Article
Fatty Acid Oxidation in Zebrafish Adipose Tissue Is Promoted by 1a,25(OH)2D3 Graphical Abstract
Authors Xuyan Peng, Guohui Shang, Wenqing Wang, ..., Jiangyan He, Yi Zhang, Zhan Yin
Correspondence
[email protected]
In Brief Peng et al. find that Cyp2r1 depletion results in 1,25(OH)2D3 deficiency, retarded growth, and excessive visceral adipose tissue (VAT) in zebrafish. 1,25(OH)2D3 regulates lipid metabolism through the regulation of Pgc1a, controlling mitochondrial biogenesis and oxidative metabolism in zebrafish VAT.
Highlights d
Zebrafish Cyp2r1 is a major but not exclusive contributor to 25(OH)D in vivo
d
Cyp2r1-deficient zebrafish exhibit decreased somatic growth and obesity
d
Vitamin D signaling promotes fatty acid oxidation by inducing Pgc1a expression
Peng et al., 2017, Cell Reports 19, 1444–1455 May 16, 2017 ª 2017 The Authors. http://dx.doi.org/10.1016/j.celrep.2017.04.066
Cell Reports
Article Fatty Acid Oxidation in Zebrafish Adipose Tissue Is Promoted by 1a,25(OH)2D3 Xuyan Peng,1,2,6 Guohui Shang,1,2,6 Wenqing Wang,3,4,6 Xiaowen Chen,1,2 Qiyong Lou,1 Gang Zhai,1 Dongliang Li,5 Zhenyu Du,5 Yali Ye,4 Xia Jin,1 Jiangyan He,1 Yi Zhang,3,4 and Zhan Yin1,7,* 1State
Key Laboratory of Freshwater Ecology and Biotechnology, Institute of Hydrobiology, Chinese Academy of Sciences, Wuhan 430072, P. R. China 2University of Chinese Academy of Sciences, Beijing 100049, P. R. China 3Center for Genome Analysis, ABLife Inc., Wuhan 430075, P. R. China 4Laboratory for Genome Regulation and Human Health, ABLife Inc., Wuhan 430075, P. R. China 5Laboratory of Aquaculture Nutrition and Environmental Health, School of Life Sciences, East China Normal University, Shanghai 200241, China 6These authors contributed equally 7Lead Contact *Correspondence:
[email protected] http://dx.doi.org/10.1016/j.celrep.2017.04.066
SUMMARY
1a,25(OH)2D3 (vitamin D3) is crucial for mineral homeostasis in mammals, but the precise effects of 1a,25(OH)2D3 in adipose tissue remain to be clarified in vivo. The initial 25-hydroxylation is catalyzed by liver microsomal cytochrome P450 2R1 (CYP2R1), which is conserved in vertebrates. To probe the physiological function(s) of 1a,25(OH)2D3 in teleosts, we generated two independent cyp2r1-deficient zebrafish lines. These mutants exhibit retarded growth and increased obesity, especially in the visceral adipose tissue (VAT). These defects could be rescued with 25(OH)D3 treatments. ChIP-PCR analyses demonstrated that pgc1a is the target of the vitamin D receptor in the liver and VAT of zebrafish. Significantly decreased protein levels of Pgc1a, impaired mitochondrial biogenesis, and free fatty acid oxidation are also observed in the cyp2r1 mutant VAT. Our results demonstrate that regulation of 1a,25(OH)2D3 during lipid metabolism occurs through the regulation of Pgc1a for mitochondrial biogenesis and oxidative metabolism within zebrafish VAT. INTRODUCTION The vitamin D3 molecule has been found in early fossils, but its initial appearance was essentially as an inactive end product of 7-dehydrocholestol or its ergosterol equivalent. 1a,25(OH)2D3 [1,25(OH)2D3], the vitamin D3 active metabolite, performs multiple physiological functions, mostly via the vitamin D receptor (VDR). Over a 100-year time frame, nutritional vitamin D (VD) deficiency; altered VDR signaling, including VDR mutations (hereditary VD-resistant rickets); and deficient production of 1,25(OH)2D3 (Cyp27b1 mutations; pseudo-VD deficiency), were all found to have the same phenotypes as rickets. This suggests
that 1,25(OH)2D3 is critical for the regulation of calcium and phosphate metabolism and the mineralization of bone (Malloy and Feldman, 2012). The VDR is found in most cells, not just those involved with bone and mineral homeostasis, suggesting widespread actions of 1,25(OH)2D3 on many other physiological processes such as chronic metabolic, cardiovascular, and neoplastic diseases (Rosen et al., 2012). Significant controversy has emerged over the past decade concerning the effects of 1,25(OH)2D3 on adipocyte physiology and lipid metabolism. Links between low serum levels of 1,25(OH)2D3 and the pathophysiology of obesity, diabetes mellitus, and metabolic syndrome have been reported (Arunabh et al., 2003; Pittas et al., 2007; Wortsman et al., 2000). However, several systematic reviews of randomized, controlled trials of VD supplementation found no support for the hypothesis that VD supplementation could reduce the risk of obesity, type 2 diabetes, metabolic syndrome, or fat content (Elamin et al., 2011; Rosen et al., 2012). Moreover, the phenotypes of decreased overall fat mass and increased energy expenditure have been demonstrated in VDR / mice (Narvaez et al., 2009). Therefore, it seems that there is an active role for VD in adipocyte physiology, but the data from many observational and longitudinal cohort studies of various populations are in contrast to many clinical trials and existent animal models. An evident gap has remained regarding the precise function of VD in adipose tissue (AT). The VD endocrine system may be highly conserved throughout vertebrate evolution, as indicated by the combined presence of highly similar, specific VDR-metabolizing enzymes belonging to the CYP P450 family (Goldstone et al., 2010; Kollitz et al., 2015). Functional VDRs have been cloned from Xenopus laevis (Li et al., 1997); zebrafish (Lin et al., 2012); many other teleosts; and even sea lamprey (Petromyzon marinus), a jawless fish (Whitfield et al., 2003). The ancient origin of the endocrine system suggests that the VD/VDR system might play a more fundamental role than its modern actions in terrestrial animals that are related to a mineralized skeleton. It has also been demonstrated that VD/VDR signaling may be involved in the
1444 Cell Reports 19, 1444–1455, May 16, 2017 ª 2017 The Authors. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
