LGR4 Acts As a Link Between the Peripheral Circadian Clock and Lipid Metabolism in Liver

F WANG, X ZHANG and others LGR4 regulates circadian rhythm 52:2 133–143 Research of lipid

LGR4 acts as a link between the peripheral circadian clock and lipid metabolism in liver

Feng Wang1,2,*, Xianfeng Zhang2,*, Jiqiu Wang2,†, Maopei Chen2, Nengguang Fan2, Qinyun Ma2, Ruixin Liu2, Rui Wang2, Xiaoying Li2, Mingyao Liu3 and Guang Ning1,2

1Laboratory for Endocrine and Metabolic Diseases, Institute of Health Sciences, Shanghai Institutes for Biological Sciences (SIBS), Chinese Academy of Sciences (CAS) and Shanghai Jiao Tong University School of Medicine (SJTUSM), Shanghai, China 2Shanghai Clinical Center for Endocrine and Metabolic Diseases, Shanghai Institute of Endocrinology and Metabolism, Shanghai Key Laboratory for Endocrine Tumors and E-Institute of Shanghai Universities, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, 197 Ruijin 2nd Road, Shanghai 200025, China 3Institute of Biosciences and Technology, Texas A&M University Health Science Center, Houston, Texas 77030, USA *(F Wang and X Zhang contributed equally to this work) Correspondence †J Wang is now at Department of Endocrine and Metabolic Diseases, Shanghai Clinical Center for Endocrine and should be addressed Metabolic Diseases, Shanghai Institute of Endocrine and Metabolic Diseases, Ruijin Hospital, Shanghai Jiaotong to J Wang University School of Medicine, Shanghai 200025, China Email [email protected]

Abstract

The circadian clock plays an important role in the liver by regulating the major aspects of Key Words energy metabolism. Currently, it is assumed that the circadian clock regulates metabolism " circadian rhythm mostly by regulating the expression of liver enzymes at the transcriptional level, but the " G protein-coupled receptor underlying mechanism is not well understood. In this study, we showed that Lgr4 " respiratory exchange ratio Journal of Molecular Endocrinology homozygous mutant (Lgr4m/m) mice showed alteration in the rhythms of the respiratory " lipid metabolism exchange ratio. We further detected impaired plasma triglyceride rhythms in Lgr4m/m mice. Although no signiﬁcant changes in plasma cholesterol rhythms were observed in the Lgr4m/m mice, their cholesterol levels were obviously lower. This phenotype was further conﬁrmed in the context of ob/ob mice, in which lack of LGR4 dampened circadian rhythms of triglyceride. We next demonstrated that Lgr4 expression exhibited circadian rhythms in the liver tissue and primary hepatocytes in mice, but we did not detect changes in the expression levels or circadian rhythms of classic clock genes, such as Clock, Bmal1 (Arntl), Pers, Rev-erbs, and Crys,inLgr4m/m mice compared with their littermates. Among the genes related to the lipid metabolism, we found that the diurnal expression pattern of the Mttp gene, which plays an important role in the regulation of plasma lipid levels, was impaired in Lgr4m/m mice and primary Lgr4m/m hepatocytes. Taken together, our results demonstrate that LGR4 plays an important role in the regulation of plasma lipid rhythms, partially through regulating the expression of microsomal triglyceride transfer protein. These data provide a possible link Journal of Molecular between the peripheral circadian clock and lipid metabolism. Endocrinology (2014) 52, 133–143

http://jme.endocrinology-journals.org Ñ 2014 Society for Endocrinology Published by Bioscientiﬁca Ltd. DOI: 10.1530/JME-13-0042 Printed in Great Britain Downloaded from Bioscientifica.com at 09/29/2021 05:19:41PM via free access Research F WANG, X ZHANG and others LGR4 regulates circadian rhythm 52:2 134 of lipid

