1

Loss of Von Hippel-Lindau (VHL) increases systemic cholesterol level through 1 targeting HIF-2α and regulation of bile acid homeostasis 2

3 Sadeesh K. Ramakrishnan1, Matthew Taylor1, Aijuan Qu4, Sung-Hoon Ahn4, 4 Madathilparambil V Suresh3, Krishnan Raghavendran3, Frank J. Gonzalez4, and 5 Yatrik M. Shah1,2 6 1Department of Molecular & Integrative Physiology, 2Department of Internal 7 Medicine, Division of Gastroenterology, and 3Department of Surgery, University of 8 Michigan Medical School, Ann Arbor MI. 4Laboratory of Metabolism, Center for 9 Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, 10 MD. 11 12 Running Title: HIF-2α regulates cholesterol homeostasis. 13 14 Key words: Systemic hypoxia, cholesterol homeostasis, bile acid homeostasis, 15 Cyp7a1, hypoxia-inducible factor. 16 17 Correspondence: Yatrik M. Shah, Department of Molecular & Integrative 18 Physiology, Department of Internal medicine, Division of Gastroenterology, 19 University of Michigan Medical School, Ann Arbor, MI 48109. Email: 20 shahy@umich.edu 21 22

MCB Accepts, published online ahead of print on 13 January 2014Mol. Cell. Biol. doi:10.1128/MCB.01441-13Copyright © 2014, American Society for Microbiology. All Rights Reserved.

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

2

Abbreviations: ARNT, aryl hydrocarbon receptor nuclear translocator; Baat, bile 23 acid CoA: amino acid N-acyltransferase; BSEP, bile salt export protein; CIH, chronic 24 intermittent hypoxia; FXR, farsenoid x receptor; HIF, hypoxia-inducible factor; 25 HMGCS, HMG-CoA synthase; Hmgr, HMG-CoA reductase; NTCP, Sodium 26 taurocholate cotransport protein; SHP, small heterodimer partner; Srebp2, sterol 27 regulatory element binding protein 2; Vhl, Von Hippel- Lindau tumor suppressor 28 protein. 29 30 Conflict of interests: The authors declare no conflict of interest. 31

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

3

Abstract 32 Cholesterol synthesis is a highly oxygen dependent process. Paradoxically, hypoxia is 33 correlated with an increase in cellular and systemic cholesterol levels and risk of 34 cardiovascular diseases. The mechanism for the increase in cholesterol during 35 hypoxia is unclear. Hypoxia signaling is mediated through hypoxia-inducible factor 36 (HIF)-1α and HIF-2α. The present study demonstrates that activation of HIF 37 signaling in the liver increases hepatic and systemic cholesterol levels due to a 38 decrease in the expression of cholesterol hydroxylases CYP7A1 and other enzymes 39 involved in bile acid synthesis. Specifically, activation of hepatic HIF-2α (but not 40 HIF-1α) led to hypercholesterolemia. HIF-2α repressed the circadian expression of 41 Rev-erbα resulting in increased expression of E4BP4, a negative regulator of Cyp7a1. 42 To understand if HIF-mediated decrease in bile acid synthesis is a physiological 43 relevant pathway by which hypoxia maintains or increases systemic cholesterol 44 levels, two hypoxic mouse models were assessed; an acute lung injury model and 45 mice exposed to 10% O2 for three weeks. In both the models cholesterol levels 46 increased with a concomitant decrease in expression of genes involved in bile acid 47 synthesis. The present study demonstrates that hypoxic activation of hepatic HIF-2α 48 leads to an adaptive increase in cholesterol levels through inhibition of bile acid 49 synthesis. 50

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

4

Introduction 51 Cholesterol is an essential nutrient important in many metabolic processes and its 52 levels are tightly regulated. Cholesterol synthesis is a highly oxygen dependent 53 process. The final step from lanosterol to cholesterol conversion requires nine 54 molecules of dioxygen (53). However, systemic hypoxia is associated with an 55 increase in cholesterol levels. Individuals living at high altitude are at high risk for 56 coronary heart disease due to elevated low-density lipoprotein and total cholesterol 57 levels caused by high-altitude hypoxia (51). In chronic hypoxic diseases, such as in 58 patients with obstructive sleep apnea, no change or an increase in serum and liver 59 cholesterol levels are observed, which can be reversed by continuous positive airway 60 pressure (33, 42, 45). Moreover, mice exposed to chronic intermittent hypoxia (CIH) 61 have a significant increase in total cholesterol levels (14, 22). These studies suggest 62 that hypoxia-signaling pathways activate an adaptive response important in 63 cholesterol homeostasis. Several mechanisms were proposed to account for hypoxia-64 induced dyslipidemia (9, 37). However, the mechanisms by which hypoxia regulates 65 cholesterol homeostasis are not clear. Hypoxic signaling is mediated by an oxygen 66 sensitive transcription factor, hypoxia-inducible factor (HIF), consisting of two 67 isoforms, HIF-1α and HIF-2α (56). Activation of HIF-1α induces glycolytic genes 68 whereas activation of HIF-2α regulates the expression of genes that are involved in 69 fatty acid synthesis, fatty acid uptake, inflammation, fibrosis and vascular tumors (17, 70 24, 37, 40, 41). 71 72

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

5

In this study, liver hypoxia-induced dyslipidemia was shown to be associated with 73 spontaneous hypercholesterolemia. Consistent with previous reports, the increase in 74 hepatic cholesterol by hypoxia was not due to alterations in cholesterol synthesis, 75 intestinal absorptive genes, or mobilization from fat stores (1, 34). Disruption of Vhl 76 led to a HIF-2α dependent decrease in cholesterol hydroxylases involved in the 77 conversion of cholesterol to bile acids. To confirm that the decrease in cholesterol 78 hydroxylase is an important mechanism to maintain cholesterol levels during 79 hypoxia, two systemic hypoxic mouse models were used; mice exposed to 80 normobaric 10% O2 for three weeks and a mouse model of lung contusion (38). Both 81 models exhibited significantly elevated levels of serum cholesterol with a parallel 82 decrease in expression of genes involved in bile acid synthesis. Taken together, this 83 study demonstrates a mechanistic basis for the adaptive maintenance of cholesterol 84 levels during hypoxia. Thus, inhibition of HIF-2α may be relevant and novel 85 therapeutic target to ameliorate hypercholesterolemia during hypoxia. 86 on A

pril 16, 2018 by guesthttp://m

cb.asm.org/

Dow

nloaded from

6

Material and Methods 87 Animals and diets 88 VhlF/F, VhlLivKO, VhlΔInt, Vhl/ArntLivKO, Vhl/Hif1αLivKO Vhl/Hif2αLivKO, VhlΔM�, 89 VhlIntKO and ODD-luc mice were described previously (17, 44, 49, 50, 55, 58). VhlΔ90 M� were generated by crossing LysM cre mice with VhlF/F mice. All mice were fed 91 ad-libitum and kept in a 12-hour dark/light cycle. Mice were fed with either regular 92 chow diet or atherogenic (35 kcal% fat, 1% cholesterol, 0.5% cholic acid; Research 93 diets, New Brunswick, NJ), high fat (~45 kcal% derived from fat; Research diets) or 94 ethanol diet (6% lieber-decarli diet, Bio-serv, Frenchtown, NJ). For systemic hypoxia 95 experiments, C57BL/6 mice in standard cages were placed in a normobaric hypoxia 96 chamber with 10% O2 for three weeks. Lung contusion (C57BL/6 mice) and in vivo 97 imaging (ODD-luc) were performed as described previously (2, 38). All the mice 98 were sacrificed at 7PM unless mentioned in the figure legend. All animal studies were 99 carried out in accordance with Association for Assessment and Accreditation of 100 Laboratory Animal Care International guidelines and approved by the University 101 Committee on the Use and Care of Animals at the University of Michigan. 102 103 Western blot analysis 104 Microsome and whole cell extracts were prepared as previously described (55). 105 Proteins were separated by SDS-PAGE. Microsomes were used for detection of 106 CYP7A1 (Abcam, Cambridge, MA) and normalized to total protein by coomassie 107 blue staining. 108 109

