Niacin fine-tunes energy homeostasis through canonical ... · 5 81 orphan receptor GPR109A (recently renamed as hydroxyl. carboxylic acid receptor 2 or HCA2) 82. was identified as

1

Niacin fine-tunes energy homeostasis through canonical 1

GPR109A signaling 2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which was notthis version posted August 1, 2018. . https://doi.org/10.1101/382416doi: bioRxiv preprint

https://doi.org/10.1101/382416

2

Abstract 21

Niacin has long been considered as a high-potency drug for beneficially treating lipid abnormalities, 22

however, its anti-atherosclerotic effects have been challenged by recent studies. Here, we 23

demonstrated that oral supplementation of niacin resulted in a significant reduction in body weight 24

and fat mass without affecting food intake in high-fat diet-fed wild-type mice, but not in 25

GPR109A-defeicient mice. Further investigation showed that niacin challenge led to a remarkable 26

inhibition of hepatic lipogenesis via a GPR109A-dependent ERK1/2/AMPK pathway. Additionally, 27

we demonstrated that niacin treatment stimulated thermogenesis in brown adipose tissue by 28

induction of thermogenic genes via GPR109A. Moreover, we observed that mice exposed to niacin 29

exhibited a dramatic decrease in intestinal absorption of fatty acids. Together, our data 30

demonstrate that acting on GPR109A, niacin shows the potential to maintain energy homeostasis 31

by fine-tuning hepatic lipogenesis, BAT/beige thermogenesis and intestinal fat absorption, 32

representing a potential approach to the treatment of lipid abnormalities. 33

Keywords: de novo lipogenesis/ GPR109A/ intestinal fat absorption/ obesity/ thermogenesis 34

35

36

37

38

39

40


https://doi.org/10.1101/382416

3

Abrreviations 41

NA, nicotinic acid; HFD, high fat diet; NCD, normal chow diet; TG, triglyceride; TC, total cholesterol; 42

FFA, free fatty acid; LDL, low-density lipoprotein; HDL, high density lipoprotein; WAT, white 43

adipose tissue; eWAT, epididymal white adipose tissue; pWAT, perirenal white adipose tissue；44

iWAT, inguinal white adipose tissue; sWAT, subcutaneous white adipose tissue; BAT, brown 45

adipose tissue; ITT, insulin tolerance test; GTT, glucose tolerance test; PDH, pyruvate 46

dehydrogenase; ACC, acetyl CoA carboxylase; ACC1, acetyl CoA carboxylase 1; DGAT2, 47

diacylglycerol acyltransferase 2; FAS: fatty acid synthase; SREBP-1c, sterol regulatory element- 48

binding transcription factor 1; SCD-1, stearoyl-CoA desaturases-1; FATP4, fatty acid transport 49

protein 4; I-FABP, intestinal-type fatty acid-binding protein; NPC1L1, Niemann-Pick C1-Like 1; 50

UCP1, uncoupling protein 1; PGC-1α, peroxisome proliferator-activated receptor gamma 51

coactivator 1-alpha; PRDM16: PR domain containing 16; AUC, area under the curves. 52

53

54

55

56

57

58

59

60


https://doi.org/10.1101/382416

4

Introduction 61

The incidence of overweight and obesity has become a global epidemic, with over 1.9 billion 62

overweight and 600 million obese individuals worldwide (Ng et al, 2014; Wang et al, 2011). Obesity 63

is caused by an abnormal excess storage of energy as lipids in adipose tissue due to a net 64

imbalance between energy intake and energy expenditure (Lowell & Spiegelman, 2000; Olsen et 65

al, 2017). Obesity is considered a major risk factor for numerous comorbidities including major 66

diseases such as cardiovascular disease, diabetes, and cancer (Haslam & James, 2005; 67

Kopelman, 2000). However, despite the recent tremendous efforts, effective antiobesity 68

pharmacotherapies are still limited. Additionally, to date, current antiobesity agents are designed to 69

reduce food-intake and appetite through the hypothalamus-based molecular pathways, but 70

unfortunately, these agents exhibit severe psychiatric or cardiovascular side effects (Cunningham 71

& Wiviott, 2014; Manning et al, 2014; Moreira et al, 2009). Therefore, the magnitude of this current 72

health epidemic has heightened the need for alternative therapeutic strategies aiming to control 73

body weight through potential targets in peripheral non-neuronal tissues. 74

75

Since discovery of decreasing plasma cholesterol levels in 1955, nicotinic acid (niacin) has been 76

used to treat dyslipidemia as an approved drug for more than half a century (Altschul et al, 1955; 77

Carlson, 2005). A number of randomized clinical trials demonstrated that niacin monotherapy has 78

been shown to substantially increase levels of HDL-C and decrease levels of triglycerides (Elam et 79

al, 2000; Guyton et al, 1998). However, its exact mechanism was not well understood until the 80


https://doi.org/10.1101/382416

5

orphan receptor GPR109A (recently renamed as hydroxyl-carboxylic acid receptor 2 or HCA2) 81

was identified as a receptor with high affinity for niacin in 2003, independently by three research 82

groups (Offermanns et al, 2011; Soga et al, 2003; Tunaru et al, 2003; Wise et al, 2003). 83

Subsequently, the ketone body β-hydroxy butyrate was identified as an endogenous ligand for 84

GPR109A (Taggart et al, 2005). In adipocytes, upon stimulation by niacin, GPR109A functions via 85

Gi/o proteins to lower intracellular cAMP production, leading to decreased protein kinase 86

A-mediated activation of hormone-sensitive lipase (HSL), which in turn reduces triglyceride 87

hydrolysis to free fatty acids (FFA) (Digby et al, 2009). This FFA hypothesis is supported by 88

evidence that niacin failed to lower FFA and TGs in GPR109A-knockout mice (Tunaru et al, 2003). 89

90

The established clinical benefits of niacin treatment on cardiovascular events have been 91

challenged by recent studies showing that niacin exerts its beneficial effects on 92

GPR109A-mediated anti-atherosclerotic activity independently from its anti-lipolytic properties 93

(Lukasova et al, 2011a; Wu et al, 2010). However, these data did not exclude the possibility that 94

niacin provides cardiovascular benefits in an anti-lipolysis-independent manner through GPR109A. 95

Niacin-mediated beneficial cardiovascular effect has been shown to be transferable with 96

GPR109A-competent bone marrow cells (Wu et al, 2010). In addition, accumulating evidence has 97

demonstrated that activation of GPR109A was able to induce anti-inflammatory effects not only on 98

atherosclerosis, also on acute ischemic stroke, arthritis, chronic renal failure and sepsis through 99

inhibition of immune cell chemotaxis and pro-inflammatory cytokine production (Chen et al, 2014; 100


https://doi.org/10.1101/382416

6

Cho et al, 2009; Digby et al, 2012; Kwon et al, 2011; Lukasova et al, 2011a; Mitrofanov et al, 2005; 101

Rahman et al, 2014; Shehadah et al, 2010; Shi et al, 2017; Wu et al, 2010). Therefore, more 102

studies are needed to thoroughly evaluate the GPR109A-mediated pleiotropic effects on 103

atherogenesis and other metabolic diseases. 104

105

GPR109A-induced inhibition in adipocyte lipolysis and fatty acid mobilization in response to 106

long-term treatment with niacin would result in excessive synthesis of TGs and obesity (Ganji et al, 107

2004). However, during breeding process of GPR109A-knockout mice we introduced from Dr. 108

Offermanns’ lab, we found that the GPR109A-knockout mice are more prone to being obese 109

compared to the wild-type mice. Therefore, we initiated this study to investigate roles of 110

niacin/GPR109A in the regulation of lipid metabolism using GPR109A-knockout mice combined 111

with high fat diet (HFD)-induced obesity mouse model. We show that upon activation by niacin, 112

GPR109A fine-tunes lipid metabolism through multipathways including inhibition of hepatocyte 113

lipogenesis and fatty acid absorption and promotion of brown adipose tissue (BAT) thermogenesis. 114

115

Results 116

GPR109A deficient animals display obesity 117

To explore GPR109A function in lipid metabolism, parallel experiments in mice with GPR109A 118

(wild-type) and without GPR109A (Gpr109a-/-

) were performed. When both wild-type and 119

Gpr109a-/-

male mice were fed with either high-fat diet (HFD) or normal-chow diet (NCD) for 12 120


https://doi.org/10.1101/382416

7

weeks, we recorded weight gain weekly. As the feeding progressed, starting from approximately 121

10-12 weeks of age, the GPR109A-knockout mice exhibited significantly increased body weight 122

compared with the wild-type mice (Figs 1A and C, and EV1A and C). In addition, we monitored 123

food intake weekly throughout experimental period, if any, not significant difference in food intake 124

between GPR109A-knockout and wild-type mice was observed (Figs 1B and EV 1B). After 12 125

weeks of normal-chow diet feeding, adipose tissue weights and organ weights were examined. 126

Our data demonstrated that GPR109A-deficient mice displayed a significant increase in adipose 127

tissue weights including epididymal, perirenal and inguinal white adipose tissue and liver weight, 128

but not in BAT or other organ weights (kidney, lung, heart and spleen) when compared to wild-type 129

mice (Figs 1E and EV1D). Moreover, H&E stained histological sections of eWAT and electron 130

microscope observation of liver and muscle revealed that GPR109A-deficient mice had a striking 131

visible increase in amount and size of lipid droplets in liver, muscle and eWAT compared with the 132

wild-type group (Fig 1D). 133

134

Niacin exerts inhibitory effects on HFD-induced obesity 135

To assess the effect of niacin on lipid metabolism, wild-type mice were fed with a 6-week high-fat 136

diet (HFD) to induce obesity. HFD supplemented with niacin (HFD + niacin, 500 mM) or vehicle 137