Introduction

Circadian rhythms are the daily oscillations of a lot of Recently, a rare nonsense mutation within the Lgr4 gene physiological functions in almost all mammalians. This was found strongly associated with low bone mineral circadian system is developed during the course of evolution density, electrolyte imbalance, late onset of menarche, to adapt to the daily changes in environmental conditions by and reduced testosterone levels in human (Styrkarsdottir anticipating sleep and activity periods (Eskin 1979, Takahashi et al. 2013). Several members of the leucine-rich repeat- 1995). These oscillations are driven by a central pacemaker containing GPCR family were found to participate in the located in the suprachiasmatic nuclei (SCN) of the hypo- regulation of the circadian rhythms of certain physiologi- thalamus (Schwartz & Gainer 1977). Similar clocks, which are cal activities. For example bursicon, the natural ligand of a synchronized by central clocks, have been found in peripheral Drosophila homolog to LGR4, DLGR2, was reported to act tissues, such as liver and adipose tissue (Welsh et al. 2004). downstream of the neuropeptide CCAP, which controls At the molecular level, central and peripheral oscillators the circadian timing of ecdysis behavior in Drosophila share a common molecular circuitry, with a battery of (Park et al. 2003, Mendive et al. 2005). LGR5, which is transcriptional activators and repressors forming feedback closely related to LGR4, was reported to be regulated by loops and generating oscillations in circadian transcription clock genes in epidermal stem cells (Lin et al. 2009, Janich factors (Ko & Takahashi 2006). These transcription factors et al. 2011). However, whether LGR4 plays a role in then exert effects on broader physiological processes circadian physiological behavior is not yet clear. through the rhythmic expression of output genes, a remar- In this study, we report a new role for LGR4 in the kable number of which are metabolic enzymes. In recent connections between circadian rhythms in gene expression decades, data on the circadian transcriptome and circadian and circadian oscillations in metabolic activity. We found m/m proteome have confirmed the concept of the mammalian that Lgr4 homozygous mutant (Lgr4 ) mice displayed a circadian clock as an important regulator of energy higher respiratory exchange ratio (RER) at night and an homeostasis (Storch et al. 2002, Reddy et al. 2006, altered circadian rhythmic pattern of energy homeostasis, Eckel-Mahan et al.2012, Vollmers et al.2012). especially regarding lipid metabolism. We further demon- As an important component of energy homeostasis, strated that LGR4 showed a rhythmic expression pattern in plasma lipid concentration is maintained within a narrow the liver, and a lack of LGR4 impaired the rhythmic range and exhibits circadian rhythm in mammals (Maillot expression of microsomal triglyceride transfer protein et al. 2005, Pan & Hussain 2007). Many molecules have (MTTP). These findings demonstrate that LGR4 appears to serve asa molecular link between the circadian oscillatorand Journal of Molecular Endocrinology been found to participate in the regulation of this process. For example, NR1D1 (REV-ERBa) participates in the circa- energy metabolism in peripheral tissue. dian modulation of SREBP signaling at the transcriptional level and is involved in cholesterol and lipid metabolism (Le Martelot et al.2009). A number of other genes, such as Materials and methods IRE1a, can influence metabolism at the posttranscriptional Animals level (Cretenet et al.2010). To gain further insights into the mechanisms that coordinate the control of the circadian Lgr4m/m mice were generated as previously described (Weng clock and diverse metabolic pathways, additional output et al.2008, Wang et al.2012). Three PCR primers were used genes need to be found. for genotyping: the common upstream primer A: 50-CCA Leucine-rich repeat-containing G protein-coupled GTC ACC ACT CTT ACA CAA TGG CTA AC-30;down- receptor 4 (LGR4/GPR48) is a member of the G protein- stream primer B: 50-ATT CCC GTA GGA GAT AGC GTC coupled receptor (GPCR) superfamily. LGR4 has been CTA-30; and downstream primer C: 50-GGT CTT TGA GCA reported to play a broad role in development, and Lgr4 CCA GAG GAC-30. Lgr4m/m mice and their WT littermates gene mutant mice display early neonatal lethality were age and gender matched throughout the experiments. (Mazerbourg et al. 2004). Our group and others have Male mice at the age of 8–10 weeks were used in the reported that LGR4 is involved in male infertility, experiments except the indirect calorimetry experiment in electrolytes homeostasis, the development of the ocular which both male and female mice were used. Lgr4 and anterior segment, and bone formation through different Leptin double mutant mice were produced by intercrossing downstream targets (Weng et al. 2008, Luo et al. Lgr4 heterozygous mutant mice with Leptin heterozygous 2009, Li et al. 2010, Wang et al. 2012, Siwko et al. 2013). mutant mice. The mice were maintained on a 12 h light:

1.5

A 1.1 WT ** m/m 1.0 1.0

0.9 (folds) RER 0.5 0.8 Dark:light ratio of RER

0.7 0.0 ZT0 ZT4 ZT8 ZT12 ZT16 ZT20 ZT24 WT m/m

1.5

B 1.1 WT * m/m 1.0 1.0

0.9 (folds) RER 0.5 0.8 Dark:light ratio of RER

0.7 0.0 ZT0 ZT4 ZT8 ZT12 ZT16 ZT20 ZT24 WT m/m

Figure 1 Lgr4m/m mice displayed an impaired circadian rhythm of the RER. The RER Lgr4m/m mice). Lights on is indicated by a white bar, and lights off is of male (A) and female (B) mice were measured in metabolic chambers indicated by a black bar (nZ4 for each genotype at each time point). m/m over 24 h (left, 24 h period; right, dark:light ratio of RER for WT and ZT, Zeitgeber time; m/m, Lgr4 ;*P!0.05; **P!0.01; error bars, S.E.M.

12 h darkness cycle and fed ad libitum. Dissected tissues manufacturer’s instructions. The absorbance ratio at were quickly frozen and stored at K80 8C. All procedures 260/280 nm of all of the RNA samples was checked using a wereapprovedbytheAnimalCareCommitteeof Nano-Drop ND-1000 spectrophotometer (Thermo Journal of Molecular Endocrinology Shanghai Jiaotong University School of Medicine. Fisher Scientiﬁc, Waltham, MA, USA), and all of the RNA samples were adjusted to the same concentration. The integrity of the RNA samples was also examined RNA isolation and RT by agarose gel electrophoresis. RT of 1 mg of RNA was Total RNA was extracted from liver samples using performed with the Reverse Transcription System the standard TRIzol reagent (Invitrogen) according to the (Promega, Madison, WI, USA).

Table 1 The circadian parameters of the plasma lipid levels in WT mice and Lgr4 gene mutant mice (Lgr4m/m) calculated using JTK_CYCLE analysis

JTK_CYCLE Circadian JTK_CYCLE P value JTK_CYCLE amplitude JTK_CYCLE phase

WT TG 0.0009 Yes 0.1848 22 Lgr4m/m TG 0.5568 No NA NA WT NEFA 0.0397 Yes 0.0315 2 Lgr4m/m NEFA 0.0040 Yes 0.1948 20 WT TC 1.0000 No 0.0983 10 Lgr4m/m TC 0.4557 No 0.1526 6 WT HDL-C 0.3968 No 0.0572 8 Lgr4m/m HDL-C 1.0000 No 0.0306 10 WT LDL-C 0.4869 No 0.1531 8 Lgr4m/m LDL-C 0.5219 No 0.1075 8

RT-PCR analysis Serum shock

RT-PCR was performed using the LC480 system (Roche, The cells were starved in high-glucose DMEM containing Penzberg, Germany) with SYBR Green Supermix (Takara, 0.5% (v/v) FBS for 12 h, followed by synchronization with Otsu, Shiga, Japan). The following RT-PCR conditions DMEM containing 50% (v/v) horse serum (tZ0) for 2 h, were applied over 50 cycles: 94 8C (30 s), 94 8C (5 s), and and the medium was then changed back to the starvation 60 8C (30 s). The primers used in this study are presented in medium (Balsalobre et al. 1998). Supplementary Table S1, see section on supplementary data given at the end of this article. Indirect calorimetry