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

7

Primary hepatocyte isolation 110 For primary hepatocytes isolation, livers were perfused continuously with 10 ml of 111 EBSS containing 0.5mM EGTA, followed by serum-free L-15 media containing 112 collagenase type II (Worthington, Invitrogen). Hepatocytes were then harvested by 113 gentle teasing and filtered through 100 micron filter. Viable hepatocytes were plated 114 in 6 well plates at a cell density of 4 X 105 in Williams E media containing 10% FBS. 115 5 hours after plating, dead cells were removed and media refreshed. For adenovirus 116 treatment, desired amount of Ad-Cyp7a1, AD-HNF4α, and AD-LRH-1 virus was 117 added after 5 hours of plating and then incubated until further analysis. 118 119 Cholesterol assay 120 For cholesterol extraction, 50 mg of liver or white adipose tissue was homogenized in 121 1 ml of 2:1 chloroform to methanol ratio. Primary hepatocytes were collected in 122 300μl methanol and then 600μl of chloroform added and vortexed. The homogenates 123 were then centrifuged and 200μl of water was added to the supernatant and vortexed. 124 The organic phase separated by centrifugation was dried overnight and resuspended 125 in isopropanol. Cholesterol measurements were performed using the Cholesterol E kit 126 (Wako Diagnostics, VA). Serum cholesterol was measured in 20μl of serum and 127 lipoprotein profiles were assessed by fast performance liquid chromatography as 128 previously described (36). 129 130 Bile acid assay 131

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

8

Bile acids from liver extracted as described by Cheng et al (7). Specific bile salts 132 species were analyzed by liquid chromatography tandem mass spectrometry (LC-133 MS/MS) performed on a PESCIEX API200 ESI triple-quadrupole mass spectrometer 134 (PerkinElmer Life Sciences) as described previously (23). 135 136 RNA extraction and qPCR analysis 137 RNA extraction and gene expression analysis using quantitative real-time PCR was 138 performed as previously described (37). Primers are included in Table S1. 139 140 Statistical analysis 141 Results are expressed as mean ± SD. P values were calculated by Independent t-test, 142 1-way anova or 2-way anova. 143

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

9

Results 144 Disruption of hepatic VHL results in hypercholesterolemia. In order to elucidate the 145 role of HIFs in cholesterol metabolism, serum cholesterol levels were assessed in 146 mice with a conditional disruption of VHL in tissues or cells important in cholesterol 147 homeostasis. Under normoxia, the Von Hippel-Lindau (Vhl) tumor suppressor protein 148 binds to hydroxylated HIFs and targets them for ubiquitination and degradation (19, 149 20). Genetic disruption of Vhl increases the expression of HIFs and HIF target genes 150 (17). Disruption of VHL in the macrophage (VhlΔM�) or the intestine (VhlΔInt) did not 151 affect serum cholesterols levels compared to wild-type littermates (VhlF/F) (Figure 152 1A). However, a temporal disruption of VHL in hepatocytes (VhlLivKO) using a 153 tamoxifen-inducible Cre recombinase (37), significantly elevated serum cholesterol 154 (Figure 1A), hepatic cholesterol (Figure 1B) and cholesterol content in isolated 155 primary hepatocytes (Figure 1C) compared to tamoxifen-treated wild-type littermates 156 (VhlF/F). Hypercholesterolemia is primarily associated with increases in VLDL and 157 LDL cholesterol, while HDL cholesterol was decreased in VhlLivKO mice (Figure1D). 158 VhlLivKO mice on a normal chow diet die within 30-days after Vhl disruption (Figure 159 1E). Feeding an atherogenic diet markedly decreased the survival of VhlLivKO mice, 160 which was not observed in VhlLivKO mice fed high fat or ethanol diets (Figure 1F). We 161 have previously reported that disruption of Vhl in the liver results in progressive liver 162 steatosis and fibrosis (37). Therefore, it is possible that the hypercholesterolemia 163 might augment the progressive liver disease in VhlLivKO mice. In order to investigate 164 the effect of atherogenic diet on serum cholesterol, VhlLivKO mice were injected with 165 tamoxifen to disrupt Vhl expression and then immediately placed on atherogenic diet 166

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

10

for one week. The atherogenic diet further augmented the serum cholesterol levels in 167 VhlLivKO mice (Figure 1F). Together, these data demonstrate that disruption of Vhl in 168 the liver leads to dysregulated cholesterol homeostasis. 169 170 Genes involved in de novo cholesterol synthesis in the liver and cholesterol 171 absorption in the intestine are not altered by hepatic Vhl disruption. Hepatic 172 cholesterol homeostasis is maintained by coordinated regulation of uptake, synthesis, 173 and oxidative metabolism to bile acids (5). In order to determine the mechanism of 174 hypercholesterolemia in VhlLivKO mice, expression of key enzymes involved in 175 cholesterol synthesis and esterification were analyzed. Expression of Cyp51, HMG-176 CoA synthase (Hmgcs), HMG-CoA reductase (Hmgr), Scd1 and Scd2 were 177 significantly repressed two weeks after Vhl disruption but not at five days (Figure 178 2A). The expression of sterol regulatory element binding protein 2 (Srebp2), a 179 transcription factor critical in cholesterol synthesis was significantly decreased at both 180 five days and two weeks following Vhl disruption (Figure 2A). These data 181 demonstrates that the cholesterol-mediated negative feedback repression on de novo 182 synthesis is still intact (4) and cholesterol synthesis is not the major mechanism by 183 which liver disruption of VHL induces hypercholesterolemia. Cholesterol is cleared 184 by the liver via LDL receptor and loss of function or genetic ablation of LDL receptor 185 results in hypercholesterolemia (16). Therefore, the dynamic changes in hepatic 186 cholesterol levels and LDL receptor expression were assessed in VhlLivKO mice. A 187 progressive and significant increase in hepatic cholesterol levels was observed 188 starting at three days following Vhl disruption (Figure 2B), however, LDL receptor 189

