PCSK9 (proprotein convertase subtilisin/kexin type 9) plays an important role in control of plasma LDL (low-density lipoprotein) cholesterol metabolism by modulating the degradation of hepatic LDL receptor. Previous studies demonstrated that PCSK9 is a target gene of the SREBP2 [SRE (sterol-regulatory element)-binding protein 2] that activates PCSK9 gene transcription through an SRE motif of the promoter. In addition to SREBP2, HNF1α (hepatic nuclear factor 1α) positively regulates PCSK9 gene transcription in hepatic cells through a binding site located 28 bp upstream from SRE. In the present study, we have identified a novel HINFP (histone nuclear factor P) recognition motif residing between the HNF1 motif and SRE that is essential for basal and sterol-regulated transcriptions of the PCSK9 promoter. Mutation of this motif lowers the basal promoter activity and abolishes the sterol-mediated repression as well as the SREBP2-induced activation of the PCSK9 promoter. We show further that the activity of SREBP2 in stimulating PCSK9 promoter activity is greatly enhanced by HINFP. Additional experiments suggest that HINFP and its cofactor NPAT (nuclear protein of the ataxia telangectasia mutated locus) form a functional complex, and NPAT may subsequently recruit the HAT (histone acetyltransferase) cofactor TRRAP (transformation/transactivation domain-associated protein) to facilitate the histone H4 acetylation of the PCSK9 promoter. Knockdown of HINFP, NPAT or TRRAP each markedly reduces the amount of acetylated histone H4 on the PCSK9 promoter region and lowers PCSK9 protein levels. Importantly, by utilizing co-immunoprecipitation assays, we have demonstrated a direct interaction between SREBP2 and HINFP and its cofactors NPAT/TRRAP. Taken together, these new findings identify HINFP as a co-activator in SREBP-mediated transactivation of PCSK9 gene expression.
Since its first discovery in 2003, PCSK9 (proprotein convertase subtilisin/kexin type 9) has been extensively studied because of its functional involvement in hypercholesterolaemia and associated cardiovascular diseases [1–6]. Gain-of-function mutations in PCSK9 are associated with hypercholesterolaemia [1,7–9], whereas loss-of-function mutations in PCSK9 result in hypocholesterolaemia in humans [10,11]. Secreted PCSK9 modulates the protein levels of hepatic LDLR [LDL (low-density lipoprotein) receptor] by inducing the intracellular degradation of the receptor, thereby influencing the levels of plasma LDL cholesterol [12,13]. Protein-binding experiments and structural studies have demonstrated that a direct binding of PCSK9 to EGF-A (epidermal growth factor-like repeat A) of LDLR is required for its degradation [14,15]. Owing to the critical function of PCSK9 in the control of protein levels of LDLR, many approaches have been taken to either block its interaction with LDLR by anti-PCSK9 antibodies  or to reduce PCSK9 expression by antisense oligonucleotides , siRNAs (small interfering RNAs)  or small molecule inhibitors such as BBR (berberine) [19,20]. BBR, a plant-derived alkaloid, has been shown to inhibit the mRNA and protein expressions of PCSK9, while increasing LDLR expression in hepatic cell lines and the liver tissue of animal models [21,22].
The transcription of PCSK9 is positively regulated by SREBPs [SRE (sterol regulatory element)-binding proteins] in response to depletion of intracellular levels of sterols [23,24] through a SRE motif of the PCSK9 promoter. In addition to the SREBP pathway, our laboratory has identified a HNF1 (hepatocyte nuclear factor 1)-binding site, a regulatory cis-acting element located 28 bp upstream from the SRE on the PCSK9 promoter, which mediates the transactivating activity of HNF1α in hepatic cells. We have shown further that HNF1α is not only required for the basal activity of PCSK9 gene transcription, but its expression is also down-regulated by BBR, leading to the suppression of PCSK9 expression in BBR-treated cells .
In order to characterize further the transcriptional regulatory network that controls PCSK9 gene expression, we put our effort into discovering new transcription factors that may directly regulate PCSK9 gene expression through binding to its promoter. In the present study we report that HINFP (histone nuclear factor P) plays a critical role in PCSK9 gene transcription through interacting with a highly conserved HINFP recognition sequence located between the HNF1-binding site and the SRE of the PCSK9 promoter.
It has been reported that HINFP and its obligated cofactor NPAT [nuclear protein of the ATM (ataxia telangiectasia mutated) locus] are principal regulators of multiple histone H4 genes , and NPAT-dependent recruitment of the TRRAP (transformation/transactivation domain-associated protein)-containing complex to histone promoters is required for transcriptional activation of histone genes during G1/S-phase transition . In addition to histone genes, several non-histone genes are the targets of HINFP . TRRAP is an ATM-related protein and belongs to the ATM superfamily [27,28]. TRRAP is a common component of many HAT (histone acetyltransferase) complexes, including the TRRAP–Tip60 complex [29–31]. It has been shown that the murine homologue of TRRAP (Trrap) plays a critical non-redundant role in cell cycle progression, protein turnover, metabolism and signal transduction, as well as DNA damage response through TRRAP-mediated recruitment of HAT to specific promoters in a cell cycle- and promoter-dependent manner [32–34]. However, to date neither HINFP nor NPAT has been implicated in the regulation of genes involved in cellular cholesterol metabolism.
