Cholesterogenic Lanosterol 14-Demethylase (CYP51) Is an Immediate Early Response Gene
http://www.100md.com
《内分泌学杂志》
Medical Center for Molecular Biology (M.F., N.T.), Center for Functional Genomics and Bio-Chips (J.A., T.R., D.R.), Institute of Biochemistry, Faculty of Medicine, University of Ljubljana, 1000 Ljubljana, Slovenia
Abstract
Lanosterol 14-demethylase (CYP51) responds to cholesterol feedback regulation through sterol regulatory element binding proteins (SREBPs). The proximal promoter of CYP51 contains a conserved region with clustered regulatory elements: GC box, cAMP-response elements (CRE-like), and sterol regulatory element (SRE). In lipid-rich (SREBP-poor) conditions, the CYP51 mRNA drops gradually, the promoter activity is diminished, and no DNA-protein complex is observed at the CYP51-SRE1 site. The majority of cAMP-dependent transactivation is mediated through a single CRE (CYP51-CRE2). Exposure of JEG-3 cells to forskolin, a mediator of the cAMP-dependent signaling pathway, provokes an immediate early response of CYP51, which has not been described before for any cholesterogenic gene. The CYP51 mRNA increases up to 4-fold in 2 h and drops to basal level after 4 h. The inducible cAMP early repressor (ICER) is involved in attenuation of transcription. Overexpressed CRE-binding protein (CREB)/CRE modulator (CREM) transactivates the mouse/human CYP51 promoters containing CYP51-CRE2 independently of SREBPs, and ICER decreases the CREB-induced transcription. Besides the increased CYP51 mRNA, forskolin affects the de novo sterol biosynthesis in JEG-3 cells. An increased consumption of lanosterol, a substrate of CYP51, is observed together with modulation of the postlanosterol cholesterogenesis, indicating that cAMP-dependent stimuli cross-talk with cholesterol feedback regulation. CRE-2 is essential for cAMP-dependent transactivation, whereas SRE seems to be less important. Interestingly, when CREB is not limiting, the increasing amounts of SREBP-1a fail to transactivate the CYP51 promoter above the CREB-only level, suggesting that hormones might have an important role in regulating cholesterogenesis in vivo.
Introduction
REGULATION OF GENES encoding enzymes of cholesterol biosynthesis and uptake is mediated generally by the negative cholesterol feedback loop and sterol regulatory element binding protein (SREBP) transcription factors. SREBPs need partners for performing their task of maintaining the cellular cholesterol level. One of the coregulatory signaling pathways is the cAMP-dependent pathway, which is characterized by the cAMP-response element binding protein (CREB)/cAMP-response element modulator (CREM)/activating transcription factor 1 family of transcription factors and protein kinase A. Induction of the cAMP pathway leads to phosphorylation of cAMP-dependent transcription factors, which bind to cAMP-responsive elements (CRE) in the promoter/regulatory regions of the cAMP-responsive genes and in this way enhance their transcription (1). The induction of transcription can be immediate (a few minutes to hours) or delayed (chronic, several hours). Genes that are rapidly and transiently induced in response to the intracellular signaling cascades are described as immediate early response genes (IEG). The induction of IEG occurs in the absence of de novo protein synthesis and thus could not be blocked by protein synthesis inhibitors. IEG encode secreted proteins, cytoplasmic enzymes, and transcription factors (2). Most of our knowledge is about IEG encoding transcription factors c-jun and c-fos in the nervous and immune systems (3). The transcription of genes that respond to the induction by cAMP-dependent signaling is later repressed. The mediator of the repression is the cAMP-inducible transcription factor inducible cAMP early repressor (ICER), the synthesis of which is mediated by cAMP (4).
The SREBP regulatory pathway is characterized by sterol regulatory element binding proteins (reviewed in Refs.5, 6, 7, 8). SREBPs are synthesized as membrane-bound inactive precursors whose processing is regulated by a lipid-sensor mechanism. Alteration in cholesterol content of membranes promotes changes in physical properties of the membrane in cholesterol (lipid)-rich conditions, which prevents the sterol cleavage activating protein (SCAP)-SREBP complex from reaching the Golgi apparatus (9). Insig was proposed to anchor the SCAP-SREBP complex in the endoplasmic reticulum in the presence of cholesterol (10). Consequently, SREBPs remain membrane-bound precursors of the endoplasmic reticulum in lipid-rich conditions. The central dogma of cholesterol homeostasis says that when cholesterol, oxysterol, and fatty acid levels are limiting (lipid-poor conditions), precursor SREBPs are transported to the Golgi apparatus and are proteolytically cleaved. The N-terminal soluble basic-helix-loop-helix-Zip portions of mature SREBPs enter the nuclei and transactivate target genes by binding to sterol regulatory elements (SRE) of the target promoters. The exception is SREBP-2gc, which is synthesized as an already soluble protein in male germ cells and can thus transactivate SREBP-dependent genes irrespective of the cell cholesterol level (11). Together with other data (12, 13), this has suggested that under certain physiological or pathophysiological conditions, the genes encoding cholesterogenic enzymes might more generally be up-regulated irrespective of the cellular cholesterol level.
Lanosterol 14-demethylase (CYP51) is an enzyme of the late portion of cholesterol biosynthesis. It represents the most evolutionarily conserved member of the cytochrome P450 (CYP) gene superfamily (14, 15, 16). The direct product of the CYP51 enzymatic reaction is follicular fluid meiosis activating sterol, one of the short-lived intermediates of cholesterol biosynthesis that regularly accumulates only in testis and ovary (17).
Herein we demonstrate for the first time that forskolin, a mediator of the cAMP-dependent signaling pathway, activates the cholesterogenic CYP51 with an immediate early time response also in the absence of mature SREBP. A single CRE is sufficient to mediate this transactivation, and ICER is involved in attenuation of transcription. The immediate early transactivation of the CYP51 gene by forskolin is the first clear demonstration that cholesterogenic genes might respond to cAMP signaling also independently of the cholesterol feedback regulation. This might influence the de novo sterol synthesis when SREBP are limiting. We used the model system of JEG-3 cells that are derived from an endocrine organ, the human placenta. These cells harbor the components of cholesterol feedback regulation as well as of the cAMP-signaling pathway. Forskolin as a naturally occurring diterpene has been used to simulate the effects of physiologically occurring hormones (18), such as epinephrine, norepinephrine, FSH, LH, ACTH, and glucagon, that are known to activate the cAMP-signaling pathway.
Materials and Methods
Construction of reporter gene plasmids
Preparation of the wild-type human CYP51 D7 chloramphenicol acetyl transferase (CAT) (–334/+316) has been described previously (12), as well as preparation of CYP51 D7 luc (–334/+316), CYP51 D7 CAT (–121/+316), and CAT reporter mutant constructs CYP51-CRE2a-mut (TGACGCGA changed to TGATTCGA), CRE2b-mut (TGACGCGA was changed to AATGGCGA), CYP51-SRE1-mut (AATCACCTCAG was changed into TTTTTTTTTT), and CYP51-GC-mut (GGGGGCGG was changed into GGGTTTTTG) (13). Both CYP51-CRE-like mutants (CRE2a-mut and CRE2b-mut) have similar attributes in transfection studies (13). The substitution mutation CAT reporter constructs CYP51-CRE3-mut and CRE2-mut were prepared using QuikChange Site-Directed Mutagenesis Kit (Stratagene, La Jolla, CA). CYP51 D7 CAT served as a template for mutagenesis and preparation of CYP51-CRE3-mut, whereas for generation of a double mutant, CRE2a-mut was used as a template.