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

11

protein levels decreased significantly only two weeks after Vhl disruption (Figure 190 2C). Loss of LDL receptor after two weeks could be attributed to the negative feed-191 back repression exerted by increased intrahepatic cholesterol levels to limit 192 cholesterol overload (32, 43). In addition, disruption of Vhl in liver did not affect the 193 expression of Npc1l1 (Figure 2D), a gene involved in intestinal cholesterol absorption 194 (27). Increasing cholesterol storage by adipose tissue improves lipid profile in mice; 195 on the other hand, cholesterol mobilization from adipose tissue is associated with 196 hypercholesterolemia in humans (46, 54). No significant difference in white adipose 197 cholesterol content between VhlLivKO and VhlF/F mice suggest that hepatic VHL 198 disruption does not affect adipose tissue cholesterol homeostasis (Figure 2E). 199 200 CYP7A1 expression is repressed by hepatic Vhl disruption. Cholesterol is cleared by 201 the liver through conversion to bile acids (12). CYP7A1 is the major rate-limiting 202 cholesterol hydroxylase involved in the classical pathway of bile acid synthesis from 203 cholesterol. In order to assess whether hypercholesterolemia is due to altered bile acid 204 homeostasis, CYP7A1 expression was assessed in VhlLivKO mice. Two weeks after 205 disruption of Vhl, expression of Cyp7a1 mRNA levels were significantly reduced in 206 the livers of VhlLivKO mice (Figure 3A). Similarly, CYP7A1 mRNA and protein levels 207 were significantly decreased in primary hepatocytes from VhlLivKO mice (Figure 3A 208 and B). To identify the mechanisms by which disruption of Vhl repress CYP7A1 209 expression, several known pathways important in the regulation of CYP7A1 were 210 assessed. Activation of JNK induces hypercholesterolemia and cholestasis (6), and is 211 a potent inhibitor of Cyp7a1 expression (30). Consistent with inflammation in the 212

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

12

livers of VhlLivKO mice (37), JNK phosphorylation was significantly increased at five 213 days and two weeks after disruption of Vhl (Figure 3C). However, JNK inhibitor (SP 214 600125) did not rescue Cyp7a1 mRNA levels in primary hepatocytes from VhlLivKO 215 mice (Figure 3D). Similarly, adenoviral-mediated overexpression of HNF-4α or 216 LRH-1, known CYP7A1 transcriptional regulators (25), did not rescue the expression 217 of Cyp7a1 mRNA levels in primary hepatocytes from VhlLivKO mice (Figure 3E). 218 219 Hepatic VHL disruption alters circadian oscillation of CYP7A1. Cholesterol 220 exhibits diurnal variation due to circadian oscillation in the expression of genes 221 involved in cholesterol metabolism, including CYP7A1 (35). In order to investigate 222 whether CYP7A1 expression and circadian regulation was altered by HIF activation, 223 VhlLivKO and littermate control mice were treated with tamoxifen for five days and 224 two weeks and killed at 7AM and 7PM. CYP7A1 mRNA and protein levels were not 225 changed at 7AM, however the circadian increase at 7PM was significantly repressed 226 in the VhlLivKO mice at five days (Figure 4A). However, circadian expression of other 227 cholesterol hydroxylases such as Cyp27a1, Cyp7b1, Hsd3b1 and Cyp3a11 were not 228 affected by disruption of Vhl (Figure 4B). Furthermore, two weeks following VHL 229 disruption, CYP7A1 mRNA and protein at 7AM, and the circadian induction at 7PM, 230 were significantly repressed (Figure 4C). Among the various key transcription factors 231 that regulate circadian genes, Rev-erbα plays an essential role in the diurnal 232 expression of Cyp7a1 (13, 26). The circadian induction of Rev-erbα is inhibited in 233 the VhlLivKO mice at both five days and two weeks following tamoxifen treatment 234 (Figure 4C). Consistent with the protein expression, the circadian induction of Rev-235

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

13

erbα but not Rev-erbβ was significantly repressed at five days after disruption of Vhl 236 (Figure 4D). Moreover, the loss of expression of Rev-erbα in VhlLivKO mice at 7PM 237 resulted in significant increase in E4bp4 mRNA levels (Figure 4D), a negative 238 regulator of Cyp7a1 (13). However, no difference in the expression of circadian genes 239 such as Bmal-1 and Per-1 was observed by disruption of Vhl in liver (Figure 4D). 240 These data suggest that hypoxia-mediated circadian dysregulation leads to decrease in 241 CYP7A1 levels. 242 243 Vhl disruption in the liver leads to decreased expression of genes critical in bile 244 acid synthesis. To address whether repression of CYP7A1 is the primary mechanism 245 for the aberrant cholesterol levels in VhlLivKO mice, CYP7A1 was overexpressed in 246 primary hepatocytes from VhlLivKO mice. Rescuing CYP7A1 did not ameliorate 247 cholesterol levels in the primary hepatocytes from VhlLivKO mice (Figure 5A and B). 248 Conversion of cholesterol to bile is a multi-enzymatic process involving numerous 249 cholesterol hydroxylases. Further analysis revealed that disruption of VHL in livers 250 resulted in repression of genes encoding other key cholesterol hydroxylases such as 251 Cyp27a1, Cyp7b1, Cyp8b1, Hsd3b, Amacr, Cyp3a11, Cyp29a1 and Slc25c1 (Figure 252 5C). Taken together, these data suggest that activation of hypoxia signaling by 253 disruption of VHL results in global decrease in cholesterol hydroxylases involved in 254 bile acid synthesis. 255 256 Vhl disruption in the liver leads to decreased bile acids. Decrease in the expression 257 and activity of cholesterol hydroxylases decreases bile acid synthesis leading to 258

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

14

hypercholesterolemia. Therefore, to understand if repression of cholesterol 259 hydroxylases by hypoxia signaling could adversely affect bile acid synthesis, the 260 hepatic bile acid pool was measured in VhlLivKO mice. Total liver bile acid pool 261 (Figure 6A) and gall bladder size were significantly reduced in VhlLivKO mice (Figure 262 6B). In addition, major hepatic bile acid species were significantly decreased, 263 including the recently identified endogenous FXR antagonist, tauro-β-muricholic acid 264 (Figure 6C) (28, 47). Bile acids inhibit its own synthesis by negative feedback 265 mechanism via activation of farsenoid X receptor (FXR) (8, 52). Consistent with the 266 decrease in its endogenous ligand, FXR target genes such as Abcg5, Abcg8, small 267 heterodimer partner (Shp), and bile salt export protein (Bsep) in the liver and Fgf15 in 268 the intestines were significantly repressed in VhlLivKO mice (Figure 6D), without 269 changes in FXR protein levels (Figure 6E). In addition, the expression of bile acid-270 CoA: amino acid N-acyltransferase (Baat) and the bile acid transporter, sodium 271 taurocholate cotransport protein (Ntcp) are significantly reduced in the livers of 272 VhlLivKO mice (Figure 6D). These data demonstrates that HIF activation disrupts bile 273 acid synthesis and signaling. 274 275 Repression of Cyp7a1 is HIF-2α dependent. Disruption of Vhl results in the 276 overexpression of both HIF-1α and HIF-2α (17). HIF-1α and HIF-2α exert distinct 277 functions by regulating different subsets of genes (18). Elevation in the serum and 278 hepatic cholesterol observed in VhlLivKO was completely abolished in mice with 279 compound disruption of VHL and HIF-2α (Vhl/Hif-2αLivKO) and VHL and ARNT 280 (Vhl/ArntLivKO) but not in mice with disruption of VHL and HIF-1α (Vhl/Hif-1αLivKO), 281

