Study of expression analysis of SIRT4 and the coordinate regulation of bovine adipocyte differentiation by SIRT4 and its transcription factors

Sirtuins, NAD+-dependent deacylases and ADP-ribosyltransferases, are critical regulators of metabolism involved in many biological processes, and are involved in mediating adaptive responses to the cellular environment. SIRT4 is a mitochondrial sirtuin and has been shown to play a critical role in maintaining insulin secretion and glucose homeostasis. As a regulator of lipid homeostasis, SIRT4 can repress fatty acid oxidation and promote lipid anabolism in nutrient-replete conditions. Using real-time quantitative PCR (qPCR) to explore the molecular mechanisms of transcriptional regulation of bovine SIRT4 during adipocyte differentiation, we found that bovine SIRT4 is expressed at high levels in bovine subcutaneous adipose tissue. SIRT4 knockdown led to decreased expression of adipogenic differentiation marker genes during adipocyte differentiation. The core promoter of bovine SIRT4 was identified in the −402/−60 bp region of the cloned 2-kb fragment containing the 5′-regulatory region. Binding sites were identified in this region for E2F transcription factor-1 (E2F1), CCAAT/enhancer-binding protein β (CEBPβ), homeobox A5 (HOXA5), interferon regulatory factor 4 (IRF4), paired box 4 (PAX4), and cAMP responsive element-binding protein 1 (CREB1) by using Electrophoretic mobility shift assay (EMSA) and luciferase reporter gene assay. We also found that E2F1, CEBPβ, and HOXA5 transcriptionally activate SIRT4 expression, whereas, IRF4, PAX4, and CREB1 transcriptionally repress SIRT4 expression. We further verified that SIRT4 knockdown could affect the ability of these transcription factors (TFs) to regulate the differentiation of bovine adipocytes. In conclusion, our results shed light on the mechanisms underlying the transcriptional regulation of SIRT4 expression in bovine adipocytes.