(HFD + water) was then treated during the following 8 weeks. As anticipated, the mice maintained 138

on the HFD exhibited a progressive increase in body weight gain. Treatment with niacin resulted in 139

a significant reduction in body weight, but weekly food intake remained unchanged as compared to 140


https://doi.org/10.1101/382416

8

vehicle-treated mice (Fig 2A, B and C). Mice fed with HFD in response to challenge with niacin 141

exhibited a significant reduction in adipose tissue weights including epididymal, perirenal and 142

inguinal white adipose tissue and liver weight, but not in BAT or other organ weights (kidney, lung, 143

heart and spleen) with respect to vehicle-treated mice (Fig 2E). Histological examination showed 144

that niacin challenge led to a marked decrease in lipid accumulation in liver, muscle and eWAT (Fig 145

2D). 146

147

Niacin has profound and unique beneficial effects on blood lipid (Carlson, 2005). We then 148

examined effects of niacin challenge on metabolic parameters in serum of experimental mice. As 149

shown in Fig 2, challenge with niacin led to a significant reduction in serum total cholesterol (TC), 150

low-density lipoprotein (LDL) and high-density lipoprotein (HDL) with a characteristic of high ratio 151

of HDL/LDL in mice fed on HFD, but triglyceride (TG) remained unchanged in respect to 152

vehicle-treated mice. The effect of niacin on TC, LDL, HDL and ratio of HDL/LDL was not observed 153

in Gpr109a-/-

mice and mice challenged with CHBA, a specific small molecule agonist for GPR81 154

(Figs 2F and EV2A). We also performed glucose tolerance test (GTT) and insulin tolerance test 155

(ITT) and documented that niacin treated mice were less glucose tolerant, but exhibited insulin 156

tolerant comparable to control mice (Fig 2G). No changes on GTT and ITT were detected in 157

niacin-treated Gpr109a-/-

mice and CHBA-challenged wild-type mice (Fig EV2B). Quantitative 158

analysis showed that niacin treatment results in a significant reduction in blood insulin (Fig 2H), 159

however, challenge of HFD-fed mice with niacin exhibited a markedly decrease in HOMA-IR 160


https://doi.org/10.1101/382416

9

values compared to vehicle-treated mice (Fig 2I), suggesting the beneficial role of niacin in insulin 161

sensitivity. 162

163

Lack of GPR109A prevents anti-obesity actions of niacin 164

To further validate the role of GPR109A in the anti-obesity action of niacin, we challenged HFD-fed 165

Gpr109a-/-

male mice with niacin. As indicated in Fig 3, Challenge of HFD-fed Gpr109a-/-

mice with 166

niacin exhibited similar weight gain, adipose tissue weight, organ weight, and food intake as 167

respect to vehicle-treated Gpr109a-/-

mice (Fig 3A-C, and E). Histological observation also showed 168

no difference in lipid accumulation between niacin-treated and vehicle-treated Gpr109a-/-

mice (Fig 169

3D). These results suggest that niacin exerts its regulatory effects on lipid metabolism through 170

GPR109A. 171

172

Niacin directly inhibits hepatocyte lipogenesis through GPR109A 173

Our data clearly showed that upon activation by niacin, GPR109A exerts its inhibitory effects on 174

lipid accumulation in adipose tissue, liver and muscle when fed on a HFD. Moreover, oil red O 175

staining and quantitative analysis revealed that HFD-fed mice displayed a significant increase in 176

hepatic triglyceride content, whereas niacin supplement led to a marked reduction in triglyceride 177

content in liver (Fig 4A and B), in contrast, no difference in TC and FFA levels was observed (Fig 178

4B). In addition, the decrease in the hepatic insulin level was observed in mice with niacin 179

challenge (Fig 4C), coincident with the result in serum. However, animals remain unaltered in food 180


https://doi.org/10.1101/382416

10

intake. We then investigate whether or not niacin plays a role in the regulation of lipogenesis via 181

GPR109A. We first used qRT-PCR to quantitatively analyze the GPR109A expression in mouse 182

liver. 183

184

It has been reported that GPR109A is highly expressed in adipocytes and activated immune cells 185

(Carlson, 2005; Soga et al, 2003; Tunaru et al, 2003; Wise et al, 2003). However, our data clearly 186

showed that when fed with HFD, mice exhibited a significant increase in lipid accumulation in liver, 187

whereas niacin supplement resulted in a significant reduction of lipid accumulation in liver. To 188

confirm the role of GPR109A in liver, we examined the expression of GPR109A in different tissues 189

and organs. High-level expression of GPR109A mRNA was detected in epididymal WAT and BAT 190

(Fig 4D). Interestingly, although relatively lower basal expression of GPR109A was detected, upon 191

challenge with HFD, GPR109A mRNA was markedly upregulated in liver, while niacin stimulation 192

showed no effect on GPR109A expression in different tissues and organs (Fig 4E and EV3A). In 193

contrast, GPR109A was completely absent in liver from Gpr109a-/-

mice (Fig EV3B and C). 194

GPR109A expression in liver was also confirmed at protein level by western blot (Fig 4F). However, 195

our data on GPR109A expression in mouse liver could not discriminate between the hepatocytes 196

and Kupffer cells. Therefore, to confirm the expression of GPR109A in hepatocytes, we further 197

analyzed the expression and function of GPR109A in HepG2 cells, a human liver carcinoma cell 198

line endogenously expressing GPR109A (Fig EV4A). As anticipated, niacin (300 μM) induced 199

remarkable phosphorylation of ERK1/2 in a time-dependent manner with a maximal activation at 5 200


https://doi.org/10.1101/382416

11

min (Fig 4B). In addition, niacin (100 μM) induced Ca2+

mobilization in HepG2 cells (Fig EV4C). 201

These data suggest that GPR109A is functionally expressed in hepatocytes. 202

203

To confirm the role of GPR109A in the regulation of de novo lipogenesis in liver, we used the 204

ELISA to quantitatively analyze the activities of key enzymes involved in fatty acid biosynthesis, 205

acetyl CoA carboxylase (ACC), acetyl CoA carboxylase-1 (ACC-1), and pyruvate dehydrogenase 206

(PDH) (Tong, 2005). As shown in Fig 4G, in wild-type mice fed with HFD, treatment with niacin 207

resulted in a significant decrease in the enzyme activities of ACC, ACC1, and PDH in liver, while 208

no change in WAT was observed. However, in Gpr109a-/-

mice, no difference in the enzyme 209

activities of PDH and ACC in liver followed treatment with niacin was detected (Fig 4H). Our data 210

also showed that niacin supplement led to no statistically significant changes in the activity of 211

diglyceride acyltransferase 2 (DGAT2) (Fig 4G), which directly catalyzes the formation of 212

triglycerides from diacylglycerol and Acyl-CoA in fat, liver and skin cells (Oelkers et al, 1998). In 213

addition, these data were further confirmed by quantitative determination of mRNA transcripts of 214

sterol regulatory element-binding protein-1c (SREBP-1c), a key transcriptional activator of 215

lipogenic genes, and its downstream target genes, fatty acid synthase (FAS), stearoyl-CoA 216

desaturases-1 (SCD-1) and acetyl CoA carboxylase-1 (ACC-1), by real time RT-PCR. As revealed 217

in Fig 4I, administration of niacin resulted in a significant decrease in mRNA levels of SREBP-1c 218

and its target genes, FAS, SCD-1 and ACC-1, in liver compared to the vehicle-treated mice. 219

220


https://doi.org/10.1101/382416

12

AMPK and ERK1/2 are involved in the GPR109A-mediated inhibition of hepatocyte 221

lipogenesis 222

Next, to elucidate the mechanism(s) involved in niacin-initiated inhibition of hepatocyte lipogenesis 223

through GPR109A, western blot analysis was performed to determine the activation of AMP kinase 224

(AMPK), extracellular signal-regulated kinase (ERK) and protein kinase B (Akt). As shown in Fig 225

5A, niacin-treated wild-type mice exhibited a markedly enhancement in phosphorylation of AMPK, 226

ERK1/2 and Akt in liver, while only phosphorylation of AMPK in muscle, but not in eWAT. However, 227

niacin-treated Gpr109a-/-

mice had no detectable differences in the activation of AMPK, ERK1/2 228

and Akt in liver (Fig 5B). The cultured HepG2 cells, a human liver carcinoma cell line 229

endogenously expressing GPR109A (Fig EV4A), were used to further examine the 230

GPR109A-mediated intracellular signaling. As shown in Fig 5C, as anticipated, niacin (300μM) 231

induced remarkable phosphorylation of AMPK, ERK1/2, and Akt in a time-dependent manner with 232

a maximal activation at 5 min. Accordingly, phosphorylation and inactivation of ACC, a 233

downstream target of AMPK, was observed (Fig 5C). Consistently, siRNA-mediated knockdown of 234

GPR109A expression led to the significant impairment of niacin-induced ERK1/2, AMPK, and ACC 235

phosphorylation and Ca2+

mobilization in HepG2 cells (Figs 5D and EV4D-F). Taken together, 236

these data suggest that niacin exerts its inhibitory effect on ACC through GPR109A-mediated 237

activation of AMPK. 238

239

We used western blot analysis combined with various inhibitors of G proteins and kinases to 240


https://doi.org/10.1101/382416

13

explore the role of ERK1/2 and/or AKT in the activation of AMPK. Our results showed that 241

inhibitors of PTX for Gαi, M119K for Gβγ, Go6983 for PKC, and U0126 for MEK exhibiting 242

inhibitory effects on ERK1/2 phosphorylation exerted the same inhibitory activities on AMPK (Fig 243