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

15

demonstrating that perturbation in cholesterol levels was HIF-2α dependent (Figure 282 7A and B). Concordant with the improvement in cholesterol homeostasis, CYP7A1 283 and Rev-erbα expression were rescued by combined disruption of VHL and HIF-2α 284 (Figure 7C). In addition, combined disruption of VHL and HIF-2α restored the 285 expression of bile acid synthesis and FXR target genes (Figure 7D and E). These data 286 demonstrates that circadian regulation of cholesterol hydroxylases by HIF-2α plays a 287 key role in cholesterol and bile homeostasis following hypoxia. 288 289 Systemic hypoxia leads to hypercholesterolemia and a decrease in bile synthesis. To 290 further demonstrate that liver hypoxia signaling is a critical regulator of cholesterol 291 homeostasis, a model of closed chest unilateral lung contusion that rapidly induces 292 systemic hypoxia was analyzed (39). Lung contusion induced a robust increase in 293 systemic hypoxia 24-hours following injury as assessed in a hypoxia reporter mouse 294 model, ODD-luc mice (Figure 8A) (2, 44). In parallel with the systemic hypoxia, 295 serum and liver cholesterol levels were significantly elevated at 24- and 48-hours 296 after the insult (Figure 8B). However, decrease in hepatic bile acid levels did not 297 reach significance raising the possibility that enterohepatic recycling of bile could 298 compensate for the short-term inhibition of bile acid synthesis (Figure 8B). Systemic 299 hypoxia decreased the levels of CYP7A1 protein in the microsomal fraction of liver 300 at 24-hours after the initiation of lung contusion (Figure 8C). In addition, mRNA 301 levels of Cyp7a1, Cyp7b1 and Cyp27a1 were significantly decreased in livers at 24- 302 and 48-hours after lung injury (Figure 8D). Further analysis revealed that the 303 circadian gene Rev-erbα is significantly repressed at 48-hours after lung injury with a 304

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

16

concomitant increase in E4bp4 expression (Figure 8D). To investigate whether 305 chronic systemic hypoxia can affect cholesterol and bile acid homeostasis, C57BL/6 306 mice were exposed to 10% O2 for three weeks. Hypoxia significantly increased the 307 serum and liver cholesterol levels and decreased hepatic bile acids (Figure 8E). 308 Moreover, the expression of cholesterol hydroxylases such as Cyp7a1, Cyp8b1, and 309 Cyp27a1, were significantly repressed (Figure 8F). Similar to the lung contusion 310 model, chronic hypoxia for three weeks significantly decreased Rev-erbα expression 311 resulting in increased expression of E4bp4 (Figure 8F). The data demonstrates that 312 activation of hypoxia signaling in liver increases hepatic and systemic cholesterol 313 levels. Moreover, the data suggests that the adaptive increase in systemic cholesterol 314 levels following hypoxia are due to a decrease in cholesterol clearance as bile acids. 315

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

17

Discussion 316 Cholesterol synthesis is a multi-enzymatic reaction, which requires energy and 317 oxygen. It is hypothesized that the evolution of cholesterol is directly correlated to 318 atmospheric O2 (15). However, during hypoxia adaptive mechanisms exists in order 319 to maintain systemic cholesterol levels. Indigenous populations to high altitude, sleep 320 apnea patients, and mouse models exposed to CIH have an increase in cholesterol 321 levels and are at higher risk for cardiovascular diseases (42, 45, 51). It is worth noting 322 that continuous positive airway pressure is the effective treatment for 323 hypercholesterolemia in obstructive sleep apnea patients (33), further raising the 324 possibility that hypoxia signaling is involved in the regulation of cholesterol 325 homeostasis. The mechanism by which hypoxia sustains or induces systemic 326 cholesterol levels is not well understood. Using liver-specific HIF overexpressing 327 (VhlLivKO) or systemic hypoxic mouse models, we demonstrate that hypoxia-mediated 328 perturbation in cholesterol homeostasis is due to a decrease in the expression of 329 cholesterol hydroxylases involved in bile synthesis. Moreover, the data suggests that 330 HIF-2α plays a critical role in this mechanism by regulating circadian expression of 331 hepatic Rev-erbα and Cyp7a1. 332 333 Hypercholesterolemia in VhlLivKO mice is not due to cholesterol biosynthesis or 334 uptake. Increase in the intrahepatic cholesterol levels precedes the changes in LDL 335 receptor levels suggesting that a defective cholesterol metabolism at the cellular level 336 as the primary cause for aberrant cholesterol levels in VhlLivKO mice. However, the 337 secondary decrease in the LDL receptor levels could further augment the increase in 338

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

18

VLDL and LDL cholesterol leading to hypercholesterolemia in VhlLivKO mice. 339 CYP7A1, the key rate-limiting enzyme in bile acid synthesis, was inhibited by 340 hypoxia and/or HIF-2α activation. Mice that overexpress CYP7A1 have increased 341 bile acid pools and lower serum cholesterol (29, 31). On the other hand, Cyp7a1-/- 342 mice have decreased bile acid synthesis. These studies are consistent with the data 343 presented in this manuscript. However, Cyp7a1-/- mice do not have any overt 344 differences in hepatic cholesterol levels compared to wild-type mice (48). Similarly, 345 adenoviral-mediated rescue of CYP7A1 alone did not ameliorate cholesterol levels in 346 primary hepatocytes from VhlLivKO mice. Bile acid metabolism is profoundly 347 repressed in isolated cultured mouse primary hepatocytes compared to the liver; 348 therefore we cannot exclude the possibility that CYP7A1 is the primary hypoxia-349 repressed gene involved in hypercholesterolemia. However, a significant decrease in 350 a battery of cholesterol hydroxylases in the liver of VhlLivKO mice suggests that global 351 repression of other cholesterol hydroxylases, in addition to CYP7A1, is the essential 352 mechanism that leads to dysregulated cholesterol levels following hypoxia. 353 354 Cholesterol levels exhibit circadian variation due to the diurnal expression of 355 CYP7A1 (35). Rev-erbα is one of the key transcription factors that regulate the 356 expression of various circadian genes. Rev-erbα knockout mice exhibit decreased bile 357 acid levels due to loss of Cyp7a1 expression (13, 26). The current study demonstrates 358 that activation of HIF signaling disrupts the circadian expression of Rev-erbα 359 resulting in increased expression of E4bp4, a negative regulator of Cyp7a1 (13). In 360 addition, systemic hypoxia due to lung contusion or chronic hypoxia exposure of 361

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

19

mice recapitulates the decrease in genes involved in bile acid synthesis and bile acid 362 levels with a concomitant increase in systemic cholesterol levels. Furthermore, 363 repression of CYP7A1 is also associated with loss of Rev-erbα and increase in E4bp4 364 in these hypoxic mouse models (13). The expressions of other cholesterol 365 hydroxylases were not repressed by HIF-2α through disrupting circadian regulation. 366 This suggests that disruption of VHL inhibits bile acid synthesis perhaps through 367 other parallel mechanisms that need to be further investigated. Together, these results 368 suggest that perturbation in cholesterol homeostasis following hypoxia is an adaptive 369 mechanism to increase systemic cholesterol levels in a HIF-2α-dependent manner. 370 The profound effect of hypoxia signaling on cholesterol metabolism is through 371 decreasing cholesterol clearance as bile acids via disrupting the circadian expression 372 of CYP7A1 and by repressing downstream cholesterol hydroxylases through a 373 currently unknown mechanism. 374 375 Atherosclerosis is a well-known inflammation and lipid-associated disease and 376 chronic intermittent hypoxia (CIH) exacerbates disease progression (10, 14, 22). An 377 important mechanism that contributes to atherosclerosis is recruitment of 378 macrophages that scavenge fatty acids resulting in the formation of foam cells. 379 Several studies have demonstrated the importance of HIF-1α in foam cell formation 380 and atherosclerosis (3, 21). Moreover, recently it was shown that HIF-1α induction of 381 Angptl4 in adipose tissues contributes to hypoxic potentiation of atherosclerosis (11). 382 However, partial disruption of HIF-1α in adipose tissue did not improve 383 hypercholesterolemia caused by CIH. The role of liver HIF-2α in hypoxia-mediated 384