Introduction
Sirtuins are a conserved family of proteins comprising NAD + -dependent deacetylases and ADP-ribosyltransferases, and thus their functions are intrinsically connected with cellular metabolism [1]. Accumulating evidence has indicated that sirtuins are important sensors of energy status and that they protect cells against metabolic stress [2]. The biological regulatory functions of sirtuins vary from metabolism to cell survival, and sirtuins are involved in a range of physiological and pathological activities, such as control of ageing [3,4], longevity pathways [5], DNA repair [6], transcriptional silencing [3], apoptosis [7], and the control of metabolic enzymes [8]. Sirtuins regulate these processes and are in turn regulated by diet and environmental stress. The mammalian sirtuin family contains seven members (SIRT1-7), which function to regulate metabolism in a non-redundant manner in many tissues [9]. Because each member exhibits distinct targets, functions, and subcellular localizations such as the nucleus (SIRT1, SIRT2, SIRT3, SIRT6, SIRT7), cytoplasm (SIRT1, SIRT2), and mitochondria (SIRT3, SIRT4, SIRT5), sirtuins can coordinate cellular responses to calorie restriction throughout the organism [10,11].
SIRT4 is a member of mitochondrial sirtuins, which, because of their dependence on NAD + as a cofactor, are metabolic sensors of cellular energy status [12]. Although it was originally reported to be an ADP-ribosyltransferase that is located in the mitochondrial matrix, SIRT4 is capable of removing several acyl-groups from lysine residues. Previous studies have reported that SIRT4 affects important biological pathways, especially those involved in metabolic diseases such as diabetes, obesity, and cancer [13][14][15][16].
SIRT4 is highly expressed in islet β cells and can interact with adenine nucleotide translocator 2/3 (ANT2/3) and insulin degrading enzyme (IDE) [17]. SIRT4, through its ADP-ribosylation action, down-regulates the enzymatic activity of glutamate dehydrogenase (GDH) an enzyme, which promotes the metabolism of glutamate and glutamine, generates ATP, and stimulates insulin secretion. Therefore, SIRT4 represses the ability of β cells to secrete insulin in response to these amino acids [18]. On the other hand, SIRT4 removes three acyl moieties from lysine residues, namely methylglutaryl (MG)-, hydroxymethylglutaryl (HMG)-, and 3-methylglutaconyl (MGc)-lysine. The metabolites leading to these post-translational modifications are intermediates in leucine oxidation, and SIRT4 plays a key role in controlling this pathway [19]. Leucine has been identified as a potent insulin secretagogue in pancreatic islets [20], and studies have shown that dysregulated leucine metabolism in SIRT4 KO mice leads to elevated basal and stimulated insulin secretion, which progressively develops into glucose intolerance and insulin resistance [21]. These findings identify a robust enzymatic activity for SIRT4, and suggest it as a crucial player in maintaining insulin secretion and glucose homeostasis.
SIRT4 acts as a key factor of adipogenesis, regulating the proliferation and differentiation of preadipocytes [22,23]. Recent studies have identified that SIRT4 acts as a critical regulator of lipid homeostasis [24], inhibits fatty acid oxidation, and promotes lipid anabolism, which indicates that SIRT4 can regulate the balance between fat oxidation and synthesis. SIRT4 also regulates lipid homeostasis by deacetylating malonyl-CoA decarboxylase (MCD). By inhibiting MCD via deacetylation and resulting in a buildup of malonyl-CoA, SIRT4 reduces carnitine palmitoyltransferase-1 (CPT1) and subsequently blocks fatty acid oxidation. SIRT4 not only inhibits fatty acid oxidation (in skeletal muscle), but also induces lipogenesis in white fatty tissue through regulation of MCD activity [25]. Moreover, SIRT4 KO mice showed increased fatty acid oxidation in various tissues such as the liver and myotubes, and were protected from obesity caused by high-fat diet, while remaining insulin resistant. Therefore, we can infer that SIRT4 regulates insulin sensitivity via energy metabolism in numerous metabolic tissues [22]. However, the molecular mechanisms associated with bovine SIRT4 expression during adipose development are still unclear.
Transcriptional regulation is a vital physiological process in all living organisms. It is orchestrated by transcription factors (TFs) and regulatory proteins working coordinately together [26]. Transcriptional regulation allows the cell or organism to respond to a variety of intracellular and extracellular signals and thus adapt accordingly. Promoter is a region of DNA fragment, which may bind RNA polymerase, TFs, and other proteins for the successful initiation of transcription directly upstream of the particular gene [27]. In general, the promoter is located near the transcription start site (TSS) of the gene, and upstream on the DNA toward the 5 region of the sense strand, and TFs can be divided in two main categories, namely activators and repressors [28]. The regulation of sequence-specific TFs such as E2F TF-1 (E2F1), CCAAT/enhancer-binding protein β (CEBPβ), homeobox A5 (HOXA5), interferon regulatory factor 4 (IRF4), paired box (PAX) 4 (PAX4), and cAMP responsive element-binding protein 1 (CREB1) is coordinated in part by the action of adipocytes [29][30][31][32][33][34]. However, the relationship between differentiation and the transcriptional regulation of the SIRT4 promoter activity still needs to be clarified.
In the present study, the mechanisms underlying the molecular regulation of the SIRT4 promoter were analyzed. In addition, we determined the relative mRNA expression pattern of bovine SIRT4 in tissues, and its subcellular localization in bovine adipocytes. Thus, the present study aimed to elucidate the transcriptional regulation of SIRT4 in bovine adipocyte differentiation.

Ethics statement
All the experimental procedures on cattle used in the present study have been approved by the Experimental Animal Manage Committee (EAMC) of Northwest A&F University (2011-31,101,684). All the operations and experimental procedures complied with the national Standards of Laboratory Animal-Guideline for ethical review of animal welfare (GB/T 35892-2018) and Guide for the Care and Use of Laboratory Animals: Eighth Edition.

Animal sources and genomic DNA isolation
Qinchuan cattle were raised with free access to food and killed in the National Beef Cattle Improvement Center (Yangling, China). Genomic DNA was extracted from cattle blood samples, using standard method [35], and stored at −20 • C until subsequent analyses.

