The Emerging Role of Zfp217 in Adipogenesis

Abstract

Zinc finger protein 217 (Zfp217), a member of the krüppel-type zinc finger protein family, plays diverse roles in cell differentiation and development of mammals. Despite extensive research on the functions of Zfp217 in cancer, pluripotency and reprogramming, its physiological roles in adipogenesis remain unknown. Our previous RNA sequencing data suggest the involvement of Zfp217 in adipogenesis. In this study, the potential function of Zfp217 in adipogenesis was investigated through bioinformatics analysis and a series of experiments. The expression of Zfp217 was found to be gradually upregulated during the adipogenic differentiation in C3H10T1/2 cells, which was consistent with that of the adipogenic marker gene Pparg2. Furthermore, there was a positive, significant relationship between Zfp217 expression and adipocyte differentiation. It was also observed that Zfp217 could not only trigger proliferative defect in C3H10T1/2 cells, but also interact with Ezh2 and suppress the downstream target genes of Ezh2. Besides, three microRNAs (miR-503-5p, miR-135a-5p and miR-19a-3p) which target Zfp217 were found to suppress the process of adipogenesis. This is the first report showing that Zfp217 has the capacity to regulate adipogenesis.

Keywords: 3T3-L1; C3H10T1/2; Zfp217; adipogenesis; high fat diet (HFD); miRNAs.

Figure 1

Zfp217 is implicated in adipogenesis based on Gene Expression Omnibus (GEO) datasets. (A) The heatmap of expression profiling of Zfp217 and marker genes in GSE76131. The resulting transcriptomic data-set includes six induction differentiation time points (0, 6, 48, 96, 192 and 384 h) in human simpson-golabi-behmel syndrome (SBGS) pre-adipocyte cells; (B,C) The heatmap of expression profiling of Zfp217 and marker genes in GSE87113. Expression profiling by RNA-seq during adipogenesis (0, 1, 2, 4 h, 2 and 7 days) in 3T3-L1 (B) and brown preadipocytes (C) in culture as indicated in the x axis, eBAT: embryo_BAT; aBAT: adult_BAT; (D) The line chart of expression profiling of Zfp217 and Pparg in diabetes mellitus (DM, the red group as indicated in the x axis), impaired glucose tolerance (IGT, the yellow group as indicated in the x axis) and normal glucose tolerance (NGT, the blue group as indicated in the x axis) adipose tissue samples. Data were obtained from GSE27951; and (E) The line chart of expression profiling of Zfp217 and marker genes in GSE20054. Expression profiling by Ezh2-flox/flox preadipocytes infected with retroviruses expressing Cre or vector alone.

Figure 2

Relative expression of Zfp217 in adipogenesis and different adipose tissues. (A) The expression profiles of Zfp217 during C3H10T1/2 adipogenesis (0, 6, 12, 24, 36, 48 h, 3, 4, 6, 8 and 10 days) (n = 3); (B) The protein level of Zfp217 and Pparg2 during C3H10T1/2 adipogenesis (0, 1, 2, 3, 4, 6, 8 and 10 days); (C) Relative expression of Pparg2 and Zfp217 using five kinds of adipose tissues of HFD/NCD. The five adipose tissues as indicated in the x axis (n = 5); and (D) The mRNA level of Zfp217 and pro-adipogenic marker genes (as indicated in the y axis) in NIH3T3, C3H10T1/2 and 3T3-L1 at non-differentiated and differentiated 6 h (n = 3). Data was standardized with β-actin and represented as means ± SEM. * p < 0.05, ** p < 0.01 versus control.

Figure 3

The effects of Zfp217 “gain-of-function” and “loss-of-function” on lipid droplet formation and triglyceride content in C3H10T1/2 and 3T3-L1. (A) Oil red O staining of Zfp217 “gain-of-function” and “loss-of-function” after 8 days of differentiation. Scale bar indicates 100 μm; and (B) Triglyceride GPO-POD assay kit (GPO: glycerol phosphate dehydrogenase; POD: peroxidase) and Bicinchoninic Acid (BCA) protein assay kit were used for analysis of the content of triglyceride. Total triglyceride was standardized with total protein and represented as means ± SEM (n = 4). * p < 0.05, ** p < 0.01 versus control.