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

20

atherosclerosis is not yet explored. CIH induces systemic inflammation and recently 385 we demonstrated that HIF-2α induces a pro-inflammatory response by increasing 386 TNF-α expression (57). The present work, together with our previous study (37), 387 demonstrates that chronic hypoxic signaling through HIF-2α in the liver may initiate 388 a pro-atherogenic environment by inducing inflammation and hypercholesterolemia. 389 The data in this study demonstrates that a decrease in cholesterol clearance as bile as 390 the primary mechanism for the adaptive increase in liver and systemic cholesterol 391 levels during hypoxia. Furthermore, HIF-2α regulation of circadian gene, Rev-erbα, 392 was demonstrated as a critical mechanism in cholesterol homeostasis. Since HIF-2α 393 has an essential role in hepatic inflammation and liver cholesterol homeostasis, HIF-394 2α could be a potential therapeutic target to improve dyslipidemia and systemic 395 inflammation in patients with increased risk of cardiovascular disease. 396 397 Acknowledgements 398 This work was supported by NIH grants (CA148828 and DK095201 Y.M.S), (HL-399 102013 K.R) and The University of Michigan Gastrointestinal Peptide Center to 400 (Y.M.S.), and the National Cancer Institute Intramural Research Program (F.J.G). 401 The authors like to thank Dr. Phillip B Hylemon (Virginia Commonwealth 402 University) for the generous gift of the AD-CYP7a1 and Dr. Yanqiao Zhang 403 (Northeast Ohio Medical University) for the AD-HNF4α and AD-LRH-1. 404

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

21

References 405 1. Albers, J. J., and E. L. Bierman. 1976. The effect of hypoxia on uptake and 406

degradation of low density lipoproteins by cultured human arterial smooth muscle 407 cells. Biochimica et biophysica acta 424:422-429. 408

2. Anderson, E. R., M. Taylor, X. Xue, A. Martin, D. S. Moons, M. B. Omary, 409 and Y. M. Shah. 2012. The hypoxia-inducible factor-C/EBPalpha axis controls 410 ethanol-mediated hepcidin repression. Molecular and cellular biology 32:4068-411 4077. 412

3. Bostrom, P., B. Magnusson, P. A. Svensson, O. Wiklund, J. Boren, L. M. 413 Carlsson, M. Stahlman, S. O. Olofsson, and L. M. Hulten. 2006. Hypoxia 414 converts human macrophages into triglyceride-loaded foam cells. Arterioscler 415 Thromb Vasc Biol 26:1871-1876. 416

4. Bricker, L. A., H. J. Weis, and M. D. Siperstein. 1972. In vivo demonstration 417 of the cholesterol feedback system by means of a desmosterol suppression 418 technique. The Journal of clinical investigation 51:197-205. 419

5. Brown, M. S., and J. L. Goldstein. 1997. The SREBP pathway: regulation of 420 cholesterol metabolism by proteolysis of a membrane-bound transcription factor. 421 Cell 89:331-340. 422

6. Chai, J., Y. He, S. Y. Cai, Z. Jiang, H. Wang, Q. Li, L. Chen, Z. Peng, X. He, 423 X. Wu, T. Xiao, R. Wang, J. L. Boyer, and W. Chen. 2012. Elevated hepatic 424 multidrug resistance-associated protein 3/ATP-binding cassette subfamily C 3 425 expression in human obstructive cholestasis is mediated through tumor necrosis 426

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

22

factor alpha and c-Jun NH2-terminal kinase/stress-activated protein kinase-427 signaling pathway. Hepatology 55:1485-1494. 428

7. Cheng, J., K. W. Krausz, F. Li, X. Ma, and F. J. Gonzalez. 2013. CYP2E1-429 dependent elevation of serum cholesterol, triglycerides, and hepatic bile acids by 430 isoniazid. Toxicol Appl Pharmacol 266:245-253. 431

8. Chiang, J. Y., R. Kimmel, C. Weinberger, and D. Stroup. 2000. Farnesoid X 432 receptor responds to bile acids and represses cholesterol 7alpha-hydroxylase gene 433 (CYP7A1) transcription. J Biol Chem 275:10918-10924. 434

9. Drager, L. F., J. Jun, and V. Y. Polotsky. 2010. Obstructive sleep apnea and 435 dyslipidemia: implications for atherosclerosis. Curr Opin Endocrinol Diabetes 436 Obes 17:161-165. 437

10. Drager, L. F., V. Y. Polotsky, and G. Lorenzi-Filho. 2011. Obstructive sleep 438 apnea: an emerging risk factor for atherosclerosis. Chest 140:534-542. 439

11. Drager, L. F., Q. Yao, K. L. Hernandez, M. K. Shin, S. Bevans-Fonti, J. Gay, 440 T. E. Sussan, J. C. Jun, A. C. Myers, G. Olivecrona, A. R. Schwartz, N. 441 Halberg, P. E. Scherer, G. L. Semenza, D. R. Powell, and V. Y. Polotsky. 442 2013. Chronic Intermittent Hypoxia Induces Atherosclerosis via Activation of 443 Adipose Angiopoietin-like 4. Am J Respir Crit Care Med 188:240-248. 444

12. Duane, W. C. 2009. Bile acids: developments new and very old. Journal of lipid 445 research 50:1507-1508. 446

13. Duez, H., J. N. van der Veen, C. Duhem, B. Pourcet, T. Touvier, C. Fontaine, 447 B. Derudas, E. Bauge, R. Havinga, V. W. Bloks, H. Wolters, F. H. van der 448

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

23

Sluijs, B. Vennstrom, F. Kuipers, and B. Staels. 2008. Regulation of bile acid 449 synthesis by the nuclear receptor Rev-erbalpha. Gastroenterology 135:689-698. 450

14. Fang, G., D. Song, X. Ye, S. Z. Mao, G. Liu, and S. F. Liu. 2012. Chronic 451 intermittent hypoxia exposure induces atherosclerosis in ApoE knockout mice: 452 role of NF-kappaB p50. Am J Pathol 181:1530-1539. 453

15. Galea, A. M., and A. J. Brown. 2009. Special relationship between sterols and 454 oxygen: were sterols an adaptation to aerobic life? Free radical biology & 455 medicine 47:880-889. 456

16. Goldstein, J. L., and M. S. Brown. 2009. The LDL receptor. Arterioscler 457 Thromb Vasc Biol 29:431-438. 458

17. Haase, V. H., J. N. Glickman, M. Socolovsky, and R. Jaenisch. 2001. Vascular 459 tumors in livers with targeted inactivation of the von Hippel-Lindau tumor 460 suppressor. Proceedings of the National Academy of Sciences of the United States 461 of America 98:1583-1588. 462