5E). Collectively, it is likely that niacin exhibits inhibitory effect on hepatic lipid synthesis through 244

activation of AMPK via Gβγ-PKC-dependent ERK1/2 signaling pathway, as the schematic model 245

shown (Fig 5F). 246

247

Niacin treatment triggers BAT thermogenesis via GPR109A 248

Obesity is an abnormal accumulation of lipid in adipose tissue caused by an imbalance between 249

energy intake and expenditure. In HFD-induced obesity mouse model, niacin supplement led to a 250

significant reduction in body weight, but no change in daily food intake, indicating that niacin is 251

more likely to promote energy expenditure. To validate this hypothesis, we first monitored the core 252

body temperature by measuring the anal temperature of mice fed an HFD and treated with vehicle 253

or niacin upon acute cold exposure. GPR109A-deficient mice showed a continuous decline in 254

body temperature and reached 32°C after acute cold exposure at 4°C for 6 hours, while wild-type 255

mice were able to maintain their body temperature at 33.5°C (Fig 6A). Moreover, wild-type mice 256

fed an HFD and treated with niacin revealed relatively but significantly higher body temperature 257

than that of vehicle-treated mice, however, no significant difference was observed between niacin- 258

and vehicle-treated group in GPR109A-deficient mice in response to acute cold exposure (Fig 6A) 259

and in wild-type mice without acute cold exposure (Fig 6B). These data collectively confirm cold 260


https://doi.org/10.1101/382416

14

tolerance in niacin-challenged wild-type mice. 261

262

Brown adipose tissue (BAT) is the major site essential to generate heat (thermogenesis) for the 263

maintenance of body temperature. Therefore, we next examined the expression of genes involved 264

in BAT thermogenesis real time RT-PCR analysis. As shown in Fig 6C, the mRNA expression of 265

uncoupling protein 1 (UCP1), a critical player in allowing electrons to be released as heat rather 266

than stored, and positive regulatory domain 16 (PRDM16), a well-known BAT transcriptional 267

regulator, were significantly up-regulated in HFD-fed wild-type mice in response to challenge of 268

niacin. This is further confirmed by observation of the ultrastructure of BAT mitochondria. 269

Transmission electron microscopy demonstrated that BAT from Gpr109a-/-

mice exhibited 270

significant reduction in number and size of mitochondria with aberrant cristae compared to 271

wild-type mice. Niacin challenge caused a remarkable increase in number and size of 272

mitochondria with improved cristae organization in HFD-fed wild-type mice (Fig 6D). Additionally, 273

we also observed that niacin showed the potential to rapidly elevate the expression of 274

beige-specific genes including Cd137, Tbx1, and Tmem26 in sWAT (Fig 6E). These data 275

collectively suggest that niacin stimulates BAT energy expenditure through GPR109A in HFD-fed 276

mice. 277

278

Additionally, we also evaluated the role of niacin/GPR109A signaling in the preadipocyte 279

differentiation. Niacin showed stimulatory effect on the differentiation-induction of brown fat 280


https://doi.org/10.1101/382416

15

precursor cells with a maximal activity at 250 μM (Fig 6F). Further real-time RT-PCR analysis 281

revealed that niacin treatment significantly up-regulated the transcript levels of UCP1, PGC-1a 282

and PRDM16 (Fig 6G). 283

284

Niacin stimulation decreases fatty acid absorption though GPR109A 285

Recent studies have demonstrated that GPR109A is expressed in colonic and intestinal epithelial 286

cells and mediates the local control of intestinal glucose absorption (Wong et al, 2015) and the 287

tumor-suppressive effects of the bacterial fermentation product butyrate in colon (Thangaraju et al, 288

2009). Accordingly, we further examined the role of GPR109A expressing in colonic and intestinal 289

epithelial cells in the control of intestinal fatty acid absorption. We collected feces from mice 290

housed individually from 5 consecutive days and measured fecal lipid contents. As indicated in Fig 291

7, the weights of feces normalized to body weight in the niacin-challenged mice when fed on HFD 292

compared to the control mice (Fig 7A). Further analysis showed that total lipid, TG, TC and FFA in 293

feces were significantly increased when HFD-fed mice challenged by niacin compared to the 294

control mice. However, total lipid in feces was reduced in HFD-fed Gpr109a-/-

mice treated with 295

niacin, no difference in TG, TC and FFA was observed between niacin-challenged mice and control 296

mice (Fig 7B). This was confirmed by RT-PCR quantitative analysis of three key proteins (FATP4, 297

I-FABP, and NCP1L1) involved in the control of fatty acid and cholesterol absorption. The mRNA 298

levels of the three genes were obviously decreased in the wild-type NA group, but not in the 299

Gpr109a-/-

mice (Fig 7C). These findings suggest that niacin is likely to exhibit the potential to 300


https://doi.org/10.1101/382416

16

inhibit the dietary fat absorption through GPR109A, thereby causing the decrease in fat deposit 301

and body weight. 302

303

Discussion 304

Since the identification of GPR109A, a Gi/o-coupled receptor, as the molecular target for niacin 305

(Digby et al, 2012; Ganji et al, 2009; Lukasova et al, 2011b), most studies have focused on its 306

pronounced hypolipidemic and cardioprotective action. Recent investigations have also shed light 307

on anti-inflammatory and anti-oxidative effects (Buhler et al, 2018; Digby et al, 2012; Gautam et al, 308

2018; Graff et al, 2016; Singh et al, 2014). However, little attention has been paid to the potential of 309

this receptor to favorably modulate lipogenesis and energy homeostasis. In the present study, we 310

explored physiological roles of niacin/GPR109A in the regulation of lipid metabolism and energy 311

homeostasis in a high-fat diet-induced murine model of obesity combined with GPR109A deficient 312

mice. The most important observation is that niacin supplementation prevented the progression of 313

dietary obesity in HFD-fed wild mice, but not in HFD-fed GPR109A deficient mice. Mechanistically, 314

we demonstrated that niacin-mediated reduction in body weight and fat mass was caused by 315

fine-tuning of hepatic de novo lipogenesis, BAT thermogenesis and intestinal fatty acid absorption 316

through GPR109A. 317

318

During the breeding process of GPR109A knockout mice introduced from Dr. Offermanns’ lab, we 319

observed that the deletion of GPR109A resulted in obvious increase in body weight and fat mass 320


https://doi.org/10.1101/382416

17

compared to wild-type mice. In contrast, lactate receptor GPR81-deficient mice gain significantly 321

less body weight than their wild-type under HFD-feeding conditions, although Gi/o-coupled GPR81 322

exerts same activity in the inhibition of lipolysis in white adipose tissue (Ahmed et al, 2010). This 323

prompted us to initiate this study to evaluate the exact role of GPR109A in the regulation of energy 324

homeostasis. A recent study indicated that niacin treatment remarkably reduced body weight in the 325

obese mice (Chen et al, 2015b). Previous investigations have demonstrated that the ketone body 326

β-hydroxybutyrate (βOHB), as an important metabolic intermediate and also a gut bacterial 327

fermentation-yielded product, which is the only known endogenous ligand for GPR109A (Taggart 328

et al, 2005), causes reductions of body weight and fat content in rats and mice (Davis et al, 1981; 329

Gao et al, 2009). Using combination of GPR109A deficient mice with HFD-induced model of obese, 330

we provided further evidence that niacin treatment led to a significant reduction in body weight and 331

fat mass through GPR109A when fed on normal chow diet or HFD. These findings suggest that 332

GPR109A has the potential to control energy homeostasis, preventing weight gain. 333

334

Our results clearly revealed that niacin challenge remarkably reduced liver weight in wild-type 335

mice fed on HFD, but not in GPR109A-deficient mice. Ultrastructural examination validated that 336

GPR109A-deficient mice had a striking visible increase in amount and size of lipid droplets in liver 337

compared with the wild-type group, and niacin treatment led to a significant reduction in amount 338

and size of lipid droplets in liver when fed on HFD. These findings suggest that GPR109A play an 339

important role in the regulation of de novo lipogenesis in liver. The liver has long been recognized 340


https://doi.org/10.1101/382416

18

as the major site of de novo lipogenesis, responsible for the conversion of excess dietary 341

carbohydrate into triglycerides. De novo lipogenesis turnover accounts for about 20% of lipid 342

turnover within adipose tissue (Strawford et al, 2004), however, De novo lipogenesis in 343

non-adipose tissues leads to ectopic lipid accumulation responsible for metabolic stress and 344

insulin resistance (Solinas et al, 2015). Although expression of GPR109A in liver remains 345

controversial, the abundance of GPR109A mRNA was detected higher in liver in lean-type than in 346

fat-type cows (Mielenz et al, 2013). GPR109A has been found to be expressed in murine liver but 347

the basal expression is low (Li et al, 2010). In the current study, we provided evidence to confirm 348

that GPR109A mRNA and protein were detected in murine liver at relatively lower level, however, 349

mice fed on HFD exhibited significantly enhanced expression of GPR109A in the liver. In addition, 350

the functional expression of GPR109A in HepG2 cells was also detected. Further biochemical 351

analysis of key enzymes involved in lipogenesis demonstrated that niacin exhibited inhibitory 352

effects on PDH, the gatekeeping enzyme of pyruvate flux into acetyl-CoA，and ACC1, the key 353

regulatory step of fatty acid synthesis (Katsurada et al, 1990; Ruderman et al, 2003). These 354

findings were verified by real-time qRT-PCR for genes encoding SREBP-1c, a key transcriptional 355

activator of lipogenic genes，and lipogenic enzymes, FAS, SCD1 and ACC1 (Rui, 2014; Shi & Burn, 356

2004; Titchenell et al, 2015). Niacin has been shown to directly inhibit the activity of diglycerol 357

acyltransferase 2 (DGAT2), the key enzyme that catalyzes the final step in triglyceride synthesis in 358

liver (Ganji et al, 2004; Kamanna et al, 2013). However, our results strongly suggest that niacin 359

exerts its inhibitory effects on hepatic lipogenesis through GPR109A-mediated signaling. 360


https://doi.org/10.1101/382416

19

361

Moreover, mechanistically, we demonstrated that Akt, ERK1/2 and AMPK were involved in 362

niacin/GPR109A-mediated inhibition of hepatic lipogenesis. AMP-activated protein kinase (AMPK), 363

a serine/threonine kinase, has been recognized as an intracellular energy sensor for maintaining 364

energy homeostasis (Shackelford & Shaw, 2009). AMPK activation is associated with the hepatic 365

inhibition of de novo lipogenesis by modulating transcription factor SREBP-1c and lipogenic key 366

enzyme ACC1 (Li et al, 2011b; Zhou et al, 2001). To gain further insight into the mechanism 367

underlying the GPR109A-mediated inhibition of hepatic lipogenesis and lipid accumulation, we 368

examined the GPR109A-midiated signaling pathways responsible for the niacin-triggered AMPK 369

activation. We demonstrated that MEK inhibitor U0126 significantly suppressed ERK1/2 370

phosphorylation and AMPK activation, suggesting that the activation of ERK1/2 is likely upstream 371

of AMPK activation. Additionally, treatment with Gi/o inhibitor PTX, Gβγ inhibitor M119K and PKC 372

inhibitor Go6983, respectively, resulted in a remarkable decrease in both ERK1/2 and AMPK 373

phosphorylation. Collectively, these findings suggest that niacin causes inhibition of hepatic 374

lipogenesis via GPR109A-dependent Gβγ/PKC/ERK1/2/AMPK signaling pathway. 375

376

It is noteworthy that mice challenged with niacin exhibited a remarkable reduction in body weight 377

and fat mass without affecting food intake, consistent with previous observation that βOHB 378

supplementation elicited body weight loss but without a decrease in food intake (Davis et al, 1981; 379