regulation of Ca2+ uptake in zebrafish (Lin et al., 2012). However, whole-transcriptome analysis after 1,25(OH)2D3 treatment in zebrafish 7 days post-fertilization (dpf) revealed that most differentially expressed transcripts had a more general spectrum of activities, with lipid metabolism as the top affected pathway (upregulation of leptin, ppars, and genes for adipocyte differentiation) (Craig et al., 2012). Together, this suggests that 1,25(OH)2D3 regulates multiple biologically diverse pathways in vertebrates, along with evolutionary routes, which have not yet been recognized. 1,25(OH)2D3 is the principal active hormonal form of vitamin D3, and it is responsible for most of VD’s biological actions. Vitamin D3 is produced from 7-dehydrocholesterol, which is catabolized to 25(OH)D3 and then 1,25(OH)2D3. In mice, CYP2R1 is one of the major enzymes responsible for the 25-hydroxylation of vitamin D3 in liver tissue, and a reduction of more than 50% of the 25(OH)D3 level in Cyp2r1 / mouse serum has been observed. Strikingly, the levels of 1,25(OH)2D3 in Cyp2r1 / mouse serum remained at a similar level as those in the wildtype (control) mice, which might be the reason for the relatively healthy Cyp2r1-deficient mice observed (Zhu et al., 2013). In teleosts, 1,25(OH)2D3 has also been demonstrated to be produced by the liver and renal tissues. Because the zebrafish is a common teleost model, the zebrafish genes cyp2r1, cyp27b1, and cyp24a1 genes have been identified. Functional zebrafish VDRs have also been demonstrated to respond to 1,25(OH)2D3 and to be involved in calcium handling (Lin et al., 2012). However, no genetic in vivo model has yet been utilized for studying the physiological functions of vitamin D3 in teleosts, including zebrafish. Moreover, the zebrafish model offers unrivaled potential to explore the basal functions of VD/VDR signaling in teleosts during evolution. Fascinated with the existing insights into the potential roles of the VD/VDR signaling involved in lipid metabolism, we used zebrafish as a model to investigate this topic. Two independent cyp2r1-deficient zebrafish lines were generated and used for this study. Unlike the equivalent levels of 1,25(OH)2D3 in both Cyp2r1 / and wild-type mice, in cyp2r1-deficient fish, the levels of 1,25(OH)2D3 in plasma (serum), liver, and AT were reduced dramatically. The cyp2r1-deficient fish exhibited obvious retarded somatic growth and increased accumulation of AT. To our surprise, we detected similar levels of plasma calcium and phosphate in both mutant and wild-type fish. The administration of 25(OH)D3 rescued the defects of somatic growth and AT content. The observed resistance of adipose stores to mobilization during starvation in cyp2r1-deficient mutants also indicated the impairment of lipid mobilization during nutrient deprivation. Using the chromatin immunoprecipitation (ChIP)-PCR technique, genomic regulatory regions of pgc1a involved in lipid metabolism, especially in mitochondrial function, were found to interact with VDR in zebrafish liver and VAT. Our qRT-PCR results also revealed a significant downregulation of the transcription of pgc1a and many mitochondrial-related genes. Lastly, there were significant reductions in the levels of mitochondrial biogenesis in cyp2r1-deficient VAT, indicated by lower levels of the Pgc1a protein, levels of other mitochondrial proteins, mitochondrial membrane potential, and free fatty acid (FFA) oxidation activity. Our work demonstrated that one of the major basal func-
tions of VD/VDR in teleosts is the promotion of lipid metabolism via regulation of mitochondrial biogenesis in AT. RESULTS Zebrafish Cyp2r1, VDRs, and VD Metabolic Enzymes The cyp2r1 gene has been identified in the zebrafish genome as an ortholog of human CYP2R1, which encodes a microsomal VD 25-hydroxylase (Cheng et al., 2003; Goldstone et al., 2010). Alignment analysis of the protein sequences of the predicted zebrafish Cyp2r1 with those of other vertebrates showed that zebrafish Cyp2r1 shares 70% and 69% sequence identities with human CYP2R1 and mouse CYP2R1, respectively. As a member of the cytochrome P450 family, zebrafish Cyp2r1 was found to have the E-class cytochrome P450 group I signature, which was identified by InterProScan 5, and a cysteine residue that is the putative ligand with a heme cofactor (Figure S1A) (Cheng et al., 2003). Phylogenetic analysis indicated that zebrafish cyp2r1 has a closer homology with those of Fugu, Xenopus, mouse, and human than with its closest homology cytochrome P450 family members in zebrafish, cyp2u1, and cyp2v1 (Figure S1B) (Goldstone et al., 2010). Taken together, these features indicate that zebrafish Cyp2r1 is an ortholog of mammalian CYP2R1 and has potential VD 25-hydroxylase activity. RT-PCR was performed to analyze the expression patterns of zebrafish cyp2r1, vdra, and vdrb during early developmental stages and in adults (Figures 1A and 1B). The maternal and embryonic expression patterns of these genes are shown in Figure 1A. In adults, cyp2r1 was mainly expressed in the liver, VAT, and muscle. The vdra and vdrb genes were universally expressed, with a higher abundance of vdra transcript than vdrb transcript (Figure 1B). The levels of the bioactive form of VD, 1,25(OH)2D3, are dynamically balanced by several enzymes involved in VD metabolism, including CYP2R1, CYP27B1, and CYP24A1, which are members of the cytochrome P450 (CYP) superfamily (Omdahl et al., 2002; Prosser and Jones, 2004; Schuster et al., 2006). The expression levels of the genes encoding these three CYP member transcripts and the total lipid contents, as well as the plasma levels of 1,25(OH)2D3, at different ages of wild-type zebrafish were measured. cyp2r1 and cyp27b1, two genes encoding the enzymes related to active VD, were highly expressed in the livers of juvenile zebrafish (1, 2, and 3 months post-fertilization [mpf]), while their expression levels were dramatically decreased during the stages after sexual maturity (4 and 6 mpf) (Figures 1C and 1D). The expression levels of cyp24a1, the gene encoding the enzyme related to deactivating 1,25(OH)2D3, were very low before the adult stage and obviously increased at 4 and 6 mpf (Figure 1E). The plasma 1,25(OH)2D3 levels were much higher at 3 mpf, with an elevation of 69%, compared with those at 2 mpf. These were significantly decreased at 4 and 6 mpf, compared with those at 3 mpf, with reductions of 75% and 80%, respectively (Figure 1F). The total lipid content was not significantly different between 1 and 2 mpf. There was a remarkable reduction in lipid content at 3 and 4 mpf, with 36.8% and 44.1% reductions, respectively, compared with that at 2 mpf. Total lipid accumulation was increased by 50% at 6 mpf, compared with that at 4 mpf
Cell Reports 19, 1444–1455, May 16, 2017 1445
Figure 1. Relative Expression Levels of Zebrafish cyp2r1, VDRs, and Three Cytochrome P450 Genes Involved in VD Metabolism and the Dynamic Patterns of Total Body Lipid Content (A and B) Expression patterns of cyp2r1, vdra, and vdrb in (A) early-stage and (B) adult tissues. (C–E) The expression levels of (C) cyp2r1, (D) cyp27b1, and (E) cyp24a1 in wild-type zebrafish livers were assayed at 1, 2, 3, 4, and 6 mpf. The expression levels of the target genes are relative to the levels at 1 mpf. (F) 1,25(OH)2D3 concentration in plasma of wild-type zebrafish at 2, 3, 4, and 6 mpf. (G) The total lipid contents of wild-type zebrafish at 1, 2, 3, 4, and 6 mpf. Error bars indicate means ± SD. The different letters above the bars indicate significant differences. See also Figure S1.