In the present study, we show that the direct interaction of HINFP with its recognition sequence (−362 to −356) embedded between the HNF1-binding site and the SRE of the PCSK9 promoter is essential for the basal and sterol-regulated transcriptions of the PCSK9 promoter. Mutation of this motif lowers the basal promoter activity and abolishes the sterol-mediated repression as well as the SREBP2-induced activation of the PCSK9 promoter. We show further that the effect of SREBP2 in stimulating PCSK9 promoter activity is greatly enhanced by HINFP. Importantly, by conducting ChIP [chromatin IP (immunoprecipitation)] and co-IP assays, we show that HINFP and its cofactor NPAT form a functional complex, and that NPAT may subsequently recruit the HAT cofactor TRRAP to the HINFP site of the PCSK9 promoter. This direct association of the HINFP complex with the PCSK9 promoter is required to maintain the status of histone H4 acetylation of the active promoter because individual knockdowns of HINFP, NPAT or TRRAP all markedly reduce the amount of acetylated H4 bound to the PCSK9 promoter region, which is accompanied by significant decreases in PCSK9 protein expression levels. Furthermore, through co-IP after transient transfections, we show that HINFP and its cofactors NPAT/TRRAP interact with SREBP2. These results provide additional evidence to support a functional role of HINFP in SREBP-mediated transactivation of PCSK9 gene expression. Combined with our previous studies showing the essential function of HNF1α in PCSK9 transcription, the present study demonstrates further that the transcription of PCSK9 requires two different co-activators, namely HNF1α and HINFP, in addition to SREBPs for maximal activation of gene expression.
MATERIALS AND METHODS
Cell culture and plasmid transfection
The hepatoma-derived cell line HepG2 was purchased from A.T.C.C. and was maintained in MEM (minimum essential medium; Mediatech) supplemented with 10% (v/v) FBS (fetal bovine serum) and 1% penicillin/streptomycin solution. HEK (human embryonic kidney)-293A cells (Invitrogen) were maintained in DMEM (Dulbecco's modified Eagle's medium; Mediatech) supplemented with 10% (v/v) FBS and 1% (v/v) penicillin/streptomycin solution. FuGENE® 6 transfection reagent (Roche Diagnostics) or jetPRIME™ transfection reagent (Polyplus-Transfection) were used to transfect plasmids into HepG2 or HEK-293A cells according to the manufacturer's instructions. pCMV6-HINFP vector, a C-terminal Myc/DDK-tagged mammalian expression vector expressing human HINFP (DDK is the same as FLAG®, which is a registered trademark of Sigma–Aldrich), and its empty vector pCMV6-Entry were purchased from OriGene Technologies. pCMV-HA-NPAT, an N-terminal HA (haemagglutinin)-tagged mammalian expression vector expressing human NPAT, was a gift from Dr Jiyong Zhao (School of Medicine and Dentistry, University of Rochester, Rochester, NY, U.S.A.). CβF-TRRAP, an N-terminal FLAG-tagged mammalian expression vector expressing full-length human TRRAP and its empty vector CβF were gifts from Dr Michael D. Cole (Departments of Pharmacology and Genetics, Dartmouth College, Hanover, NH, U.S.A.). pGL3-basic and pRL-TK vectors were purchased from Promega. The PCSK9 promoter luciferase reporter vector D4 and its mutation vectors were described previously . HINFP-binding site mutation vector (HINFP-mu) was generated using the QuikChange site-directed mutagenesis kit (Stratagene) and the following oligonucleotides: sense, 5′-CAGATAGGATCcTaCtATGGGGCTCTGG-3′; and antisense, 5′-CCAGAGCCCCATaGtAgGATCCTATCTG-3′. The lower case letters are mutated nucleotides and the sequence alteration was verified by DNA sequencing.
Generation of plasmids
The plasmid pcDNA3.1-2×FLAG-SREBP2 obtained from Addgene was used to generate HA- and Myc-tagged vectors expressing the N-terminus of SREBP2. The nucleotide sequence encoding amino acids 14–481 of human SREBP2 was PCR amplified with the following primers containing an EcoRI site (underlined) at the 5′ end and a NotI site (bold) at the 3′ end: 5′-GCCCGAATTCGGGAGACCCTCACGGAGCTGGG-3′ and 5′-CCCCGCGGCCGCCCGTGAGCGGTCTACCATGC-3′. The PCR product was then inserted into pCMV-HA or pCMV-Myc vectors (Clontech) respectively, at EcoRI and NotI sites. The final sequence of vectors and the gene products were validated by sequencing and Western blotting respectively. To generate the pGL3 luciferase reporter for human ATM promoter, an ATM promoter construct pLightSwitch-ATM was obtained from SwitchGear Genomics. A 984 bp promoter region was PCR amplified with primers containing a KpnI site (underlined) at the 5′ end and a XhoI site (bold) at the 3′ end of the PCR product. The primer sequences were 5′-GCTTGGTACCGGGGTCCTAATTAAGTGTGGAGG-3′ and 5′-AGATCTCGAGTGCACTCGGAAGGTCAAAGTAG-3′. The PCR product was cloned into pGL3-basic vector at KpnI and XhoI sites. The new cloned vector was verified by sequencing and luciferase assay respectively.
siRNA against human HINFP (catalogue number AM16708), Silencer® negative control siRNA (catalogue number AM4635) and siPORT™ NeoFX™ transfection reagent were purchased from Applied Biosystems. The siRNA assay was performed according to the manufacturer's instructions. A total of 0.5 μl/well of 2 μM siRNA in 96-well plates (1.5×104 cells/well) or 12.5 μl/well of 2 μM siRNA in six-well plates (0.5×106 cells/well) were used for each transfection. Plasmids that expressed shRNA (small hairpin RNA) against human HINFP, NPAT or TRRAP were constructed by using a BLOCK-iT™ U6 RNAi Entry Vector Kit (catalogue number K4945-00; Invitrogen) according to the manufacturer's instructions. shRNA expression vectors were transfected into HepG2 cells using jetPRIME™ transfection reagent. The knockdown efficiency of shRNA plasmids was validated by Western blotting.