Human CYP51 C10 CAT (–471/+316) was amplified by the cloned Pfu polymerase (Stratagene) using sense 5'-TTTAG TTGGA GTGGG ACGGG TGACC-3' and antisense 5'-ACCTC GAACT GTGGC ACCTC ACCCT TCTCC-3' primers from the cosmid 121G12 containing the entire human CYP51 gene (19). The fragment was cloned into the SmaI-digested pCATbasic plasmid (Promega, Madison, WI).
Mouse Cyp51 CAT construct (–527/+104) was amplified by the cloned Pfu polymerase (Stratagene) using sense 5'-CCTTG AACCG ACGCG TGCCC AGAGG TGACG TCTCC-3' and antisense 5'-CCTTG GCATC AAGAT CTCAT GGCCT GTCCG AGCAC C-3' primers from the BAC 51921 clone, containing the entire mouse Cyp51 gene. This region of the Cyp51 promoter was cloned into the pCR-TOPO vector (Invitrogen, Carlsbad, CA) digested with MluI and BglII. The –527/+124 region of the mouse Cyp51 promoter was then cloned into the gel-purified pCAT basic from the MluI- and BglII-digested human CYP51 D7 CAT. The –527/+124 region of the mouse Cyp51 promoter is analogous to the –334/+316 region of the human CYP51 promoter. All human and mouse CYP51 CAT and luciferase reporter constructs were verified by sequencing.
Cell cultures
Human choriocarcinoma JEG-3 and human hepatocarcinoma HepG2 cells were cultured in DMEM (Sigma, Taufkirchen, Germany) containing 5% bovine calf serum and 1% L-glutamine in a 5% CO2 incubator at 37 C. In some experiments, lipid-rich medium (COPUFA) has been used (DMEM with 1% BSA, 10 μg/ml cholesterol, 1 μg/ml 25-hydroxycholesterol, and 0.15 mM linoleic acid) or 10% delipidated serum (Sigma). In experiments with delipidated serum, the 10% bovine calf serum medium was used as control (12). Cells were grown to 90% confluency in T150 tissue culture flasks and were split 1:4 into 60-cm2 dishes 24 h before transfections into normal or lipid-rich medium (with COPUFA).
Transfections and determination of reporter gene activity
For transfection studies, CAT and firefly luciferase reporter systems have been used. All transfections using CAT reporters were performed in human choriocarcinoma JEG-3 cells, whereas with human CYP51 D7 luciferase reporter, human hepatoma HepG2 and JEG-3 cells were transfected. Transfections and analysis of CAT activity have been performed as described previously (12, 13). In experiments with deletion CAT constructs, 3 μg pSV CREM has been used. In experiments with mouse Cyp51 CAT construct, where interactions between SREBP and forskolin have been studied, 3 μg pCMV SREBP-1a has been applied. In all other cotransfection studies, 200 ng pRSV CREB, pSV CREM, pCMV-SREBP1a, and pSV-specificity protein 1 (pSV-Sp1) has been applied. -Galactosidase has been used for normalization of the transfection efficiency with CAT reporters. pCAT basic was used as a DNA carrier to 20 μg.
Transfections of HepG2 or JEG-3 cells were performed with the firefly luciferase human CYP51 D7 reporter construct. Cells were transfected with 2 μg human CYP51 D7 luciferase reporter, 500 ng pSV -galactosidase plasmid, and 200 ng pSV CREM, pRSV CREB, or pSV ICERII, pSV Sp1, and pCMV SREBP-1a expression vectors or as indicated in the figure. pCAT basic was used as the carrier DNA to 3 μg. -Galactosidase was used for normalization of the transfection efficiency also in this case. The assay for determining the -galactosidase activity was performed as described (20). Firefly luciferase activity was analyzed with the commercial kit for luciferase (Promega) and Turner TD-20/20 luminometer. All transfection experiments were performed at least three times with two petri dishes for each experimental condition. In all transfection experiments, reporter activity has been calculated by the following formula: reporter activity = (normalized reporter gene activity)/(reporter activity in untreated cells). Reporter activity in untreated cells represents unit 1 and is shown as a black bar in column 1 of each diagram. The average value, SEM, and P values from two-tailed t test were calculated with the Excel program (Microsoft Corp., Redmond, WA). pSV CREM and pSV ICERII were a gift of Dr. P. Sassone-Corsi (Centre National de la Recherche Scientifique, Strasbourg, France). pRSV CREB and pSV Sp1 originate from the laboratory of Dr. M. R. Waterman (Vanderbilt University, Nashville, TN). It was confirmed previously that the results of transfections of human CYP51 CAT and firefly luciferase reporter genes are comparable (Aimovi, J., M. Fink, and D. Rozman, unpublished).
Nuclear extract preparation, expression, and purification of recombinant CREM protein and gel shift analysis
Nuclear extracts were isolated from JEG-3 cells that were grown in normal and COPUFA media as described above. Medium was changed every 24 h. Nuclear extracts were isolated from four petri dishes after 72 h, as described previously (12).
Purification of CREM expressed in bacteria has been described previously (21). Oligonucleotides used to generate double-stranded fragments containing human, mouse, and rat CYP51-CRE and human CYP51-SRE1 are presented in Table 1. Gel shift experiments were performed as described previously (12, 13).
RNA isolation and quantification of CYP51 mRNA by real-time PCR and Northern analysis
Total cellular RNA was isolated from JEG-3 cells. Forty-eight hours before the beginning of the experiment, cells were split 1:5 and were subjected to normal or lipid-rich medium (with COPUFA). Media were replaced every 24 h. After 15 min and 0.5, 1.5, 2, 4, 8, 12, 16, 24, 36, 48, 50, 52, 56, and 60 h, the medium was aspirated. Cells were washed twice with RNase-free PBS buffer and harvested with a rubber policeman, and total RNA was isolated with TRI reagent (Sigma). At 48 h, the medium was changed into fresh COPUFA medium, and cycloheximide (CHX) was added (10 μg/ml final concentration). Fifteen minutes after the addition of CHX, forskolin (25 μM final concentration) was added. Cells were harvested after 15 and 30 min and 1, 1.5, 2, 4, 8, and 12 h, and RNA was isolated as described above.
Total RNA from two 55-mm petri dishes was combined in each experiment. RNA concentration and quality were determined by RNA 6000 Nano Assay with Agilent 2100 Bioanalyzer (Agilent Technologies, Waldbronn, Germany). At least two different samples of each RNA sample were isolated and investigated for each time point.
One microgram of total RNA was converted into cDNA in a 20-μl reaction mixture using a SuperScript II reverse transcriptase (Invitrogen) with random primers (Promega). The reaction mixture was treated with DNase I (Sigma) to remove the contaminating DNA. The quality of each cDNA was tested in PCR with -actin primers.