Gao et al, 2009). It suggests that energy expenditure is likely to contribute to niacin-induced body 380


https://doi.org/10.1101/382416

20

weight loss. Results derived from cold response experiments showed that thermogenic function is 381

enhanced in wild-type mice challenged with niacin, but not in GPR109A knockout mice. Real time 382

RT-PCR analysis demonstrated that treatment with niacin increased the expression of 383

thermogenic genes UCP1, a thermogenic protein exclusively expressed in brown/beige 384

adipocytes (Cannon & Nedergaard, 2004; Gesta et al, 2007)，PGC-1α，a master regulator of 385

mitochondrial biogenesis (Cohen et al, 2014), and PRDM16, a well-known BAT transcriptional 386

regulator (Harms & Seale, 2013) in mice fed on HFD, and in MDI-induced brown fat precursor cells. 387

Interestingly, we found that mitochondria in BAT of GPR109A knockout mice exhibited smaller in 388

size with abnormal cristae structures compared to wild-type mice, and niacin challenge showed 389

significant improvement in mitochondrial size and cristae structure of mice fed on HFD. 390

Additionally, niacin stimulated a beige fat thermogenic program of the sWAT. These data strongly 391

suggest that niacin treatment triggers BAT and/or browning WAT thermogenic activity in mice 392

through GPR109A. 393

394

Importantly, GPR109A has been shown to functionally express in colonic and intestinal epithelial 395

cells (Thangaraju et al, 2009; Wong et al, 2015). Niacin has been found to locally control glucose 396

uptake at the level of the small intestine though GPR109A (Wong et al, 2015). This prompts us to 397

examine whether or not GPR109A is involved in the control of uptake of sterols and fatty acids in 398

the level of the small intestine. To our surprise, niacin supplementation caused a significant 399

increase of total lipids, triglycerides, cholesterols, and free fatty acids (FFAs) in the feces of 400


https://doi.org/10.1101/382416

21

wild-type mice fed on HFD, but no change was observed in HFD-fed GPR109A-deficient mice. 401

Further gene expression analysis showed that genes encoding FATP4, I-FABP and NPC1L1 402

exhibited a significant reduction in expression in the intestinal tissues. It is established that 403

polytopic transmembrane protein NPC1L1 is essential for intestinal absorption of cholesterol 404

(Altmann et al, 2004), whereas proteins FATP4 and I-FABP play a role in dietary fat absorption and 405

fatty acid transport, respectively (Besnard et al, 2002; Hall et al, 2005). Taken together, these 406

findings suggest that in addition to the control of glucose uptake, GPR109A expressed in the small 407

intestine is likely involved in the local regulation of intestinal absorption of triglycerides, 408

cholesterols, and free fatty acids. 409

410

Consistent with previous studies (Chang et al, 2006; Chen et al, 2015a; Garg & Grundy, 1990; 411

Goldberg & Jacobson, 2008), we also observed that the HFD-induced obese mice treated with 412

niacin exhibited an obvious reduction in serum insulin concentration and glucose tolerance 413

compared to the vehicle-treated mice. However, our data demonstrated that treatment with niacin 414

led to a significant improvement in insulin resistance in HFD-fed obese mice (HOMA-IR). 415

GPR109A has been shown to be functionally expressed in both human and murine islet beta-cells 416

(Li et al, 2011a; Wang et al, 2016). Chen et al. have shown that niacin exerts its inhibitory actions 417

on β-cell function via GPR109A-mediated signaling, but without affecting islet architecture (Chen 418

et al, 2015a). Nevertheless, a recent study demonstrated that GPR109A expression was 419

down-regulated dramatically in islet β-cells of type 2 diabetic patients as well as diabetic db/db 420


https://doi.org/10.1101/382416

22

mice, suggesting that the important role of GPR109A signaling in the prevention of diabetes (Wang 421

et al, 2016). Moreover, an early study observed that a short-term niacin analogue administration 422

resulted in an improved glucose tolerance in prediabetic or diabetic patients (Fulcher et al, 1992). 423

In addition, our data clearly demonstrated that niacin challenge caused a significant reduction in 424

hepatic de novo lipogenesis and ectopic lipid accumulation in liver, adipose tissue and muscle of 425

mice fed on HFD, leading to a beneficial effect on insulin resistance. The effects of niacin on 426

glucose homeostasis remain controversial, and the exact roles of GPR109A in the regulation of 427

insulin resistance now deserve thorough evaluation. 428

429

In conclusion, the data presented in this report have demonstrated a novel role of niacin in the 430

regulation of energy homeostasis by fine-tuning of hepatic de novo lipogenesis, BAT/beige 431

thermogenesis and intestinal fat absorption via GPR109A-mediated signaling. Our findings 432

provide a critical clue to the potential new therapeutic strategies that could treat obesity, diabetes 433

and cardiovascular disease. As the debate continues about the favorable effects of 434

niacin/GPR109A on the development and progression of atherosclerosis and its complications, it 435

will be vital to further perform a more comprehensive assessment of niacin/GPR109A in the 436

regulation of energy homeostasis in addition to anti-inflammation. 437

438

439

440


https://doi.org/10.1101/382416

23

Materials and Methods 441

Cell culture and treatment 442

HepG2 cells (ATCC) were maintained in Dulbecco’s modified Eagle’s medium (DMEM) (Hyclone, 443

USA) containing 10% heat-inactivated fetal bovine serum (FBS) (GIBCO, USA) with 1% 444

penicillin-streptomycin solution. All cells were incubated at 37°C in a humidified chamber with 5% 445

CO2. For the western blot analysis, the HepG2 cells were seeded in 24-well plates. For examining 446

the intracellular signaling pathway, all cells were starved for 5h in serum-free medium to reduce 447

the background before the agonist activation. And after stimulation with niacin (Sigma, 300μM), 448

cells were lysed by RIPA buffer. If inhibitors were required, they were added to the cell before 449

agonist stimulation. All samples were stored at -20°C until use. 450

451

Small Interfering RNAs (SiRNAs) Transfection 452

Small-interfering RNA (siRNA) against GPR109A and a non-specific scrambled siRNA (siRNA-NC) 453

were chemically synthesized by Dharmacon RNA Technologies (Lafayette). The sequences of 454

siRNA were 5’ UAUGUGAGGCGUUGGGACU-3’ for GPR109A and 5’-UCAAAUAACCAUUCCAA 455

GA-3’ for negative control. Cells at 60% confluence were transfected with siRNA using 456

LipofectamineTM 3000 reagent (Invitrogen, USA), according to the manufacturer’s specifications. 457

Media were replaced with DMEM containing 10% heat-inactivated FBS 6 h later. After the siRNA 458

transfection, the cells were split for the indicated assay the following day. 459

460


https://doi.org/10.1101/382416

24

Brown fat precursor cells extraction and culture 461

Brown fat precursor cells were isolated from interscapular BAT of newborn wild-type mice as 462

previously described (Cannon & Nedergaard, 2001). BAT was stripped from mice and transplanted 463

into 6 cm cell culture dish with PBS immediately. Then cut the tissue into small pieces with elbow 464

surgery. Disperse the broken tissue with 5mL isolation buffer containing Collagenase I in 37°C 465

incubator for 30 min. After digestion, tissue remnants were removed by filtration through a 70μm 466

nylon mesh (Corning, USA) and centrifuged at 200g for 5 min. Re-suspended with 5mL isolation 467

buffer without collagenase and filtered through a 40μm nylon mesh (Corning, USA) and 468

centrifuged at 200g for 5 min. Lastly, re-suspended cells with DMEM containing 10% FBS and 469

inoculated into 6-well plate. When cells reached to 80%-90% density, began to induce 470

differentiation. 471

472

Brown fat precursor cells differentiation 473

For differentiation, two days after cells confluence (DAY 0), the medium was replaced with DMEM 474

containing 10% FBS with inducers (MDI) (1.0μM dexamethasone, 10μg/mL insulin, and 0.5mM 475

3-isobutyl-1-methylxanthine) for 2 days. Two days after MDI (DAY 2), the medium was changed to 476

DMEM supplemented with 10% FBS, 10μg/mL insulin. Two days later (DAY 4), cells were 477

maintained in DMEM with 10% FBS for other 4-6 days. Full differentiation is usually achieved by 478

DAY 8. Adipogenesis was monitored by morphological examination with Bodipy 493/503 479

(Molecular probes) staining. 480


https://doi.org/10.1101/382416

25

Animal studies 481

All mice were housed in groups of 4–6 per cage under constant temperature (23-25°C) with a 12 h 482

light/12 h dark cycle. Animals were given ad libitum access to water and food. GPR109A 483

heterozygous mice on C57BL/6 background were a generous gift from Dr. Offermanns’ lab (Bad 484

Nauheim, Germany). Homozygous GPR109A knock out and their littermate wild-type mice were 485

generated by matting heterozygous mice and genotyped using PCR. Male mice were used for all 486

the experiments described in this study. And 4-weeks male C57BL/6 mice were purchased from 487

the company of SLAC Laboratory Animal (Shanghai, China), and after a week of adaptation, mice 488

were fed a HFD (60% kcal from fat) (D12492, Research Diets) for generating diet-induced obese 489

(DIO) model and a normal chow diet (10% kcal, SLAC, Shanghai, China) in the control group. After 490

6 weeks on HFD, mice were divided randomly into two groups. One of group received niacin (50 491

mM) (Sigma-Aldrich, USA) dissolved in drinking water, and the other received vehicle (water) for 492