Plasma lipid analysis The consumption of O2 and production of CO2 were determined using the Comprehensive Laboratory Animal Blood was collected in EDTA-coated tubes. All samples were Monitoring System (Columbus Instruments, Columbus, maintained on ice until being centrifuged at 960 g for 10 min. OH, USA), according to the manufacturer’s instructions. Serum lipids were measured using commercial kits according Data were recorded every 10 min. The RER was calculated to the manufacturer’s instructions. Specially, LabAssay NEFA as the ratio of the volume of CO2 produced to the volume kit (294-63601) and total triglyceride kit (290-63701) were of O2 consumed. obtained from Wako (Osaka, Japan). Total cholesterol, HDL-C, and LDL-C kit (KH-G-C-005) was obtained from Statistical analysis Shanghai Kehua Bio-engineering Co., Ltd (Shanghai, China). Values are expressed as the meanGS.E.M.One-wayANOVA was applied, followed by a t-test. Circadian parameters were Isolation of mouse hepatocytes analyzed using a new version of JTK_CYCLE software as Primary hepatocytes were isolated from mice by hepatic described previously (Hughes et al. 2010, Miyazaki et al.2011). portal collagenase perfusion as described previously (Hengstler et al. 2000). Brieﬂy, the whole liver was ﬁrst perfused with Hank’s Balanced Saline buffer (HBSS) in situ Results and then with collagenase solution (1% BSA and 0.05 Loss of LGR4 changes the rhythm of the RER collagenase in HBSS) for 10 min in plate. Dispersed cells were resuspended and seeded onto a plate. The hepato- To determine whether LGR4 plays a role in the circadian Journal of Molecular Endocrinology cytes were grown in high-glucose DMEM supplemented regulation of metabolism, the RER was measured in male with 10% (v/v) fetal bovine serum (FBS). (Fig. 1A) and female (Fig. 1B) Lgr4m/m mice and their WT

ABCTG TC NEFA 1.5 WT 1.5 WT 1.5 WT m/m * m/m m/m

1.0 1.0 1.0

0.5 0.5 0.5 Concentration (eq/l) Concentration (mmol/l) Concentration (mmol/l) 0.0 0.0 0.0 ZT0 ZT4 ZT8 ZT12 ZT16 ZT20 ZT0 ZT4 ZT8 ZT12 ZT16 ZT20 ZT0 ZT4 ZT8 ZT12 ZT16 ZT20

D LDL E HDL 2.0 0.8 WT WT m/m m/m 0.7 1.5 0.6 1.0 0.5 0.5 0.4 Concentration (mmol/l) Concentration (mmol/l) 0.0 0.3 ZT0 ZT4 ZT8 ZT12 ZT16 ZT20 ZT0 ZT4 ZT8 ZT12 ZT16 ZT20

Figure 2 Diurnal variations in plasma lipids in WT and Lgr4m/m mice. Plasma was levels were measured as described. Each time point represents the meanG obtained from male mice fed ad libitum at the indicated times. The time S.E.M.; nZ3–4 for each genotype at each time point. TG, total triglycerides; reﬂects circadian hours, with the lights on at ZT0 and off at ZT12. Total TC, total cholesterol; ZT, Zeitgeber time; m/m, Lgr4m/m;*P!0.05. triglyceride (A), total cholesterol (B), NEFA (C), LDL-C (D), and HDL-C (E)

A 2.5 TG than that of their WT littermates during the dark phase (P!0.010) and showed no difference during the light 2.0 OB DKO phase of the day, suggesting that the lack of LGR4 altered the circadian rhythm of lipid metabolism, and Lgr4m/m mice 1.5 consumed lower amounts of lipids, but more sugar * compared with WT mice (Hawley et al.2012). 1.0 **

0.5 Concentration (mmol/l) Loss of LGR4 changes the lipid rhythm in mice

0.0 To reveal further details regarding the diurnal variation of ZT0 ZT4 ZT8 ZT12 ZT16 ZT20 lipid metabolism, we next measured the plasma lipid levels in WT and Lgr4m/m mice at different time points. Table 1 provides the results of the statistical analysis with the B TC 15 circadian parameters of the plasma lipid levels. As shown m/m OB in Fig. 2A, the Lgr4 mice exhibited higher plasma *** DKO triglyceride levels at some time points and lost the rhythmic *** *** 10 *** pattern compared with WT mice. Lgr4m/m mice also *** *** presented a changed phase in their plasma non-esterified fatty acid (NEFA) levels, reflected by lower plasma NEFA 5 levels during the light phase and higher levels in the dark phase in Lgr4m/m mice in comparison with WT mice (Fig. 2B Concentration (mmol/l) and Table 1). The cholesterol levels in Lgr4m/m mice, 0 including those of total cholesterol (Fig. 2C), LDL-C ZT0 ZT4 ZT8 ZT12 ZT16 ZT20 (Fig. 2D), and HDL-C (Fig. 2E), did not show significant changes in their rhythmic patterns compared with WT mice,

Figure 3 with the exception of some slight alterations at certain time Diurnal variations in plasma lipids in ob/ob mice and Lgr4–Leptin double- points (Table 1). These results demonstrate that loss of LGR4 mutant mice. Plasma was obtained from male mice fed ad libitum at the results in an impaired plasma triglyceride rhythmic pattern. indicated times. The time reflects circadian hours, with the lights on at ZT0 and off at ZT12. Total triglyceride (A) and total cholesterol (B) levels were It is known that lipid levels and rhythmic patterns Journal of Molecular Endocrinology measured as described. The data represent the meanGS.E.M.; nZ7–8 for are greatly altered in obese mice (Hems et al. 1975). each genotype at each time point. TG, total triglyceride; TC, total To determine whether LGR4 can exert functions on cholesterol; ZT, Zeitgeber time; DKO, Lgr4–Leptin double mutant; *P!0.05; **P!0.01; ***P!0.001. the circadian rhythms of plasma lipids in obese mice, we measured the plasma lipid levels in Leptin-deficient mice littermates over 24 h using indirect calorimetry. Consistent (ob/ob) and Leptin and Lgr4 double-mutant (DKO) mice. with the findings of a previous report (Tu et al.2005), the Similar to the results obtained in Lgr4m/m mice and WT RER is higher in the dark phase than in the light phase in WT mice, the total plasma triglyceride levels loss circadian mice, suggesting the existence of a circadian rhythm rhythms which were found in the WT mice (Fig. 3A and in substrate utilization for energy source during the day, Table 2), while the total plasma cholesterol levels were more glucose in the dark phase, and more lipid usage in the lower and showed no significant change in their rhythmic light phase. However, the RER of Lgr4m/m mice was higher pattern (Fig. 3B and Table 2).

