Part 4 Adrenoceptor agonists §1. , receptor agonists §2. receptor agonists §3. receptor agonists.
Niacin fine-tunes energy homeostasis through canonical ... · 5 81 orphan receptor GPR109A...
Transcript of Niacin fine-tunes energy homeostasis through canonical ... · 5 81 orphan receptor GPR109A...
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
40
Figure 2. Niacin exerts inhibitory effects on HFD-induced obesity. 912
A,B Body weight (A) and food intake (B) were recorded weekly (n=16-17). 913
C Representative images of fat mass were shown. 914
D Representative SEM images in liver (left panels; scale bars, as shown in figure) and muscle 915
(middle panels; scale bars, as shown in figure), and H&E staining of epididymal white 916
adipose tissue (eWAT) (right panels; scale bars, 100 μm). 917
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
A,B Body weight (A) and food intake (B) were recorded weekly (n=10-11). 930
C Representative images of fat mass were shown. 931
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
41
D Representative SEM images in liver (left panels; scale bars, as shown in figure) and muscle 932
(middle panels; scale bars, as shown in figure), and H&E staining of epididymal white 933
adipose tissue (eWAT) (right panels; scale bars, 100 μm). 934
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
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
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
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
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
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
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
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
1
Main figures
Figure 1
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
2
Figure 2
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
3
Figure 3
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
4
Figure 4
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
5
Figure 5
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
6
Figure 6
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
7
Figure 7
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