RNA isolation and real-time RT (reverse transcription)–PCR
Total RNA was extracted from HepG2 cells using the Quick-RNA™ miniprep kit (catalogue number R1055; Zymo Research). For each sample, 2 μg of total RNA was reverse transcribed with the high-capacity cDNA reverse transcription kit (catalogue number 4368813; Applied Biosystems). Real-time PCR was performed on an ABI PRISM® 7900-HT Sequence Detection System (Applied Biosystems) with Power SYBR® Green PCR master mix (Applied Biosystems). The primers used in real-time PCR were: 5′-AGGGGAGGACATCATTGGTG-3′ and 5′-CAGGTTGGGGGTCAGTACC-3′ for PCSK9; 5′-AATCCTGAGTGGTTTTATCGGC-3′ and 5′-CTCGAAGTTTACTGCGGTCCT-3′ for HINFP; and 5′-ATGGGGAAGGTGAAGGTCG-3′ and 5-GGGGTCATTGATGGCAACAATA-3′ for GAPDH (glyceraldehyde-3-phosphate dehydrogenase). Each sample was run in triplicate. All values are reported as means±S.D. of the triple measurement of each cDNA sample.
Luciferase reporter assay
The Dual-Luciferase® Reporter Assay System (Promega) was used according to the manufacturer's instructions. pGL3-basic constructs containing either wild-type or mutant PCSK9 promoter were individually transfected into HepG2 cells along with Renilla luciferase (pRL-TK) reporter as an internal transfection efficiency control. Briefly, 1.5×104 HepG2 cells/well were seeded on to 96-well plates (Corning) the day before transfection. Firefly luciferase plasmid (100 ng per well) plus Renilla luciferase plasmid (5 ng per well) were co-transfected using FuGene® 6 or jetPRIME™ transfection reagents. At 2 days after transfection, cells were lysed with 1× passive lysis buffer, and the firefly and Renilla luciferase activities were measured using a 96-well plate reader (SpectraMax® L microplate luminometer, Molecular Devices).
Western blotting was performed with whole cell lysates prepared from HepG2 or HEK-293A cells as described previously . Briefly, 50 μg of total protein extract for each sample was loaded into PAGEr® Gold Precast gels (Lonza Rockland). Proteins were transferred on to nitrocellulose membranes and incubated with the indicated antibodies. The membranes were developed and visualized using ECL (enhanced chemiluminescence) Plus Western blotting reagent (GE Healthcare) and Kodak Image Station 4000R (Kodak Molecular Imaging Systems). Polyclonal goat anti-HINFP antibody (sc-49818 X), polyclonal goat anti-TRRAP antibody (sc-5405), monoclonal mouse anti-c-Myc antibody (sc-40) and polyclonal rabbit anti-HA antibody (sc-805) were purchased from Santa Cruz Biotechnology; monoclonal mouse anti-β-actin antibody (A1978) and monoclonal mouse anti-FLAG antibody (F3165) were purchased from Sigma; polyclonal rabbit anti-PCSK9 antibody was a gift from Dr Sahng Wook Park (College of Medicine, Yonsei University, Seoul, Korea); and monoclonal mouse anti-NPAT antibody and polyclonal rabbit anti-NPAT antibody were gifts from Dr Jiyong Zhao.
ChIP assays were performed as described previously [35,36]. Briefly, the cultured HepG2 cells were cross-linked by the addition of formaldehyde to a final concentration of 1.42% (w/v) for 15 min at room temperature (23–26°C). Cross-linking was quenched with 125 mM glycine for 5 min at room temperature, and the cells were washed twice with ice-cold PBS before scraping and centrifugation (2000 g for 5 min at 4°C). Then the cell pellets were resuspended and lysed with IP buffer containing protease inhibitors . The cell lysates were sonicated (on ice using a VirSonic 550 sonicator with the following settings: 3 rounds, each with 1 s on and 1 s off pulses for a total of 10 s with a 2 min rest between rounds) to shear chromatin DNA to an average length of approximately 500–1000 bp. Monoclonal mouse anti-NPAT antibody and polyclonal goat anti-TRRAP antibody were incubated with 500 μl of IP buffer containing 1 mg of sheared chromatin in an ultrasonic water bath for 30 min at 4°C. For each sample, 20 μl of washed Protein G–agarose bead slurry (catalogue number 16-266; Millipore) was used to precipitate the bound DNA–NPAT or DNA–TRRAP complexes. The same amount of normal mouse or goat IgG was used for the mock IP. Owing to the difficulty of finding an anti-HINFP antibody that can suitably immunoprecipitate the endogenous HINFP, the mammalian expression vector pCMV6-HINFP, expressing a C-terminal FLAG-tagged human fulllength HINFP, was transiently transfected into HepG2 cells and the empty vector pCMV6-Entry was used as a control. Subsequently, 1 mg of sheared chromatin in 0.5 ml of IP buffer was incubated with 20 μl of washed anti-FLAG® M2–agarose (catalogue number A2220; Sigma–Aldrich) to capture the DNA–HINFP–FLAG complex. To detect the acetylated status of acetyl-histone H3 and H4 on the PCSK9 promoter after shRNA knockdown of HINFP, NPAT or TRRAP, 2 μg of anti-acetyl-histone H3 antibody (catalogue number 06-599; Millipore) or 5 μl of anti-acetyl-histone H4 antibody (catalogue number 06-866; Millipore) were incubated with 50 μg of sheared chromatin in a total volume of 500 μl of IP buffer. The bound and the input DNA were analysed by PCR with primers that amplify a 199-bp fragment of the human PCSK9 promoter region from −490 to −292, relative to the ATG start codon, or a 226-bp fragment within exon 12 of the human PCSK9 gene as a negative control . The sequences of ChIP primers were as follows: PCSK9 promoter forward, 5′-TCCAGCCCAGTTAGGATTTG-3′ and PCSK9 promoter reverse, 5′-CGGAAACCTTCTAGGGTGTG-3′; PCSK9 exon 12 forward, 5′-TGGGGCTGAGCTTTAAAATG-3′ and PCSK9 exon 12 reverse, 5′-TCGACCTGTTTGAATGGTGA-3′.