8-9 weeks as a control. Groups of age- and weight-matched animals were used in all experiments. 493

Food consumption and body weight were recorded weekly. After the modeling period, mice were 494

fasted overnight (12 h) and anesthetized with 3% isoflurane in a dedicated chamber, and the 495

weights of major organ including fat pads and liver were recorded. Whole blood was collected and 496

processed to isolate serum and tissues were flash frozen in liquid nitrogen and stored at -80°C for 497

biochemical and histological analyses. All animal procedures were conducted in accordance with 498

the Guide for the Care and Use of Laboratory Animals (United States National Institutes of Health). 499

The protocol was approved by the research ethics committee. 500


https://doi.org/10.1101/382416

26

501

Body temperature in cold response 502

To test resistance to cold exposure, mice were individually caged and exposed to 4°C for 6 hours 503

with free access to water. Anal temperature was monitored at the beginning and hourly after the 504

start of cold exposure. 505

506

Serum collection and parameters determination 507

Blood were collected form the orbital sinus, and samples were incubated at room temperature for 508

30 min to allow clotting. Serum samples were obtained after centrifuging at 3,500 rpm at 4°C for 20 509

min. Finally the supernatant (serum) was collected and stored at -80°C. Serum total triglyceride 510

(TG), total cholesterol (TC), high density lipoprotein (HDL), and low density lipoprotein (LDL) were 511

measured by using biochemical testing kits (Roche, Switzerland). Free fatty acids (FFA) were 512

assayed by commercial kits (Nanjing Jiancheng, China). And plasma insulin levels were 513

determined by a rat/mouse insulin enzyme-linked immunosorbent assay (ELISA) kit (Millipore 514

Billerica, USA) according to the manufacturer’s instruction. Homeostasis model assessment of 515

insulin resistance (HOMA-IR) was calculated using the following equation: HOMA-IR = (fasting 516

plasma insulin (mIU/L) × fasting plasma glucose (mmol/L)/22.5. 517

518

Fecal and liver lipids analysis 519

Fecal lipids were extracted as previously described (Folch et al, 1957). Feces were collected from 520


https://doi.org/10.1101/382416

27

mice housed individually from 5 consecutive days. Feces were dried for 1 h at 70°C, and 521

one-hundred milligram aliquots of feces were weighed, incubated with 2 ml of chloroform-methanol 522

(2:1,v/v) for 30 min at 60°C with constant agitation, and then centrifuged（5 min at 5,000 rpm). 523

Water (1 mL) was added to the supernatant, and following vortex, phase separation was induced 524

by low-speed centrifugation (2,000 rpm for 10 min). The lower chloroform phase was then 525

removed and transferred to a new tube, and the sample was dried under nitrogen. Then samples 526

were re-suspended in 500 μL chloroform, evaporated to dryness, and finally re-suspended in 527

500μL 100% ethanol. For analysis of liver lipids, 100-mg aliquots of tissue were homogenized with 528

2 mL chloroform-methanol and then agitated overnight on an orbital shaker at 4°C. The 529

homogenate was then centrifuged (5 min at 5,000 rpm), 0.9% NaCl solution was added to the 530

liquid phase, and the samples were vortexed. Phase separation was induced by centrifugation 531

(2,000 rpm for 10 min), and the bottom phase was removed to a new tube and then processed as 532

above. The levels of TC, TG, and FFA in the fecal and liver lipid extracts were measured using 533

commercial kits following the manufacturer’s instructions and normalized to feces or tissue 534

weights. 535

536

Glucose and insulin tolerance tests 537

For glucose tolerance tests (GTTs), mice were performed after an overnight fast. Glucose (Sigma, 538

USA) was injected (intraperitoneal [i.p.] injection of a 40% solution, 1.5g/kg body weight), and tail 539

vein blood glucose levels were measured at 0, 30, 60, 90, and 120 minutes after injection with a 540


https://doi.org/10.1101/382416

28

Onetouch Ultra blood glucose monitoring system (LifeScan, USA). For insulin tolerance tests 541

(ITTs), mice were fasted for 4 h and then intraperitoneally injected with human insulin (0.75 U/kg 542

body weight; NovoRapid, Novo Nordisk). Glucose concentrations in blood were measured after 0, 543

30, 60, 90, and 120 min. Area under the curves (AUC) for GTT and ITT were calculated by 544

trapezoidal approximation, using GraphPad Prism 6.0. 545

546

RNA isolation and quantitative real-time PCR analysis 547

Total RNA was extracted by using an RNAiso reagent kit (Takara, Japan). Reverse transcription 548

was performed on each RNA sample (1μg) by using PrimeScript RT reagent kit with gDNA eraser 549

(Takara) with a final reaction volume of 20μL. For qRT-PCR analysis, the reaction mixture was run 550

in an iQ5 real time PCR machine (Bio-Rad Laboratories, Hercules, CA, USA) instrument using 551

SYBR Premix Ex Taq II (Takara, Japan). All expression levels were normalized to the 552

corresponding GAPDH mRNA levels and analyzed using the 2-△△CT

method. The qRT-PCR cycles 553

were as follows: Step 1, preparative denaturation (30 s at 95 °C); Step 2, 40 cycles of denaturation 554

(5 s at 95 °C) and annealing (30 s at 60 °C); Step 3, dissociation following the manufacturer’s 555

protocol. PCR primers used in the real-time-PCR analysis are listed in Table S1. 556

557

Semi-quantification PCR 558

The total RNA was extracted by using an RNAiso reagent kit. The cDNA are synthesized using 559

PrimeScript RT reagent kit. PCR was performed in 20μL reaction mixture according to the 560


https://doi.org/10.1101/382416

29

instruction of KOD-Plus (TOYOBO, Japan). Primers used for GPR109A were indicated at table 1. 561

PCR was performed using an initial denaturation at 94 °C for 5 minutes, followed by 28 cycles at 562

94 °C for 30 seconds, 53 °C for 30 seconds, and 68 °C for 30 seconds. After 28 cycles, an 563

additional elongation step was performed at 68 °C for 10 minutes. Amplified PCR products were 564

run on 1.2% agarose gels containing ethidium bromide. 565

566

Histological analysis 567

To examine hepatic steatosis, liver samples were embedded in Tissue-Tek (SA-KURA, USA), 568

frozen and sectioned (10 μm) and stored at -80 °C until use. The sections were stained with 569

hematoxylin and Oil red O (Sigma Aldrich, USA) for lipid deposition. 570

For histology, adipose tissues were fixed in 10% neutral-buffered formalin, followed by dehydration 571

with graded ethanol (80-100%) and embedding in paraffin. Then the tissues were sliced into 6μm 572

pieces using microtome, de-paraffinized in xylene, passed through 80-100% ethanol. And 573

hematoxylin and eosin (H&E) staining was performed using the standard techniques. 574

For transmission electron microscopy (TEM) analysis, liver and muscle tissues were cut into 1 575

mm3 fragments and fixed by immersion in 2.5% glutaraldehyde in phosphate buffer (0.1 M, pH 7.4) 576

overnight at 4°C. After that, they were washed in 0.1 M phosphate buffer for three times. Then, 577

they were post-fixated with 1% osmium tetroxide (OsO4) in 0.1 M phosphate buffer for 2 hours at 578

room temperature. Subsequently, they were rinsed with 0.1 M phosphate buffer for three times 579

again. Tissues were dehydrated through graded alcohols (30, 50, 70, 80, 90, 95 and 100%) for 30 580


https://doi.org/10.1101/382416

30

min. Last they were embedded in Epon 812. Ultrathin sections (70-90 nm in thickness) of 581

Epon-embedded samples were placed on copper grids and stained with uranyl acetate and lead 582

citrate. Sections were examined with a Hitachi model H-7650 electron microscope (Hitachi, 583

Japan). 584

585

Measurement of enzymes activities 586

To determine the activities of enzymes involving in lipogenesis in liver and white adipose tissue, 587

the enzyme source fraction was prepared as follows. Briefly, a 10% (w/v) homogenate was 588

prepared in phosphate buffered solution (pH 7.2-7.4) and then centrifuged at 2000-3000 rpm for 589

20 min and the supernatant was transferred into a new tube. Samples were stored at -80 °C until 590

use. All of enzymes, such as pyruvate dehydrogenase (PDH), acetyl CoA carboxylase (ACC), 591

acetyl CoA carboxylase 1 (ACC1), and diacylglycerol acyltransferase 2 (DGAT2), were measure 592

by ELISA (Jin Ma, China). 593

594

Intracellular calcium measurement 595

The fluorescent Ca2+

indicator fura-2 was employed to monitor changes of calcium. HepG2 cells 596

were harvested with a Nonenzymatic Cell Dissociation Solution (M&C Gene Technology, China), 597

washed twice with Hanks’ solution, and resuspended at a density of 5×106 cells/mL in Hanks’ 598

balanced salt solution containing 0.025% bovine serum albumin. The cells were then loaded with 3 599

μM Fura-2-AM (Dojindo Laboratories) for 30 min at 37 °C. Cells were washed twice in Hanks’ 600


https://doi.org/10.1101/382416

31

solution, and then resuspended in Hanks’ solution at a concentration of 1×107 cells/mL. Cells were 601

stimulated with the indicated concentrations of niacin and Ca2+

flux was measured using excitation 602

wavelengths of 340 and 380 nm in a fluorescence spectrometer. 603

604

Western blot analysis 605

Proteins were extracted from tissues or cultured cells by homogenizing or scraping in ice-cold 606

RIPA buffer (50mM Tris (pH 7.4), 150mM NaCl, 1% Triton X-100, 1% sodium deoxycholate, 0.1% 607

SDS, 2 mM EDTA) (Beyotime, China) containing 1mM sodium orthovanadate and protease 608

inhibitors (Complete Mini, Roche). Tissue samples were transferred to micro-centrifuge tubes and 609

rotated on a rocking platform for 40 min at 4°C after homogenization on ice. The homogenate was 610

cleared by centrifugation at 4°C for 20 min at 12,000 rpm and the supernatant was transferred into 611

a new tube. Protein concentration in the supernatant was determined using the BCA Protein Assay 612