Real-time PCR was performed with the ABI PRISM 7900 HT sequence detection system (Applied Biosystems, Foster City, CA) applying TaqMan technologies with the assay-by-design human CYP51 primers, sense 5'-CAG GTT GGC TGC CTT TGC-3' and antisense 5'-CTT GAT TTC CCG ATG AGC TCT GT-3', and probe CCC TGC GTC TGA AAC T labeled with FAM. The expression of the CYP51 was normalized to 18S rRNA content using a TaqMan rRNA control reagent kit. Each 20-μl reaction mixture contained 1 μl of 100x diluted cDNA template, 10 μl TaqMan Universal PCR Master Mix, 8 μl water and 1 μl of 20x control 18S rRNA or investigated CYP51 primers and probe and was amplified as follows: after incubation at 50 C for 2 min and denaturating at 95 C for 10 min, 50 cycles were performed at 95 C for 15 sec and 60 C for 1 min. The level of each cDNA was determined by the comparative CT method as described in User Bulletin 2, 1997. All procedures followed the Applied Biosystems protocols. For each time point, real-time PCR was conducted in triplicate from at least two independently isolated RNA samples. All calculations (average CT values, CT, average values, SEM, and P values from two-tailed t test) were calculated with the Excel program (Microsoft).
Northern blot analysis was performed by standard procedures (20). 32P-labeled 449-bp probe from the coding region of the human CYP51 gene (exons 1–3) was prepared with Prime-It II random primer labeling kit (Stratagene). Filters were hybridized for 2 h with QuikHyb hybridization solution (Stratagene), washed twice for 15 min at room temperature (0.1% SDS, 0.3 M NaCl, 0.03 M sodium citrate), once for 30 min at 60 C (0.1% SDS, 0.3 M NaCl, 0.03 M sodium citrate), and exposed for 16–48 h.
Immunoblot analysis
Western blot analysis was performed with 20 μg of nuclear proteins isolated from JEG-3 cells after forskolin treatment for different times and analyzed using an anti-CREM polyclonal antibody (21) and visualized by the SuperSignal West Pico chemiluminescent substrate (Pierce, Rockford, Il).
Metabolic labeling and sterol analysis
JEG-3 cells were split 1:2 the day before the beginning of the experiment. Cells were split 1:6 into 75-cm2 cell culture flasks at d 0, and 10 ml of normal medium was added. On d 1, medium was changed to normal medium or lipid-rich medium (with COPUFA). After 48 h (d 3), ketoconazole (1 μM final concentration) and/or forskolin was added (25 μM final concentration) together with the mixture of 80 μCi [3H]sodium acetate (NEN Life Science Products, Boston, MA) (10 mCi/ml, 3.1 Ci/mmol) and 0.2 μmol cold sodium acetate per 1 ml medium.
Cells were returned to the cell culture incubator for an additional 8 h. After trypsinization and homogenization, the concentration of total cellular proteins was determined in the homogenate and sterols extracted from 0.8 ml of the homogenate in glass vials. Ten microliters of ergosterol in ethanol (0.5 mg/ml), 100 μl of 0.3 M NaH2PO4 (pH 1.0), and 3 ml of extraction solution (75% n-heptane, 25% isopropanol) were added. The vials were incubated for at least 2 h on a shaker in the dark at room temperature and centrifuged (2000 x g for 10 min at room temperature). The organic phase was transferred and dried under vacuum, washed with 2 ml n-heptane, and centrifuged. Supernatants were transferred into fresh tubes and dried again. Samples were dissolved in the weight quantity of HPLC solvent. HPLC analysis was performed as described previously (22). The cholesterol fraction contains up to 20% lathosterol. Postlanosterol intermediates 7-dehydrocholesterol, zymosterol, and desmosterol eluted at identical retention time under applied separation conditions. Sterol standards (cholesterol, lanosterol, 7-dehydrocholesterol, zymosterol, and desmosterol) were from Steraloids (Newport, RI).
The average values, SEM, and P values from two-tailed t test were calculated with the Excel program (Microsoft).
Results
CYP51 mRNA expression pattern in lipid-repressed and cAMP-induced conditions follows the immediate early response
The cAMP-signaling pathway is one of the most investigated and in biological systems one of the most frequently used pathways mediating cellular responses to different physiological stimuli, but its involvement in cholesterol homeostasis has been suggested only recently. The aim of our work was to investigate the molecular mechanisms of cAMP-mediated transcription of cholesterogenic CYP51 that participates in late phases of cholesterol biosynthesis. CYP51 is transactivated by transcription factors of the SREBP family (12, 13) in a similar manner as other cholesterogenic genes. In accordance with the cholesterol feedback regulation, the amount of nuclear SREBP is highest when COPUFA are at a minimum (7, 23) and is decreased in lipid-rich media (23). Figure 1A shows that lipid-rich medium with COPUFA, containing 25-hydroxycholesterol, cholesterol, and linoleic acid, prevents the formation of a specific SREBP-dependent transactivation complex at the CYP51 promoter. A 22-bp CYP51-SRE1 element of the human CYP51 promoter binds SREBP specifically (Fig. 1A). Addition of SREBP-1a antibodies prevented binding of nuclear proteins to the human CYP51-SRE1 (compare lanes 1 and 2). No binding to the CYP51-SRE1 was detected in nuclear extracts from the COPUFA medium (lane 3). Results indicate that lipid-rich medium abolishes SREBP binding to CYP51-SRE1. Figure 1B shows a gradual decrease in CYP51 mRNA in lipid-rich medium (0–12 h), which was studied in detail by real-time PCR (Fig. 1C). The level of CYP51 mRNA is reduced by 60% in 2 h and 90% in 12–16 h and is later stabilized at about 30% for the entire time of the experiment. The data of Fig. 1 thus show that no SREBP are bound to the CYP51-SRE in COPUFA medium and that this results in a marked decrease of CYP51 mRNA. If under such conditions, cells are treated with forskolin, a mediator of cAMP response, the CYP51 mRNA level raises again, in a transient fashion. Forty-eight hours of COPUFA treatment represents also time 0 of forskolin addition. Forskolin, in the presence of COPUFA, increased the CYP51 mRNA in 1.5–2 h (49.5–50 h in Fig. 2A). After this time, the CYP51 mRNA progressively decreased to the basal level in 4–6 h (52–54 h in Fig. 2A). The addition of CHX, an inhibitor of protein synthesis, not only abolishes the decrease in CYP51 mRNA accumulation with time but allows the increase to continue (Fig. 2A). These results show that CYP51 transcription is induced by cAMP-dependent agents in an immediate-early fashion and that for such activation, de novo protein synthesis is not needed. However, de novo protein synthesis is required for attenuation of transcription.
To monitor whether the forskolin-induced CYP51 transcription affects also sterol synthesis, JEG-3 cells have been metabolically labeled with [3H]acetate, a C-2 precursor of cholesterol. Addition of forskolin to JEG-3 cells leads to a statistically significant decrease in the CYP51 enzyme substrate lanosterol after addition of forskolin, whereas the quantity of postlanosterol intermediates and cholesterol did not change significantly (Fig. 2B), which could be because of the well-known forskolin-induced consumption of cholesterol in the synthesis of steroid hormones (24, 25). Data indicate that the forskolin-mediated activation of CYP51 transcription leads to an increased activity of the CYP51 protein followed by an increased expenditure of lanosterol. Addition of COPUFA medium reduced the cholesterol quantity 10-fold, and the lanosterol quantity almost 6-fold, reaching the detection limit of the method (data not shown).