Table 2 The circadian parameters of the plasma lipid levels in ob/ob (OB) mice and Leptin and Lgr4 double-mutant (DKO) mice calculated using JTK_CYCLE analysis

JTK_CYCLE Circadian JTK_CYCLE P value JTK_CYCLE amplitude JTK_CYCLE phase

K OB TG 3.15!10 8 Yes 0.3486 0 DKO TG 0.0952 No 0.2009 2 OB TC 0.1327 No NA 0 K DKO TC 0.0655 No 8.35!10 5 2

2.0 Bmal12.5 Clock

2.0 1.5 WT WT m/m 1.5 m/m 1.0 1.0 0.5 0.5 Relative expression level Relative expression level 0.0 0.0

ZT0 ZT4 ZT8 ZT12 ZT16 ZT20 ZT0 ZT4 ZT8 ZT12 ZT16 ZT20

15 Per15 Per2

WT 4 WT 10 m/m m/m 3

2 5 1 Relative expression level

0 Relative expression level 0

ZT0 ZT4 ZT8 ZT12 ZT16 ZT20 ZT0 ZT4 ZT8 ZT12 ZT16 ZT20

Rev-erba Rev-erbb 30 40

WT WT 30 20 m/m m/m

10 Journal of Molecular Endocrinology 10 Relative expression level Relative expression level 0 0

ZT0 ZT4 ZT8 ZT12 ZT16 ZT20 ZT0 ZT4 ZT8 ZT12 ZT16 ZT20

Cry1 Cry2 4 5

4 3 WT WT 3 2 m/m m/m 2 1 1 Relative expression level Relative expression level 0 0

ZT0 ZT4 ZT8 ZT12 ZT16 ZT20 ZT0 ZT4 ZT8 ZT12 ZT16 ZT20

Figure 4 m/m Expression pattern of clock genes in WT and Lgr4 mice. Male mice fed ad The data represent the meanGS.E.M.; nZ3–4 for each genotype at each libitum were sacriﬁced at indicated time points. mRNA levels of the time point. ZT, Zeitgeber time; m/m, Lgr4m/m. clock genes were measured by real-time PCR and normalized to Gapdh.

Loss of LGR4 does not affect the central clock genes ZT16, indicating a circadian rhythm as analyzed using the in the liver JTK_CYCLE (PZ0.0057, Fig. 5A). Lgr4 expression level in Lgr4m/m mice was very low and the amplitude was The circadian rhythms of physiological activities are dampened (Fig. 5B). To determine whether Lgr4 controlled by clock genes. To examine whether the expression also exhibits circadian rhythms in vitro,we phenotype found in Lgr4m/m mice is a result of altered synchronized mice primary hepatocytes with 50% horse rhythms of components of the circadian clock, we next serum and then collected mRNA samples every 4 h over investigated the daily expression pattern of circadian genes a period of 48 h. As shown in Fig. 5B, Lgr4 presented a in the liver tissue in WT and Lgr4m/m mice. As shown in rhythmic expression pattern based on analysis with the Fig. 4, the circadian rhythms of all of the circadian genes in JTK_CYCLE (PZ0.0056, Fig. 5C). We also measure the WT mice were similar to those presented in previous reports expression levels of Bmal1 (Arntl) as positive control, we (Kume et al.1999, Preitner et al.2002, Yang et al.2006), but also found it presented a rhythmic expression pattern no signiﬁcant changes were found between WT and (PZ0.0296 Fig. 5D). Lgr4m/m mice (Fig. 4). These results suggest that LGR4 does not directly regulate clock genes. LGR4 ablation impairs the rhythmic expression of MTTP

LGR4 expression shows circadian rhythms in vivo To determine the molecular mechanism underlying the and in vitro impaired triglyceride rhythm observed in Lgr4m/m mice, we next measured the circadian rhythms of genes related Previous studies show a lot of genes involved in circadian to triglyceride metabolism in liver tissue using real-time rhythms as downstream output genes of the circadian PCR. Interestingly, out of genes involved in the metab- clock (Benito et al. 2010, Tong et al. 2010), so we examined olism and synthesis of triglycerides, such as Srebp1c, Lpl, whether the expression level of LGR4 shows a circadian and so on, we found that MTTP, which plays a role in the rhythm in following study. We measured the Lgr4 mRNA absorption of lipids and the circadian rhythms of plasma levels in the liver tissue of WT mice and Lgr4m/m mice lipid levels (Preitner et al. 2002, Pan & Hussain 2007, Pan during a 24 h period. As shown in Fig. 5A, Lgr4 expression et al. 2010), displayed the most signiﬁcant changes in its of WT mice was higher overall during the light phase than circadian rhythms among the genes we examined in the dark phase, presenting a peak at ZT4 and a nadir at Lgr4m/m mice at mRNA level (Fig. 6A). MTTP protein m/m AB showed diurnal variations in the WT but not in Lgr4

Journal of Molecular Endocrinology 2.0 Lgr4 0.6 livers, although analyzing the quantitative signals of 1.5 western blotting failed to detect circadian rhythms at 0.4 1.0 protein level (Fig. 6B). We next analyzed the expression 0.5 0.2 rhythms of Mttp in primary hepatocytes, and similar