EMSA (electrophoretic mobility-shift assay)
EMSA was performed using a LightShift® chemiluminescent EMSA kit purchased from Pierce according to the manufacturer's instructions. Briefly, 50 fmol of 5′-biotin end-labelled dsDNA (double-stranded DNA) probes were incubated with 0.2 μg of purified recombinant HINFP protein in binding buffer containing 80 mM Hepes, pH 7.5, 500 mM KCl, 1 mM ZnCl2, 5 mM MgCl2, 10 mM DTT, 0.8 mM EDTA, 2.5% (v/v) glycerol, 50 ng/μl poly(dI-dC) and 0.05% Nonidet P40 in a total of 20 μl of reaction mixture for 20 min at room temperature. 5′-Biotin end-labelled single-stranded sense oligonucleotides and unlabelled sense and antisense oligonucleotides were synthesized by Elim Biopharmaceuticals. Two sets of probes were used in the EMSA assays: short probes and long probes. The short probes contained only the HINFP-binding sequences, whereas the long probes encompassed the HNF1- and HINFP-binding sequences and SRE. The sense sequences of short probes were: PCSK9 wild-type, 5′-GTTTAATCAGATAGGATCGTCCGATGGG-3′; PCSK9 mutant, 5′-GTTTAATCAGATAGGATCcTaCtATGGG-3′; and ATM wild-type, 5′-CAATACAAGCCGGGCTACGTCCGAGGGT-3′. The sense sequences of long probes were: PCSK9-WT, 5′-GTTCCGTTAATGTTTAATCAGATAGGATCGTCCGATGGGGCTCTGGTGGCGTGATCTGCG-3′; HNF1-mu, 5′-GTTCCGTTcgTGTTgccTCAGATAGGATCGTCCGATGGGGCTCTGGTGGCGTGATCTGCG-3′; HINFP-mu, 5′-GTTCCGTTAATGTTTAATCAGATAGGATCcTaCtATGGGGCTCTGGTGGCGTGATCTGCG-3′; and SRE-mu, 5′-GTTCCGTTAATGTTTAATCAGATAGGATCGTCCGATGGGGCTCTGGatcCGTGATCTGCG-3′. The lower case letters are mutated nucleotides. The underlines indicate binding elements. Purified FLAG-tagged recombinant HINFP protein was used in EMSA assay. To purify the HINFP protein, pCMV6-HINFP or pCMV6-Entry vectors were transfected into HEK-293A cells and the protein was purified using anti-FLAG® M2–agarose (Sigma–Aldrich) according to the manufacturer's instructions.
To analyse protein–protein interactions, IP was employed. Plasmids expressing two different tagged proteins to be tested for interactions were co-transfected into HEK-293A cells by using GenJet® transfection reagent (Signagen® Laboratories). At 48 h after transfection, cells were lysed in lysis buffer containing 250 mM NaCl, 25 mM Tris/HCl, pH 7.4, 1 mM EDTA, 1% Nonidet P40 and protease inhibitors . Mock transfections with empty vectors were performed in parallel as a control. Anti-FLAG® M2–agarose (Sigma–Aldrich), anti-HA–agarose (Sigma–Aldrich), Protein A– or Protein G–agarose (Millipore) and a specific antibody were mixed with 200 μg of cell lysates at 4°C overnight. After incubation, the agarose beads were washed three times with lysis buffer. All proteins were released from agarose beads by boiling in 20 μl of 1× Laemmli sample buffer and then subjected to SDS/PAGE and Western blotting analysis with the indicated antibodies.
For luciferase reporter assays, each experiment was representative of at least three independent experiments with a minimum of three samples per condition. Real-time RT–PCR results were summarized from at least three independent assays. Significant differences between control and treatment groups or between wild-type and mutated plasmids were assessed by one-way ANOVA with post-hoc test of Bonferroni's multiple comparison test or by a two-tailed Student's t test. Values of P<0.05 were considered statistically significant, where * represents P<0.05; ** represents P<0.01; and *** represents P<0.001.
PCSK9 promoter contains a functional HINFP-binding site that is required for sterol-mediated regulation
The region of the PCSK9 promoter that is necessary for sterol-regulated transcription is contained within an approximately 100 bp region (−430 to −337, relative to the translation start codon), which includes a GC box for Sp1 (specificity protein 1) binding, an HNF1 recognition motif and an SRE. Nucleotide sequences within these motifs are highly conserved among human, mouse, rat, and hamster PCSK9 promoters, whereas sequences outside these motifs are relatively less conserved. Interestingly, we noticed a stretch of nearly identical sequences on PCSK9 promoters of different species that lies between the HNF1 site and SRE. Analysis of this sequence stretch by MatInspector software revealed a putative HINFP-binding sequence (GTCCGAT) that is 9 bp upstream from SRE and 11 bp downstream of the HNF1-binding site (Figure 1A). To explore its potential function, using the wild-type PCSK9 promoter construct D4 as the template, we generated a mutant luciferase reporter (HINFP-mu) in which three nucleic acids of the HINFP motif were mutated. We transfected HepG2 cells with plasmid D4 or its mutation vectors, including HNF1 site mutation (HNF1-mu), SRE mutation (SRE-mu) and HINFP site mutation (HINFP-mu) along with a Renilla luciferase reporter (pRL-TK) as an internal transfection control. The results in Figure 1(B) show that mutation of the HINFP site markedly reduced the luciferase activity of the mutant construct (HINFP-mu) compared with the wild-type (D4), and this decrease in promoter activity was as severe as that caused by mutating the HNF1 site (HNF1-mu) or SRE (SRE-mu).
PCSK9 promoter contains a functional HINFP-binding site between the HNF1 site and SRE
PCSK9 transcription is tightly regulated by intracellular cholesterol levels through the binding of SREBP2 to the SRE of the PCSK9 promoter [23,24]. Given that the putative HINFP-binding site is in close proximity to the SRE, it might be involved in cholesterol-mediated regulation. To test this possibility, we transfected HepG2 cells with D4 and the mutant constructs for 1 day, and then cultured transfected cells in medium depleted of cholesterol [10% (v/v) LPDS (lipoprotein-depleted serum)] or in medium supplemented with cholesterol [10% (v/v) LPDS plus cholesterol] for 24 h prior to the isolation of cell lysates. The promoter-less luciferase control vector pGL3-basic was included in this experiment as a negative control. The results of normalized luciferase activity in Figure 1(C) show that the HINFP site mutation not only reduced the promoter basal activity, but also exhibited the resistance to cholesterol-mediated suppression in a manner closely resembling the SRE mutation as well as the HNF1 site mutation.