To show that the decrease in lanosterol quantity is a result of an increased consumption of lanosterol and not a decreased production of lanosterol because of a forskolin-mediated inhibition of an enzyme of the prelanosterol cholesterol biosynthesis, a CYP51 inhibitor, ketoconazole, was applied (Fig. 2B). Treatment with ketoconazole increased significantly the quantity of lanosterol (black bars) and 24,25-dehydrolanosterol (not shown), whereas the postlanosterol intermediates have been below the level of detection both in the normal and in the COPUFA medium. The appearance of 24,25-dehydrolanosterol, another possible substrate of CYP51 in cholesterol biosynthesis, indicates that ketoconazole completely blocked CYP51 and that cholesterol biosynthesis did not take place by an alternative pathway. This is in accordance with the belief that a block in any of the cholesterol biosynthesis steps results in the absence of all downstream pathway intermediates. ICER is a cAMP-inducible repressor that is encoded by the last five exons of the CREM gene and is transcribed from the intronic CREM promoter P2 upon cAMP induction (4). ICER lacks the transactivation domain but contains the DNA binding domain, thus representing a cAMP-dependent repressor that binds to identical cAMP-responsive elements as the activator forms of CREB and CREM. The addition of forskolin results in a gradual increase of the cAMP-inducible repressor CREM isoforms ICER II and ICER II, whereas significantly higher molecular weight activator CREM isoforms (CREM, -, -, and -S) do not fluctuate much during the treatment (Fig. 3A), which is consistent with the lack of inducibility of non-ICER CREM transcripts (4). Overexpression of pSV ICER repressed the human CYP51 D7 luciferase reporter gene activity, whereas overexpression of pRSV CREB induced the CYP51 D7 luciferase reporter (Fig. 3). ICER represses the CREB-induced transactivation of CYP51, because lower luciferase activity was determined when ICER and CREB were coexpressed in comparison with the overexpression of CREB alone (P < 0.05). This is consistent with the CYP51 mRNA pattern of Fig. 2A.
Evaluation of CRE in promoters of mammalian CYP51 genes
Mammalian CYP51 promoters are evolutionarily conserved and contain three cAMP-responsive elements, CRE1, CRE2, and CRE3, which all differ from the consensus CRE. Only the human and mouse CYP51-CRE1 are capable of binding the recombinant CREM (Fig. 4A, columns 1 and 2), whereas the rat CYP51-CRE1 failed in binding this cAMP-dependent transcription factor (Fig. 4A, column 3). The sequences of individual CRE-like elements (Fig. 4B) show that the downstream portion of the CYP51-CRE1 palindrome is not conserved. In contrast to CRE1, CYP51-CRE2 and CRE3 elements are conserved and are identical in human/pig and in mouse/rat CYP51 promoters (Fig. 4B).
To explore further the importance of individual CRE, the human CYP51 promoter deletion CAT constructs have been transfected into JEG-3 cells together with overexpressed CREM. Results demonstrate that the shortest construct D7 exhibits a very low basal activity (5% of D7) and does not respond to overexpression of CREM (Fig. 5B, column 1). This indicates that the –121/+316 human CYP51 promoter is not sufficient for basal transcription and is unable to respond to cAMP-dependent stimuli despite the presence of CRE3.
Interestingly, the longest CYP51 promoter construct C10 (Fig. 5B, column 3) also exhibits a lower basal activity (25% of D7) and shows a weaker response to CREM despite the presence of three CRE-like elements in comparison with D7, containing CRE2 and CRE3 only (Fig. 5B, column 2).
The –471/–334 region including CYP51-CRE1 seems not to be important for the cAMP-dependent transactivation of hCYP51, or the repressor element might be present in this region, because no additional activation has been observed in the CYP51 C10 construct that contains CRE1 in addition to CRE2 and CRE3 (26, 27).
To evaluate further the hypotheses based on deletion promoter-reporter construct, promoters with mutant CRE2 and CRE3 regulatory elements have been transfected together with overexpressed CREM or CREB. Although the mutation of CRE2 abolished both basal expression and the cAMP-dependent transactivation, CRE3 mutation retained approximately half of CREM-dependent and over 60% of CREB-dependent transactivation (Fig. 5C). As expected, the double mutation exhibits a similar response as the CRE2-mutant alone. The cumulative results of Fig. 5, B and C, suggest that the human CYP51-CRE2 is the principal CRE-like element responsible for cAMP-dependent transactivation. Interestingly, CRE3 has a strong effect on the basal activity of the CYP51 reporter, suggesting potential interactions with the basal transcriptional machinery.
cAMP-dependent transactivation of human and mouse CYP51 and cross-talk with the sterol regulatory pathway
The –334/+316 human and –527/+104 mouse CYP51 promoters, both containing CRE2 and CRE3 elements, have been applied in functional promoter studies in JEG-3 and HepG2 cells with overexpressed CREM and CREB. CAT activity of the human and mouse CYP51 promoters in normal medium was taken as basal, and all modifications, such as change of the medium or overexpression of transcription factors, were compared with the CYP51 promoter-CAT reporter activities observed in the normal medium.
Generally, both human and mouse CYP51-CAT constructs show a similar pattern of response (transactivation) to overexpression of CREM and CREB in individual cell lines. In normal medium, the overexpression of CREM and CREB results in up to 3-fold activation of hCYP51 in HepG2 cells and 4- to 12-fold activation in JEG-3 cells (Fig. 6A). The mouse Cyp51 promoter is also transactivated by overexpressed CREM and CREB, albeit to a lower extent compared with the human CYP51 promoter. The observed weaker response of the mouse Cyp51 promoter in a human cell line nuclear environment might be explained by the requirement for an optimal species-specific transactivation complex. Several gene-specific transcription factors as well as factors associated with the general transcriptional machinery differ between human and mouse in amino acid sequence and consequently also in the three-dimensional structure. For example, transcription factor SREBP-1a is 80.4% identical between mouse (NP_035610) and human (NP_004167), whereas CREM seems to be more conserved and shows a 94.2% identity between mouse (P27699) and human (Q03060). Numbers in parentheses represent GenBank identification numbers.
The transactivation capacity of CREM and CREB was also monitored in lipid-rich medium COPUFA (Fig. 6B). As shown previously (Fig. 1A), COPUFA prevents binding of SREBP to CYP51-SRE. Figure 6B shows that treatment of cells with COPUFA results in a decrease of human (7-fold in HepG2 and 6-fold in JEG-3 cells) and mouse (20-fold in JEG-3 cells) CYP51-CAT reporter activities. Overexpression of CREM and CREB results in reactivation of CYP51 promoters of both species in lipid-rich medium, 1.5-fold (CREM) or 3.5-fold (CREB) above the basal level for the human promoter.
