at 5 µm, stained with Mac-2 antibody (CL8942AP, Cedarlane), a 129

macrophage marker. Sections were evaluated via an Olympus BX34 photomicroscope 130

(Olympus, Melville, NY) and images were taken at 10x magnification via an Olympus 131

SC30 Optical Microscope Accessory CMOS color camera. Objective quantification of 132

6

macrophage infiltration was done by determining the positive Mac-2 stained area per 133

10x field of view using Image J software (NIH public domain; National Institutes of 134

Health, Bethesda, MD). The average of six 10x fields of view was used per animal. 135

Adipocyte size was calculated based on 100 adipocytes/animal obtained from the same 136

six 10x fields as performed previously (19). Briefly, cross-sectional areas of the 137

adipocytes were obtained from perimeter tracings using Image J software. All 138

procedures were performed by an investigator who was blinded to the groups. 139

RNA Extraction and Quantitative Real-time RT-PCR 140

Epididymal white and interscapular brown AT as well as aorta samples were 141

homogenized in TRIzol solution using a tissue homogenizer (TissueLyser LT, Qiagen, 142

Valencia, CA). Total RNA was isolated according to the Qiagen’s RNeasy lipid tissue 143

protocol and assayed using a Nanodrop spectrophotometer (Thermo Scientific, 144

Wilmington, DE) to assess purity and concentration. The MU DNA Core confirmed our 145

RNA extraction produced optimal RNA integrity (all RQN’s>8.0). First-strand cDNA was 146

synthesized from total RNA using the High Capacity cDNA Reverse Transcription kit 147

(Applied Biosystems, Carlsbad, CA). Quantitative real-time PCR was performed as 148

previously described using the ABI StepOne Plus sequence detection system (Applied 149

Biosystems). Primer sequences were designed using the NCBI Primer Design tool and 150

have been previously published (19). All primers were purchased from IDT (Coralville, 151

IA). A 20 μL reaction mixture containing 10 μL iTaq UniverSYBR Green SMX (BioRad, 152

Hercules, CA) and the appropriate concentrations of gene-specific primers plus 4 μL of 153

cDNA template were loaded in a single well of a 96-well plate. All PCR reactions were 154

performed in duplicate under thermal conditions as follows: 95ºC for 10 min, followed by 155

7

40 cycles of 95ºC for 15 s and 60ºC for 45 s. A dissociation melt curve analysis was 156

performed to verify the specificity of the PCR products. GAPDH was used as house-157

keeping control gene. GAPDH cycle threshold (CT) was not different among the groups 158

of animals. mRNA expression values are presented as 2ΔCT whereby ΔCT = GAPDH CT 159

- gene of interest CT. mRNA levels were normalized to the wild-type control diet group. 160

Western blotting 161

Triton X-100 tissue lysates were used to produce Western blot-ready Laemmli 162

samples. Protein samples (10 µg/lane) were separated by SDS-PAGE, transferred to 163

polyvinylidene difluoride membranes, and probed with primary antibodies. eNOS 164

(#610296, 1:500) and phospho-specific eNOS at Ser1177 (#61239, 1:250) and Thr495 165

(#612706, 1:500) antibodies were purchased from BD Biosciences. Akt (#4691, 1:500), 166

phospho-specific Akt at Ser473 (#4060, 1:250), Mac-2 (#127335, 1:1000), iNOS 167

(#13120, 1:500) and GAPDH (#5174, 1:1000) antibodies were purchased from Cell 168

Signaling. OXPHOS (#ab110413, 1:2000) cocktail antibody was purchased from 169

Abcam. Nitrotyrosine (#5411; 1:500) antibody was purchased from Millipore. Intensity of 170

individual protein bands were quantified using FluoroChem HD2 (AlphaView, version 171