(Figure 1G). This suggests an existing inverse correlation between plasma levels of 1,25(OH)2D3 and body lipid contents. Targeted Disruption of the cyp2r1 Gene Caused Retarded Growth in Zebrafish To investigate the functions of 1,25(OH)2D3 in zebrafish, the cyp2r1 gene was disrupted using the TALENs (transcription activator-like effector nucleases) technique. The TALEN target sites are shown in Figure 2A. Two independent mutant lines, mutant line 1 (M1) and mutant line 2 (M2), were obtained with 8- and 2-bp deletions, respectively (Figure 2B). The mutations resulted in premature stop codons that produced truncated proteins of 72 amino acids (aas) and 74 (aas) for Cyp2r1 in M1 and M2, respectively (Figure 2C). The typical AluI digestion patterns of the mutants (cyp2r1 / ) and of the heterozygous (cyp2r1+/ )
1446 Cell Reports 19, 1444–1455, May 16, 2017
and control (cyp2r1+/+) siblings in the F2 generation are shown in Figure 2D. The transcript levels of cyp2r1 were significantly increased in the livers of mutants, compared to those of the control siblings (Figure 2E). The expression levels of cyp27a1 and cyp27b1 were also significantly increased, whereas those of cyp24a1 were obviously decreased in the livers of mutants (Figure S2A). The changes in the transcription levels of the P450 cytochrome family members represent the typical negative-feedback regulation of diminished 1,25(OH)2D3 in cyp2r1-deficient fish (Lechner et al., 2007; Sundaram et al., 2014), which was confirmed by the measurements of the 1,25(OH)2D3 contents in fish plasma, liver, and VAT (Table 1). As shown in Table 1, a clear decrease in the 1,25(OH)2D3 levels in VAT, liver, and plasma was observed in the mutants, compared with those of control siblings with reductions of 87% in VAT, 53% in liver, and 68% in plasma.
Figure 2. cyp2r1 Gene Depletion in Zebrafish by TALENs (A) The binding site of engineered TALENs on the cyp2r1 gene exon (E)1 (exons are in boxes). The underlined blue font indicates the sequences of the two targeting arms of TALENs, and the red font shows the restriction enzyme AluI cutting site. (B) Mutation confirmation, as shown by the sequencing results of the transcripts of cyp2r1 gene of M1 and M2. (C) A diagram representative of WT and two truncated mutant Cyp2r1 proteins. (D) An agarose gel electrophoresis image with the PCR products following AluI digestion of the cyp2r1 locus of F2 fish. +/+, WT; +/ , heterozygous; / , homozygous. (E) The relative expression levels of cyp2r1 in the adult livers of mutant lines and their control siblings. **p < 0.01, compared with the control group. (F and G) Comparison of the morphological features of the cyp2r1 / fish and wild-type controls at (F) 5 dpf and (G) 100 dpf. See also Figure S2 and Tables S1 and S2.
Though no obvious defects were observed in the mutants in early developmental stages (Figure 2F), a retarded growth phenotype of a stunted body was exhibited in the adult mutants (Figure 2G). The survival rate was similar to those of the control siblings, but the body weights and standard lengths of the mutants decreased obviously after 25 dpf (Table S1). The classical role of VD is maintenance of calcium and phosphate homeostasis (Holick, 2011). Though some mineral contents were decreased, Ca content in our mutants was similar to that in the size-matched control fish (Table S2). The plasma Ca and P levels were similar in the mutants compared with the age-matched wild-type fish (Figures S2B and S2C). The assays of Alcian Blue staining for cartilage and Calcein staining for bone observation were utilized at the larval stage. Microcomputed tomography (microCT) was used for bone scanning at the adult stage. There were no obvious differences of the cartilage and bone integrality and shapes observed in the mutants based on the assays (Figures S2D–S2G). Cyp2r1 Deletion Caused Excessive VAT in Zebrafish An inverse correlation between lipid contents and 1,25(OH)2D3 levels along the aging was observed (Figures 1F and 1G); therefore, the features of lipid metabolism in cyp2r1-deficient fish were analyzed. The Nile-Red-stained visceral fat of larvae fish was viewed using both stereoscope (Figure 3A, a1 and a2) and confocal microscope (Figure 3A, a3 and a4) with size-matched mutant fish and wild-type control fish. In contrast to the size-
matched control larvae fish, the adipocyte numbers of VAT in the mutants were significantly increased (Figure 3B). The frequency of adipocyte areas was similar between larval fish of the two genotypes (Figure 3C), as well as in the adults (Figures 3F and 3G). The expansion of AT, especially VAT, was also observed in adult mutants compared to the size-matched controls (Figures 3D and 3E). Additionally, the total lipid contents of the male and female mutants were significantly increased compared to the age-matched control siblings, with elevations of 63% and 41%, respectively (Figure 3H). These results indicate that insufficient 1,25(OH)2D3 levels caused by cyp2r1 deletion resulted in excessive VAT accumulation, which was mainly due to the elevated numbers of adipocytes but not the expansion of adipocyte areas. Furthermore, compared with the age-matched control siblings, the mutants exhibited decreased lean content (percent dry weight) and water content (percent wet weight) (Table S3). Although the total lipids were accumulated excessively, the livers exhibited triglyceride levels in the mutants that were similar to those in age-matched control siblings (Figure S3). 25(OH)D3 Treatment Rescued the Defects of the Mutants 25(OH)D3, as the direct product of CYP2R1, was used to rescue the defects of the mutants. After 25(OH)D3 treatment, the mutants were no longer stunted, compared with the vehicle-treated control siblings (Figure 4A). Statistical analysis was performed on the weight, standard length, and total lipid content of agematched siblings between 25(OH)D3- and vehicle-treated F2generation fish. The weight and standard length of 25(OH)D3treated mutants were no longer lower and were even slightly higher than those of the vehicle-treated control siblings (Figures 4B and 4C). The total lipid content of 25(OH)D3-treated mutants
Cell Reports 19, 1444–1455, May 16, 2017 1447
Table 1. 1,25(OH)2D3 and FFA Concentration cyp2r1+/+
Test
Sample or Treatment
1,25(OH)2D3
adipose tissue (pg/mg)
17.48 ± 1.93a
liver (pg/mg)
94.77 ± 5.38a
plasma (pg/mL) FFA in plasma (mmol/L)
cyp2r1
441.5 ± 134 629.9 ± 81.6
300.17 ± 20.36
/
+ 25(OH)D3
17.65 ± 0.82a
45.02 ± 4.99b
947.79 ± 56.78
starvation
cyp2r1
2.33 ± 0.80b a
chow diet
/
114.57 ± 8.17c b
819.7 ± 79** 1171.5 ± 310.9**
1068.78 ± 60.49a ND ND
Different superscript letters indicate a significant difference across rows. **p < 0.01, compared with the control group (cyp2r1+/+). ND, no detection. See also Figures S4 and S5.