18. Hu, C. J., L. Y. Wang, L. A. Chodosh, B. Keith, and M. C. Simon. 2003. 463 Differential roles of hypoxia-inducible factor 1alpha (HIF-1alpha) and HIF-464 2alpha in hypoxic gene regulation. Molecular and cellular biology 23:9361-9374. 465

19. Ivan, M., K. Kondo, H. Yang, W. Kim, J. Valiando, M. Ohh, A. Salic, J. M. 466 Asara, W. S. Lane, and W. G. Kaelin, Jr. 2001. HIFalpha targeted for VHL-467 mediated destruction by proline hydroxylation: implications for O2 sensing. 468 Science 292:464-468. 469

20. Jaakkola, P., D. R. Mole, Y. M. Tian, M. I. Wilson, J. Gielbert, S. J. Gaskell, 470 A. von Kriegsheim, H. F. Hebestreit, M. Mukherji, C. J. Schofield, P. H. 471

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

24

Maxwell, C. W. Pugh, and P. J. Ratcliffe. 2001. Targeting of HIF-alpha to the 472 von Hippel-Lindau ubiquitylation complex by O2-regulated prolyl hydroxylation. 473 Science 292:468-472. 474

21. Jiang, G., T. Li, Y. Qiu, Y. Rui, W. Chen, and Y. Lou. 2007. RNA interference 475 for HIF-1alpha inhibits foam cells formation in vitro. European journal of 476 pharmacology 562:183-190. 477

22. Jun, J., C. Reinke, D. Bedja, D. Berkowitz, S. Bevans-Fonti, J. Li, L. A. 478 Barouch, K. Gabrielson, and V. Y. Polotsky. 2010. Effect of intermittent 479 hypoxia on atherosclerosis in apolipoprotein E-deficient mice. Atherosclerosis 480 209:381-386. 481

23. Kim, I., S. H. Ahn, T. Inagaki, M. Choi, S. Ito, G. L. Guo, S. A. Kliewer, and 482 F. J. Gonzalez. 2007. Differential regulation of bile acid homeostasis by the 483 farnesoid X receptor in liver and intestine. Journal of lipid research 48:2664-484 2672. 485

24. Kim, J. W., I. Tchernyshyov, G. L. Semenza, and C. V. Dang. 2006. HIF-1-486 mediated expression of pyruvate dehydrogenase kinase: a metabolic switch 487 required for cellular adaptation to hypoxia. Cell Metab 3:177-185. 488

25. Kir, S., Y. Zhang, R. D. Gerard, S. A. Kliewer, and D. J. Mangelsdorf. 2012. 489 Nuclear receptors HNF4alpha and LRH-1 cooperate in regulating Cyp7a1 in vivo. 490 J Biol Chem 287:41334-41341. 491

26. Le Martelot, G., T. Claudel, D. Gatfield, O. Schaad, B. Kornmann, G. Lo 492 Sasso, A. Moschetta, and U. Schibler. 2009. REV-ERBalpha participates in 493 circadian SREBP signaling and bile acid homeostasis. PLoS biology 7:e1000181. 494

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

25

27. Levy, E., S. Spahis, D. Sinnett, N. Peretti, F. Maupas-Schwalm, E. Delvin, M. 495 Lambert, and M. A. Lavoie. 2007. Intestinal cholesterol transport proteins: an 496 update and beyond. Current opinion in lipidology 18:310-318. 497

28. Li, F., C. Jiang, K. W. Krausz, Y. Li, I. Albert, H. Hao, K. M. Fabre, J. B. 498 Mitchell, A. D. Patterson, and F. J. Gonzalez. 2013. Microbiome remodelling 499 leads to inhibition of intestinal farnesoid X receptor signalling and decreased 500 obesity. Nature communications 4:2384. 501

29. Li, T., J. M. Francl, S. Boehme, and J. Y. Chiang. 2013. Regulation of 502 cholesterol and bile acid homeostasis by the CYP7A1/SREBP2/miR-33a axis. 503 Hepatology. 504

30. Li, T., A. Jahan, and J. Y. Chiang. 2006. Bile acids and cytokines inhibit the 505 human cholesterol 7 alpha-hydroxylase gene via the JNK/c-jun pathway in human 506 liver cells. Hepatology 43:1202-1210. 507

31. Li, T., M. Matozel, S. Boehme, B. Kong, L. M. Nilsson, G. Guo, E. Ellis, and 508 J. Y. Chiang. 2011. Overexpression of cholesterol 7alpha-hydroxylase promotes 509 hepatic bile acid synthesis and secretion and maintains cholesterol homeostasis. 510 Hepatology 53:996-1006. 511

32. Ma, P. T., G. Gil, T. C. Sudhof, D. W. Bilheimer, J. L. Goldstein, and M. S. 512 Brown. 1986. Mevinolin, an inhibitor of cholesterol synthesis, induces mRNA for 513 low density lipoprotein receptor in livers of hamsters and rabbits. Proceedings of 514 the National Academy of Sciences of the United States of America 83:8370-8374. 515

33. Michailidis, V., P. Steiropoulos, E. Nena, N. Papanas, E. Maltezos, and D. 516 Bouros. 2011. Continuous positive airway pressure treatment: effect on serum 517

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

26

lipids in patients with obstructive sleep apnoea. Open Cardiovasc Med J 5:231-518 238. 519

34. Mukodani, J., Y. Ishikawa, and H. Fukuzaki. 1990. Effects of hypoxia on 520 sterol synthesis, acyl-CoA:cholesterol acyltransferase activity, and efflux of 521 cholesterol in cultured rabbit skin fibroblasts. Arteriosclerosis 10:106-110. 522

35. Noshiro, M., E. Usui, T. Kawamoto, H. Kubo, K. Fujimoto, M. Furukawa, S. 523 Honma, M. Makishima, K. Honma, and Y. Kato. 2007. Multiple mechanisms 524 regulate circadian expression of the gene for cholesterol 7alpha-hydroxylase 525 (Cyp7a), a key enzyme in hepatic bile acid biosynthesis. Journal of biological 526 rhythms 22:299-311. 527

36. Qu, A., Y. M. Shah, S. K. Manna, and F. J. Gonzalez. 2012. Disruption of 528 endothelial peroxisome proliferator-activated receptor gamma accelerates diet-529 induced atherogenesis in LDL receptor-null mice. Arterioscler Thromb Vasc Biol 530 32:65-73. 531

37. Qu, A., M. Taylor, X. Xue, T. Matsubara, D. Metzger, P. Chambon, F. J. 532 Gonzalez, and Y. M. Shah. 2011. Hypoxia-inducible transcription factor 2alpha 533 promotes steatohepatitis through augmenting lipid accumulation, inflammation, 534 and fibrosis. Hepatology 54:472-483. 535

38. Raghavendran, K., B. A. Davidson, J. D. Helinski, C. J. Marschke, P. 536 Manderscheid, J. A. Woytash, R. H. Notter, and P. R. Knight. 2005. A rat 537 model for isolated bilateral lung contusion from blunt chest trauma. Anesthesia 538 and analgesia 101:1482-1489. 539