3.4.0.0), and expressed as ratio to control band GAPDH. 172

Statistical Analysis 173

A 2x2 (genotype x diet) analysis of variance (ANOVA) was used to evaluate the 174

effects of eNOS deficiency and Western diet for all dependent variables. When a 175

significant genotype by diet interaction existed, LSD post-hoc test was used for pair-176

wise comparisons. All data are presented as mean ± standard error (SE). For all 177

8

statistical tests, the alpha level was set at 0.05. All statistical analyses were performed 178

with SPSS V23.0. 179

RESULTS 180

As shown in Fig 1, independent of genotype, mice fed a Western diet had larger 181

body weights, greater adiposity and consumed more kilocalories per day compared to 182

control diet-fed mice (p<0.05). eNOS knockout mice weighted less, had reduced 183

adiposity, and consumed fewer kilocalories per week compared to wild-type mice 184

(p<0.05). Table 1 summarizes the fasting blood characteristics. Independent of 185

genotype, compared to control-fed mice, Western diet-fed mice had greater plasma total 186

cholesterol, low density lipoprotein, high density lipoprotein, alanine aminotransferase, 187

insulin, and HOMA-IR, and reduced plasma nitrite + nitrate (p<0.05). Despite being 188

leaner, eNOS knockout mice had greater plasma NEFAs compared to wild-type mice 189

(p<0.05). 190

Figure 2A illustrates representative histological images of epididymal white AT 191

stained with Mac-2, a macrophage marker, and quantification is presented in panel B. 192

Western diet-fed mice had greater Mac-2 staining in white AT than mice fed a control 193

diet (p<0.05), independent of genotype. Similarly, adipocyte size was greater in Western 194

diet-fed mice compared to control diet-fed mice (p<0.05) and smaller in eNOS knockout 195

mice compared to wild-type mice (Fig 2C, p<0.05). 196

Gene expression data in epididymal white AT and interscapular brown AT are 197

summarized in Figure 3A and B, respectively. In white AT, mRNA levels related to 198

inflammation [e.g., MCP-1, TNF-α, CCL5], immune cell infiltration [e.g., CD68, ITGAM, 199

EMR1 and CD11c], oxidative stress [e.g., p22phox and p47phox] were all elevated with 200

9

Western diet (p<0.05) and not affected by eNOS ablation (p>0.05). Adiopokine leptin 201

mRNA was increased and adiponectin mRNA was decreased with Western diet 202

(p<0.05) but not affected by loss of eNOS (p>0.05). In brown AT, mRNA levels related 203

to inflammation [MCP1, TNF-α, CCR2, and IL-6], immune cell infiltration [e.g., CD68, 204

ITGAM, and CD11c], oxidative stress [e.g., p22phox and p47phox], endoplasmic 205

reticulum stress [e.g., DDIT3], anti-clotting protein PAI-1, and the adipokine leptin were 206

all elevated with Western diet (p<0.05) and, again, not affected by eNOS abrogation 207

(p>0.05). 208

eNOS protein content in epididymal white AT was undetectable in eNOS 209

knockout mice (Fig 4, p<0.05). No differences in total eNOS, p-eNOS at Ser1177, or p-210

eNOS at Thr495 were observed with Western diet in wild-type mice (p>0.05). iNOS 211

protein content was also unaltered with Western diet or eNOS ablation (p>0.05). In 212

addition, there were no effects of diet or genotype on Akt and p-Akt at Ser473 (p>0.05). 213

Similar to findings obtained by immunohistochemistry (Fig 2) and gene expression (Fig 214

3), Mac-2 protein content was greater in Western diet-fed mice (Fig 4, p<0.05), 215

independent of genotype. Furthermore, consistent with findings in gene expression (Fig 216

3), we found a Western diet-induced increase in nitrotyrosine, a marker of oxidative 217

stress (p<0.05). Western diet reduced mitochondrial oxidative phosphorylation complex 218

I and IV protein content in both genotypes (p<0.05) and complex II protein content in 219

eNOS knockout (p<0.05). No effects of diet or genotype were observed in complexes III 220

or V (p>0.05). 221

In stark contrast with AT data (Fig 3), aortic gene expression markers of immune 222

cell infiltration [CD68, ITGAM, EMR1, and CD11c], oxidative stress [p22phox and 223