The effect of forskolin on the SREBP-1a-mediated transactivation is shown in Fig. 6C. Lanes 1–3 show the effect of decreasing amounts of lipids (COPUFA, normal medium, and delipidated serum) on the CYP51-CAT reporter transactivation. Overexpression of nuclear SREBP-1a (lanes 4 and5) amplifies transactivation. No major difference is observed between cells grown in normal serum and delipidated serum. When forskolin is added to the cells with overexpressed SREBP-1a, the transactivation is diminished (compare last two lanes). This suggests interactions and/or competition for binding sites of CRE-2 and SRE-1 factors, because the core elements (CRE-2 octamer and SRE-1 decamer) are separated by only 12 bp (13). The titration assay, where different concentrations of SREBP-1a and CREB expression plasmids were cotransfected to JEG-3 cells together with the human CYP51-luciferase reporter (Fig. 6D), was used to evaluate potential interactions between CRE-2 and SRE-1 binding factors. At low CREB concentrations (0–20 ng CREB expression plasmid), increasing amounts of SREBP-1a increase the transactivation of the CYP51 promoter. The transactivation is highest when 600 ng SREBP-1a plasmid and no CREB are expressed. Similarly, at low SREBP-1a concentrations (0–20 ng SREBP-1a expression plasmid), increasing amounts of CREB increase the transactivation of the CYP51 promoter. The maximal CREB-dependent transactivation of human CYP51 is lower compared with the SREBP-dependent transactivation. Interestingly, increasing amounts of CREB decrease the maximal SREBP-dependent transactivation, suggesting that CRE-dependent transcription factors bind with a higher affinity to the CYP51-CRE2 site. When CREB is not limiting (at least 200 ng of the expression plasmid), even increasing amounts of SREBP-1a cannot transactivate the promoter above the CREB-only maximal level. This suggests that at high concentrations the CYP51-CRE2-bound CREB might be able to displace SREBP-1a from the CYP51-SRE1. Alternatively, CREB protein interaction with SREBP-1a might lead to a less potent transactivation complex, resulting in less efficient transcription of the CYP51 gene. Similar results were obtained for CREM; however, 600 ng of the pRSV CREM expression plasmid did not yet saturate the promoter (not shown). It seems possible that at physiological conditions that result in high concentrations of cAMP-dependent transcription factors, the cholesterogenic CYP51 responds as a cAMP-dependent gene irrespective of the SREBP signaling pathway. Additional studies are required to evaluate this hypothesis and to depict the endogenous concentrations of interacting CRE-binding factors and SREBP.
Relative importance of the CYP51 promoter elements for the cAMP-mediated transactivation
Data above show CRE2 as most important for the cAMP-dependent transactivation of CYP51. This element lies within the –282/–225 promoter region that also contains GC-box and SRE-1 elements (13). In the normal medium, all three mutants (CRE2, SRE, and GC-box) exhibit a lower basal activity compared with the wild-type human CYP51-CAT reporter (Fig. 7, A and B). The highest effect of the mutation is observed with mutated GC-box where only 11% of the basal activity was retained, whereas CRE2b and SRE1 mutations retained about 45% of the basal activity. CRE2b-mut showed the weakest response with only 12% of the wild-type reporter activity after overexpression of CREB. Fifty percent of CAT activity remains after overexpression of CREM with the SRE-1 mutant and 37% after overexpression of CREB. GC-box-mut preserved only 10% of the CREM-dependent transactivation and 30% of the CREB-dependent transactivation.
The data in Fig. 7, A and B, demonstrate that the three DNA regulatory elements (GC-box, CRE2, and SRE1) from the –282/–225 region of the human CYP51 promoter contribute to the basal as well as cAMP-dependent CYP51 gene transactivation in normal medium that contains SREBP from bovine serum. Although an intact GC-box seems to be most important for the basal transcription, CRE2 is essential for the CREM/CREB-dependent transcription, whereas SRE1 seems to have a lower impact. As established previously, SRE1 has a major role in SREBP-dependent transactivation of CYP51 (13).
COPUFA diminished the activity of the wild-type human CYP51-CAT reporter to 16% (Fig. 7C). The activities of CRE2 mut and GC-box-mut were diminished to 50% of the initial value in normal medium. The CYP51 SRE1-mut reporter activity is diminished to 45% of the wild type in the presence of SREBP from the normal, bovine serum-supplied medium. Additionally, our previous results have shown clearly that in the presence of overexpressed SREBP-1a, the CYP51 SRE1-mut activity is diminished to 10% of the wild type (13). The human CYP51 SRE1-mut has a similar response as the wild-type promoter in sterol-rich media, particularly with CREM-dependent transactivation, suggesting activation by CREM in the absence of SREBP-dependent transactivation complex.
Overexpression of Sp1 does not affect the majority the CYP51 transactivation in normal media or in the lipid-rich conditions (Figs. 7, B and C gray bars). This observation is in accordance with previously published work (13); however, it seems to be controversial with dramatic effects observed with mutations of the GC-box. This seeming discrepancy can be explained by the ubiquitous nature of Sp1 that is likely not limiting in JEG-3 cells. Alternatively, Sp1 might serve as a competence factor. For example, overexpression of a competence factor may result in no basal induction of promoter activity, but its binding to response elements may still be required to potentiate transactivation by other factors.
Discussion
Being a part of the cholesterol biosynthesis pathway, CYP51 transcription is regulated primarily by transcription factors SREBP. Cholesterol/SREBP-dependent regulation of CYP51 has been demonstrated previously by several examples (8, 12, 13, 28, 29, 30, 31). Unsaturated fatty acids and 25-hydroxycholesterol work synergistically in lowering the nuclear, transcriptionally active forms of SREBP-1 and SREBP-2 proteins (32) by inhibiting the SREBP precursor cleavage and its translocation to the nucleus. A repressing effect on SREBP-regulated genes has been detected already by 16 h and has been preserved for 48 h at the level of transcription (33, 34, 35, 36, 37). The CYP51 transcription is reduced by over 50% within 4 h in the lipid-rich COPUFA medium. This quick repression could be explained with the short half-life of the nuclear forms of SREBP. Wang et al. (38) reported that reduction at the level of nuclear forms of SREBP is observed already 30 min after addition of cholesterol and 25-hydroxycholesterol. Degradation of SREBP requires protein synthesis and can be blocked by the inhibitor of protein synthesis CHX. Because in the presence of CHX the degradation of SREBP is blocked, the transcription of SREBP target genes is slightly up-regulated (39). A similar observation has been noticed also in our research (Fig. 2A, 52 h).