Real-time quantitative PCR analysis of gene expression patterns
Fifteen tissue types (testis, subcutaneous fat, liver, lung, brain, kidney, longissimus thoracis, heart, spleen, large intestine, small intestine, omasum, rumen, abomasum, and reticulum) were obtained from three adult Qinchuan cattle. Total RNA was extracted from the tissues using a Total RNA kit (Tiangen, Beijing, China), and cDNA was generated using a PrimeScript TM RT Reagent kit with gDNA Eraser (Perfect Real Time; TaKaRa, Dalian, China). Real-time quantitative PCR (qPCR) was performed using a SYBR Green PCR Master Mix kit (TaKaRa, Dalian, China), the 7500 real time PCR System and the sequence detecting software SDS V 1.4.0 (Applied Biosystems, U.S.A.). All the procedures were performed in accordance with the manufacturer's instructions. All the primers used for qPCR are listed in Table 1. β-actin was included as the endogenous control gene. The relative expression levels of the target mRNAs were analyzed using the 2 − C t method [36].

Immunofluorescence assays
Bovine adipocytes were fixed with 4% paraformaldehyde for 20 min at room temperature and permeabilized with 0.1% Triton X-100 for 10 min. Non-specific binding was blocked by treating permeabilized bovine adipocytes with 5% BSA for 30 min. These cells were then incubated with primary antibody against SIRT4, 2 μg/ml (ab10140; Abcam, U.K.), at 37 • C for 2 h, and subsequently incubated with 2% BSA and Cy3-conjugated Donkey Anti-Rabbit IgG (Sangon Biotech, China) at 1:100 dilution for 30 min. All washes were performed using 1× PBS. An anti-fade solution containing 50 ng/ml DAPI (Solarbio, China) was also used to stain the nuclei. Finally, images of the immunofluorescence staining were captured with an Olympus IX71 microscope (Olympus Optical, Tokyo, Japan).

Promoter cloning and generation of luciferase reporter constructs
To clone the bovine SIRT4 promoter region, gene-specific primers (SIRT4-PF/PR, Table 1) were designed to amplify a 2-kb promoter region upstream of the TSS. PCR amplifications were performed using genomic DNA, from Qinchuan cattle blood, as a template and KOD DNA Polymerase (Toyobo, Osaka, Japan). After digestion with the restriction enzymes Kpn I and Xho I (TaKaRa, Dalian, China), the 2-kb bovine SIRT4 promoter fragment, was ligated on to the pGL3-basic vector and named as pGL3-P1. Using pGL3-P1 as a template and with specific primers (Table 1) designed to incorporate Kpn I and Xho I, plasmids with unidirectional deletions in the promoter sequence, namely pGL3-P2, P3, P4, P5, P6, P7, P8, and P9, were generated by PCR amplification.