10

p47phox], endoplasmic reticulum stress [DDIT3, HSPA1A, HSP5A (p=0.072)], and anti-224

clotting protein PAI-1 were significantly upregulated in eNOS knockout compared to 225

wild-type mice (Fig 5, p<0.05, unless stated otherwise). Also, in contrast to the results 226

observed in AT, no diet effects were found in the aorta (p>0.05). 227

DISCUSSION 228

We found that increased white AT inflammation with Western diet-induced 229

obesity was not accompanied by changes in eNOS phosphorylation at Ser1177 and 230

Thr495, the two phosphorylation sites that mainly regulate eNOS activity. Furthermore, 231

ablation of eNOS did not increase AT inflammation in either white or brown AT. 232

Collectively, these findings do not support the hypothesis that reduced NO production 233

from eNOS contributes to obesity-related AT inflammation. 234

Our findings are in discordance with results from Handa et al. (5), a study also 235

using the eNOS knockout mouse model. In that previous study, the authors reported 236

that eNOS deficiency, but surprisingly not 4 weeks of high-fat diet feeding, promoted 237

inflammation in epididymal AT (5). Our results indicate that markers of inflammation in 238

white AT were upregulated by 18 weeks of Western diet-induced obesity, but not by 239

eNOS deficiency. We do not have a clear explanation for the discrepancies in results 240

between these two studies. One possibility could be the use of different strains of 241

eNOS knockout mice; however, the strain used in that previous publication (5) was not 242

reported. Another explanation could be related to the use of different diets. Indeed, in 243

Handa et al. (5) obesity was induced with a diet containing 60% kcal from fat, 20% kcal 244

from carbohydrate, and 20% kcal from protein (5.24 kcal/g of food); whereas in our 245

study the Western diet contained 44.9% kcal from fat, 35.1% kcal carbohydrate, and 246

11

20% kcal protein, with 1% cholesterol (4.68 kcal/g of food). The control diets were 247

comparable between studies in macro-nutrient composition (10% kcal from fat, 70% 248

kcal from carbohydrate, and 20% kcal from protein; 3.85 kcal/g of food) but levels of 249

sucrose were markedly different (34.5% kcal in Handa et al. vs. 3.5% kcal in our study). 250

Another important difference between study designs to highlight is that our mice were 251

given the respective diets for 18 weeks, whereas in Handa et al. study the diet 252

intervention was 4 weeks. To determine if the lack of effect of eNOS deficiency was 253

limited to white AT, we examined the same genes in interscapular brown AT, a fat depot 254

that is more vascularized. Similar to white AT, we found no evidence of increased 255

inflammation in brown AT from eNOS knockout mice. Interestingly, however, the lack of 256

inflammation could be specific to AT, as we have preliminary data suggesting that there 257

are robust differences in inflammatory markers in the liver of eNOS knockout mice 258

compared to wild-type mice (RSR, unpublished observations). As will be discussed 259

later, inflammation was also detected in vascular tissue of eNOS knockout mice. 260

The present findings are congruent with findings from a genome-wide microarray 261

analysis of visceral white AT from eNOS knockout versus wild-type female mice (13). In 262

that study, no significant differences between eNOS knockout and wild-type mice were 263

reported in regards to inflammatory genes. In support of the notion that reduced 264

production of NO does not contribute to AT inflammation, we recently found that 265

expression of inflammatory genes and markers of immune cell infiltration in white and 266

brown AT were, by and large, unchanged with chronic administration of L-NAME in 267

drinking water in both lean and obese rats (12). However, because L-NAME inhibits all 268

NOS isoforms, we were unable to exclusively test the role of eNOS as source of NO. 269