Relative expression level 0.0

Amplitude (95% Conf. Int) results were obtained. Mttp mRNA levels showed a 0.0 m/m ZT0 ZT4 ZT8 ZT12 ZT16 ZT20 WT Lgr4 circadian periodicity of w28 h in the hepatocytes of WT CD 4 Lgr42.5 Bmal1 mice (PZ0.0059), but no rhythm was detected in the 3 2.0 hepatocytes of Lgr4m/m mice (PZ0.2891, Fig. 6B and 1.5 2 Table 3), which may explain the dampened circadian 1.0 1 m/m 0.5 rhythms of plasma triglyceride in Lgr4 mice compared Relative expression level 0 Relative expression level 0.0 with their WT littermates. Consistent with the overall 0 h 8 h 16 h 24 h 32 h 40 h 48 h 0 h 8 h 16 h 24 h 32 h 40 h 48 h lower but unaltered rhythms of plasma cholesterol seen in Lgr4m/m mice, we also detected a lower level of genes of the Figure 5 Lgr4 shows a rhythmic expression pattern in vivo and in vitro. (A) Male mice SREBP2 pathway but did not find significant changes in fed ad libitum were sacrificed at the indicated times. The levels of Lgr4 the circadian rhythm of these genes in Lgr4m/m mice mRNA in the liver tissue were detected via real-time PCR. The data (Fig. 6C, D, and E). represent the meanGS.E.M.(nZ3–4 for each genotype at each time point). (B) Amplitude of Lgr4 mRNA in the liver tissue with 95% CIs as error bars was plotted. (C and D) Primary hepatocytes from WT mice were synchronized with a serum shock at time 0 h, as described. Cells were Discussion harvested every 4 h at the indicated time points. The Lgr4 mRNA level (C) and Bmal1 (Arntl) mRNA level (D) were measured via real-time PCR GPCRs play critical roles in most biological processes and (P!0.05). ZT, Zeitgeber time. represent the most important group of drug targets

A Mttp WT D Srebp2 2.0 m/m 2.0 * WT ** m/m ** 1.5 * 1.5

1.5 1.0

0.5 0.5 Relative expression level Relative expression level 0.0 0.0

ZT0 ZT4 ZT8 ZT0 ZT4 ZT8 ZT12 ZT16 ZT20 ZT12 ZT16 ZT20

B E Hmgcr 4 WT MTTP GAPDH m/m WT 3 *

Lgr4m/m 2

ZT0 ZT4 ZT8 ZT0 ZT4 ZT8 ZT12ZT16ZT20 ZT12 ZT16 ZT20 1

Relative expression level 0

ZT0 ZT4 ZT8 ZT12 ZT16 ZT20 C Mttp F Hmgcs 2.5 4 WT

Journal of Molecular Endocrinology 2.0 m/m Lgr4 3 WT 1.5 m/m 2 1.0 1 0.5 Relative expression level

Relative expression level 0.0 0

0h 8h 16h 24h 32h 40h 48h ZT0 ZT4 ZT8 ZT12 ZT16 ZT20

Figure 6 Expression pattern of genes related to lipid metabolism in mice and expression level in primary hepatocytes (C) was also measured using real- primary hepatocytes. Primary hepatocytes were isolated and undergo time PCR. Genes related to cholesterol metabolism in mice (D, E, and F) serum shock as described. Male mice fed ad libitum were sacriﬁced at the were then measured by real-time PCR. The data represent the meanGS.E.M.; indicated time points. mRNA was extracted from primary hepatocytes and nZ3–4 for each genotype at each time point; ZT, Zeitgeber time; m/m, liver tissue, transcribed into cDNA as described. The mRNA levels (A) and Lgr4m/m;*P!0.05; **P!0.01. protein level (B) of Mttp in mouse liver was measured. Mttp mRNA

(Klabunde & Hessler 2002), but little is known about their through regulating cAMP levels in the SCN (Mertens role in the regulation of circadian rhythms. In Drosophila, et al. 2005). In mammals, cryptochrome proteins can the receptor for the neuropeptide, pigment-dispersing directly interact with the Gsa subunit of GPCRs to inhibit factor (PDF), exerts functions on circadian rhythms the rhythmic accumulation of cAMP in peripheral tissues