Cell isolation, culture, and transfection
Bovine adipocytes were isolated from inguinal adipose tissue collected from 3-day-old post-natal Qinchuan fetal cattle. The inguinal adipose tissue was minced and digested at 37 • C for 1.5 h with 1 mg/ml collagenase type I (Invitrogen, U.S.A.), followed by filtration through a nylon mesh (200 μm pore size). The cells were collected by centrifugation at 1500 g for 10 min and washed thrice with Dulbecco's modified Eagle's medium: Nutrient Mixture F-12 (DMEM/F12; GIBCO, U.S.A.) and suspended in DMEM/F12 medium supplemented with 10% FBS (PAN-Biotech, South America) and antibiotics (100 IU/ml penicillin and 100 μg/ml streptomycin). Subsequently, the cells were seeded on to 100 mm × 20 mm cell culture dishes at a density of 5 × 10 4 cells/cm 2 and incubated at 37 • C in a humidified atmosphere with 5% CO 2 . After 12 h, fresh DMEM/F12 medium supplemented with 10% FBS and antibiotics (henceforth, this culture medium will be referred to as DMEM/F12 medium and only the other supplementations will be mentioned) was added to the adhered cells. To verify if the isolated bovine cells were actually undifferentiated adipocytes, the medium was replaced with adipogenic differentiation inducing culture medium (DMEM/F12 medium supplemented with 0.5 mM 3-isobutyl-1-methylxanthine (IBMX), 1 μM dexamethasone (DEX), and 5 μg/ml insulin) to induce differentiation and incubated for 4 days. We found that with induction both the number and the diameter of the lipid droplets increased, which confirmed that this procedure isolated undifferentiated bovine adipocytes. Transfections were carried out in 3T3-L1 and bovine adipocytes maintained in DMEM/F-12 medium at 37 • C with 100% humidity and 5% CO 2 . Cells were grown overnight in 24-well plates to a confluency of 80%. Each well was treated with the transfection mixture comprising 800 ng of the expression construct (pGL3-P1, P2, P3, P4, P5, P6, P7, P8, and P9), 10 ng of pRL-TK normalizing vector, 2 μl of X-tremeGENE HP DNA transfection reagent (Roche, U.S.A.), and 100 μl of Opti-MEM (GIBCO; Invitrogen) and incubated for 6 h (in triplicate). The pGL3-basic vector  The restriction enzyme sites (Kpn I and Xho I) are indicated by the italicized letters in the forward (F) and reverse (R) primers used for promoter cloning and generating overexpression systems. The underlined bases are core putative TF binding sites. Abbreviation: EMSA, electrophoretic mobility shift assay.
served as a negative control. Six hours after transfection, adipogenic differentiation was induced in the 3T3-L1 and bovine adipocytes using the adipogenic differentiation inducing culture medium and incubated for 42 h. Cell lysates were collected 48 h post-transfection and used to measure the relative transcriptional activity of each fragment using the Dual-Luciferase Reporter Assay System (Promega, Madison, WI, U.S.A.) according to the manufacturer's instructions. Relative luciferase activities were determined using a NanoQuant Plate TM (TECAN, infinite M200PRO).
In order to elucidate the regulation of bovine adipocyte differentiation by SIRT4 and its TFs, differentiation was induced in the undifferentiated bovine adipocytes after transfection with siRNA. To this end, undifferentiated adipocytes were grown in six-well plates until they reached a confluency of 90% (with DMEM/F-12 medium at 37 • C with 100% humidity and 5% CO 2 ). At this stage, they were transiently transfected with siRNA, 50 nM siRNA was added to each well. Adipogenic differentiation was induced, 6 h after transfection, in culture batches meant to be differentiated by replacing the DMEM/F12 medium with adipogenic differentiation inducing culture medium and incubated for 4 days before the total RNA was harvested. While, the medium of the culture batches meant to be left undifferentiated was replaced with DMEM/F-12 medium only and then incubated for 4 days before harvesting of the total RNA. si-NC served as a control group.

Site-directed mutagenesis
To carry out site-directed mutagenesis, putative TF binding sites were first analyzed using the Genomatix suite. The identified putative TF binding sites for E2F1, CEBPβ, HOXA5, IRF4, PAX4, and CREB1 were then mutated using the corresponding primers (Table 1) and the QuickChange Site-Directed Mutagenesis Kit (Stratagene, La Jolla, CA, U.S.A.). The procedure was carried out in accordance with the manufacturer's instructions. PCR conditions used were also in accordance with the instructions in the kit. The products of the above site-directed mutagenesis procedure were treated with Dpn I, and then amplified using XL10-Gold competent cells (Stratagene); the resulting vectors were sequenced for confirmation.

Electrophoretic mobility shift assays
A LightShift Chemiluminescent EMSA Kit (Thermo Fisher Corp., Waltham, MA, U.S.A.) was used for this assay in accordance with the manufacturer's protocol with some modifications. All DNA probes (Table 1)  . Finally, the DNA-protein complexes were separated on a 6% non-denaturing polyacrylamide gel by electrophoresis, using 0.5× Tris-borate-EDTA buffer, for 1 h. Images were captured using the molecular imager ChemiDoc TM XRS+ system (Bio-Rad, Hercules, California, U.S.A.).

Statistical analysis
Statistical calculations were performed using the Statistical Analysis System (SAS) v8.0 (SAS Institute, Cary, NC). Statistical significance was determined using one-way ANOVA test. Results for gene expression and luciferase assay were based on three independent experiments. Data are expressed as mean + − S.D.; P<0.05 was considered significant (* denotes significance at P<0.05 and ** denotes significance at P<0.01).