12

In the present study, the lack of effect of eNOS deficiency on white and brown AT 270

inflammation was in marked contrast with findings in the aorta where loss of eNOS 271

produced increased expression of inflammatory genes and markers of immune cell 272

infiltration. Our data expand on previous work demonstrating the atheroprotective role of 273

vascular NO. Indeed, previous research demonstrates that inhibition of NOS with L-274

NAME upregulates inflammatory and pro-oxidant genes, increases endothelial cell 275

adhesion, and promotes atherosclerotic lesions (1, 4, 6, 12). Further, there is evidence 276

that eNOS deficient mice display aberrant vascular remodeling in response to external 277

carotid artery ligation (15), and mice with eNOS/apoE double knockout reveal 278

accelerated atherosclerosis, aortic aneurysm formation, and ischemic heart disease (8, 279

9). 280

A novel finding of the present study was the observation in the aorta that eNOS 281

deficiency was distinctly associated with increased expression of endoplasmic reticulum 282

(ER) stress markers, including DDIT3, HSP5A, and HSPA1A mRNA (genes encoding 283

for CHOP, GRP78, and HSP72, respectively). ER stress has been implicated in the 284

etiology and progression of atherosclerosis (2, 11, 16, 18) and these data suggest that 285

reduced eNOS-derived NO production promotes ER stress; however, more research is 286

needed to elucidate the intracellular signaling mechanisms underlying the link between 287

reduced NO and ER stress. 288

This study’s findings should be considered in light of its limitations. First, our 289

study did not establish whether the reported changes in adipose tissue mRNA and 290

protein levels are attributable to alterations in the phenotype of adipocytes and/or 291

resident immune cells within AT. It should be noted that expression of some 292

13

inflammatory genes is markedly different when assessed in the stromal vascular 293

fractions versus whole tissue (20). As such, future research using FACS analysis is 294

needed to examine the differential inflammatory effects of obesity and eNOS ablation on 295

adipose tissue immune cells (i.e., macrophages, T-cells) versus adipocytes. Second, we 296

have attempted to measure nitrate + nitrite levels in AT homogenates as a surrogate 297

measure of NO; however, the levels were below the detection limits of the assay. Thus, 298

it remains unknown the extent to which AT NO bioavailability is affected by Western diet 299

and eNOS ablation. Our finding that markers of oxidative stress were increased with 300

Western diet suggests that NO bioavailabilty was likely reduced due to NO scavenging 301

by superoxide radicals. Third, despite eNOS deficient mice being smaller, leaner, and 302

revealing a decrease in adipocyte size, they displayed an equivalent degree of AT 303

inflammation. One interpretation of this finding is that, relative to the degree of adiposity, 304

the eNOS knockout mice are more inflamed (i.e., more inflammation per unit of adipose 305

tissue). To examine the effect of eNOS ablation independent of adiposity differences, 306

we performed a secondary analysis which involved the comparison of the 4 leanest 307

mice from the wild-type Western diet group (weight = 41.0±2.0 g; 37.9±4.4 %fat) versus 308

the 4 heaviest mice from the eNOS knockout Western diet group (weight=40.2±1.7 g; 309

38.5±1.2 %fat). We found no significant differences in any of the markers of 310

inflammation between these two subgroups (p values ranged from 0.233 to 0.934), thus 311

suggesting that the lack of genotype effect on AT inflammation may be unrelated to 312

differences in body weight or adiposity. However, to more precisely evaluate the 313

influence of adiposity, future studies should experimentally match body weight and 314

adiposity between wild-type and eNOS knockout mice using a pair-feeding paradigm. 315