was significantly decreased, even with a reduction of 22%, compared with those of the vehicle-treated control siblings (Figure 4D). Otherwise, the insufficient 1,25(OH)2D3 levels in the mutants were restored (Table 1). These results demonstrate that the bioactive form of VD3 is essential for the regulation of lipid metabolism and somatic growth in zebrafish. Lipolysis Processes Were Not Affected in the Mutants Previous studies in mammals have demonstrated that, during starvation, adipocytes activate lipolysis, the breakdown of triglycerides into FFAs, and deliver FFAs into the mitochondrion to drive fatty acid oxidation to produce ATP for survival (Eaton, 2002; Finn and Dice, 2006; Kerner and Hoppel, 2000; O’Neill et al., 2013). A significant elevation of plasma FFA levels was observed in the mutants, compared with those of control siblings during both chow diets and starvation, with increases of 85.7% and 85.9%, respectively (Table 1). On the other hand, the extent of the increase in plasma FFA levels in response to the starvation challenge compared to that of chow-diet-fed mutants was 42.9%, similar to that of the control siblings (42.7%), which suggests that a normal lipolysis process is involved in the lipid mobilization of the mutants. However, mutants still had much more AT than the control siblings after starvation treatments for 10 or 35 days, as probed by Nile Red staining (Figure S4), indicating that the lipid mobilization process was hindered in the mutants during the starvation. Since our cyp2r1 mutant zebrafish exhibit similar phenotypes with gh1 (growth hormone) mutants (McMenamin et al., 2013), growth hormone (GH) signaling activation was examined (Figure S5). In contrast to control siblings, the amount of GH1 protein in the pituitary and the expression levels of ghra in the liver and VAT were dramatically increased (Figures S5A and S5C), the phosphorylation levels of Stat5 in the liver was clearly diminished (Figure S5B), and the expression levels of igf1 were increased in both VAT and livers of the mutants (Figure S5C). These indicated that the status of GH signaling was ambiguous in our mutants, and the mechanism was unknown. VD Receptor Bound the Endogenous Promoter of pgc1a As PGC1a is the master regulator of mitochondrial biogenesis (Ventura-Clapier et al., 2008), which is critical for the FFA oxidation process, we further explored the potential roles of VD/VDR signaling in PGC1a regulation. We performed ChIP-PCR assays and demonstrated that VDR interacts with the promoter region of pgc1a in both VAT (Figure 5A) and liver (data not shown). The transcript and protein levels of pgc1a were significantly dimin-
1448 Cell Reports 19, 1444–1455, May 16, 2017
ished in the VAT of the mutants, compared to that of the control siblings, and the reductions were restored by 25(OH)D3 treatment (Figures 5B and 5C), while in the livers, there was an elevation in the amount of Pgc1a in the mutants (Figure S6). We found the putative DBDs (DNA-binding domains) of VDR and RXR (retinoid X receptor) in the promoter region of pgc1a (Carlberg and Campbell, 2013). To evaluate whether the pgc1a promoter region could mediate transcriptional activation through a synergetic effect of VDR and 1,25(OH)2D3 and whether the putative DBDs of VDR and RXR were essential for this activation in vivo, a series of luciferase reporter constructs were generated (Figure 5D). Our experiments demonstrate that 1,25(OH)2D3 treatment could effectively augment pgc1a:luciferase activity but could not stimulate the mutant pgc1a: luciferase activity (Figure 5E). These results demonstrate that the putative DBDs of VDR and RXR were both necessary for pgc1a activation by 1,25(OH)2D3. These suggest that the stimulation of pgc1a by 1,25(OH)2D3 is mediated by VDR-RXR heterodimers, which is a common feature of VDR signaling. Mitochondrial Function Was Impaired in cyp2r1 / Fish VAT Next, the transcript levels of two PGC1a target transcription factors, nrf-1 (nuclear respiratory factor) and erra (estrogen-related receptor), through which PGC1a controls mitochondrial biogenesis and lipid oxidation (Scarpulla, 2011), were assayed (Figure 6A). The downregulated expression of these two genes in the mutants was rescued by 25(OH)D3 treatment. PPARg, which plays a central role in the functions of AT (Sharma and Staels, 2007), was upregulated in the VAT of mutants. The transcript levels of some mitochondria-related genes, such as ucp1 (uncoupling protein 1), mrpl15 (mitochondrial ribosomal protein L15), mrpl2 (mitochondrial ribosomal protein L2), mrps10 (mitochondrial ribosomal protein S10), mtfp1 (mitochondrial fission process 1), mcu (mitochondrial calcium uniporter) and slc25a20 (solute carrier family 25 [carnitine/acylcarnitine translocase], member 20) were significantly decreased, and these reductions were all restored by 25(OH)D3 treatment (Figure 6A). Finally, we tested several mitochondria-related parameters in the VAT and livers of mutants and control siblings. There was approximately 50% reduction in the relative mitochondrial gene levels in the VAT of mutants, indicating a decreased number of mitochondria in the VAT of mutants (Figure 6B), whereas no difference was observed in the livers (Figure S7A). In addition, the protein levels of two mitochondrial protein markers, Mtco1 (mitochondrial cytochrome c oxidase subunit 1, encoded by
Figure 3. Excessive VAT Developed in cyp2r1-Deficient Fish (A) The neutral lipids of juvenile mutants and sizematched controls (both with a standard length of 7.6 mm) were stained with Nile Red and viewed with stereoscope (a1 and a2) and confocal microscope (a3 and a4). (B and C) Shown here: (B) visceral adipocyte numbers and (C) area frequency of VAT in the confocal view (a3 and a4). (D and E) Shown here: (D) Nile Red fluorescence and (E) cross-sections through trunks of adult mutants and their size-matched controls show increased subcutaneous AT (arrows) and excessive VAT (arrowheads) in the mutants. Scale bars, 1 mm. (F) H&E-stained sections of VAT of adult mutants and their size-matched controls. Scale bars, 100 mm. (G) Frequency distribution of adipocyte cell-surface area. (H) The total lipid contents of males and females of both mutants and age-matched controls at 100 dpf. cyp2r1+/+, n = 5; cyp2r1 / , n = 4. Error bars indicate means ± SD. **p < 0.01; N.S., no statistically significant difference. See also Figure S3 and Table S3.
decreased in the mutants (Figure S7E). These results demonstrate mitochondrial dysfunction and an impaired biogenesis process in the VAT, but not in the livers, of cyp2r1 mutants, which might be the reason for the excessive VAT without hepatic steatosis. Taken together, cyp2r1 deletion caused a VD deficiency in zebrafish, which impaired mitochondrial lipid oxidation via the regulation of Pgc1a, a key regulator of mitochondrial biogenesis and oxidative metabolism. DISCUSSION
mtDNA) and Cox IV (cytochrome c oxidase IV, encoded by nuclear DNA), were significantly decreased in the VAT (Figure 6C), while they were both increased in the livers of mutants (Figure S7B). The loss of the mitochondrial membrane potential was significantly increased in the VAT of mutants (Figure 6D), while it was decreased in the livers (Figure S7C), indicating that mitochondrial function was impaired in the VAT, but not in the liver, of the mutants. This conclusion was further supported by the oxidation of [1-14C] palmitic acid in mitochondria (Figures 6E and S7D). The whole body oxygen consumption rate was
AT is the body’s largest energy reservoir and a major source of metabolic fuel. Stored lipids in AT are important for the survival of an organism when food supplies are limited or when energy consumption suddenly spikes. AT is an elementary feature of life that is under tight control. The regulation of lipid storage and mobilization must be finely tuned to precisely balance energy homeostasis in organisms (Zimmermann et al., 2004). Excessive AT in animals could be the result of defective lipid mobilization, including the lipolytic, FFA transportation, and FFA oxidation processes (McMenamin et al., 2013; Zimmermann et al., 2004). For example, the excessive storage of lipids in zebrafish VAT and impaired lipid mobilization in gh1-deficient mutants is detrimental to somatic growth (McMenamin et al., 2013). Among the many functions attributed
Cell Reports 19, 1444–1455, May 16, 2017 1449
Figure 4. 25(OH)D3 Rescue Experiments (A–D) Shown here: (A) the morphological features, (B) body weight, (C) standard length, and (D) total lipid contents of wild-type control, cyp2r1 mutant, and 25(OH)D3-treated cyp2r1 mutant fish at 100 dpf. N = 11. Error bars indicate means ± SD. The different letters above the bars indicate a significant difference at p < 0.01.