Kit (Probe Gene, China). Equal amounts of protein samples were subjected to 613

SDS-polyacrylamide gel electrophoresis and transferred to PVDF membranes (Millipore, USA). 614

Membranes were blocked with 5% BSA in TBS containing 0.1% Tween-20 (TBST) for 1 h at room 615

temperature, and incubated overnight at 4°C with primary antibodies against p-ERK1/2 (Cat# 616

4370), ERK1/2 (Cat# 9102), p-AKT (Ser473) (Cat# 4058), AKT (Cat# 4691), p-AMPKα (Thr 172) 617

(Cat# 2535), AMPKα (Cat# 5832), p-ACC (Ser79) (Cat# 11818), ACC (Cat# 3676), β-tubulin (Cat# 618

2128) (Cell Signaling Technology, USA), GPR109A (Santa Cruz Biotechnology, USA, Cat# 619

sc-134583) respectively. The following day, the membranes were washed with TBST and 620


https://doi.org/10.1101/382416

32

incubated with a horseradish peroxidase-conjugated secondary antibody (1: 2,000) for 2 hours at 621

room temperature. Proteins were visualized with an enhanced chemiluminescence (ECL) reagent 622

(Beyotime, China) by the Tanon 5200 Chemiluminescent Imaging System (Tanon, China). The 623

target proteins were quantified by densitometric analysis with Image J software and normalized to 624

β-tubulin level. 625

626

Statistical analysis and data processing 627

For animal studies, n values indicate the number of animals per cohort. For animal studies, sample 628

sizes were based on previous studies and expertise. All data of cells shown are representative of 629

at least three independent experiments. All results are expressed as mean ± SEM. The GraphPad 630

Prism 6 software (San Diego, CA) was used for statistical analyses. Unpaired two-tailed Student’s 631

t-test was used for comparisons between two groups for all figures. Differences were considered 632

significant when P < 0.05. All P-values for main figures and EV figures can be found in Appendix 633

Tables S2–S12. 634

635

Conflict of interest 636

The authors declare that they have no conflict of interest. 637

638

References 639

Ahmed K, Tunaru S, Tang C, Muller M, Gille A, Sassmann A, Hanson J, Offermanns S (2010) An Autocrine 640

Lactate Loop Mediates Insulin-Dependent Inhibition of Lipolysis through GPR81. Cell Metab 11: 311-319 641


https://doi.org/10.1101/382416

33

642

Altmann SW, Davis HR, Zhu LJ, Yao XR, Hoos LM, Tetzloff G, Iyer SPN, Maguire M, Golovko A, Zeng M, Wang 643

LQ, Murgolo N, Graziano MP (2004) Niemann-Pick C1 like 1 protein is critical for intestinal cholesterol 644

absorption. Science 303: 1201-1204 645

646

Altschul R, Hoffer A, Stephen JD (1955) Influence of nicotinic acid on serum cholesterol in man. Archives of 647

biochemistry and biophysics 54: 558-559 648

649

Besnard P, Niot I, Poirier H, Clement L, Bernard A (2002) New insights into the fatty acid-binding protein 650

(FABP) family in the small intestine. Mol Cell Biochem 239: 139-147 651

652

Buhler S, Frahm J, Tienken R, Kersten S, Meyer U, Huber K, Danicke S (2018) Effects of energy supply and 653

nicotinic acid supplementation on serum anti-oxidative capacity and on expression of oxidative 654

stress-related genes in blood leucocytes of periparturient primi- and pluriparous dairy cows. Journal of 655

animal physiology and animal nutrition 102: e87-e98 656

657

Cannon B, Nedergaard J (2001) Cultures of adipose precursor cells from brown adipose tissue and of clonal 658

brown-adipocyte-like cell lines. Methods in molecular biology (Clifton, NJ) 155: 213-224 659

660

Cannon B, Nedergaard J (2004) Brown adipose tissue: function and physiological significance. Physiological 661

reviews 84: 277-359 662

663

Carlson LA (2005) Nicotinic acid: the broad-spectrum lipid drug. A 50th anniversary review. Journal of 664

internal medicine 258: 94-114 665

666

Chang AM, Smith MJ, Galecki AT, Bloem CJ, Halter JB (2006) Impaired beta-cell function in human aging: 667

response to nicotinic acid-induced insulin resistance. The Journal of clinical endocrinology and metabolism 668

91: 3303-3309 669

670

Chen H, Assmann JC, Krenz A, Rahman M, Grimm M, Karsten CM, Kohl J, Offermanns S, Wettschureck N, 671

Schwaninger M (2014) Hydroxycarboxylic acid receptor 2 mediates dimethyl fumarate's protective effect in 672

EAE. The Journal of clinical investigation 124: 2188-2192 673

674

Chen L, So WY, Li SY, Cheng Q, Boucher BJ, Leung PS (2015a) Niacin-induced hyperglycemia is partially 675

mediated via niacin receptor GPR109a in pancreatic islets. Mol Cell Endocrinol 404: 56-66 676

677

Chen LH, So WY, Li SYT, Cheng QN, Boucher BJ, Leung PS (2015b) Niacin-induced hyperglycemia is partially 678

mediated via niacin receptor GPR109a in pancreatic islets. Mol Cell Endocrinol 404: 56-66 679

680

Cho KH, Kim HJ, Rodriguez-Iturbe B, Vaziri ND (2009) Niacin ameliorates oxidative stress, inflammation, 681


https://doi.org/10.1101/382416

34

proteinuria, and hypertension in rats with chronic renal failure. American journal of physiology Renal 682

physiology 297: F106-113 683

684

Cohen P, Levy JD, Zhang Y, Frontini A, Kolodin DP, Svensson KJ, Lo JC, Zeng X, Ye L, Khandekar MJ, Wu J, 685

Gunawardana SC, Banks AS, Camporez JP, Jurczak MJ, Kajimura S, Piston DW, Mathis D, Cinti S, Shulman GI 686

et al (2014) Ablation of PRDM16 and beige adipose causes metabolic dysfunction and a subcutaneous to 687

visceral fat switch. Cell 156: 304-316 688

689

Cunningham JW, Wiviott SD (2014) Modern obesity pharmacotherapy: weighing cardiovascular risk and 690

benefit. Clinical cardiology 37: 693-699 691

692

Davis JD, Wirtshafter D, Asin KE, Brief D (1981) Sustained intracerebroventricular infusion of brain fuels 693

reduces body weight and food intake in rats. Science 212: 81-83 694

695

Digby JE, Lee JM, Choudhury RP (2009) Nicotinic acid and the prevention of coronary artery disease. 696

Current opinion in lipidology 20: 321-326 697

698

Digby JE, Martinez F, Jefferson A, Ruparelia N, Chai J, Wamil M, Greaves DR, Choudhury RP (2012) 699

Anti-inflammatory effects of nicotinic acid in human monocytes are mediated by GPR109A dependent 700

mechanisms. Arteriosclerosis, thrombosis, and vascular biology 32: 669-676 701

702

Elam MB, Hunninghake DB, Davis KB, Garg R, Johnson C, Egan D, Kostis JB, Sheps DS, Brinton EA (2000) 703

Effect of niacin on lipid and lipoprotein levels and glycemic control in patients with diabetes and peripheral 704

arterial disease: the ADMIT study: A randomized trial. Arterial Disease Multiple Intervention Trial. Jama 705

284: 1263-1270 706

707

Folch J, Lees M, Sloane Stanley GH (1957) A simple method for the isolation and purification of total lipides 708

from animal tissues. The Journal of biological chemistry 226: 497-509 709

710

Fulcher GR, Walker M, Catalano C, Agius L, Alberti KG (1992) Metabolic effects of suppression of 711

nonesterified fatty acid levels with acipimox in obese NIDDM subjects. Diabetes 41: 1400-1408 712

713

Ganji SH, Qin S, Zhang L, Kamanna VS, Kashyap ML (2009) Niacin inhibits vascular oxidative stress, 714

redox-sensitive genes, and monocyte adhesion to human aortic endothelial cells. Atherosclerosis 202: 715

68-75 716

717

Ganji SH, Tavintharan S, Zhu D, Xing Y, Kamanna VS, Kashyap ML (2004) Niacin noncompetitively inhibits 718

DGAT2 but not DGAT1 activity in HepG2 cells. Journal of lipid research 45: 1835-1845 719

720

Gao Z, Yin J, Zhang J, Ward RE, Martin RJ, Lefevre M, Cefalu WT, Ye J (2009) Butyrate improves insulin 721


https://doi.org/10.1101/382416

35

sensitivity and increases energy expenditure in mice. Diabetes 58: 1509-1517 722

723

Garg A, Grundy SM (1990) Nicotinic acid as therapy for dyslipidemia in non-insulin-dependent diabetes 724

mellitus. Jama 264: 723-726 725

726

Gautam J, Banskota S, Shah S, Jee JG, Kwon E, Wang Y, Kim DY, Chang HW, Kim JA (2018) 727

4-Hydroxynonenal-induced GPR109A activation elicits bipolar responses, Galphai-mediated 728

anti-inflammatory effects, and Gbetagamma-mediated cell death. 729

730

Gesta S, Tseng YH, Kahn CR (2007) Developmental origin of fat: tracking obesity to its source. Cell 131: 731

242-256 732

733

Goldberg RB, Jacobson TA (2008) Effects of niacin on glucose control in patients with dyslipidemia. Mayo 734

Clinic proceedings 83: 470-478 735

736

Graff EC, Fang H, Wanders D, Judd RL (2016) Anti-inflammatory effects of the hydroxycarboxylic acid 737

receptor 2. British journal of pharmacology 65: 102-113 738

739

Guyton JR, Goldberg AC, Kreisberg RA, Sprecher DL, Superko HR, O'Connor CM (1998) Effectiveness of 740

once-nightly dosing of extended-release niacin alone and in combination for hypercholesterolemia. The 741