14

These effects of eNOS deficiency on body weight and composition may be due to 316

central effects resulting in reduced food and caloric intake (Fig 1F). In our previous 317

chronic L-NAME treatment study in rats it was also noted that NOS inhibition diminished 318

food intake and adiposity (12). 319

In conclusion, we found that increased white AT inflammation with Western diet-320

induced obesity was not accompanied by changes in eNOS phosphorylation at Ser1177 321

and Thr495, two sites of phosphorylation that largely control eNOS activation. In 322

addition, we found no evidence of increased AT inflammation in mice with genetic 323

deletion of eNOS. Therefore, obesity and insulin resistance-associated AT inflammation 324

appears to be unrelated to eNOS-derived NO production. 325

326

ACKNOWLEDGMENTS 327

The authors thank Grace Meers, Rebecca Scroggins, Nick Fleming, Youngmin 328

Park, and T’Keaya Gaines for excellent technical assistance. 329

330

GRANTS 331

This work was supported by a University of Missouri Life Sciences Fellowship (R. 332

D. Sheldon), Veterans Health Administration grant CDA2 1BX001299 (R. S. Rector), 333

and National Institute of Health grants K01 HL-125503 and R21 DK-105368 (J. Padilla). 334

This work was completed with resources and the use of facilities at the Harry S Truman 335

Memorial Veterans Hospital. 336

337

338

15

REFERENCES 339

1. Cayatte A, Palacino J, Horten K, and Cohen R. Chronic inhibition of nitric 340 oxide production accelerates neointima formation and impairs endothelial function in 341 hypercholesterolemic rabbits. Arterioscler Thromb 14: 1994. 342 2. Chistiakov DA, Sobenin IA, Orekhov AN, and Bobryshev YV. Role of 343 endoplasmic reticulum stress in atherosclerosis and diabetic macrovascular 344 complications. BioMed research international 2014: 610140, 2014. 345 3. Ginsberg HN, and MacCallum PR. The obesity, metabolic syndrome, and type 346 2 diabetes mellitus pandemic: Part I. Increased cardiovascular disease risk and the 347 importance of atherogenic dyslipidemia in persons with the metabolic syndrome and 348 type 2 diabetes mellitus. Journal of the cardiometabolic syndrome 4: 113-119, 2009. 349 4. Gomez-Guzman M, Jimenez R, Sanchez M, Romero M, O'Valle F, Lopez-350 Sepulveda R, Quintela A, Galindo P, Zarzuelo M, Bailon E, Delpon E, Perez-351 Vizcaino F, and Duarte J. Chronic (-)-epicatechin improves vascular oxidative and 352 inflammatory status but not hypertension in chronic nitirc oxide-deficient rats. Br J Nutr 353 106: 1337-1348, 2011. 354 5. Handa P, Tateya S, Rizzo N, Cheng A, Morgan-Stevenson V, Han C, Clowes 355 A, Daum G, O'Brien K, Schwartz M, Chait A, and Kim F. Reduced vascular nitric 356 oxide-cGMP signaling contributes to adipose tissue inflammation during high-fat 357 feeding. Arterioscler Thromb Vasc Biol 31: 2827-2835, 2011. 358 6. Hossain M, Qadri S, and Liu L. Inhibition of nitric oxide synthesis enhances 359 leukocyte rolling and adhesion in human microvasculature. J Inflamm 9: 28, 2012. 360 7. Jacobi D, Stanya KJ, and Lee CH. Adipose tissue signaling by nuclear 361 receptors in metabolic complications of obesity. Adipocyte 1: 4-12, 2012. 362 8. Kuhlencordt P, Gyurko R, Han F, Scherrer-Crosbie M, Aretz T, Hajjar R, 363 Picard M, and Huang P. Accelerated atherosclerosis, aortic aneurysm formation, and 364 ischemic heart disease in apolipoprotein E/endothelial nitric oxide synthase double-365 knockout mice. Circulation 104: 2001. 366 9. Kuhlencordt P, Padmapriya P, Rutzel S, Schodel J, Hu K, Schafer A, Huang 367 P, Ertl G, and Bauersachs J. Ezetimibe potently reduces vascular inflammation and 368 arteriosclerosis in eNOS-deficient ApoE ko mice. Atherosclerosis 202: 48-57, 2009. 369 10. Laroux F, Pavlick K, Hines I, Kawachi S, Harada H, Bharwani S, Hoffman J, 370 and Grisham M. Role of nitric oxide in inflammation. Acta Physiol Scand 173: 113-118, 371 2001. 372 11. McAlpine CS, and Werstuck GH. The development and progression of 373 atherosclerosis: evidence supporting a role for endoplasmic reticulum (ER) stress 374 signaling. Cardiovascular & hematological disorders drug targets 13: 158-164, 2013. 375 12. Padilla J, Jenkins NT, Thorne PK, Lansford KA, Fleming NJ, Bayless DS, 376 Sheldon RD, Rector RS, and Laughlin MH. Differential regulation of adipose tissue 377 and vascular inflammatory gene expression by chronic systemic inhibition of NOS in 378 lean and obese rats. Physiological reports 2: e00225, 2014. 379 13. Razny U, Kiec-Wilk B, Polus A, Wator L, Dyduch G, Partyka L, Bodzioch M, 380 Tomaszewska R, and Wybranska I. The adipose tissue gene expression in mice with 381 different nitric oxide availability. J Physiol Pharmacol 61: 607-618, 2010. 382