Gene expression pattern and subcellular localization of SIRT4 in bovine adipocytes
The spatial expression pattern of bovine SIRT4 was studied, and the results showed highest expression in the testis followed by subcutaneous fat (Figure 1A; P<0.05). We then measured mRNA levels of SIRT4 during bovine adipocyte differentiation. We found that SIRT4 expression increased before the seventh day of differentiation, which was followed by a decline over the remainder of differentiation ( Figure 1B; P<0.05). To further explore the effect of SIRT4 on the differentiation of bovine adipocytes, the expression levels of SIRT4 were effectively reduced by siRNA during differentiation. Results showed that the expression of marker genes that promote bovine adipocyte differentiation was significantly inhibited after SIRT4 knockdown ( Figure 1C; P<0.05). The cytoplasmic subcellular localization of bovine SIRT4 was confirmed by immunofluorescence experiments; SIRT4 was found to localize in adipocyte mitochondria ( Figure 1D). These results suggested that SIRT4 might be tightly associated with adipose development.

Isolation and sequence analysis of the core promoter of bovine SIRT4
We further explored the activity of potential cis-acting elements and identified the key sequence required for their activity. We constructed a series of reporter vectors with progressively larger deletions, from the 5 end of the promoter. Results of luciferase reporter assays using 3T3-L1 and bovine adipocytes identified a core promoter in the −402 bp/−60 bp region of SIRT4 (Figure 2A,B; P<0.05). In addition, the analysis using the MethPrimer program discovered a CpG island located in the −378 bp/+18 bp region, which was found to be highly similar to the core promoter in the −402 bp/−60 bp region ( Figure 2C). We further analyzed regulatory elements in the SIRT4 −2077 bp/+44 bp promoter region using the Genomatix suite. Six motifs including TF binding sites for E2F1, CEBPβ, HOXA5, IRF4, PAX4, and CREB1 were identified in the core promoter region ( Figure 2D).

Identification of TFs that function as transcriptional activators or repressors in the SIRT4 core promoter region
To identify the functions of TFs in the regulation of bovine SIRT4, we constructed a series of DNA plasmids with 4-bp point mutations in the TF binding motifs, and transfected these into 3T3-L1 and bovine adipocytes. As shown, mutating IRF4 site in the pGL−402/+44 construct and the PAX4 and CREB1 sites in the pGL−160/+44 construct resulted in a significant increase in SIRT4 promoter activity; in contrast, mutating the E2F1, CEBPβ, and HOXA5 sites in the pGL−402/+44 construct resulted in a significant decrease in SIRT4 promoter activity ( Figure 3A; P<0.05). Next, siRNAs targetting E2F1, CEBPβ, HOXA5, IRF4, PAX4, and CREB1 were designed, synthesized, and transfected into bovine adipocytes to detect the knockdown efficiency of respective genes. Results indicated that the expression levels of these TFs were effectively reduced ( Figure 3B; P<0.05). We then utilized these siRNAs to determine the effect of knocking down the TFs involved in SIRT4 expression. E2F1, CEBPβ, and HOXA5 knockdown down-regulated SIRT4 mRNA expression, whereas ablation of IRF4, PAX4, and CREB1 up-regulated SIRT4 mRNA levels ( Figure  3C; P<0.05). To further validate the results, siRNAs targetting the TFs and pGL−402/+44 or pGL−160/+44 were co-transfected, which obviously altered activity of the SIRT4 promoter. The transcriptional activity was significantly decreased when E2F1, CEBPβ, or HOXA5 were knocked down, whereas it was increased when IRF4, PAX4, or CREB1 were suppressed ( Figure 3D; P<20.05).