In this work, we describe for the first time an immediate early transactivation of the cholesterogenic CYP51 by a cAMP mediator, forskolin. Reports from the late 1980s have indicated that the low-density lipoprotein receptor (36, 40) and 3-hydroxy-3-methylglutaryl coenzyme A reductase (36, 37) mRNA levels are elevated by cAMP-dependent agents; however, the mechanisms have not been investigated. The immediate early cAMP-dependent response described in this work does not depend on the cellular lipid level, showing that at least some cholesterogenic genes can become transactivated by other stimuli, also irrespectively of the cholesterol feedback regulation. When phosphorylated CREB binds to CRE, it regulates immediate early genes, such as c-fos and junB (41). Together with other immediate early genes, the expression of ICER (42, 43) mRNA is up-regulated after 1.5 h of forskolin treatment, reaching maximal levels after 3.5 h. This rise in expression of ICER correlates well with the decrease in c-fos and junB levels (44) as well as with time and concentration dependency of the repression of CRE-mediated luciferase gene expression after the treatment of CHO cells with forskolin (45). Our results show that the CYP51 gene responds to activation by the cAMP-signaling pathway in the same immediate-early time-dependent manner. The highest level of CYP51 mRNA is detected 1.5–2 h after addition of forskolin to JEG-3 cells grown in lipid-rich medium with COPUFA (Fig. 2A) as well as cells grown in normal medium (data not shown) and returns to basal level after 4 h. The early response does not require protein synthesis in contrast to the delayed cAMP response (46, 47). In our case, the addition of protein synthesis inhibitor CHX did not prevent the induction of CYP51 transcription, suggesting that CYP51 transactivation does not depend on de novo protein synthesis. However, the attenuation of CYP51 transcription likely requires de novo synthesis of ICER. It seems that the declination of CYP51 transcription is a result of the synthesis of ICER, which binds to CRE in the promoters of cAMP-responsive genes and represses their transcription (4, 48). ICER is transcribed from a cAMP-dependent intronic promoter of the CREM gene. It binds identical CRE as the activator CREM and CREB isoforms. However, ICER does not contain a transactivation domain. It displaces the activator transcription factors from CRE in the cAMP-dependent promoters and thus acts as a transcriptional repressor (4). As shown in Fig. 3A, ICER synthesis was induced in a time-dependent manner also in our experiments, in accordance with the attenuated CYP51 mRNA level (Fig. 2A). Overexpression of ICER presumably repressed the CREB-mediated transcription of CYP51 D7 luciferase reporter construct (Fig. 3B), suggesting competitive displacement of CREB by ICER. Even if mammalian CYP51 promoters contain three potential CREs, it seems that only CYP51-CRE2 is responsible for the immediate early response of CYP51, including the cAMP-mediated transactivation as well as repression. This conclusion arises from the following observations: 1) CRE2 is evolutionarily the most conserved CRE-like element in the four mammalian species; 2) mutation of CRE2 abolished the cAMP-dependent response of the human CYP51 gene; 3) the longest promoter-reporter construct, containing CRE1, CRE2, and CRE3, has a weaker cAMP-dependent response compared with the construct containing CRE2 and CRE3 only; and 4) the shortest promoter-reporter construct containing only CRE3 does not respond to cAMP-dependent transactivation. Interestingly, the shortest –121/+316 D7 construct, which contains the downstream promoter element and the initiator element of the four studied mammals (49), shows less than 10% of the basal activity compared with the –334/+316 promoter. This suggests that in the case of mammalian TATA-box lacking CYP51 promoters, the downstream promoter element and initiator element seem not to serve as the RNA pol II complex binding sites.
These results suggest that CYP51 belongs to the group of immediate early response genes. Importantly, forskolin, by modulating the transcription of CYP51, modulates also the precholesterol sterol synthesis in JEG-3 cells ex vivo. Although the quantity of the de novo synthesized lanosterol seems to decrease, the quantity of postlanosterol intermediates of cholesterol biosynthesis did not change or even increased in conditions when cells have been treated with forskolin. This suggests an increased consumption of lanosterol to form postlanosterol intermediates. Reasons for a statistically nonsignificant increase in postlanosterol intermediates might be a result either of the low activity of enzymes of the postlanosterol phases of cholesterol biosynthesis or of an increased expenditure of the synthesized cholesterol. It is well known that forskolin increases the expenditure of cholesterol because genes that encode enzymes involved in production of steroid hormones are highly up-regulated by the cAMP-signaling pathway (24, 25). Identification of all possible steroid metabolites that are derived from cholesterol after the forskolin-mediated immediate early activation in JEG-3 cells is beyond the scope of this work.
Our data show for the first time that a cholesterogenic gene CYP51 can be activated through CYP51-CRE2 in an immediate early fashion after forskolin treatment, avoiding the cholesterol feedback repression. Either CREB or CREM can mediate this transactivation and ICER is responsible for attenuation of transcription. The forskolin stimulus leads to a short-term transactivation of cholesterogenic CYP51 that is reflected also at the level of de novo synthesis of cholesterogenic sterol intermediates.
In accordance with our data, others have indicated that the cholesterol feedback regulation might be bypassed in some physiological or pathophysiological conditions, such as tumor development (50, 51), regeneration of tissues after injury (52, 53, 54), response to cytokines (52, 53, 54, 55, 56), or maturation of male germ cells (12). The high level of cholesterol in the cell presumably protects cells against stress, whereas an increased cholesterol biosynthesis is one of the cellular responses to stress (57, 58, 59). However, the biochemical details that would allow understanding of these phenomena have been lacking so far. Our work shows that the CYP51 gene is transcriptionally activated in sterol-repressed conditions by either CREB or CREM. Although CRE2 and GC-box DNA elements are essential for cAMP-dependent transactivation, the SRE1 site seems to be less important. Transactivation of CYP51 through CRE2 represents the first clear evidence that the cholesterol feedback repression of cholesterogenic genes can be bypassed by transcriptional activation through another signaling pathway. Additionally, our work shows for the first time that a gene participating in cholesterol biosynthesis (CYP51) is regulated in an immediate early fashion, which influences the sterol profile of the cell. This opens new venues of research to better understand cholesterol-independent roles and cross-talk of the cholesterol feedback regulation with other signaling pathways.
Acknowledgments
We thank Dr. P. Sassone-Corsi (Centre National de la Recherche Scientifique, Strasbourg, France), for providing the pSV CREM, pSV ICER expression vectors and anti-CREM antibodies. Thanks also to Dr. M. R. Waterman (Vanderbilt University, Nashville, TN) for providing pRSV CREB and pSV Sp1 expression vectors. Sincere thanks to Dr. S. Svetina and Dr. U. Batista (Institute of Biophysics, Faculty of Medicine, University of Ljubljana) for sharing the cell culture laboratory and to Dr. V. urin erbec, dr. V. Galvani, and Dr. R. Rupreht (Blood Transfusion Center of Slovenia, Ljubljana, Slovenia) for sharing the 7900 HT sequence detection system.
Footnotes
This work was supported by the Ministry of Science of Slovenia, Grants J1-1380, SLO-USA 0002, and P1–0527, by the Fogarty International Research Collaboration Award/National Institutes of Health Grant 1 RO3 TW01174-01, and by International Centre for Genetic Engineering and Biotechnology Grant CRP/SLO0001. T.R. was supported by a fellowship from the Ministry of Science of Slovenia.
First Published Online August 25, 2005
Abbreviations: CAT, Chloramphenicol acetyl transferase; CHX, cycloheximide; COPUFA, lipid-rich medium with cholesterol, 25-hydroxycholesterol and linoleic acid; CRE, cAMP-response element; CREB, CRE-binding protein; CREM, CRE modulator; CYP, cytochrome P450; CYP51, lanosterol 14-demethylase; ICER, inducible cAMP early repressor; IEG, immediate early response gene; Sp1; specificity protein 1; SCAP, sterol cleavage activating protein; SRE, sterol regulatory element; SREBP, SRE-binding protein.
Accepted for publication August 19, 2005.