Next, we examined the specificity of HINFP as well as SREBP2 in transactivating the promoter activity of PCSK9 by comparing their activities on the PCSK9 promoter with pGL3-ATM and pGL3-LDLR234 reporter constructs . Based on our sequence analysis results using the MatInspector program and literature reports , the proximal ATM promoter contains a consensus HINFP-binding site but it lacks a SRE sequence; conversely, the promoter of LDLR has the classical SRE but it does not contain HINFP-binding sites . These three luciferase reporter constructs were co-transfected with plasmids expressing either HINFP (pCMV-HINFP) or the active form of SREBP2 (pCMV-HA-SREBP2) into HEK-293A cells along with the Renilla reporter pRL-SV40. The results in Figure 1(D) show that transfection of pCMV-HINFP increased PCSK9 and ATM promoter activities without affecting LDLR promoter, and transfection of pCMV-HA-SREBP2 had no effect on ATM promoter but it increased the promoter activities of PCSK9 and LDLR, thereby demonstrating the promoter specificities of these transactivators.
To examine the direct effects of HINFP site mutation on HINFP-mediated as well as SREBP2- and HNF1α-mediated transactivation of the PCSK9 promoter, we co-transfected the wild-type and mutated promoter constructs into HEK-293A cells along with individual expression vectors for HINFP, HNF1α and SREBP2 (Supplementary Figure S1A at http://www.BiochemJ.org/bj/443/bj4430757add.htm). The wild-type PCSK9 promoter activity was increased 14-fold, 8-fold and 2.1-fold over mock-transfected cells by co-transfection of HNF1α, HINFP or SREBP2 respectively. Mutation of the HINFP site greatly reduced the stimulatory effect of HINFP from 8-fold to 3.1-fold of the value of mock transfected cells, lowered the activity of HNF1α from 14-fold to 11-fold, and ablated the transactivation activity of SREBP2. In a similar pattern, mutations of the HNF1 site and SRE reduced the activities of their specific interacting transcription factors and attenuated the activities of the transcription factors for the adjacent sequence motifs, which were in line with our previous study . The negative effect of HINFP-mu on SREBP2-mediated transactivation of PCSK9 promoter activity was further demonstrated in HepG2 cells (Supplementary Figure S1B).
We explored further the function of HINFP as a co-activator of SREBP2 in PCSK9 transcription by examining the dose-dependent effects of pCMV-HINFP on PCSK9 promoter activity in the presence or absence of pCMV-SREBP2. The results in Figure 1(E) show that the SREBP2-stimulated PCSK9 promoter activity was increased by the exogenous expression of HINFP in a pCMV-HINFP plasmid dose-dependent manner. In addition, we examined the effects of increasing concentrations of pCMV-HINFP on PCSK9 promoter activities of cells cultured in cholesterol-depleted medium [10% (v/v) LPDS] or in medium supplemented with cholesterol [10% (v/v) LPDS plus cholesterol]. The results show that the transactivating activity of HINFP on the PCSK9 promoter was attenuated by cholesterol (see Supplementary Figure S2 at http://www.BiochemJ.org/bj/443/bj4430757add.htm). Taken together, these results identify the HINFP recognition sequence as a critical cis-acting element that works in concert with SRE to mediate the sterol regulation of the PCSK9 promoter activity by interacting with HINFP protein, which functions as a co-activator of SREBP2 in transactivation of the PCSK9 promoter.
HINFP binds to the PCSK9 promoter in vitro and in vivo
Next, we performed an EMSA assay to examine the direct binding of the HINFP protein to this newly identified sequence motif of the PCSK9 promoter by using biotinylated probes and purified HINFP protein. A GelCode™ Blue-stained gel showed a prominent band with an approximate molecular mass of 74 kDa from the immunoprecipitates of pCMV-HINFP transfected cells but not from the empty vector-transfected cells (see Supplementary Figure S3A at http://www.BiochemJ.org/bj/443/bj4430757add.htm). The identity of the protein was further validated by Western blotting with an anti-FLAG antibody (Supplementary Figure S3B). Figure 2(A) shows the results of EMSA that detected a single shifted band (Figure 2A, lane 2) compared with the reaction mixture containing only the biotinylated probe (Figure 2A, lane 1). Competition experiments showed that a 100-fold molar excess of unlabelled wild-type probe blocked the formation of the probe–HINFP complex (Figure 2A, lane 3), whereas 100-fold molar amounts of unlabelled mutant probe failed to compete (Figure 2A, lane 4), thereby demonstrating the binding specificity. In this experiment, we also included a biotin-labelled oligonucleotide probe that contains a consensus HINFP-binding site present in the ATM promoter . We detected a stronger interaction of HINFP protein with the ATM probe (Figure 2A, lane 6) than that with the PCSK9 probe (Figure 2A, lane 2) and the binding was specific as it was completely inhibited by the excess of the unlabelled ATM probe. Importantly, the binding of HINFP protein to the ATM probe was markedly blocked, although not completely, by the HINFP–PCSK9 wild-type probe, but it was not inhibited at all by the mutated HINFP–PCSK9 probe.
HINFP binds to its cognate site on PCSK9 promoter in vitro and in vivo
We examined further the impact of SRE mutation or HNF1 site mutation on the in vitro binding affinity of HINFP to the PCSK9 promoter (Figure 2B). Utilizing a biotinylated probe that contains the PCSK9 promoter sequence from −391 to −336 encompassing HNF1, HINFP and SRE sites, and unlabelled probes of the same region, we showed that the direct binding of HINFP protein to the labelled probe was not affected by SRE mutation or the HNF1 site mutation, because these probes competed as effectively as the wild-type probe. These results clearly demonstrated the direct interaction of HINFP to the HINFP site of the PCSK9 promoter under in vitro conditions.