Figure 4

Zfp217 suppresses cell proliferation and interacts with Ezh2, which represses Wnt signaling genes, to facilitate adipogenesis. (A) DNA synthesis of C3H10T1/2 cells was measured by 5-ethynyl-2′-deoxyuridine (EdU) incorporation assay after the indicated transfection. The EdU staining (red dots) represents the population of newborn cells. Hoechst 33342 staining (blue dots) was used to label cell nuclei. Scale bar indicates 100 μm; (B) EdU incorporation quantitative analysis. Five microscopic fields were randomly selected. EdU positive cells (%) were quantified by Image J, which was calculated by the formula: EdU positive cell (100%) = red dots/(red dots + blue dots) × 100%. All data were obtained from three reproducible experiments; (C) In the mammalian two-hybrid system, the Zfp217-Ezh2 interaction is confirmed by significantly higher luciferase activity in cells transfected with pACT-Zfp217 and pBIND-Ezh2, compared to cells transfected with one expression vector and one empty vector or two empty vectors. Results were displayed as firefly luciferase activity normalized to renilla luciferase activity and represented as means ± SEM (n = 4); (D) Schematic representation of the Mammalian Two-Hybrid System. The pG5luc Vector contains five GAL4 binding sites upstream of a minimal TATA box, which in turn is upstream of the firefly luciferase gene. The interaction between the two test proteins, expressed as GAL4-X and VP16-Y fusion constructs. An interaction between proteins X and Y brings the VP16 and GAL4 domains into close proximity, and results in an increase in luciferase expression over the negative controls; (E) The mRNA levels of Wnt6 and Wnt10b at non-induction, two days and four days post-adipogenic induction after treatment with pcDNA3.1-Zfp217 or empty pcDNA3.1 vector (n = 3); and (F) The mRNA levels of Cebpa and Pparg2 at four days post-adipogenic induction after treatment with pcDNA3.1-Zfp217 or empty pcDNA3.1 vector/Zfp217 siRNA or siRNA control (n = 3). Reverse transcriptase-quantitative polymerase chain reaction (qRT-PCR) data were standardized with β-actin and represented as means ± SEM. * p < 0.05, ** p < 0.01 versus control.

Figure 5

The miRNAs of targeting Zfp217. (A) The prediction of miRNAs which targeting Zfp217. The brilliant blue ring represents the predicted miRNAs in human (Homo sapiens, hsa). The pink ring represents the predicted miRNAs in mouse (Mus musculus, mmu). The gray blue ring represents the predicted miRNAs in both human and mouse. Venn diagram was output to statistical result; (B) Conservation of the miRNA binding site in the Zfp217-3′ unstranslated (UTR) region. The miRNA seed match region is highlighted in red; (C,D) Two pmirGLO vector constructs, containing either the Zfp217-3′ UTR-wt or the Zfp217-3′ UTR-mut with corresponding miRNA seed region, were transfected into BHK-21 cells either alone or in combination with negative control or each miRNA mimics. Renilla luciferase activity was used to normalize firefly luciferase activity. Data represents means ± SEM (n = 4); and (E,F) The expression levels of Zfp217 was measured by Western blot and qRT-PCR for different treated groups as indicated. The qRT-PCR data represents means ± SEM (n = 3). * p < 0.05, ** p < 0.01 versus control.

Figure 5

The miRNAs of targeting Zfp217. (A) The prediction of miRNAs which targeting Zfp217. The brilliant blue ring represents the predicted miRNAs in human (Homo sapiens, hsa). The pink ring represents the predicted miRNAs in mouse (Mus musculus, mmu). The gray blue ring represents the predicted miRNAs in both human and mouse. Venn diagram was output to statistical result; (B) Conservation of the miRNA binding site in the Zfp217-3′ unstranslated (UTR) region. The miRNA seed match region is highlighted in red; (C,D) Two pmirGLO vector constructs, containing either the Zfp217-3′ UTR-wt or the Zfp217-3′ UTR-mut with corresponding miRNA seed region, were transfected into BHK-21 cells either alone or in combination with negative control or each miRNA mimics. Renilla luciferase activity was used to normalize firefly luciferase activity. Data represents means ± SEM (n = 4); and (E,F) The expression levels of Zfp217 was measured by Western blot and qRT-PCR for different treated groups as indicated. The qRT-PCR data represents means ± SEM (n = 3). * p < 0.05, ** p < 0.01 versus control.

Figure 6

C3H10T1/2 (A) and 3T3-L1 (B) cells were stained with Oil Red O for different treated groups as indicated. One out of three independent but comparable experiments is shown. Scale bar indicates 100 μm; and (C) Schematic diagrams illustrating the mechanism of inducing adipogenesis by Zfp217. Zfp217 positively regulates adipogenesis by suppressing cell proliferation and interacting with Ezh2 which represses Wnt signaling genes. Zfp217 is directly targeted by miR-503-5p, miR-135a-5p and miR-19a-3p which impair adipocyte differentiation.