to 1,25(OH)2D3, its role in the control of lipid and energy metabolism has been recognized as having potential implications for chronic disorders, including obesity, diabetes, and nonalcoholic fatty liver disease (Arunabh et al., 2003; Pittas et al., 2007; Wortsman et al., 2000), although its actions are unclear because of conflicting reports (Elamin et al., 2011; Rosen et al., 2012). Based on the expression profile of the members of the cytochrome P450 family related to vitamin D3 metabolism along the ages of zebrafish, plasma 1,25(OH)2D3 levels in fish exhibit an inverse correlation with fish adiposity (Figures 1C–1G), which indicates the involvement of VD in lipid metabolism. Our zebrafish model offers unrivaled potential for both genetic manipulation and visualizing vertebrate tissue; thus, we can use them to compare lipid fluctuations in both 1,25(OH)2D3-deficient fish and their wild-type siblings in vivo. In this study, we demonstrated that cyp2r1-deficient fish have excess VAT relative to body size. Lower 1,25(OH)2D3 contents also resulted in elevated plasma FFA levels, reduced mitochondrial biogenesis, and FFA oxidation defects in VAT (Figure 6; Table 1). Moreover, an interaction between zebrafish VDR and pgc1a, the sensor for FFA involving fatty acid oxidation, was observed. Reduced amounts of Pgc1a and two mitochondrial component proteins were accompanied by defective mitochondrial membrane potential in the VAT (Figures 5 and 6). It has been implied that vitamin D3 may have initially evolved as a calcium regulatory hormone in teleosts, as indicated by
1450 Cell Reports 19, 1444–1455, May 16, 2017
evolutionary and physiological studies. It has been reported that vitamin D3/VDR signaling could upregulate trpv6 transcription in zebrafish (Lin et al., 2012). Compared with that of the control siblings, approximately 32% of the 1,25(OH)2D3 content remained in cyp2r1-deficient fish plasma, which might be the reason for similar plasma levels of calcium and phosphate in both control and mutant fish (Figure S2; Table 1). There was no significant difference between the transcriptional levels of trpv6 in the gills at the larval (5 dpf) and adult (100 dpf) stages in cyp2r1-deficient fish and control fish (data not shown). The liver plays a central role in the maintenance of systemic lipid homeostasis, while AT is specialized in energy storage. Stored triglycerides contribute to adipocytes in AT hypertrophy and are effectively mobilized in situations of energy deficit, such as fasting. However, in both the fed and fasted states, cycles of hydrolysis, re-esterification of triglycerides, and FFA oxidation occur in adipocytes in AT. The major expression sites of zebrafish cyp2r1 are liver, AT, muscle, and ovary (Figure 1B), while VDR is expressed ubiquitously in zebrafish. This finding indicates that the major production sites of 1,25(OH)2D3 include the liver, AT, and muscle. However, no evident hepatic steatosis was observed in histopathological analyses (data not shown), even though there was severe AT accumulation in the mutant fish. In regard to mitochondrial biogenesis, pgc1a transcription, expression levels
Figure 5. VD/VDR Signaling Regulates pgc1a Transcription (A) ChIP-PCR showed VDR associated with the promoter region of pgc1a. The assay was performed with the VDR antibody or control rabbit immunoglobulin G (IgG). The input indicates the nonimmunoprecipitated tissue lysate. (B and C) The decreased (B) transcript and (C) protein levels of pgc1a in the VAT of the mutants were both restored by 25(OH)D3 immersion. (D) A diagram of the zebrafish wild-type (WT) and the putative DBDs of VDR/RXR mutant form (M1, M2, and M3) of pgc1a promoter luciferase reporter. The ChIP target site with VDR is shown in the gray box ( 293 bp to 55 bp). The putative DBDs of RXR and VDR are shown with green and red boxes, respectively. (E) 1,25(OH)2D3 treatment augments the activity of WT promoter of pgc1a but not the mutant forms in zebrafish embryos. Error bars indicate means ± SD; **p < 0.01 versus control. The different letters above the bars indicate a significant difference. NS, no statistically significant difference. vdr-ab, VDR antibody. See also Figure S6.
of mitochondrial mRNAs, mitochondrial protein contents, and mitochondrial membrane potential levels, major alterations between those of the cyp2r1-deficient and control fish have been observed in VAT (Figures 5 and 6). However, a similarly defective phenotype was not observed in cyp2r1-deficient liver tissue (Figure S7). This might be due to the different sets of VDR co-factors present in the liver or AT cells. Moreover, a greater reduction rate in the 1,25(OH)2D3 contents of the VAT, compared with those in the liver or plasma in cyp2r1-deficient fish, was observed (Table 1). This might be one of the reasons why AT is the major target organ in cyp2r1-deficient fish. The mechanisms of VDR regulation of pgc1a and lipid oxidation remain unexplored.
With approximately 32% of the plasma 1,25(OH)2D3 levels remaining in the cyp2r1-deficient fish, compared with those of the control siblings, we can speculate that another potential enzyme (or enzymes) may exist, as in mice (Zhu et al., 2013). For instance, elevated expression levels of cyp27a1 have been seen in cyp2r1deficient livers (Figure S2A); it has previously been suggested as another viable potential VD3 25-hydroxylase in animal liver (Zhu et al., 2013). Approximately 68%, 53%, and 87% depletion of 1,25(OH)2D3 contents in the tissues of the plasma, liver, and VAT, respectively; excessive accumulation of VAT; and severe defects of somatic growth have been observed. Previously, both stimulation and inhibition of adipogenesis by 1,25(OH)2D3 have been reported in cell-culture models, including 3T3-L1 cells
Cell Reports 19, 1444–1455, May 16, 2017 1451
Figure 6. Impaired Mitochondrial Biogenesis and Functions in cyp2r1-Deficient VAT (A) Transcript levels of two Pgc1a target genes, nrf-1 and erra; ppary; and some mitochondrial-related genes such as ucp1, mrpl15, mrpl2, mrps10, mtfp1, mcu, and slc25a20 in the VAT of 25(OH)D3- or vehicle-treated mutants and control siblings. (B and C) Relative mtDNA levels (B) and amounts of the mitochondrial proteins Mtco1 and CoxIV (C) were decreased in the VAT of mutants. (D) Impaired mitochondrial membrane potential in the VAT of the mutants. The relative loss of mitochondrial membrane potential is indicated by the ratio of JC-1 monomers and aggregates. (E) Capacities of mitochondria to oxidize [1-14C] palmitic acid in the VAT of mutants and their control siblings. Error bars indicate means ± SD; **p < 0.01 versus controls. The different letters above the bars indicate significant difference. See also Figure S7.