To examine the in vivo interaction of HINFP with the PCSK9 promoter, we performed a ChIP assay in HepG2 cells. As the anti-HINFP antibody does not work in IP assays, we expressed HINFP–FLAG in HepG2 cells and used the anti-FLAG antibody to immunoprecipitate the HINFP-bound chromatins. The empty vector pCMV-entry transfected HepG2 cells were included in this experiment as the mock control. Isolated DNA fragments were used as templates to perform the PCR reaction with one set of primers that amplify a 199-bp fragment of the human PCSK9 promoter region from −490 to −292, encompassing the HINFP-binding site, and another set of primers to amplify a 226-bp fragment within exon 12 of the human PCSK9 gene where the HINFP-binding sequence is absent. Figure 2(C) shows that a strong binding signal of HINFP–FLAG to the PCSK9 promoter was detected using immunoprecipitated materials obtained from pCMV-HINFP-transfected cells, but very little binding signal was detected using the ChIP materials of the empty vector-transfected HepG2 cells (mock control). Furthermore, this strong binding was not observed in exon 12 of the PCSK9 gene. We also did not observe specific signals using a control antibody (anti-IgG) (results not shown). These results demonstrated the direct binding of HINFP to the PCSK9 promoter under in vivo conditions.
Silencing HINFP decreases PCSK9 promoter activity, mRNA and protein expression
From the results shown in Figures 1 and 2, we may conclude that HINFP plays an important role in PCSK9 gene transcription. To obtain additional evidence, we examined the effects of HINFP knockdown on PCSK9 promoter activity, mRNA and protein expression. Transfection of HepG2 cells with the HINFP siRNA (si-HINFP) led to an approximately 50% reduction of the mRNA level of HINFP as compared with that of cells transfected with a control siRNA (si-control) that contains a non-specific scrambled sequence (Figure 3A). A significant reduction of HINFP protein was observed in specific siRNA-transfected cells compared with the control siRNA (Figure 3D, top panel). To examine the specific effects of si-HINFP, we examined the promoter activities of PCSK9, ATM and LDLR in HepG2 cells transfected with si-HINFP or control siRNA. The results showed that si-HINFP lowered the PCSK9 and ATM promoter activities by ~60% and ~40% respectively. In contrast, LDLR promoter activity was not affected by knocking down HINFP, thereby providing validation to the on-target effect of si-HINFP (Figure 3B). Similar extents of diminution in PCSK9 mRNA levels (Figure 3C) were observed in si-HINFP-transfected cells compared with that of control siRNA. PCSK9 protein is synthesized as a ~72 kDa precursor that undergoes autocatalytic cleavage to produce the 63 kDa mature protein. Figure 3(D) shows that abundances of both the precursor and the mature form of PCSK9 in HepG2 total cell lysates were reduced by si-HINFP transfection to ~40% of that of si-control. Taken together, these results further demonstrate the functional role of HINFP in PCSK9 gene expression.
siRNA-mediated knockdown of HINFP decreases PCSK9 promoter activity, mRNA and protein levels in HepG2 cells
HINFP cofactor NPAT and HAT cofactor TRRAP contribute to PCSK9 transcription
It has been reported that HINFP/NPAT functions as a transcriptional activation complex to regulate its target genes  and NPAT recruits TRRAP/TIP60 histone acetyltransferase complex to histone gene promoters to co-ordinate the transcriptional activation of multiple histone genes during the G1/S-phase transition . To examine whether NPAT and TRRAP are also involved in HINFP-regulated transcription of PCSK9, a gene involved in the cholesterol biosynthetic pathway, we first investigated whether HINFP and NPAT can form a protein complex in the context of HepG2 cells by conducting co-IP assays using HepG2 total cell lysates that transiently expressed HINFP–FLAG and HA–NPAT. The cell lysates prepared from HepG2 cells transfected with empty vectors were used as negative controls for non-specific interactions. Figure 4(A) shows that IP with anti-FLAG antibody precipitated NPAT (lane 1, lower panel) along with HINFP (lane 1, upper panel). Vice versa, IP with the anti-HA antibody pulled down NPAT (lane 3, lower panel) together with HINFP (lane 3, upper panel). Very little signal was detected in IP samples of control lysates (lanes 2 and 4). To examine the interaction of NPAT with TRRAP, we performed IP assays using cell lysates prepared from HepG2 cells expressing HA–NPAT and FLAG–TRRAP. Again, total protein lysates prepared from HepG2 cells transfected with their corresponding empty vectors were used as negative controls. The results of IP and Western blotting are presented in Figure 4(B). The anti-FLAG antibody pulled down TRRAP (lane 1, upper panel) and NPAT (lane 1, lower panel). Likewise, the anti-HA antibody precipitated NPAT (lane 3, lower panel) and TRRAP (lane 3, upper panel). Altogether, these co-IP experiments demonstrated that these proteins interact with each other in HepG2 cells, which is in agreement with previous observations made in other cell types [25,26].
HINFP-PCSK9 sequence motif may interact with a transactivating complex containing HINFP, NPAT and TRRAP in HepG2 cells
In an attempt to address the specific question as to whether these proteins are recruited to the HINFP site of the PCSK9 promoter in HepG2 cells, we performed ChIP assays to detect the in vivo bindings of NPAT and TRRAP to the chromatin region containing the HINFP site of the PCSK9 promoter using a mouse anti-NPAT monoclonal antibody and a goat anti-TRRAP polyclonal antibody. In these experiments, we used mouse IgG and goat IgG as controls for non-specific binding. Strong binding of NPAT and TRRAP to the HINFP–PCSK9 sequence was detected, whereas no binding signals were observed in the DNA region of exon 12 of the PCSK9 gene (Figure 4C). These results strongly support the notion that NPAT and TRRAP participate in the HINFP-mediated transactivation of PCSK9 gene expression.