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

27

39. Raghavendran, K., R. H. Notter, B. A. Davidson, J. D. Helinski, S. L. Kunkel, 540 and P. R. Knight. 2009. Lung contusion: inflammatory mechanisms and 541 interaction with other injuries. Shock 32:122-130. 542

40. Rankin, E. B., J. Rha, M. A. Selak, T. L. Unger, B. Keith, Q. Liu, and V. H. 543 Haase. 2009. Hypoxia-inducible factor 2 regulates hepatic lipid metabolism. 544 Molecular and cellular biology 29:4527-4538. 545

41. Rankin, E. B., J. Rha, T. L. Unger, C. H. Wu, H. P. Shutt, R. S. Johnson, M. 546 C. Simon, B. Keith, and V. H. Haase. 2008. Hypoxia-inducible factor-2 547 regulates vascular tumorigenesis in mice. Oncogene 27:5354-5358. 548

42. Robinson, G. V., J. C. Pepperell, H. C. Segal, R. J. Davies, and J. R. 549 Stradling. 2004. Circulating cardiovascular risk factors in obstructive sleep 550 apnoea: data from randomised controlled trials. Thorax 59:777-782. 551

43. Russell, D. W., T. Yamamoto, W. J. Schneider, C. J. Slaughter, M. S. Brown, 552 and J. L. Goldstein. 1983. cDNA cloning of the bovine low density lipoprotein 553 receptor: feedback regulation of a receptor mRNA. Proceedings of the National 554 Academy of Sciences of the United States of America 80:7501-7505. 555

44. Safran, M., W. Y. Kim, F. O'Connell, L. Flippin, V. Gunzler, J. W. Horner, 556 R. A. Depinho, and W. G. Kaelin, Jr. 2006. Mouse model for noninvasive 557 imaging of HIF prolyl hydroxylase activity: assessment of an oral agent that 558 stimulates erythropoietin production. Proceedings of the National Academy of 559 Sciences of the United States of America 103:105-110. 560

45. Sanner, B. M., N. Fluerenbrock, A. Kleiber-Imbeck, J. B. Mueller, and W. 561 Zidek. 2001. Effect of continuous positive airway pressure therapy on infectious 562

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

28

complications in patients with obstructive sleep apnea syndrome. Respiration 563 68:483-487. 564

46. Saraswathi, V., L. Gao, J. D. Morrow, A. Chait, K. D. Niswender, and A. H. 565 Hasty. 2007. Fish oil increases cholesterol storage in white adipose tissue with 566 concomitant decreases in inflammation, hepatic steatosis, and atherosclerosis in 567 mice. J Nutr 137:1776-1782. 568

47. Sayin, S. I., A. Wahlstrom, J. Felin, S. Jantti, H. U. Marschall, K. Bamberg, 569 B. Angelin, T. Hyotylainen, M. Oresic, and F. Backhed. 2013. Gut microbiota 570 regulates bile acid metabolism by reducing the levels of tauro-beta-muricholic 571 acid, a naturally occurring FXR antagonist. Cell Metab 17:225-235. 572

48. Schwarz, M., D. W. Russell, J. M. Dietschy, and S. D. Turley. 1998. Marked 573 reduction in bile acid synthesis in cholesterol 7alpha-hydroxylase-deficient mice 574 does not lead to diminished tissue cholesterol turnover or to 575 hypercholesterolemia. Journal of lipid research 39:1833-1843. 576

49. Shah, Y. M., S. Ito, K. Morimura, C. Chen, S. H. Yim, V. H. Haase, and F. J. 577 Gonzalez. 2008. Hypoxia-inducible factor augments experimental colitis through 578 an MIF-dependent inflammatory signaling cascade. Gastroenterology 134:2036-579 2048, 2048 e2031-2033. 580

50. Shah, Y. M., T. Matsubara, S. Ito, S. H. Yim, and F. J. Gonzalez. 2009. 581 Intestinal hypoxia-inducible transcription factors are essential for iron absorption 582 following iron deficiency. Cell Metab 9:152-164. 583

51. Sherpa, L. Y., Deji, H. Stigum, V. Chongsuvivatwong, O. Luobu, D. S. 584 Thelle, P. Nafstad, and E. Bjertness. 2011. Lipid profile and its association with 585

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

29

risk factors for coronary heart disease in the highlanders of Lhasa, Tibet. High Alt 586 Med Biol 12:57-63. 587

52. Sinal, C. J., M. Tohkin, M. Miyata, J. M. Ward, G. Lambert, and F. J. 588 Gonzalez. 2000. Targeted disruption of the nuclear receptor FXR/BAR impairs 589 bile acid and lipid homeostasis. Cell 102:731-744. 590

53. Summons, R. E., A. S. Bradley, L. L. Jahnke, and J. R. Waldbauer. 2006. 591 Steroids, triterpenoids and molecular oxygen. Philosophical transactions of the 592 Royal Society of London. Series B, Biological sciences 361:951-968. 593

54. Swaner, J. C., and W. E. Connor. 1975. Hypercholesterolemia of total 594 starvation: its mechanism via tissue mobilization of cholesterol. The American 595 journal of physiology 229:365-369. 596

55. Taylor, M., A. Qu, E. R. Anderson, T. Matsubara, A. Martin, F. J. Gonzalez, 597 and Y. M. Shah. 2011. Hypoxia-inducible factor-2alpha mediates the adaptive 598 increase of intestinal ferroportin during iron deficiency in mice. Gastroenterology 599 140:2044-2055. 600

56. Wang, G. L., B. H. Jiang, E. A. Rue, and G. L. Semenza. 1995. Hypoxia-601 inducible factor 1 is a basic-helix-loop-helix-PAS heterodimer regulated by 602 cellular O2 tension. Proceedings of the National Academy of Sciences of the 603 United States of America 92:5510-5514. 604

57. Xiang Xue, S. R., Erik Anderson, Matthew Taylor, Ellen M. Zimmermann, 605 Jason R. Spence, Sha Huang, Joel K. Greenson, Yatrik M. Shah. 2013. 606 Endothelial PAS Domain Protein 1 Activates the Inflammatory Response in the 607 Intestinal Epithelium to Promote Colitis in Mice. Gastroenterology. 608

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

30

58. Xue, X., M. Taylor, E. Anderson, C. Hao, A. Qu, J. K. Greenson, E. M. 609 Zimmermann, F. J. Gonzalez, and Y. M. Shah. 2012. Hypoxia-inducible 610 factor-2alpha activation promotes colorectal cancer progression by dysregulating 611 iron homeostasis. Cancer research 72:2285-2293. 612