References
Fimia GM, Sassone-Corsi P 2001 Cyclic AMP signalling. J Cell Sci 114:1971–1972
Sng JC, Taniura H, Yoneda Y 2004 A tale of early response genes. Biol Pharm Bull 27:606–612
Herdegen T, Leah JD 1998 Inducible and constitutive transcription factors in the mammalian nervous system: control of gene expression by Jun, Fos and Krox, and CREB/ATF proteins. Brain Res Rev 28:370–490
Molina CA, Foulces NS, Lalli E, Sassone-Corsi P 1993 Inducibility and negative autoregulation of CREM: an alternative promoter directs the expression of ICER, an early repressor. Cell 75:875–886
Brown MS, Goldstein JL 1997 The SREBP pathway: regulation of cholesterol metabolism by proteolysis of a membrane-bound transcription factor. Cell 89:331–340
Goldstein JL, Brown MS 1990 Regulation of the mevalonate pathway. Nature 343:425–430
Horton JD, Goldstein JL, Brown MS 2002 SREBPs: activators of the complete program of cholesterol and fatty acid synthesis in the liver. J Clin Invest 109:1125–1131
Sakakura Y, Shimano H, Sone H, Takahashi A, Inoue K, Toyoshima H, Suzuki S, Yamada N 2001 Sterol regulatory element-binding proteins induce an entire pathway of cholesterol synthesis. Biochem Biophys Res Commun 286:176–183
Espenshade PJ, Li WP, Yabe D 2002 Sterols block binding of COPII proteins to SCAP, thereby controlling SCAP sorting in ER. Proc Natl Acad Sci USA 99:11694–11699
Loewen CJ, Levine TP 2002 Cholesterol homeostasis: not until the SCAP lady INSIGs. Curr Biol 12:R779–R781
Wang H, Liu F, Millette CF, Kilpatrick DL 2002 Expression of a novel, sterol-insensitive form of sterol regulatory element binding protein 2 (SREBP2) in male germ cells suggests important cell- and stage-specific functions for SREBP targets during spermatogenesis. Mol Cell Biol 22:8478–8490
Rozman D, Fink M, Fimia GM, Sassone-Corsi P, Waterman MR 1999 Cyclic adenosine 3',5'-monophosphate (cAMP)/cAMP-responsive element modulator (CREM)-dependent regulation of cholesterogenic lanosterol 14-demethylase (CYP51) in spermatids. Mol Endocrinol 13:1951–1962
Halder S, Fink M, Waterman MR, Rozman D 2002 A cAMP responsive element binding site is essential for sterol regulation of the human lanosterol 14-demethylase gene (CYP51). Mol Endocrinol 16:1853–1863
Aoyama Y, Noshiro M, Gotoh O, Imaoka S, Funae Y, Kurosawa N, Horiuchi T, Yoshida Y 1996 Sterol 14-demethylase P450 (P45014DM) is one of the most ancient and conserved P450 species. J Biochem 119:926–933
Bellamine A, Mangla AT, Nes WD, Waterman MR 1999 Characterization and catalytic properties of the sterol 14-demethylase from Mycobacterium tuberculosis. Proc Natl Acad Sci USA 96:8937–8942
Rezen T, Debeljak N, Kordis D, Rozman D 2004 New aspects on lanosterol 14-demethylase and cytochrome P450 evolution: lanosterol/cycloartenol diversification and lateral transfer. J Mol Evol 59:51–58
Byskov AG, Andersen CY, Nordholm L, Thogersen H, Guoliang X, Wassman O, Guddal JVAE, Roed T 1995 Chemical structure of sterols that activate oocyte meiosis. Nature 374:559–562
Seamon KB, Padgett W, Daly JW 1981 Forskolin: unique diterpene activator of adenylate cyclase in membranes and in intact cells. Proc Natl Acad Sci USA 78:3363–3367
Rozman D, Strmstedt, M., Tsui LC, Scherer SW, Waterman MR 1996 Structure and mapping of the human lanosterol 14-demethylase gene (CYP51) encoding the cytochrome P450 involved in cholesterol biosynthesis: comparison of exon/intron organization with other mammalian and fungal CYP genes. Genomics 38:371–381
Sambrook J, Fritsch EF, Maniatis TM 1989 Molecular cloning: a laboratory manual. New York: Cold Spring Harbor Laboratory Press
Delmas V, van der Horn F, Mellstrm B, Jegou B, Sassone-Corsi P 1993 Induction of CREM activator proteins in spermatids: downstream targets and implications for haploid germ cell differentiation. Mol Endocrinol 7:1502–1514
Fon Tacer K, Haugen TB, Baltsen M, Debeljak N, Rozman D 2002 Tissue-specific transcriptional regulation of the cholesterol biosynthetic pathway leads to accumulation of testis meiosis-activating sterol (T-MAS). J Lipid Res 43:82–89
Worgall TS, Sturley SL, Seo T, Osborne TF, Deckelbaum RJ 1998 Polyunsaturated fatty acids decrease expression of promoters with sterol regulatory elements by decreasing levels of mature sterol regulatory element-binding protein. J Biol Chem 273:25537–25540
Christenson LK, Strauss 3rd JF 2001 Steroidogenic acute regulatory protein: an update on its regulation and mechanism of action. Arch Med Res 32:576–586
Omura T, Morohashi K 1995 Gene regulation of steroidogenesis. J Steroid Biochem Mol Biol 53:19–25
Zhan HC, Liu DP, Liang CC 2001 Insulator: from chromatin domain boundary to gene regulation. Hum Genet 109:471–478
Bell AC, Felsenfeld G 1999 Stopped at the border: boundaries and insulators. Curr Opin Genet Dev 9:191–198
Ness GC, Gertz KR, Holland RC 2001 Regulation of hepatic lanosterol 14-demethylase gene expression by dietary cholesterol and cholesterol-lowering agents. Arch Biochem Biophys 395:233–238
Rodriguez C, Martinez-Gonzalez J, Sanchez-Gomez S, Badimon L 2001 LDL downregulates CYP51 in porcine vascular endothelial cells and in the arterial wall through a sterol regulatory element binding protein-2-dependent mechanism. Circ Res 88:268–274
Sonoda Y, Amano C, Endo M, Sato Y, Sekigawa Y, Fukuhara M 1995 Lanosterol 14-demethylase (cytochrome P-45014DM): modulation of its enzyme activity by cholesterol feeding. Biol Pharm Bull 18:1009–1011
Trzaskos JM, Bowen WD, Shafiee A, Fischer RT, Gaylor JL 1984 Cytochrome P-450-dependent oxidation of lanosterol in cholesterol biosynthesis. J Biol Chem 259:13402–13412
Thewke DP, Panini SR, Sinensky M 1998 Oleate potentiates oxysterol inhibition of transcription from sterol regulatory element-1-regulated promoters and maturation of sterol regulatory element-binding proteins. J Biol Chem 273:21402–21407
Hannah VC, Ou J, Luong A, Goldstein JL, Brown MS 2001 Unsaturated fatty acids down-regulate SREBP isoforms 1a and 1c by two mechanisms in HEK-293 cells. J Biol Chem 276:4365–4372
Ellsworth JL, Lloyd DB, Carlstrom AJ, Thompson JF 1995 Protein binding to the low density lipoprotein receptor promoter in vivo is differentially affected by gene activation in primary human cells. J Lipid Res 36:383–392