HINFP, NPAT or TRRAP are all required for Histone H4 acetylation of the PCSK9 promoter and PCSK9 protein expression
Next, we sought to further characterize the action mechanism of the HINFP–NPAT–TRRAP complex in PCSK9 transcription. Histone acetylation is associated with transcriptional regulation . It has been shown that the recruitment of the TRRAP–TIP60 HAT complex by NPAT induces the acetylation of histone H4 on its target promoters . We carried out ChIP assays to examine the changes in the acetylation status of histone H3 (Ac-H3) and H4 (Ac-H4) of the PCSK9 promoter after individual knockdowns of these proteins by utilizing the shRNA approach. We first generated plasmid-based shRNAs targeting HINFP, NPAT or TRRAP and transfected these shRNA-expressing plasmids into HepG2 cells. The knockdown efficiencies of individual plasmids were validated by Western blotting (see Supplementary Figure S4 at http://www.BiochemJ.org/bj/443/bj4430757add.htm). On the basis of that result, the most effective shRNA constructs, including shHINFP-C, shNPAT-A and shTRRAP-D, were used to assess the effects of cellular depletion of these factors on the histone acetylation status of the PCSK9 promoter in HepG2 cells. The shRNA construct shLacZ was included in the experiment as a negative control. The results of ChIP assays showed that acetylation of H3 of the PCSK9 promoter was not changed in HepG2 cells transfected with shHINFP, shNPAT or shTRRAP compared with shLacZ-transfected cells. In contrast, the amount of Ac-H4 of the PCSK9 promoter decreased markedly by the cellular depletion of HINFP, NPAT or TRRAP (Figure 5A, left-hand panel). To demonstrate the specific changes of Ac-H3 and Ac-H4 associated with the PCSK9 promoter, we examined the status of Ac-H3 and Ac-H4 on the PCSK9 exon 12 region in control and gene-specific shRNA-transfected cells. The amounts of Ac-H3 and Ac-H4 bound to the PCSK9 exon 12 region were very low, and there were no significant differences between Ac-H3 and Ac-H4 bound to this non-promoter region, regardless of shRNA transfections (Figure 5A, right-hand panel).
HINFP, NPAT and TRRAP are required for histone H4 acetylation of the PCSK9 promoter
Finally, we examined the effects of these shRNAs on PCSK9 protein expression. Figure 5(B) shows the result of one representative Western blot, and the reductions of PCSK9 levels by shRNA transfection in three separate transfection experiments were quantified by densitometry analysis and are presented as means±S.D. These results showed that individual transfections of HepG2 cells with shHINFP, shNPAT and shTRRAP all produced a negative impact on the expression level of PCSK9 protein as compared with shLacZ, thereby further suggesting that NPAT and TRRAP participate in the HINFP-mediated PCSK9 transactivation as a multi-factorial complex that positively contributes to the maximal transcription of PCSK9.
Interaction of SREBP2 and HINFP functional complex
The above results of various functional assays are highly suggestive of a possible mutual interaction of SREBP2 and the HINFP functional complex, including HINFP, NPAT and TRRAP. To examine the direct protein–protein interactions, we performed co-IP experiments. HEK-293A cells were co-transfected with HA-tagged SREBP2 and FLAG-tagged HINFP, and then cell lysates were used for IP with a monoclonal anti-HA antibody. After extensive washing, the immunoprecipitates were resolved by SDS/PAGE (4–20% gel) and analysed for the presence of HA–SREBP2 and HINFP–FLAG. In this and subsequent IP experiments, cell lysates transfected with HA, FLAG or Myc tag empty vectors were included in parallel for the control of non-specific interactions/detections. As shown in Figure 6(A), after IP with anti-HA beads, the HA–SREBP2 signal was detected (Figure 6A, lane 1, lower panel). Importantly, HINFP–FLAG (Figure 6A, lane 1, upper panel) was co-immunoprecipitated with HA–SREBP2, whereas no bands were detected in mock-transfected cell lysates (Figure 6A, lane 2). The HA–SREBP2 was detected by the anti-HA antibody as a doublet, which is consistent with the results of others who have shown that mature SREBP2 migrates on SDS/PAGE as a cluster of bands in the 55–70 kDa range . In the inverse experiments, lysates were subjected to IP with an anti-FLAG antibody (Figure 6A, lanes 3 and 4). As predicted, HINFP–FLAG was readily detectable. In addition, HA–SREBP2 was also clearly detected. These results show that SREBP2 and HINFP can physically interact in vivo. Next, we performed co-IP experiments using cell lysates co-transfected with Myc–SREBP2 and HA–NPAT (Figure 6B) or with HA–SREBP2 and FLAG–TRRAP (Figure 6C). We observed a direct interaction of SREBP2 with either HA–NPAT or TRRAP. Again, no signal was detected in lysates transfected with empty vectors by Western blots. Taken together, these IP assays demonstrate that SREBP2 can directly interact with HINFP and its cofactors.
Physical interactions of SREBP2 and HINFP, NPAT and TRRAP
PCSK9 has emerged as a novel therapeutic target for the management of hypercholesterolaemia. Recent results of clinical trials of anti-PCSK9 antibody have shown great efficacies of anti-PCSK9 antibody in LDL-C reduction in the absence of significant side effects, which further support PCSK9 as a therapeutic target of hypercholesterolaemia . A clear understanding of the transcriptional network that controls PCSK9 gene expression in liver cells is important for developing new interventions to block its synthesis in order to enhance LDL clearance from the circulation.
In the present study, we have discovered a new cis-regulatory element, the HINFP-binding site, which is clustered with SRE- and HNF1-binding motifs in a highly conserved region of the PCSK9 promoter. Luciferase reporter assays showed that mutation of any of these three sites greatly reduced the PCSK9 promoter activity and abolished sterol-mediated repression, indicating that they are all essential for PCSK9 gene transcription and regulation by intracellular sterols.