TFs bind to the core promoter region of SIRT4
Electrophoretic mobility shift assays (EMSAs) were then performed to assess binding of E2F1, CEBPβ, HOXA5, IRF4, PAX4, and CREB1 to the SIRT4 core promoter. As shown (lane 2, Figure 4A), bovine adipocyte nuclear protein binds the 5 -biotin labeled E2F1 probes and forms one predominant complex. Competition assays determined that the mutant probe had a negligible effect on this complex (lane 3, Figure 4A). At the same time, the specific E2F1/DNA interaction was prevented by competition from excess unlabeled DNA (lane 4, Figure 4A). The last lane indicates that the complex was super-shifted upon incubation with an anti-E2F1 antibody (lane 5, Figure 4A). CEBPβ, HOXA5, IRF4, PAX4, and CREB1 probes yielded similar results ( Figure 4B,C,D,E, and F). In the all EMSAs, we found that the upshifted bands were diminished with antibody addition. These results indicated that the TFs bind to the core promoter region of the SIRT4 in vitro.

TFs and SIRT4 gene cooperatively regulated the relative expression of adipocyte differentiation marker genes
To confirm whether the degree to which the bovine adipocytes had differentiated was appropriate for the study, we first ascertained the relative expression levels of marker genes in undifferentiated and differentiated cultures of adipocytes; results revealed that the expression of the marker genes changed significantly in the cultures induced to differentiate ( Figure 5A; P<0.05). The differences in expression of the six TFs were also ascertained in undifferentiated and differentiated bovine adipocytes ( Figure 5B; P<0.05).
In order to detect the regulation of differentiation in bovine adipocytes by SIRT4 and TFs, we knocked down the SIRT4 and TFs using siRNAs. The results showed that these TFs have a significant regulatory effect on the expression of marker genes during adipocyte differentiation. In particular, interference of E2F1, CEBPβ, and HOXA5 significantly down-regulated the expression of adipogenesis marker genes, which promote adipocyte differentiation ( Figure  5C-E; P<0.05). However, the expression of adipogenic differentiation marker genes was significantly up-regulated with the interference of IRF4, PAX4, and CREB1 ( Figure 5F-H; P<0.05). Since these TFs bind to the core promoter region of SIRT4, and SIRT4 also has a significant regulatory effect on adipocyte differentiation, we assumed that TFs coordinate with SIRT4 to regulate adipocyte differentiation. In order to verify this hypothesis, we knocked down SIRT4 with interference of each TF separately by siRNAs during adipocyte differentiation, and then detected the expression level of the differentiation marker genes. The results showed that TFs and SIRT4 interact coordinately in promoting adipocyte differentiation. E2F1, CEBPβ, HOXA5, and SIRT4 all increase the expression of adipogenic marker genes. Moreover, SIRT4 synergized the effects of E2F1, CEBPβ, and HOXA5 on adipocyte differentiation Based on above results, we concluded that SIRT4 is a vital downstream effector of these TFs in not only the pathways regulating adipocyte differentiation, but also in pathways regulating fatty acid oxidation, lipid deposition, and adipocyte growth.