and brown adipocytes (Blumberg et al., 2006; Ricciardi et al., 2015). Lean phenotypes, resistance to high-fat diet-induced obesity, decreased energy expenditure, increased oxygen consumption, and CO2 production have been seen in both of the VDR / and Cyp27b1 / mice in which the VD/VDR signaling was depleted completely (Narvaez et al., 2009; Wong et al., 2011; Zhu et al., 2013). Here, mutants had about 32% of the plasma 1,25(OH)2D3 and similar plasma levels of calcium and phosphate as control fish, which might be the cause of the discrepancies between our results and those reflecting a complete loss of VD/VDR signaling with severe hypocalcemia, such as rickets, seen in VDR / or Cyp27b1 / mice on regular rodent chow. Besides, with no definition of typical white and brown AT in teleosts (Minchin and Rawls, 2011), the mechanism underlying lipid metabolism in fish VAT and the link between ucp expression and non-shivering thermogenesis or energy expenditure remained unexplored. On the other hand, our data are in agreement with numerous epidemiological studies that have reported an inverse relationship between obesity and circulating
1452 Cell Reports 19, 1444–1455, May 16, 2017
levels of 25(OH)D3 (Arunabh et al., 2003; Pittas et al., 2007; Wortsman et al., 2000). It has been also reported that VD supplementation limited weight gain induced by high-fat diet in mice, which paralleled increased pgc1a expression, mitochondrial metabolism, and lipid oxidation in the brown AT (Marcotorchino et al., 2014). Our present results clearly demonstrate the role of zebrafish VD/ VDR signaling in promoting the transcription of pgc1a and FFA oxidation. The potential crosstalk between ligand-bound VDR and other nuclear receptors critical in lipid metabolism, such as PPAR and RXR, is still unknown. Despite vast observational and epidemiological evidence suggesting an association among VD status, obesity, and metabolic syndrome, all of the current systematic reviews of clinical trial data are unable to demonstrate a statistically positive improvement with VD supplemental treatment (Elamin et al., 2011; George et al., 2012). We speculate that there must have been changes in the pharmacodynamic relationship of the VD/VDR system throughout vertebrate evolution that may have been driven by changes in the protein-protein interactions between VDR and its coregulators. However, based on observations in our experiments, only 1,25(OH)2D3, the active form of vitamin D3, could effectively improve conditions when the plasma levels of 1,25(OH)2D3 decreased. The low expression levels of cyp2r1 in adult animals might not be able to convert enough of the inactive form of vitamin D3 (or cholecalciferol) supplemented in all the randomized clinic trials (Elamin et al., 2011; George et al., 2012). Our present data, based on the zebrafish model, revealed that 1,25(OH)2D3 is an important regulator of lipid metabolism in animals; understanding the roles of this hormone is applicable to the worldwide obesity epidemic.
EXPERIMENTAL PROCEDURES Zebrafish Maintenance Our zebrafish were maintained at 28.5 C in the system water with a light:dark cycle of 14 hr:10 hr (h) according to the zebrafish book. The fish were fed three times daily with newly hatched brine shrimp. The embryos were obtained by natural spawning and kept in egg water at 28.5 C. The developmental stages were determined by hours post-fertilization (hpf) or mpf (Kimmel et al., 1995). All animal procedures of this research were approved by the Animal Research and Ethics Committee of the Institute of Hydrobiology of the Chinese Academy of Sciences (Approval ID: IHB 2013724). Cyp2r1 Knockout by TALENs The sequence-specific TALEN effector pairs targeting the zebrafish cyp2r1 gene (NC_007118.6, NCBI) were designed according a previous report (Cermak et al., 2011) and assembled using the Golden Gate TALEN Kit purchased from Addgene. Two DNA-target TALEN arms flanked the AluI endonuclease domain, which was used for the detection of nucleotide changes. PCR was performed for genotyping using the primers (Table S4), and the products were analyzed by the AluI digestion and DNA sequencing. Finally, two independent cyp2r1 mutant lines were obtained. Most of the assays were performed with the zebrafish of the M1 unless specifically indicated. To avoid the sexual differences in metabolism, male fish were chosen in this study. Nile Red Staining and Sections Nile Red (N3013, Sigma) was dissolved in acetone at 1 mg/mL as the stock solution. Mutants and size-matched control zebrafish were immersed in system water containing 0.1 mg/mL Nile Red for 1 hr for larval fish or overnight for adults at 28 C in the dark. Images were obtained using an Olympus SZX16 FL Stereo Microscope at an excitation wavelength of 488 nm. For larvae, the VAT Nile Red fluorescence was clearly viewed with a Zeiss ISM 710 confocal microscope. The mutants and size-matched control fish were fixed to generated cross-sections with a vibratome following a previous report (McMenamin et al., 2013). For H&E staining, the VAT was isolated from the mutants and sizematched control fish, fixed in 10% formalin, embedded in paraffin, cut into 8-mm sections, and stained with H&E. The adipocyte areas were analyzed using the ImageJ software. Total Lipid Measurement The fish were sampled from the males and females of mutants and agematched control siblings (100 dpf). Total lipid contents (percent dry weight) were measured using the Folch procedure as previously described (Flynn et al., 2009). Rescue Experiments by 25(OH) D3 Immersion 25-hydroxyvitamin D3 monohydrate (catalog no. 17938, Sigma) was dissolved in 95% ethanol at 6 mM as the stock solution. A working solution of 25(OH)D3 at 6 nM was prepared by diluting the stock solution with system water. F2 embryos were incubated in the system water with or without 25(OH)D3 (6 nm) from the first feeding stage (5 dpf) to the sampled day (100 dpf), and the media were renewed daily. To avoid degradation and isomerization of 25(OH)D3 by light, the experiments were performed in a dark place. 1,25(OH)2 D3 Levels and FFA Concentration Plasma, livers and VAT of mutant fish and control siblings at 100 dpf were collected. 1,25(OH)2D3 (Sigma) and d6 (26,26,26,27,27,27)-1,25(OH)2D3 (Medical Isotopes) were used. The plasma (100 mL) and tissue (100 mg) contents of 1,25(OH)2D3 were extracted and measured according to previous reports (Hedman et al., 2014; van den Ouweland et al., 2010), using ultraperformance liquid chromatography-triple quadrupole mass spectrometry (UPLC-TQMS) (ACQUITY UPLC, Waters). Quantitation was performed using calibration curves generated with standards. Calibrators and samples were all triplicated. The plasma FFA levels were measured with the Free Fatty Acid Fluorometric Assay Kit (Item No. 700310, Cayman) according to the manufacturer’s procedures.