American journal of cardiology 82: 737-743 742

743

Hall AM, Wiczer BM, Herrmann T, Stremmel W, Bernlohr DA (2005) Enzymatic properties of purified 744

murine fatty acid transport protein 4 and analysis of acyl-CoA synthetase activities in tissues from FATP4 745

null mice. Journal of Biological Chemistry 280: 11948-11954 746

747

Harms M, Seale P (2013) Brown and beige fat: development, function and therapeutic potential. Nature 748

medicine 19: 1252-1263 749

750

Haslam DW, James WP (2005) Obesity. Lancet 366: 1197-1209 751

752

Kamanna VS, Ganji SH, Kashyap ML (2013) Recent advances in niacin and lipid metabolism. Current opinion 753

in lipidology 24: 239-245 754

755

Katsurada A, Iritani N, Fukuda H, Matsumura Y, Nishimoto N, Noguchi T, Tanaka T (1990) Effects of 756

nutrients and hormones on transcriptional and post-transcriptional regulation of acetyl-CoA carboxylase in 757

rat liver. European journal of biochemistry 190: 435-441 758

759

Kopelman PG (2000) Obesity as a medical problem. Nature 404: 635-643 760

761


https://doi.org/10.1101/382416

36

Kwon WY, Suh GJ, Kim KS, Kwak YH (2011) Niacin attenuates lung inflammation and improves survival 762

during sepsis by downregulating the nuclear factor-kappaB pathway. Critical care medicine 39: 328-334 763

764

Li HM, Zhang M, Xu ST, Li DZ, Zhu LY, Peng SW, Chen GQ, Martin PM, Ganapathy V, Wei CJ (2011a) Nicotinic 765

acid inhibits glucose-stimulated insulin secretion via the G protein-coupled receptor PUMA-G in murine 766

islet beta cells. Pancreas 40: 615-621 767

768

Li X, Millar JS, Brownell N, Briand F, Rader DJ (2010) Modulation of HDL metabolism by the niacin receptor 769

GPR109A in mouse hepatocytes. Biochemical pharmacology 80: 1450-1457 770

771

Li Y, Xu S, Mihaylova MM, Zheng B, Hou X, Jiang B, Park O, Luo Z, Lefai E, Shyy JY, Gao B, Wierzbicki M, 772

Verbeuren TJ, Shaw RJ, Cohen RA, Zang M (2011b) AMPK phosphorylates and inhibits SREBP activity to 773

attenuate hepatic steatosis and atherosclerosis in diet-induced insulin-resistant mice. Cell Metab 13: 774

376-388 775

776

Lowell BB, Spiegelman BM (2000) Towards a molecular understanding of adaptive thermogenesis. Nature 777

404: 652-660 778

779

Lukasova M, Malaval C, Gille A, Kero J, Offermanns S (2011a) Nicotinic acid inhibits progression of 780

atherosclerosis in mice through its receptor GPR109A expressed by immune cells. The Journal of clinical 781

investigation 121: 1163-1173 782

783

Lukasova M, Malaval C, Gille A, Kero J, Offermanns S (2011b) Nicotinic acid inhibits progression of 784

atherosclerosis in mice through its receptor GPR109A expressed by immune cells. Journal of Clinical 785

Investigation 121: 1163-1173 786

787

Manning S, Pucci A, Finer N (2014) Pharmacotherapy for obesity: novel agents and paradigms. Therapeutic 788

advances in chronic disease 5: 135-148 789

790

Mielenz M, Kuhla B, Hammon HM (2013) Abundance of adiponectin system and G-protein coupled 791

receptor GPR109A mRNA in adipose tissue and liver of F2 offspring cows of Charolais x German Holstein 792

crosses that differ in body fat accumulation. Journal of dairy science 96: 278-289 793

794

Mitrofanov VA, Ovchinnikova NM, Belova SV, Fedotova MV, Gladkova EV (2005) Inflammatory 795

degeneration of joint tissue in adjuvant arthritis after intraarticular treatment with the mixture of silver 796

drug and nicotinic acid. Bulletin of experimental biology and medicine 140: 702-704 797

798

Moreira FA, Grieb M, Lutz B (2009) Central side-effects of therapies based on CB1 cannabinoid receptor 799

agonists and antagonists: focus on anxiety and depression. Best practice & research Clinical endocrinology 800

& metabolism 23: 133-144 801


https://doi.org/10.1101/382416

37

802

Ng M, Fleming T, Robinson M, Thomson B, Graetz N, Margono C, Mullany EC, Biryukov S, Abbafati C, Abera 803

SF, Abraham JP, Abu-Rmeileh NM, Achoki T, AlBuhairan FS, Alemu ZA, Alfonso R, Ali MK, Ali R, Guzman NA, 804

Ammar W et al (2014) Global, regional, and national prevalence of overweight and obesity in children and 805

adults during 1980-2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet 384: 806

766-781 807

808

Oelkers P, Behari A, Cromley D, Billheimer JT, Sturley SL (1998) Characterization of two human genes 809

encoding acyl coenzyme A:cholesterol acyltransferase-related enzymes. The Journal of biological chemistry 810

273: 26765-26771 811

812

Offermanns S, Colletti SL, Lovenberg TW, Semple G, Wise A, AP IJ (2011) International Union of Basic and 813

Clinical Pharmacology. LXXXII: Nomenclature and Classification of Hydroxy-carboxylic Acid Receptors 814

(GPR81, GPR109A, and GPR109B). Pharmacological reviews 63: 269-290 815

816

Olsen MK, Johannessen H, Cassie N, Barrett P, Takeuchi K, Kulseng B, Chen D, Zhao CM (2017) Steady-state 817

energy balance in animal models of obesity and weight loss. Scandinavian journal of gastroenterology 52: 818

442-449 819

820

Rahman M, Muhammad S, Khan MA, Chen H, Ridder DA, Muller-Fielitz H, Pokorna B, Vollbrandt T, Stolting 821

I, Nadrowitz R, Okun JG, Offermanns S, Schwaninger M (2014) The beta-hydroxybutyrate receptor HCA2 822

activates a neuroprotective subset of macrophages. Nature communications 5: 3944 823

824

Ruderman NB, Saha AK, Kraegen EW (2003) Minireview: malonyl CoA, AMP-activated protein kinase, and 825

adiposity. Endocrinology 144: 5166-5171 826

827

Rui L (2014) Energy metabolism in the liver. Comprehensive Physiology 4: 177-197 828

829

Shackelford DB, Shaw RJ (2009) The LKB1-AMPK pathway: metabolism and growth control in tumour 830

suppression. Nature reviews Cancer 9: 563-575 831

832

Shehadah A, Chen J, Zacharek A, Cui Y, Ion M, Roberts C, Kapke A, Chopp M (2010) Niaspan treatment 833

induces neuroprotection after stroke. Neurobiology of disease 40: 277-283 834

835

Shi Y, Burn P (2004) Lipid metabolic enzymes: emerging drug targets for the treatment of obesity. Nature 836

reviews Drug discovery 3: 695-710 837

838

Shi Y, Lai X, Ye L, Chen K, Cao Z, Gong W, Jin L, Wang C, Liu M, Liao Y, Wang JM, Zhou N (2017) Activated 839

niacin receptor HCA2 inhibits chemoattractant-mediated macrophage migration via 840

Gbetagamma/PKC/ERK1/2 pathway and heterologous receptor desensitization. Scientific reports 7: 42279 841


https://doi.org/10.1101/382416

38

842

Singh N, Gurav A, Sivaprakasam S, Brady E, Padia R, Shi H, Thangaraju M, Prasad PD, Manicassamy S, Munn 843

DH, Lee JR, Offermanns S, Ganapathy V (2014) Activation of Gpr109a, receptor for niacin and the 844

commensal metabolite butyrate, suppresses colonic inflammation and carcinogenesis. Immunity 40: 845

128-139 846

847

Soga T, Kamohara M, Takasaki J, Matsumoto S, Saito T, Ohishi T, Hiyama H, Matsuo A, Matsushime H, 848

Furuichi K (2003) Molecular identification of nicotinic acid receptor. Biochemical and biophysical research 849

communications 303: 364-369 850

851

Solinas G, Boren J, Dulloo AG (2015) De novo lipogenesis in metabolic homeostasis: More friend than foe? 852

Molecular metabolism 4: 367-377 853

854

Strawford A, Antelo F, Christiansen M, Hellerstein MK (2004) Adipose tissue triglyceride turnover, de novo 855

lipogenesis, and cell proliferation in humans measured with 2H2O. American journal of physiology 856

Endocrinology and metabolism 286: E577-588 857

858

Taggart AK, Kero J, Gan X, Cai TQ, Cheng K, Ippolito M, Ren N, Kaplan R, Wu K, Wu TJ, Jin L, Liaw C, Chen R, 859

Richman J, Connolly D, Offermanns S, Wright SD, Waters MG (2005) (D)-beta-Hydroxybutyrate inhibits 860

adipocyte lipolysis via the nicotinic acid receptor PUMA-G. The Journal of biological chemistry 280: 861

26649-26652 862

863

Thangaraju M, Cresci GA, Liu K, Ananth S, Gnanaprakasam JP, Browning DD, Mellinger JD, Smith SB, Digby 864

GJ, Lambert NA, Prasad PD, Ganapathy V (2009) GPR109A Is a G-protein-Coupled Receptor for the 865

Bacterial Fermentation Product Butyrate and Functions as a Tumor Suppressor in Colon. Cancer Res 69: 866

2826-2832 867

868

Titchenell PM, Chu QW, Monks BR, Birnbaum MJ (2015) Hepatic insulin signalling is dispensable for 869

suppression of glucose output by insulin in vivo. Nature communications 6 870

871

Tong L (2005) Acetyl-coenzyme A carboxylase: crucial metabolic enzyme and attractive target for drug 872

discovery. Cellular and molecular life sciences : CMLS 62: 1784-1803 873

874

Tunaru S, Kero J, Schaub A, Wufka C, Blaukat A, Pfeffer K, Offermanns S (2003) PUMA-G and HM74 are 875

receptors for nicotinic acid and mediate its anti-lipolytic effect. Nature medicine 9: 352-355 876

877

Wang N, Guo DY, Tian X, Lin HP, Li YP, Chen SJ, Fu YC, Xu WC, Wei CJ (2016) Niacin receptor GPR109A 878

inhibits insulin secretion and is down-regulated in type 2 diabetic islet beta-cells. Gen Comp Endocr 237: 879