Table 3 The circadian parameters of the Mttp expression clockwork. Taking these findings together, they suggest that levels in primary hepatocyte of WT mice and Lgr4 gene mutant LGR4 may participate in the circadian modulation of mice (Lgr4m/m) calculated using JTK_CYCLE analysis metabolism. In plasma, most lipids, such as triglycerides, phospho- JTK_CYCLE Circadian JTK_CYCLE lipids, and cholesterol, are transported by forming Gene P value JTK_CYCLE amplitude protein–lipid complexes known as lipoproteins. Triglycer- WT Mttp 0.0059 Yes 0.11566651 ides are transported to the plasma by lipoproteins Lgr4m/m Mttp 0.2891 No 0.23140949 containing ApoB, which requires the chaperone protein MTTP (Shelness & Sellers 2001, Hussain et al. 2003). Pan (Zhang et al. 2010). In this study, we found that the daily and colleagues reported that MTTP is rhythmic and is rhythms of the RER and plasma lipid levels were impaired important for the daily variations in plasma lipid levels in Lgr4m/m mice, and we further observed that the (Pan & Hussain 2007, Pan et al. 2010). We found a similar expression levels of LGR4 were rhythmic in both liver expression pattern of MTTP in WT mice, which was tissue and cultured primary hepatocytes. We also detected impaired in the absence of LGR4. This finding suggests altered MTTP expression rhythms in the absence of LGR4, that the loss of MTTP rhythms due to the lack of LGR4 which may be the reason for the altered plasma activity at least partially explains the arrhythmic plasma m/m triglyceride circadian rhythm in Lgr4m/m mice. lipid phenotype observed in Lgr4 mice, although the Lipid homeostasis has long been known to exhibit detailed mechanisms underlying this phenomenon are circadian rhythms (Seaman et al. 1965, Schlierf & Dorow still unknown. We compared the gene expression profiles m/m 1973). Rodents exhibit nocturnal feeding behavior, and of liver tissue of WT and Lgr4 mice (data not shown) thus, the plasma lipid levels in mice were high at night and and found that genes involved in the regulation of Mttp, m/m low during the day (Pan & Hussain 2007). We consistently such as Pgc1a, Rxra, Foxa2, were all upregulated in Lgr4 detected similar circadian rhythms in plasma triglyceride mice. These genes can all upregulate the expression of Mttp levels in WT mice and ob/ob mice in this study, but these (Kang et al. 2003, Wolfrum & Stoffel 2006). This may also rhythms were impaired in the absence of LGR4. It is worth suggest that Lgr4 might be involved in the negative noting that the plasma cholesterol level was lower with regulation of Mttp. Further studies are needed to fully unaltered circadian rhythm in Lgr4m/m mice, which was understand the mechanisms involved in the crosstalk different from plasma triglyceride. between the clock, LGR4, and MTTP. It has recently been suggested that plasma lipid rhythms In summary, this study established a molecular link Journal of Molecular Endocrinology between circadian physiology and plasma lipid metabolism. are a result of the interplay between the circadian clock and Although it is possible that other unknown mechanisms metabolism. Almost all of components of the circadian clock, might be involved, our results suggest that LGR4 regulates such as CLOCK (DeBruyne et al.2007), BMAL1 (Shimba et al. the circadian rhythms of plasma lipids through MTTP. Lack 2011), CRYs, PERs(Grimaldi et al.2010), and REV-ERBs(Cho of LGR4 causes an arrhythmic plasma lipid phenotype in et al.2012), have been reported to regulate the plasma lipid mice. Considering that LGR4 is a member of the GPCR rhythms. To obtain further insights into the crosstalk, family, this study provides a potential means of regulating numerous screens were performed to find circadian clock metabolism through the regulation of the circadian output genes (Storch et al. 2002, Reddy et al.2006, functions. Eckel-Mahan et al. 2012). Eckel-Mahan and colleagues established a database entitled ‘CircadiOmics’ that provides a consolidated model of how the daily metabolome, Supplementary data This is linked to the online version of the paper at http://dx.doi.org/10.1530/ transcriptome, and proteome might work together JME-13-0042. (Eckel-Mahan et al.2012). In this database (http:// circadiomics.igb.uci.edu/), we found that the LGR4 Declaration of interest expression level is high during the light phase and low during The authors declare that there is no conflict of interest that could be the dark phase in liver tissue, which is consistent with the perceived as prejudicing the impartiality of the research reported. results of this study. However, the loss of the rhythmic activity of LGR4 did not affect the rhythms of clockwork Funding This work was supported by grants from the National Natural Science components in peripheral tissue. This suggests that LGR4 is Foundation of China (No. 81100601; No. 81270867; No. 81030011; regulatedbythecircadianclockandmaybeanoutputgeneof No. 30725037; No. 81100634; No. 81270931), Shanghai Pujiang Program