16

14. Reaven GM. Insulin resistance: the link between obesity and cardiovascular 383 disease. The Medical clinics of North America 95: 875-892, 2011. 384 15. Rudic R, Shesely E, Maeda N, Smithies O, Segal S, and Sessa W. Direct 385 evidence for the importance of endothelium-derived nitric oxide in vascular remodeling. 386 J Clin Invest 101: 731-736, 1998. 387 16. Scull CM, and Tabas I. Mechanisms of ER stress-induced apoptosis in 388 atherosclerosis. Arterioscler Thromb Vasc Biol 31: 2792-2797, 2011. 389 17. Sheldon RD, Padilla J, Jenkins NT, Laughlin MH, and Rector RS. Chronic 390 NOS inhibition accelerates NAFLD progression in an obese rat model. American journal 391 of physiology Gastrointestinal and liver physiology 308: G540-549, 2015. 392 18. Tabas I. The Role of Endoplasmic Reticulum Stress in the Progression of 393 Atherosclerosis. Circulation research 107: 839-850, 2010. 394 19. Wainright KS, Fleming NJ, Rowles JL, Welly RJ, Zidon TM, Park YM, Gaines 395 TL, Scroggins RJ, Anderson-Baucum EK, Hasty AH, Vieira-Potter VJ, and Padilla 396 J. Retention of sedentary obese visceral white adipose tissue phenotype with 397 intermittent physical activity despite reduced adiposity. American journal of physiology 398 Regulatory, integrative and comparative physiology 309: R594-602, 2015. 399 20. Xu H, Barnes GT, Yang Q, Tan G, Yang D, Chou CJ, Sole J, Nichols A, Ross 400 JS, Tartaglia LA, and Chen H. Chronic inflammation in fat plays a crucial role in the 401 development of obesity-related insulin resistance. J Clin Invest 112: 1821-1830, 2003. 402

403

404

405

17

FIGURE LEGENDS 406

Figure 1. Body weight (A), percent body fat (B), epididymal AT weight (C), 407 retroperitoneal AT weight (D), interscapular brown AT weight (E), and kilocalories 408 consumed per week (F) in wild-type and eNOS knockout mice fed a control diet 409 versus Western diet. Reported body weight, percent fat, and fat pad weights are at 410 time of sacrifice. Kcal consumed/week values are calculated as an average over the 18-411 week feeding period. Data are expressed as means ± SE. D, main effect of diet; G, 412 main effect of genotype; DxG, diet by genotype interaction. Significant p values (<0.05) 413 are highlighted in bold. N=7 to 8/group. 414

Figure 2. Representative histology images (10X) of visceral white (i.e., epididymal) 415 AT stained for Mac-2 (A), fold differences in Mac-2 positive stained area (B), and 416 average adipocyte size (C) in wild-type and eNOS knockout mice fed a control 417 diet versus Western diet. Data are expressed as means ± SE. D, main effect of diet; 418 G, main effect of genotype; DxG, diet by genotype interaction. Significant p values 419 (<0.05) are highlighted in bold. N=5/group. 420