ChIP-PCR ChIP was conducted in the zebrafish VAT at 100 dpf according to a previous report (Haim et al., 2013). The VDR antibody (ab3508, Abcam) and a negative control antibody (normal rabbit immunoglobulin G [IgG]) were used. Purified DNA fragments were analyzed by PCR and sequencing. The primer sequences are listed in Table S4. Luciferase Reporter The zebrafish pgc1a proximal promoter (2,761 bp), including the VDR ChIP site, was amplified with the primers (Table S4). The promoter fragment was cloned into the pGL3-Basic luciferase reporter plasmid (Promega), and PCR-mediated site-directed mutagenesis was performed to generate deletions of putative DBDs of VDR and RXR (Carlberg and Campbell, 2013). The pGL3-basic-pgc1a-luc (wild-type or mutant) and pRL-SV40 (as internal control) plasmids were microinjected into embryos. The injected embryos were incubated in egg water containing 1,25(OH)2D3 (catalog no. D1530, Sigma) (10 nM) or vehicle. The luciferase activity was assayed at 24 hpf using the Dual Luciferase Reporter Assay System (Promega). The luciferase activity was normalized relative to the Renilla activity of the control (vehicle). The assays were performed in triplicate. Mitochondrial Membrane Potential and Relative mtDNA Levels 50-mg livers or VAT of mutants and control siblings at 100 dpf were used to perform mitochondrial isolation using the Tissue Mitochondria Isolation Kit (Beyotime) according to the manufacturer’s procedures. The mitochondrial membrane potential was studied using the MitoProbe JC-1 Assay Kit for Flow Cytometry (Life Technologies) following the protocol provided by the manufacturer. CCCP (carbonyl cyanide 3-chlorophenylhydrazone) was applied as a positive control. The relative loss of mitochondrial membrane potential was analyzed by studying the ratio of JC-1 monomers to JC-1-aggregates on a BD FACSVerse (BD Biosciences). The experiments were performed in triplicate. The total DNA was extracted from the livers or VAT of mutants and control siblings at 100 dpf using the E.Z.N.A. Tissue DNA Kit (Omega). RNA was eliminated with RNase A. The relative mtDNA levels were determined using the qRT-PCR method with the SYBR Green Realtime PCR Master Mix (Toyobo) using mt-nd1 and mt-nd6 as the mitochondrial gene targets and the neb and polg1 genes as the references for nuclear DNA content. The data were calculated with neb as the reference, and the data from control fish were normalized. The primers used are shown in Table S4 and partially referenced to Artuso et al. (2012). The experiments were performed in triplicate. FFA Beta-Oxidation The mitochondrial FFA beta-oxidation was performed as previously reported (Demizieux et al., 2002). The VAT and livers sampled from mutants and control siblings at 100 dpf were homogenized. [1-14C] palmitic acid (PerkinElmer) was used. The radioactivity was determined using Tri-Carb 4910TR Liquid Scintillation Analyzer (PerkinElmer). Statistical Analysis Statistical analysis was performed using Student’s t test in the Statistical Product and Service Solutions (SPSS) software. The results are expressed as the means ± SD, and a probability of p < 0.05 was considered to be significant. See also the Supplemental Experimental Procedures. SUPPLEMENTAL INFORMATION Supplemental Information includes Supplemental Experimental Procedures, seven figures, and four tables and can be found with this article online at http://dx.doi.org/10.1016/j.celrep.2017.04.066. AUTHOR CONTRIBUTIONS Z.Y., X.P., and G.S. designed the experiments, analyzed the data, and wrote the manuscript. X.P. and G.S. performed the experiments. W.W. performed the ChIP-PCR experiments. X.C., Q.L., G.Z., D.L., Z.D., Y.Y., X.J., J.H., and Y.Z. co-analyzed and discussed the results.
Cell Reports 19, 1444–1455, May 16, 2017 1453
ACKNOWLEDGMENTS
lites: 1,25 dihydroxyvitamin D2&3 measurement using a novel derivatization agent. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 953-954, 62–67.
This work was supported by the National Natural Science Foundation of China (No. 31530077), the National Basic Research Program of China (973 Program, 2014CB138602) and the Pilot Program A Project from the Chinese Academy of Sciences (XDA08010405).
Holick, M.F. (2011). Vitamin D: evolutionary, physiological and health perspectives. Curr. Drug Targets 12, 4–18.
Received: October 26, 2016 Revised: February 28, 2017 Accepted: April 24, 2017 Published: May 16, 2017
Kimmel, C.B., Ballard, W.W., Kimmel, S.R., Ullmann, B., and Schilling, T.F. (1995). Stages of embryonic development of the zebrafish. Dev. Dyn. 203, 253–310.
REFERENCES Artuso, L., Romano, A., Verri, T., Domenichini, A., Argenton, F., Santorelli, F.M., and Petruzzella, V. (2012). Mitochondrial DNA metabolism in early development of zebrafish (Danio rerio). Biochim. Biophys. Acta 1817, 1002–1011. Arunabh, S., Pollack, S., Yeh, J., and Aloia, J.F. (2003). Body fat content and 25-hydroxyvitamin D levels in healthy women. J. Clin. Endocrinol. Metab. 88, 157–161. Blumberg, J.M., Tzameli, I., Astapova, I., Lam, F.S., Flier, J.S., and Hollenberg, A.N. (2006). Complex role of the vitamin D receptor and its ligand in adipogenesis in 3T3-L1 cells. J. Biol. Chem. 281, 11205–11213.
Kerner, J., and Hoppel, C. (2000). Fatty acid import into mitochondria. Biochim. Biophys. Acta 1486, 1–17.
Kollitz, E.M., Zhang, G., Hawkins, M.B., Whitfield, G.K., Reif, D.M., and Kullman, S.W. (2015). Molecular cloning, functional characterization, and evolutionary analysis of vitamin D receptors isolated from basal vertebrates. PLoS ONE 10, e0122853. Lechner, D., Ka´llay, E., and Cross, H.S. (2007). 1a,25-dihydroxyvitamin D3 downregulates CYP27B1 and induces CYP24A1 in colon cells. Mol. Cell. Endocrinol. 263, 55–64. €ppner, H., and Demay, M.B. (1997). Cloning and charLi, Y.C., Bergwitz, C., Ju acterization of the vitamin D receptor from Xenopus laevis. Endocrinology 138, 2347–2353. Lin, C.-H., Su, C.-H., Tseng, D.-Y., Ding, F.-C., and Hwang, P.-P. (2012). Action of vitamin D and the receptor, VDRa, in calcium handling in zebrafish (Danio rerio). PLoS ONE 7, e45650.
Carlberg, C., and Campbell, M.J. (2013). Vitamin D receptor signaling mechanisms: integrated actions of a well-defined transcription factor. Steroids 78, 127–136.
Malloy, P.J., and Feldman, D. (2012). Genetic disorders and defects in vitamin D action. Rheum. Dis. Clin. North Am. 38, 93–106.
Cermak, T., Doyle, E.L., Christian, M., Wang, L., Zhang, Y., Schmidt, C., Baller, J.A., Somia, N.V., Bogdanove, A.J., and Voytas, D.F. (2011). Efficient design and assembly of custom TALEN and other TAL effector-based constructs for DNA targeting. Nucleic Acids Res. 39, e82.
Marcotorchino, J., Tourniaire, F., Astier, J., Karkeni, E., Canault, M., Amiot, M.-J., Bendahan, D., Bernard, M., Martin, J.-C., Giannesini, B., and Landrier, J.F. (2014). Vitamin D protects against diet-induced obesity by enhancing fatty acid oxidation. J. Nutr. Biochem. 25, 1077–1083.
Cheng, J.B., Motola, D.L., Mangelsdorf, D.J., and Russell, D.W. (2003). De-orphanization of cytochrome P450 2R1: a microsomal vitamin D 25-hydroxilase. J. Biol. Chem. 278, 38084–38093.
McMenamin, S.K., Minchin, J.E., Gordon, T.N., Rawls, J.F., and Parichy, D.M. (2013). Dwarfism and increased adiposity in the gh1 mutant zebrafish vizzini. Endocrinology 154, 1476–1487.
Craig, T.A., Zhang, Y., McNulty, M.S., Middha, S., Ketha, H., Singh, R.J., Magis, A.T., Funk, C., Price, N.D., Ekker, S.C., and Kumar, R. (2012). Research resource: whole transcriptome RNA sequencing detects multiple 1a,25-dihydroxyvitamin D(3)-sensitive metabolic pathways in developing zebrafish. Mol. Endocrinol. 26, 1630–1642.