and Specialized Research Fund for the Doctoral Program of Higher Ko CH & Takahashi JS 2006 Molecular components of the mammalian Education (No. 20110073120104). circadian clock. Human Molecular Genetics 15 R271–R277. (doi:10.1093/ hmg/ddl207) Kume K, Zylka MJ, Sriram S, Shearman LP, Weaver DR, Jin X, Maywood ES, Hastings MH & Reppert SM 1999 mCRY1 and mCRY2 are essential References components of the negative limb of the circadian clock feedback loop. Cell 98 193–205. (doi:10.1016/S0092-8674(00)81014-4) Balsalobre A, Damiola F & Schibler U 1998 A serum shock induces circadian Le Martelot G, Claudel T, Gatfield D, Schaad O, Kornmann B, Sasso GL, gene expression in mammalian tissue culture cells. Cell 93 929–937. Moschetta A & Schibler U 2009 REV-ERBa participates in circadian (doi:10.1016/S0092-8674(00)81199-X) SREBP signaling and bile acid homeostasis. PLoS Biology 7 e1000181. Benito J, Hoxha V, Lama C, Lazareva AA, Ferveur JF, Hardin PE & (doi:10.1371/journal.pbio.1000181) Dauwalder B 2010 The circadian output gene takeout is regulated by Li XY, Lu Y, Sun HY, Wang JQ, Yang J, Zhang HJ, Fan NG, Xu J, Jiang JJ, Pdp1epsilon. PNAS 107 2544–2549. (doi:10.1073/pnas.0906422107) Liu RY et al. 2010 G protein-coupled receptor 48 upregulates estrogen Cho H, Zhao X, Hatori M, Yu RT, Barish GD, Lam MT, Chong LW, receptor a expression via cAMP/PKA signaling in the male reproductive Ditacchio L, Atkins AR, Glass CK et al. 2012 Regulation of circadian tract. Development 137 151–157. (doi:10.1242/dev.040659) behaviour and metabolism by REV-ERB-a and REV-ERB-b. Nature 485 Lin KK, Kumar V, Geyfman M, Chudova D, Ihler AT, Smyth P, Paus R, 123–127. (doi:10.1038/nature11048) Takahashi JS & Andersen B 2009 Circadian clock genes contribute to Cretenet G, Le Clech M & Gachon F 2010 Circadian clock-coordinated the regulation of hair follicle cycling. PLoS Genetics 5 e1000573. 12 Hr period rhythmic activation of the IRE1a pathway controls lipid (doi:10.1371/journal.pgen.1000573) metabolism in mouse liver. Cell Metabolism 11 47–57. (doi:10.1016/ Luo J, Zhou W, Zhou X, Li D, Weng J, Yi Z, Cho SG, Li C, Yi T, Wu X et al. j.cmet.2009.11.002) 2009 Regulation of bone formation and remodeling by G-protein- DeBruyne JP, Weaver DR & Reppert SM 2007 Peripheral circadian coupled receptor 48. Development 136 2747–2756. (doi:10.1242/ oscillators require CLOCK. Current Biology 17 R538–R539. (doi:10.1016/ dev.033571) j.cub.2007.05.067) Maillot F, Garrigue MA, Pinault M, Objois M, Theret V, Lamisse F, Eckel-Mahan KL, Patel VR, Mohney RP, Vignola KS, Baldi P & Sassone-Corsi P Hoinard C, Antoine JM, Lairon D & Couet C 2005 Changes in plasma 2012 Coordination of the transcriptome and metabolome by the triacylglycerol concentrations after sequential lunch and dinner in circadian clock. PNAS 109 5541–5546. (doi:10.1073/pnas.1118726109) healthy subjects. Diabetes & Metabolism 31 69–77. (doi:10.1016/ Eskin A 1979 Identification and physiology of circadian pacemakers. S1262-3636(07)70169-6) Introduction. Federation Proceedings 38 2570–2572. (doi:10.1146/ Mazerbourg S, Bouley DM, Sudo S, Klein CA, Zhang JV, Kawamura K, annurev.ph.40.030178.002441) Goodrich LV, Rayburn H, Tessier-Lavigne M & Hsueh AJ 2004 Leucine- Grimaldi B, Bellet MM, Katada S, Astarita G, Hirayama J, Amin RH, rich repeat-containing, G protein-coupled receptor 4 null mice exhibit Granneman JG, Piomelli D, Leff T & Sassone-Corsi P 2010 PER2 controls intrauterine growth retardation associated with embryonic and lipid metabolism by direct regulation of PPARg. Cell Metabolism 12 perinatal lethality. Molecular Endocrinology 18 2241–2254. (doi:10.1210/ 509–520. (doi:10.1016/j.cmet.2010.10.005) me.2004-0133) Hawley SA, Fullerton MD, Ross FA, Schertzer JD, Chevtzoff C, Walker KJ, Mendive FM, Van Loy T, Claeysen S, Poels J, Williamson M, Hauser F, Peggie MW, Zibrova D, Green KA, Mustard KJ et al. 2012 The ancient Grimmelikhuijzen CJ, Vassart G & Vanden Broeck J 2005 Drosophila drug salicylate directly activates AMP-activated protein kinase. Science molting neurohormone bursicon is a heterodimer and the natural FEBS Letters 336 918–922. (doi:10.1126/science.1215327) agonist of the orphan receptor DLGR2. 579 2171–2176. (doi:10.1016/j.febslet.2005.03.006) Journal of Molecular Endocrinology Hems DA, Rath EA & Verrinder TR 1975 Fatty acid synthesis in liver and Mertens I, Vandingenen A, Johnson EC, Shafer OT, Li W, Trigg JS, De Loof A, adipose tissue of normal and genetically obese (ob/ob) mice during the Schoofs L & Taghert PH 2005 PDF receptor signaling in Drosophila 24-hour cycle. Biochemical Journal 150 167–173. contributes to both circadian and geotactic behaviors. Neuron 48 Hengstler JG, Ringel M, Biefang K, Hammel S, Milbert U, Gerl M, Klebach M, 213–219. (doi:10.1016/j.neuron.2005.09.009) Diener B, Platt KL, Bottger T et al. 2000 Cultures with cryopreserved Miyazaki M, Schroder E, Edelmann SE, Hughes ME, Kornacker K, Balke CW hepatocytes: applicability for studies of enzyme induction. & Esser KA 2011 Age-associated disruption of molecular clock Chemico-Biological Interactions 125 51–73. (doi:10.1016/S0009-2797 expression in skeletal muscle of the spontaneously hypertensive rat. (99)00141-6) PLoS ONE 6 e27168. (doi:10.1371/journal.pone.0027168) Hughes ME, Hogenesch JB & Kornacker K 2010 JTK_CYCLE: an efficient Pan X & Hussain MM 2007 Diurnal regulation of microsomal triglyceride nonparametric algorithm for detecting rhythmic components in transfer protein and plasma lipid levels. Journal of Biological Chemistry genome-scale data sets. Journal of Biological Rhythms 25 372–380. 282 24707–24719. (doi:10.1074/jbc.M701305200) (doi:10.1177/0748730410379711) Pan X, Zhang Y, Wang L & Hussain MM 2010 Diurnal regulation of MTP Hussain MM, Shi J & Dreizen P 2003 Microsomal triglyceride transfer and plasma triglyceride by CLOCK is mediated by SHP. Cell Metabolism protein and its role in apoB-lipoprotein assembly. Journal of Lipid 12 174–186. (doi:10.1016/j.cmet.2010.05.014) Research 44 22–32. (doi:10.1194/jlr.R200014-JLR200) Park JH, Schroeder AJ, Helfrich-Forster C, Jackson FR & Ewer J 2003 Targeted Janich P, Pascual G, Merlos-Suarez A, Batlle E, Ripperger J, Albrecht U, ablation of CCAP neuropeptide-containing neurons of Drosophila Cheng HY, Obrietan K, Di Croce L & Benitah SA 2011 The circadian causes specific defects in execution and circadian timing of ecdysis molecular clock creates epidermal stem cell heterogeneity. Nature 480 behavior. Development 130 2645–2656. (doi:10.1242/dev.00503) 209–214. (doi:10.1038/nature10649) Preitner N, Damiola F, Lopez-Molina L, Zakany J, Duboule D, Albrecht U & Kang S, Spann NJ, Hui TY & Davis RA 2003 ARP-1/COUP-TF II determines Schibler U 2002 The orphan nuclear receptor REV-ERBa controls hepatoma phenotype by acting as both a transcriptional repressor of circadian transcription within the positive limb of the mammalian microsomal triglyceride transfer protein and an inducer of CYP7A1. circadian oscillator. Cell 110 251–260. (doi:10.1016/S0092-8674 Journal of Biological Chemistry 278 30478–30486. (doi:10.1074/ (02)00825-5) jbc.M304201200) Reddy AB, Karp NA, Maywood ES, Sage EA, Deery M, O’Neill JS, Wong GK, Klabunde T & Hessler G 2002 Drug design strategies for targeting Chesham J, Odell M, Lilley KS et al. 2006 Circadian orchestration of G-protein-coupled receptors. Chembiochem 3 928–944. (doi:10.1002/ the hepatic proteome. Current Biology 16 1107–1115. (doi:10.1016/ 1439-7633(20021004)3:10!928::AID-CBIC928O3.0.CO;2-5) j.cub.2006.04.026)