Figure 3. Visceral white (i.e., epididymal) (A) and brown (i.e. interscapular) (B) AT 421 gene expression in wild-type and eNOS knockout mice fed a control diet versus 422 Western diet. Data are expressed as means ± SE. D Denotes main effect of diet 423 (p<0.05). No significant main effects of genotype or interactions were found (all p>0.05). 424 N=7 to 8/group. 425

Figure 4. Visceral white (i.e., epididymal) AT protein content in wild-type and 426 eNOS knockout mice fed a control diet versus Western diet. Data are expressed as 427 means ± SE. D, main effect of diet; G, main effect of genotype; DxG, diet by genotype 428 interaction. Significant p values (<0.05) are highlighted in bold. N=7 to 8/group. For the 429 nitrotyrosine blot, an average of all bands (between 75kda and 25kda) was calculated. 430

Figure 5. Aorta gene expression in wild-type and eNOS knockout mice fed a 431 control diet versus Western diet. Data are expressed as means ± SE. G, denotes 432 main effect of genotype (p<0.05). No significant main effects of diet or interactions were 433 found (all p>0.05). N=4 to 8/group. 434

435

436

437

438

439

440

441

18

Table 1. Fasting blood characteristics. 442

Variable Wild-type eNOS knockout Control diet Western

diet Control diet Western diet Two-Way

ANOVA Total cholesterol, mg/dl

145.4±6.9 215.6±25.7 134.6±4.9 247.1±25.6 D: p<0.001 G: p=0.590 DxG: p=0.276

LDL cholesterol, mg/dl

7.8±0.6 16.9±2.4 6.0±0.4 20.4±5.3 D: p=0.001 G: p=0.776 DxG: p=0.400

HDL cholesterol, mg/dl

64.3±2.0 96.0±10.5 65.3±3.9 105.9±4.2 D: p<0.001 G: p=0.354 DxG: 0.452

Triglycerides, mg/dl

51.1±4.8 42.0±3.3 59.9±4.1 52.1±9.7 D: p=0.194 G: p=0.148 DxG: p=0.913

NEFA, mmol/l 0.73±0.05 0.69±0.07 0.80±0.05 0.92±0.10 D: p=0.585 G: p=0.042 DxG: p=0.285

ALT, U/l 38.3±4.8 210.6±50.3 42.3±5.3 250.8±72.3 D: p<0.001 G: p=0.631 DxG: p=0.694

AST, U/l 226.9±19.6 288.7±52.9 273.0±50.0 321.9±38.4 D: p=0.189 G: p=0.342 DxG: p=0.876

Insulin, ng/ml 0.8±0.1 2.4±0.5 1.2±0.2 3.2±1.2 D: p=0.017 G: p=0.415 DxG: p=0.786

Glucose, mg/dl 290.5±25.5 309.0±30.4 316.1±14.6 315.5±46.4 D: p=0.786 G: p=0.625 DxG: p=0.770

HOMA-IR 17.5±2.5 57.3±12.4 27.9±5.8 74.9±30.6 D: p=0.021 G: p=0.436 DxG: p=0.840

Nitrate + Nitrite, pmoles/µl

20.6±3.1 11.6±1.4 16.4±2.6 11.2±2.4 D: p=0.011 G: p=0.382 DxG: p=0.455

Abbreviations: LDL, low density lipoprotein; HDL, high density lipoprotein; NEFA; non-443 esterified fatty acids; ALT, alanine aminotransferase; AST, aspartate aminotransferase; 444 HOMA-IR, homeostatic model assessment insulin resistance index. Data are expressed 445 as means ± SE.D, main effect of diet; G, main effect of genotype; DxG, diet by genotype 446 interaction. Significant p values (<0.05) are highlighted in bold. N=7 to 8/group. 447

448

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5