In order to examine the direct binding of HINFP to this sequence motif on the PCSK9 promoter, we carried out EMSA and ChIP assays. EMSA assays clearly showed that the protein HINFP specifically binds to the PCSK9 probe. HINFP is a sequence-specific transcription factor. The consensus binding motifs of HNIFP have been detected in promoter regions of both histone and non-histone genes, although the binding affinities vary among its target genes, with binding sites containing the core sequence 5′-GTCCG-3′ . The affinity of HINFP for the ATM probe is reported to be less strong than that for the histone H4 gene probe . The results of the present study show that the affinity of HINFP to the PCSK9 probe is even weaker than that to the ATM probe, as shown by the intensity of the ATM probe–HINFP complex, which is much stronger than that of the PCSK9–HINFP complex probe, and the excessive amount of the PCSK9 probe could not completely inhibit the binding of HINFP to the ATM probe. This might be due to the differences in the flanking sequences of the core binding motif of different probes.
HINFP, also known as MIZF [MBD2 (methyl-CpG-binding domain protein)-interacting zinc finger protein), can either positively regulate its target genes  or repress transcription of its target genes through interaction with the MBD2 protein [41,42]. In the case of the present study, HINFP is a transactivator for PCSK9 gene expression, as siRNA silencing of HINFP decreased PCSK9 mRNA and protein levels. It has been shown that NPAT is an obligate partner of HINFP in the regulation of the histone H4 gene [25,43]. The gene regulation by HINFP/NPAT may further involve TRRAP. One study has reported that the TRRAP-containing HAT Tip60 interacts with NPAT and associates with the histone H4 promoter in an NPAT-dependent manner . This induces histone H4 acetylation at the target promoter and activates gene transcription. The results of the present study show that NPAT and TRRAP form a complex in HepG2 cells, and individual knockdowns of HINFP, NPAT or TRRAP by shRNA transfections in HepG2 cells all decrease the levels of histone H4 acetylation at the PCSK9 promoter. These results are consistent with previous reports showing that TRRAP recruits HAT complexes with affinity towards histone H4 [30,44–46]. In agreement with the decrease in acetylated histone H4 at the PCSK9 promoter, the expression levels of PCSK9 protein were reduced in cells depleted of HINFP, NPAT or TRRAP through gene-specific shRNA transfections as compared with the mock transfection.
In the present study, initially we examined the interactive relationship between the HINFP-binding motif and the SRE of the PCSK9 promoter. Disruption of the HINFP site nearly abolished the cholesterol-mediated suppression and the SREBP2-mediated activation of the PCSK9 promoter. The dependency of SRE to an intact HINFP site is similar to our previous study of HNF1 mutation in which the regulation of PCSK9 transcription by SRE/SREBP2 was virtually abrogated by mutating the HNF1 site. We further showed that the mutation of the HINFP binding motif not only attenuated the inducing activity by HINFP on the PCSK9 promoter, but also severely compromised the trans-activating activities of SREBP2 and, to a lesser extent, HNF1α in HEK-293 cells. In HepG2 cells, the strong inducing activity of SREBP2 on the PCSK9 promoter was nearly abolished by mutations of HINFP site as well as the SRE and HNF1 sites (Supplementary Figure S1B). Vice versa, the mutation in SRE motif also reduced the stimulatory activity of HINFP on the PCSK9 promoter (Supplementary Figure S1A). Although our EMSA results suggested that the in vitro binding of the purified HINFP protein to the PCSK9 probe is not diminished by the mutation in SRE (Figure 2B), probably due to the non-overlapping manner of these two sites, the fact that the intact sequences of HINFP and SRE sites are required for the promoter activation induced by SREBP2 or HINFP suggests an interactive relationship of these two proteins under in vivo conditions. Our subsequent investigations of IPs have provided strong evidence for such an interaction (Figure 6). In cells co-transfected with SREBP2 with HINFP, we detected direct protein–protein interactions between these two transactivators. In addition to HINFP, through separate co-transfections we also detected the presence of NPAT or TRRAP with SREBP2 in immunoprecipitates. Although further investigations are required to clearly delineate the interrelationship between SREBP2 and HINFP and its cofactors, the present study has demonstrated a co-ordinated regulation of PCSK9 gene expression by SREBP2 and HINFP at the promoter level and at the level of direct protein interactions.
It has been well documented that SREBP2 has a dominant activator role in the transcription of PCSK9 and other genes involved in cellular cholesterol biosynthesis [23,24,47,48]. However, in all SREBP-regulated promoters studied to date, additional co-regulatory transcription factors are required for sterol-regulated activation of transcription. Sp1, NF-Y (nuclear factor Y) and CREB (cAMP-response-element-binding protein) are co-activators of SREBP2 in a number of sterol-regulated promoters [49,50]. With regard to partnership with HINFP, we have performed promoter sequence analysis of several other SREBP2 target genes, with putative HINFP-binding sites on promoters of FASN (fatty acid synthase) and HMG-CoA (3-hydroxy-3-methylglutaryl-CoA) reductase being revealed by the MatInspector program. Whether these genes are regulated by HINFP will require experimental investigation. Nevertheless, to the best of our knowledge, PCSK9 is the first promoter that has been demonstrated to require HINFP as the co-regulatory transcription factor to enhance SREBP2-activated gene transcription. The identification of HINFP as a co-activator of SREBP2 expands our current knowledge on sterol-regulated activation of transcription in the cellular process of cholesterol homoeostasis.
ataxia telangectasia mutated
electrophoretic mobility-shift assay
fetal bovine serum
human embryonic kidney
histone nuclear factor P
hepatic nuclear factor 1
methyl-CpG-binding domain protein
minimum essential medium
nuclear protein of the ATM locus
proprotein convertase subtilisin/kexin type 9
small hairpin RNA
small interfering RNA
specificity protein 1
sterol regulatory element
transformation/transactivation domain-associated protein
Hai Li performed the experiments and contributed to the experimental design and the writing of the paper. Jingwen Liu initiated the study, designed and supervised the experiments and drafted the paper.
This study was supported by the Department of Veterans Affairs (Office of Research and Development, Medical Research Service) and by the National Center of Complementary and Alternative Medicine [grant numbers 1RO1 AT002543-01A1 and 1R21AT003195-01A2].