613 614

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

31

Figure legends 615 Figure 1 Disruption of hepatic VHL in the liver results in hypercholesterolemia and 616 decreased survival. (A) Serum cholesterol levels in VhlF/F, VhlLivKO, VhlΔM�, and VhlΔInt 617 mice two weeks after tamoxifen treatment. (B) Cholesterol content measured in the liver 618 and (C) primary hepatocytes from VhlF/F and VhlLivKO mice. (D) Lipoprotein profile was 619 assessed in the serum from VhlF/F and VhlLivKO mice two weeks after tamoxifen treatment. 620 (E) Survival curve of VhlF/F and VhlLivKO mice fed with regular chow, high fat, ethanol or 621 atherogenic diet. (F) Serum cholesterol levels measured in VhlF/F and VhlLivKO mice on 622 chow or atherogenic diet for one week. Five to seven mice were assessed per each 623 treatment group. Each bar graph represents the mean value ± SD. (**) vs VhlF/F, p<0.01; 624 (#) vs regular diet, p<0.01. 625 626 Figure 2 Cholesterol synthesis, absorption and mobilization are not affected by the 627 hepatic disruption of VHL. (A) qPCR analysis of cholesterol synthesis genes in VhlF/F 628 and VhlLivKO livers five days or two weeks following tamoxifen treatment. (B) Hepatic 629 cholesterol and (C) LDL receptor expression analyzed and quantitated (bottom inset) in 630 the livers of VhlLivKO. (D) qPCR analysis of Npc1l1 in the small intestine of VhlF/F and 631 VhlLivKO mice two weeks after tamoxifen treatment. mRNA was normalized to β-actin. 632 (E) White adipose tissue cholesterol from VhlF/F and VhlLivKO mice two weeks after 633 tamoxifen treatment. Five to nine mice were assessed per each treatment group. Each bar 634 graph represents the mean value ± SD. (**) p <0.01 compared to VhlF/F mice. 635 636

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

32

Figure 3 Loss of VHL in liver results in repression of CYP7A1 expression. (A) qPCR 637 analysis examining the expression of Cyp7a1 mRNA levels in the livers and primary 638 hepatocytes of VhlLivKO mice. mRNA was normalized to β-actin. Four to six mice were 639 assessed per each treatment group. (B) Western analysis of CYP7A1 in primary 640 hepatocytes from VhlF/F and VhlLivKO two weeks after tamoxifen treatment. (C) Western 641 analysis examining the phosphorylation of JNK in the liver of VhlLivKO five or two weeks 642 after tamoxifen treatment. (D) qPCR analysis examining the expression of Cyp7a1 643 mRNA levels in the primary hepatocytes of VhlLivKO treated with JNK inhibitor, SP 644 600125 or (E) with recombinant adenovirus expressing HNF-4α or LRH-1 and mRNA 645 was normalized to β-actin. (*) vs VhlF/F, p <0.05. 646 647 Figure 4 Loss of VHL disrupts circadian expression of CYP7A1. (A) qPCR and 648 Western blot analysis of CYP7A1 in microsome fraction and (B) qPCR analysis of 649 cholesterol hydroxylases at 7AM and 7PM in VhlF/F and VhlLivKO livers, five days after 650 Vhl disruption. (C) qPCR and Western blot analysis of CYP7A1 at 7AM and 7PM in 651 VhlF/F and VhlLivKO livers, two weeks after Vhl disruption. mRNA was normalized to β-652 actin and microsome fraction was normalized by coomassie blue staining. Four to six 653 mice assessed per each treatment group. (D) Western blot analysis examining the 654 circadian expression of Rev-erbα in VhlF/F and VhlLivKO livers, five days or two weeks 655 after Vhl disruption. (E) qPCR analysis of circadian genes in VhlF/F and VhlLivKO livers, 656 five days after Vhl disruption. mRNA was normalized to β-actin. Each bar graph 657 represents the mean value ± SD. (*) vs VhlF/F, p<0.05; (**) vs VhlF/F, p<0.01; (#) vs 7am, 658 p<0.01. 659

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

33

Figure 5 Adenoviral rescue of CYP7A1 did not ameliorate cholesterol levels in 660 primary hepatocytes. (A) Western blot analysis of CYP7A1 and (B) cholesterol assay 661 in primary hepatocytes treated with recombinant adenovirus expressing Cyp7a1 in 662 primary hepatocytes from VhlF/F and VhlLivKO mice. (C) qPCR analysis of cholesterol 663 hydroxylases in VhlF/F and VhlLivKO livers, two weeks after Vhl disruption. mRNA was 664 normalized to β-actin. Each bar graph represents the mean value ± SD. (*) vs VhlF/F, 665 p<0.05; (**) vs VhlF/F, p<0.01; (#) vs 7am, p<0.01. 666 667 Figure 6 Disruption of liver VHL decreases bile acid and FXR signaling. (A) Hepatic 668 bile acids in VhlF/F and VhlLivKO mice two weeks after tamoxifen treatment. (B) Gross 669 morphology of the liver and gall bladder in VhlF/F and VhlLivKO mice two weeks after 670 tamoxifen treatment. (C) Bile acid species quantitation in livers of VhlF/F and VhlLivKO 671 mice two weeks after tamoxifen treatment. (D) FXR target genes were assessed by qPCR 672 in the liver and ileum of VhlF/F and VhlLivKO mice two weeks following tamoxifen 673 treatment. Expression was normalized to β-actin. Five to eight mice were assessed per 674 each treatment group. Each bar graph represents the mean value ± SD. (E) Western blot 675 analysis of FXR in livers of VhlF/F and VhlLivKO mice, two weeks after tamoxifen 676 treatment. (*) vs VhlF/F, p <0.05; (**) vs VhlF/F, p <0.01. 677 678 Figure 7 Liver HIF-2α induces hypercholesterolemia and inhibits bile acids 679 synthesis. (A) Serum and (B) liver cholesterol analyzed in VhlF/F, VhlLivKO, Vhl/Hif1αF/F, 680 Vhl/Hif1αLivKO, Vhl/Hif2αF/F, and Vhl/Hif2αLivKO mice two weeks following tamoxifen 681 treatment. (C) Western blot analysis examining the expression of CYP7A1 and Rev-erbα 682

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

34

in the microsome fraction and nuclear lysates, and qPCR analysis for (D) bile acid 683 synthesis genes and (E) FXR target genes in livers of VhlF/F, VhlLivKO, Vhl/Hif2αF/F and 684 Vhl/Hif2αLivKO mice two weeks after tamoxifen treatment. Expression was normalized to 685 β-actin. Five to nine mice assessed per each treatment group. Each bar graph represents 686 the mean value ± SD. (**) vs VhlF/F, p <0.01. 687 688 Figure 8 Chronic hypoxia and lung contusion increases cholesterol levels and 689 decreases in bile acid synthesis in mice. (A) In vivo imaging in Odd-luc mice 690 demonstrating systemic hypoxia 24-hours following lung contusion. (B) Serum 691 cholesterol, liver cholesterol and total liver bile acid measured at 5-, 24-, 48- and 96- 692 hours after induction of lung injury. (C) Western blot analysis of CYP7A1 in the 693 microsome fractions 24-hours after lung injury. Expression was normalized to coomassie 694 blue staining. (D) qPCR analysis in livers at 5-, 24- and 48-hours after lung injury. 695 Expression was normalized to β-actin. Three to six mice assessed per each treatment 696 group. (E) Serum cholesterol, liver cholesterol and hepatic total bile acids measured in 697 C57BL/6 mice exposed to 10% O2 for three weeks. (F) qPCR analysis of bile acid 698 synthesis and circadian genes in mice exposed to 10% O2 for three weeks. Expression 699 was normalized to β-actin. Four mice were assessed per each treatment group. Each bar 700 graph represents the mean value ± SD. (*) vs control, p <0.05; (**) vs control, p <0.01. 701

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

on April 16, 2018 by guest

http://mcb.asm

.org/D

ownloaded from