Discussion
Energy metabolism is an important aspect of metabolism; metabolism of glucose and lipid is critical to maintain cell energy homeostasis. In the physiological and biochemical processes of cells, glucose and lipid metabolism is primarily regulated by caloric restriction, oxidative stress, or other changes in energy [12]. Therefore, the focus of biological research has been on the main molecules that regulate metabolism in cell survival and homeostasis. Amongst these regulators, sirtuins have been the new research hotspots because of their important roles in regulating and maintaining glucose and lipid homeostases. Moreover, the biological functions of sirtuins have been confirmed to vary vastly from being involved in metabolism to being core regulators of cell survival [12,37].
Mammalian SIRT4 is localized to the mitochondria and plays an important role in regulating mitochondrial gene expression and fatty acid oxidation in liver and adipocytes [25,38,39]. The decreased expression level of SIRT4 usually leads to a significant increase in the expression of enzymes involved in fatty acid oxidation metabolism [40]. SIRT4 liver-specific KO mice display an increased expression of PPARα target genes associated with fatty acid catabolism [22]. SIRT4 tissue-specific KO mouse models could also be critical providing a comprehensive view of SIRT4 functions in fatty acid metabolism. In the present study, we found high mRNA expression of SIRT4 in bovine subcutaneous fat; SIRT4 expression increased before the seventh day of differentiation and then decreased with the differentiation of bovine adipocytes; further, the expression of marker genes that promote bovine adipocyte differentiation were significantly inhibited by SIRT4 knockdown. These results suggested that SIRT4 might have an important role in regulating the development of bovine adipose tissue [17,18].
The E2F TF family plays a crucial role in modulation of tumor suppressor proteins and in the control of cell cycle; it has also been identified as a target of the transforming proteins of small DNA tumor viruses [41,42]. E2F1 has an additional cyclin-binding domain, and plays a key role in metabolism beyond the control of the cell cycle in non-proliferating cells [43,44]. CCAAT-enhancer binding proteins (C/EBPs) are key TFs that control cellular proliferation, differentiation, metabolism, inflammation, and numerous other responses, particularly in adipocytes [30,45]. CEBPβ are transiently induced during the early stages of adipocyte differentiation [30], and the ectopic expression of CEBPβ in 3T3-L1 promotes adipogenesis, even in the absence of adipogenic stimuli [46,47]. Expression of Homeobox genes is spatially and temporally regulated during embryonic development. HOXA5 encodes a DNA-binding TF, which may regulate gene expression, morphogenesis, and differentiation [31]. HOXA5 is controlled by DNA methylation [48], and has been shown to up-regulate the tumor suppressor p53 and AKT1 by down-regulation of PTEN [49]. Homeobox genes also have a functional role in determining adipose tissue expansion and body fat distribution, and it has been reported that HOXA5 promotes adipose differentiation in mice [31,50]. Interferon regulatory factors are proteins, which regulate transcription of interferons and play key roles in the JAK-STAT signaling pathway [51]. IRF4 has been implicated in acute leukemia [52], and strongly associated with pigmentation traits such as sensitivity of skin to sun exposure, freckles, blue eyes, and brown hair color [53]. IRF4 was also identified as an important regulator of adipogenesis and adipose lipid handling; it also regulates isoproterenol-stimulated glycerol release in adipose tissue both ex vivo and in vivo [54,55]. PAX family plays critical roles during fetal development and cancer growth. PAX4 is involved in pancreatic islet development, mouse studies have demonstrated PAX4 is a key mediator of the generation of insulin-producing β-cells during embryonic development [56,57]. CREB1 is a member of the leucine zipper family of DNA-binding proteins, it stimulates transcription by binding the cAMP response element, a DNA sequence present in many gene promoters [58]. CREB1 is phosphorylated by several protein kinases, and induces transcription of genes in response to hormonal stimulation of the cAMP pathway [59]. CREB1 has already been confirmed to be a well-documented TF, which is involved in adipocyte differentiation and lipid metabolism [60,61].
To elucidate the regulatory mechanism of bovine SIRT4 expression, we analyzed the 5 regulatory region of this gene using online prediction software, and found putative binding sites for the following TFs, namely E2F1, CEBPβ, HOXA5, IRF4, PAX4, and CREB1 at the −402/−60 bp core promoter region of the SIRT4. We hypothesized that these TFs might play major roles in regulating the transcriptional activity of SIRT4, and TFs can coordinate with SIRT4 to regulate adipocyte differentiation. We found that E2F1, CEBPβ, and HOXA5 are transcriptional activators of SIRT4 and IRF4, PAX4, and CREB1 transcriptional repressors of SIRT4. We also showed that SIRT4 could affect the ability of these TFs in regulating the adipocyte differentiation. These results confirm the coordinate regulation by SIRT4 and TFs on adipose differentiation.

Conclusion
In summary, we revealed high SIRT4 expression in bovine adipose tissue, which increased before the seventh day of adipocyte differentiation. During adipocyte differentiation, the expression of marker genes that promote adipocyte differentiation was significantly inhibited after SIRT4 knockdown. We identified the core promoter of the bovine SIRT4 gene and determined that SIRT4 is regulated by various TFs: E2F1, CEBPβ, and HOXA5 as transcriptional activators, whereas IRF4, PAX4, and CREB1 as transcriptional repressors. Moreover, we elucidated the regulatory mechanisms of TFs in adipocyte differentiation. Further, we verified that SIRT4 knockdown could affect the ability of TFs in regulating the differentiation of bovine adipocytes. These data provide a foundation to better understand the transcriptional regulation and biological function of bovine SIRT4 in adipocyte differentiation ( Figure 6).