98-108 880

881


https://doi.org/10.1101/382416

39

Wang YC, McPherson K, Marsh T, Gortmaker SL, Brown M (2011) Health and economic burden of the 882

projected obesity trends in the USA and the UK. Lancet 378: 815-825 883

884

Wise A, Foord SM, Fraser NJ, Barnes AA, Elshourbagy N, Eilert M, Ignar DM, Murdock PR, Steplewski K, 885

Green A, Brown AJ, Dowell SJ, Szekeres PG, Hassall DG, Marshall FH, Wilson S, Pike NB (2003) Molecular 886

identification of high and low affinity receptors for nicotinic acid. The Journal of biological chemistry 278: 887

9869-9874 888

889

Wong TP, Chan LKY, Leung PS (2015) Involvement of the Niacin Receptor GPR109a in the Local Control of 890

Glucose Uptake in Small Intestine of Type 2 Diabetic Mice. Nutrients 7: 7543-7561 891

892

Wu BJ, Yan L, Charlton F, Witting P, Barter PJ, Rye KA (2010) Evidence that niacin inhibits acute vascular 893

inflammation and improves endothelial dysfunction independent of changes in plasma lipids. 894

Arteriosclerosis, thrombosis, and vascular biology 30: 968-975 895

896

Zhou G, Myers R, Li Y, Chen Y, Shen X, Fenyk-Melody J, Wu M, Ventre J, Doebber T, Fujii N, Musi N, 897

Hirshman MF, Goodyear LJ, Moller DE (2001) Role of AMP-activated protein kinase in mechanism of 898

metformin action. The Journal of clinical investigation 108: 1167-1174 899

900

901

Figure legends 902

Figure 1. GPR109A deficient animals display obesity. 903

A,B Body weight (A) and food intake (B) were recorded weekly (n=10-11). 904

C Representative images of fat mass were shown. 905

D Representative SEM images in liver (left panels; scale bars, as shown in figure) and muscle 906

(middle panels; scale bars, as shown in figure), and H&E staining of epididymal white 907

adipose tissue (eWAT) (right panels; scale bars, 100 μm). 908

E Fat pads and organ weights were recorded after mice anesthetized (n=10-11). 909

Data information: All the data are presented as means ±SEM. In (A, B, E), data were analyzed by 910

unpaired two-tailed student’s t-test. *P < 0.05; **P <0.01; ***P < 0.001; ****P < 0.0001. 911


https://doi.org/10.1101/382416

40

Figure 2. Niacin exerts inhibitory effects on HFD-induced obesity. 912






E Fat pads and organ weights were recorded after mice anesthetized (n=16-17). 918

F Serum lipid parameters, including total triglyceride (TG), total cholesterol (TC), high density 919

lipoprotein (HDL), and low density lipoprotein (LDL) were measured using biochemical 920

testing kits (n=7-8). 921

G Glucose tolerance test (GTT) and insulin tolerance test (ITT) results and the corresponding 922

AUC for the blood glucose levels (n=5-6). 923

H Serum insulin level during GTT was measured by ELISA (n=5). 924

I HOMA-IR index was calculated as stated in the “Method” section (n=5). 925

Data information: All the data are presented as means ±SEM. In (A, B, E-I), data were analyzed by 926

unpaired two-tailed student’s t-test. *P < 0.05; **P <0.01; ***P < 0.001; ****P < 0.0001. 927

928

Figure 3. Lack of GPR109A prevents anti-obesity actions of niacin. 929




https://doi.org/10.1101/382416

41




E Fat pad and organ weights were recorded after mice anesthetized (n=10-11). 935

Data information: All the data are presented as means ±SEM. In (A, B, E), data were analyzed by 936

unpaired two-tailed student’s t-test. ns, not significant. 937

938

Figure 4. Niacin directly inhibits hepatocyte lipogenesis through GPR109A. 939

A Representative images of Oil Red O staining of livers. Scale bars, 100 μm. 940

B Liver lipid content, including the levels of TG, TC, and FFA, were measured (n=5-6). 941

C Insulin levels were measured by ELISA (n=5). 942

D GPR109A gene expression was measured by qRT-PCR in liver, intestine, muscle, eWAT, 943

and BAT of the mice fed normal chow diet (NCD), high-fat diet (HFD), and HFD with 50 mM 944

niacin (NA) respectively. Gene expression in muscle served as the statistical controls 945

(n=8-9). 946

E Expression of GPR109A was measured in liver isolated from the WT mice fed NCD, HFD 947

with or without NA by quantitative analyses (n=8-9). 948

F The relative gene expression of GPR109A in liver was calculated by western blot (left panel) 949

and quantitative analyses (right panel). The β-tubulin was used as a loading control. 950

G Comparing the activities of enzymes involving in lipogenesis (PDH, ACC, ACC1, and 951


https://doi.org/10.1101/382416

42

DGAT2) of WT mice fed HFD with or without NA in liver (upper panels) and eWAT (lower 952

panels) (n=5). 953

H The activities of PDH and ACC in liver were measured in Gpr109a-/-

mice fed HFD (n=7,9). 954

I The mRNA expression analysis of genes involved in de novo lipogenesis in liver, and the 955

data of WT (Vehicle) group served as the statistical controls (n=8). 956

Data information: All the data are presented as means ±SEM. In (B, C, E-I), data were analyzed by 957

unpaired two-tailed student’s t-test. *P < 0.05, **P <0.01, ***P < 0.001. 958

959

Figure 5. AMPK and ERK1/2 are involved in the GPR109A-mediated inhibition of hepatocyte 960

lipogenesis. 961

A Phosphorylated of ERK1/2, AKT and AMPK were measured in the liver, muscle, and eWAT 962

from the WT mice by western blot and quantitative analyses (n=5). 963

B Phosphorylated of ERK1/2, AKT and AMPK were measured in the liver from Gpr109a-/-

mice. 964

(n=5). 965

C Activation of ERK1/2, AKT, AMPK and ACC in serum-starved HepG2 cells stimulated with 966

niacin (300 μM) for the indicated time periods. 967

D HepG2 cells were transfected with GPR109A siRNA for 36 h and stimulated with niacin (300 968

μM) for 5 min. Phosphorylation of ERK1/2, AKT, AMPK and ACC was accessed. 969

E Activation of ERK1/2, AKT, AMPK, and ACC in serum-starved HepG2 cells pretreated with 970

PTX (100 ng/mL) for 16 hr or M119K (10 μM), Go6983 (10 μM), or U0126 (1 μM) for 1h 971


https://doi.org/10.1101/382416

43

followed by a challenge with niacin (300 μM) for 5 min was accessed by western blot. And 972

β-tubulin was used as a loading control. 973

F Schematic model of suppression of ACC via GPR109A. 974

Data information: All the data are presented as means ±SEM. In (C,D,E), All data of cells shown 975

are representative of at least three independent experiments. In (A,B), data were analyzed by 976

unpaired two-tailed student’s t-test. *P < 0.05, **P <0.01. 977

978

Figure 6. Niacin treatment triggers BAT thermogenesis via GPR109A. 979

A Mice were exposed to 4°C for 6 hours. Monitor the anal temperature every hour (n=5-6). 980

B Monitor the anal temperature weekly under the room temperature (n=5). 981

C The mRNA expression of UCP1, PGC-1α and PRDM16 in brown adipose tissue (BAT) 982

from the WT mice (n=11-15). 983

D Electron micrographs of the BAT at the end of the experiment. Scale bar: 1μm. 984

E The mRNA expression of brown or beige genes (UCP1, PRDM16, PGC-1α, Cd137, Tbx1, 985

and Tmem26) in eWAT and sWAT (n=10-11). 986

F Brown fat precursor cells were differentiated in MDI with or without niacin for 8 days. Visible 987

lipid accumulations were observed by BODIPY 493/503 staining. Scale bars, 100 μm. 988

G The mRNA expression of UCP1, PRDM16, and PGC-1α in MDI-induced brown fat precursor 989

cells with the indicated concentration of niacin for 8 days. 990

Data information: All the data are presented as means ±SEM. In (F,G), All data of cells shown are 991


https://doi.org/10.1101/382416

44

representative of at least three independent experiments. In (A-C, E, G), data were analyzed by 992

unpaired two-tailed student’s t-test. *P < 0.05, **P <0.01, ***P < 0.001, ****P < 0.0001. 993

994

Figure 7. Niacin stimulation decreases fatty acid absorption though GPR109A. 995

A Fecal mass were normalized to body weight (mg/g). Fecal lipids were extracted and the 996

fecal lipids concentration was expressed as milligram of lipids per gram of feces weight. 997

Total triglyceride, total cholesterol, and free fatty acid (FFA) were measured in WT mice 998

(n=7,9). 999

B Feces lipid indicators were measured in Gpr109a-/-

mice (n=6). 1000

C Total mRNA isolated from intestines and examined the mRNA expression of indicated 1001

genes involved in fatty acid absorption (fatty acid transport protein 4 (FATP4) and 1002

intestinal-type fatty acid-binding protein (I-FABP)) and cholesterol absorption 1003

(Niemann-Pick C1-Like 1 (NPC1L1)) by qRT-PCR. 1004

Data information: All the data are presented as means ±SEM. In (F,G), All data of cells shown are 1005

representative of at least three independent experiments. Data were analyzed by unpaired 1006

two-tailed student’s t-test. *P < 0.05, **P <0.01. 1007


https://doi.org/10.1101/382416

1

Main figures

Figure 1


https://doi.org/10.1101/382416

2

Figure 2


https://doi.org/10.1101/382416

3

Figure 3


https://doi.org/10.1101/382416

4

Figure 4


https://doi.org/10.1101/382416

5

Figure 5


https://doi.org/10.1101/382416

6

Figure 6


https://doi.org/10.1101/382416

7

Figure 7


https://doi.org/10.1101/382416

Documents

Niacin fine-tunes energy homeostasis through canonical ... · 5 81 orphan receptor GPR109A (recently renamed as hydroxyl. carboxylic acid receptor 2 or HCA2) 82. was identified as