Minchin, J.E., and Rawls, J.F. (2011). In vivo analysis of white adipose tissue in zebrafish. Methods Cell Biol. 105, 63–86.
Demizieux, L., Degrace, P., Gresti, J., Loreau, O., Noe¨l, J.-P., Chardigny, J.-M., Se´be´dio, J.-L., and Clouet, P. (2002). Conjugated linoleic acid isomers in mitochondria: evidence for an alteration of fatty acid oxidation. J. Lipid Res. 43, 2112–2122. Eaton, S. (2002). Control of mitochondrial b-oxidation flux. Prog. Lipid Res. 41, 197–239. Elamin, M.B., Abu Elnour, N.O., Elamin, K.B., Fatourechi, M.M., Alkatib, A.A., Almandoz, J.P., Liu, H., Lane, M.A., Mullan, R.J., Hazem, A., et al. (2011). Vitamin D and cardiovascular outcomes: a systematic review and meta-analysis. J. Clin. Endocrinol. Metab. 96, 1931–1942. Finn, P.F., and Dice, J.F. (2006). Proteolytic and lipolytic responses to starvation. Nutrition 22, 830–844. Flynn, E.J., Trent, C.M., and Rawls, J.F. (2009). Ontogeny and nutritional control of adipogenesis in zebrafish (Danio rerio). J Lipid Res. 50, 1641–1652. George, P.S., Pearson, E.R., and Witham, M.D. (2012). Effect of vitamin D supplementation on glycaemic control and insulin resistance: a systematic review and meta-analysis. Diabet. Med. 29, e142–e150. Goldstone, J.V., McArthur, A.G., Kubota, A., Zanette, J., Parente, T., Jo¨nsson, M.E., Nelson, D.R., and Stegeman, J.J. (2010). Identification and developmental expression of the full complement of cytochrome P450 genes in zebrafish. BMC Genomics 11, 643. Haim, Y., Tarnovscki, T., Bashari, D., and Rudich, A. (2013). A chromatin immunoprecipitation (ChIP) protocol for use in whole human adipose tissue. Am. J. Physiol. Endocrinol. Metab. 305, E1172–E1177. Hedman, C.J., Wiebe, D.A., Dey, S., Plath, J., Kemnitz, J.W., and Ziegler, T.E. (2014). Development of a sensitive LC/MS/MS method for vitamin D metabo-
1454 Cell Reports 19, 1444–1455, May 16, 2017
Narvaez, C.J., Matthews, D., Broun, E., Chan, M., and Welsh, J. (2009). Lean phenotype and resistance to diet-induced obesity in vitamin D receptor knockout mice correlates with induction of uncoupling protein-1 in white adipose tissue. Endocrinology 150, 651–661. O’Neill, H.M., Holloway, G.P., and Steinberg, G.R. (2013). AMPK regulation of fatty acid metabolism and mitochondrial biogenesis: implications for obesity. Mol. Cell. Endocrinol. 366, 135–151. Omdahl, J.L., Morris, H.A., and May, B.K. (2002). Hydroxylase enzymes of the vitamin D pathway: expression, function, and regulation. Annu. Rev. Nutr. 22, 139–166. Pittas, A.G., Lau, J., Hu, F.B., and Dawson-Hughes, B. (2007). The role of vitamin D and calcium in type 2 diabetes. A systematic review and meta-analysis. J. Clin. Endocrinol. Metab. 92, 2017–2029. Prosser, D.E., and Jones, G. (2004). Enzymes involved in the activation and inactivation of vitamin D. Trends Biochem. Sci. 29, 664–673. Ricciardi, C.J., Bae, J., Esposito, D., Komarnytsky, S., Hu, P., Chen, J., and Zhao, L. (2015). 1,25-Dihydroxyvitamin D3/vitamin D receptor suppresses brown adipocyte differentiation and mitochondrial respiration. Eur. J. Nutr. 54, 1001–1012. Rosen, C.J., Adams, J.S., Bikle, D.D., Black, D.M., Demay, M.B., Manson, J.E., Murad, M.H., and Kovacs, C.S. (2012). The nonskeletal effects of vitamin D: an Endocrine Society scientific statement. Endocr. Rev. 33, 456–492. Scarpulla, R.C. (2011). Metabolic control of mitochondrial biogenesis through the PGC-1 family regulatory network. Biochim. Biophys. Acta 1813, 1269–1278. €ssler, M., Schuster, I., Egger, H., Herzig, G., Reddy, G.S., Schmid, J.A., Schu and Vorisek, G. (2006). Selective inhibitors of vitamin D metabolism–new concepts and perspectives. Anticancer Res. 26 (4A), 2653–2668.
Sharma, A.M., and Staels, B. (2007). Peroxisome proliferator-activated receptor g and adipose tissue—understanding obesity-related changes in regulation of lipid and glucose metabolism. J. Clin. Endocrinol. Metab. 92, 386–395. Sundaram, K., Sambandam, Y., Tsuruga, E., Wagner, C.L., and Reddy, S.V. (2014). 1a,25-dihydroxyvitamin D3 modulates CYP2R1 gene expression in human oral squamous cell carcinoma tumor cells. Horm. Cancer 5, 90–97. van den Ouweland, J.M., Beijers, A.M., Demacker, P.N., and van Daal, H. (2010). Measurement of 25-OH-vitamin D in human serum using liquid chromatography tandem-mass spectrometry with comparison to radioimmunoassay and automated immunoassay. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 878, 1163–1168. Ventura-Clapier, R., Garnier, A., and Veksler, V. (2008). Transcriptional control of mitochondrial biogenesis: the central role of PGC-1a. Cardiovasc. Res. 79, 208–217. Whitfield, G.K., Dang, H.T., Schluter, S.F., Bernstein, R.M., Bunag, T., Manzon, L.A., Hsieh, G., Dominguez, C.E., Youson, J.H., Haussler, M.R., and Marchalonis, J.J. (2003). Cloning of a functional vitamin D receptor from the lamprey
(Petromyzon marinus), an ancient vertebrate lacking a calcified skeleton and teeth. Endocrinology 144, 2704–2716. Wong, K.E., Kong, J., Zhang, W., Szeto, F.L., Ye, H., Deb, D.K., Brady, M.J., and Li, Y.C. (2011). Targeted expression of human vitamin D receptor in adipocytes decreases energy expenditure and induces obesity in mice. J. Biol. Chem. 286, 33804–33810. Wortsman, J., Matsuoka, L.Y., Chen, T.C., Lu, Z., and Holick, M.F. (2000). Decreased bioavailability of vitamin D in obesity. Am. J. Clin. Nutr. 72, 690–693. Zhu, J.G., Ochalek, J.T., Kaufmann, M., Jones, G., and Deluca, H.F. (2013). CYP2R1 is a major, but not exclusive, contributor to 25-hydroxyvitamin D production in vivo. Proc. Natl. Acad. Sci. USA 110, 15650–15655. Zimmermann, R., Strauss, J.G., Haemmerle, G., Schoiswohl, G., BirnerGruenberger, R., Riederer, M., Lass, A., Neuberger, G., Eisenhaber, F., Hermetter, A., and Zechner, R. (2004). Fat mobilization in adipose tissue is promoted by adipose triglyceride lipase. Science 306, 1383–1386.
Cell Reports 19, 1444–1455, May 16, 2017 1455