Schlierf G & Dorow E 1973 Diurnal patterns of triglycerides, free fatty acids, regulates metabolic hormone fibroblast growth factor 21 (FGF21) blood sugar, and insulin during carbohydrate-induction in man and during circadian cycles and feeding. Journal of Biological Chemistry 285 their modification by nocturnal suppression of lipolysis. Journal of 36401–36409. (doi:10.1074/jbc.M110.172866) Clinical Investigation 52 732–740. (doi:10.1172/JCI107235) Tu Y, Thupari JN, Kim EK, Pinn ML, Moran TH, Ronnett GV & Kuhajda FP Schwartz WJ & Gainer H 1977 Suprachiasmatic nucleus: use of 14C-labeled 2005 C75 alters central and peripheral gene expression to reduce food deoxyglucose uptake as a functional marker. Science 197 1089–1091. intake and increase energy expenditure. Endocrinology 146 486–493. (doi:10.1126/science.887940) (doi:10.1210/en.2004-0976) Seaman GV, Engel R, Swank RL & Hissen W 1965 Circadian periodicity in Vollmers C, Schmitz RJ, Nathanson J, Yeo G, Ecker JR & Panda S 2012 some physicochemical parameters of circulating blood. Nature 207 Circadian oscillations of protein-coding and regulatory RNAs in 833–835. (doi:10.1038/207833a0) a highly dynamic mammalian liver epigenome. Cell Metabolism 16 Shelness GS & Sellers JA 2001 Very-low-density lipoprotein assembly and 833–845. (doi:10.1016/j.cmet.2012.11.004) secretion. Current Opinion in Lipidology 12 151–157. (doi:10.1097/ Wang J, Li X, Ke Y, Lu Y, Wang F, Fan N, Sun H, Zhang H, Liu R, Yang J et al. 00041433-200104000-00008) 2012 GPR48 increases mineralocorticoid receptor gene expression. Shimba S, Ogawa T, Hitosugi S, Ichihashi Y, Nakadaira Y, Kobayashi M, Journal of the American Society of Nephrology 23 281–293. (doi:10.1681/ Tezuka M, Kosuge Y, Ishige K, Ito Y et al. 2011 Deficient of a clock gene, ASN.2011040351) brain and muscle Arnt-like protein-1 (BMAL1), induces dyslipidemia Welsh DK, Yoo SH, Liu AC, Takahashi JS & Kay SA 2004 Bioluminescence and ectopic fat formation. PLoS ONE 6 e25231. (doi:10.1371/journal. imaging of individual fibroblasts reveals persistent, independently pone.0025231) phased circadian rhythms of clock gene expression. Current Biology 14 Siwko S, Lai L, Weng J & Liu M 2013 Lgr4 in ocular development and 2289–2295. (doi:10.1016/j.cub.2004.11.057) glaucoma. Journal of Ophthalmology 2013 987494. (doi:10.1155/2013/ Weng J, Luo J, Cheng X, Jin C, Zhou X, Qu J, Tu L, Ai D, Li D, Wang J et al. 987494) 2008 Deletion of G protein-coupled receptor 48 leads to ocular anterior Storch KF, Lipan O, Leykin I, Viswanathan N, Davis FC, Wong WH & segment dysgenesis (ASD) through down-regulation of Pitx2. PNAS 105 Weitz CJ 2002 Extensive and divergent circadian gene expression in 6081–6086. (doi:10.1073/pnas.0708257105) liver and heart. Nature 417 78–83. (doi:10.1038/nature744) Wolfrum C & Stoffel M 2006 Coactivation of Foxa2 through Pgc-1b Styrkarsdottir U, Thorleifsson G, Sulem P, Gudbjartsson DF, Sigurdsson A, promotes liver fatty acid oxidation and triglyceride/VLDL secretion. Jonasdottir A, Oddsson A, Helgason A, Magnusson OT, Walters GB et al. Cell Metabolism 3 99–110. (doi:10.1016/j.cmet.2006.01.001) 2013 Nonsense mutation in the LGR4 gene is associated with several Yang X, Downes M, Yu RT, Bookout AL, He W, Straume M, Mangelsdorf DJ human diseases and other traits. Nature 497 517–520. (doi:10.1038/ & Evans RM 2006 Nuclear receptor expression links the circadian nature12124) clock to metabolism. Cell 126 801–810. (doi:10.1016/j.cell. Takahashi JS 1995 Molecular neurobiology and genetics of circadian 2006.06.050) rhythms in mammals. Annual Review of Neuroscience 18 531–553. Zhang EE, Liu Y, Dentin R, Pongsawakul PY, Liu AC, Hirota T, Nusinow DA, (doi:10.1146/annurev.ne.18.030195.002531) Sun X, Landais S, Kodama Y et al. 2010 Cryptochrome mediates Tong X, Muchnik M, Chen Z, Patel M, Wu N, Joshi S, Rui L, Lazar MA & circadian regulation of cAMP signaling and hepatic gluconeogenesis. Yin L 2010 Transcriptional repressor E4-binding protein 4 (E4BP4) Nature Medicine 16 1152–1156. (doi:10.1038/nm.2214)

Received in ﬁnal form 3 December 2013 Accepted 17 December 2013 Accepted Preprint published online 18 December 2013 Journal of Molecular Endocrinology