Mascaro C, Nadal A, Hegardt FG, Marrero PF, Haro D 2000 Contribution of steroidogenic factor 1 to the regulation of cholesterol synthesis. Biochem J 350:785–790
Takagi K, Strauss 3rd JF 1989 Control of low density lipoprotein receptor gene expression in steroidogenic cells. Can J Physiol Pharmacol 67:968–973
Golos TG, Strauss 3rd JF 1988 8-Bromoadenosine cyclic 3',5'-phosphate rapidly increases 3-hydroxy-3-methylglutaryl coenzyme A reductase mRNA in human granulosa cells: role of cellular sterol balance in controlling the response to tropic stimulation. Biochemistry 27:3503–3606
Wang X, Sato R, Brown MS, Hua X, Goldstein JL 1994 SREBP-1, a membrane-bound transcription factor released by sterol-regulated proteolysis. Cell 77:53–62
Xu J, Teran-Garcia M, Park JH, Nakamura MT, Clarke SD 2001 Polyunsaturated fatty acids suppress hepatic sterol regulatory element-binding protein-1 expression by accelerating transcript decay. J Biol Chem 276:9800–9807
Golos TG, August AM, Strauss 3rd JF 1986 Expression of low density lipoprotein receptor in cultured human granulosa cells: regulation by human chorionic gonadotropin, cyclic AMP, and sterol. J Lipid Res 27:1089–1096
Ma W, Hatzis C, Eisenach JC 2003 Intrathecal injection of cAMP response element binding protein (CREB) antisense oligonucleotide attenuates tactile allodynia caused by partial sciatic nerve ligation. Brain Res 988:97–104
Sassone-Corsi P 1998 Molecular clocks: mastering time by gene regulation. Nature 392:871–874
Sassone-Corsi P 1998 Coupling gene expression to cAMP signalling: role of CREB and CREM. Int J Biochem Cell Biol 30:27–38
Mao D, Warner EA, Gurwitch SA, Dowd DR 1998 Differential regulation and transcriptional control of immediate early gene expression in forskolin-treated WEHI7.2 thymoma cells. Mol Endocrinol 12:492–503
Kemp DM, George SE, Kent TC, Bungay PJ, Naylor LH 2002 The effect of ICER on screening methods involving CRE-mediated reporter gene expression. J Biomol Screen 7:141–148
Simpson ER, Waterman MR 1988 Regulation of the synthesis of steroidogenic enzymes in adrenal cortical cells by ACTH. Annu Rev Physiol 50:427–440
Waterman MR, Simpson ER 1989 Regulation of steroid hydroxylase gene expression is multifactorial in nature. Recent Prog Horm Res 45:533–563
Thommesen L, Norsett K, Sandvik AK, Hofsli E, Laegreid A 2000 Regulation of inducible cAMP early repressor expression by gastrin and cholecystokinin in the pancreatic cell line AR42J. J Biol Chem 275:4244–4250
Debeljak N, Fink M, Rozman D 2003 Many facets of mammalian lanosterol 14-demethylase from the evolutionarily conserved cytochrome P450 family CYP51. Arch Biochem Biophys 409:159–171
Rao KN 1995 The significance of the cholesterol biosynthetic pathway in cell growth and carcinogenesis (review). Anticancer Res 15:309–314
Hentosh P, Yuh SH, Elson CE, Peffley DM 2001 Sterol-independent regulation of 3-hydroxy-3-methylglutaryl coenzyme A reductase in tumor cells. Mol Carcinog 32:154–166
Bocchetta M, Bruscalupi G, Castellano F, Trentalance A, Komaromy M, Fong LG, Cooper AD 1993 Early induction of LDL receptor gene during rat liver regeneration. J Cell Physiol 156:601–609
Feingold KR, Pollock AS, Moser AH, Shigenaga JK, Grunfeld C 1995 Discordant regulation of proteins of cholesterol metabolism during the acute phase response. J Lipid Res 36:1474–1482
Zager RA, Shah VO, Shah HV, Zager PG, Johnson AC, Hanson S 2002 The mevalonate pathway during acute tubular injury: selected determinants and consequences. Am J Pathol 161:681–692
Ruan XZ, Varghese Z, Fernando R, Moorhead JF 1998 Cytokine regulation of low-density lipoprotein receptor gene transcription in human mesangial cells. Nephrol Dial Transplant 13:1391–1397
Ruan XZ, Varghese Z, Powis SH, Moorhead JF 2001 Dysregulation of LDL receptor under the influence of inflammatory cytokines: a new pathway for foam cell formation. Kidney Int 60:1716–1725
Cutler RG, Kelly J, Storie K, Pedersen WA, Tammara A, Hatanpaa K, Troncoso JC, Mattson MP 2004 Involvement of oxidative stress-induced abnormalities in ceramide and cholesterol metabolism in brain aging and Alzheimer’s disease. Proc Natl Acad Sci USA 101:2070–2075
Gesquiere L, Loreau N, Minnich A, Davignon J, Blache D 1999 Oxidative stress leads to cholesterol accumulation in vascular smooth muscle cells. Free Radic Biol Med 27:134–145
Shack S, Gorospe M, Fawcett TW, Hudgins WR, Holbrook NJ 1999 Activation of the cholesterol pathway and Ras maturation in response to stress. Oncogene 18:6021–6028(Martina Fink, Jure Aimovi, Tadeja Reen, )
Abstract
Lanosterol 14-demethylase (CYP51) responds to cholesterol feedback regulation through sterol regulatory element binding proteins (SREBPs). The proximal promoter of CYP51 contains a conserved region with clustered regulatory elements: GC box, cAMP-response elements (CRE-like), and sterol regulatory element (SRE). In lipid-rich (SREBP-poor) conditions, the CYP51 mRNA drops gradually, the promoter activity is diminished, and no DNA-protein complex is observed at the CYP51-SRE1 site. The majority of cAMP-dependent transactivation is mediated through a single CRE (CYP51-CRE2). Exposure of JEG-3 cells to forskolin, a mediator of the cAMP-dependent signaling pathway, provokes an immediate early response of CYP51, which has not been described before for any cholesterogenic gene. The CYP51 mRNA increases up to 4-fold in 2 h and drops to basal level after 4 h. The inducible cAMP early repressor (ICER) is involved in attenuation of transcription. Overexpressed CRE-binding protein (CREB)/CRE modulator (CREM) transactivates the mouse/human CYP51 promoters containing CYP51-CRE2 independently of SREBPs, and ICER decreases the CREB-induced transcription. Besides the increased CYP51 mRNA, forskolin affects the de novo sterol biosynthesis in JEG-3 cells. An increased consumption of lanosterol, a substrate of CYP51, is observed together with modulation of the postlanosterol cholesterogenesis, indicating that cAMP-dependent stimuli cross-talk with cholesterol feedback regulation. CRE-2 is essential for cAMP-dependent transactivation, whereas SRE seems to be less important. Interestingly, when CREB is not limiting, the increasing amounts of SREBP-1a fail to transactivate the CYP51 promoter above the CREB-only level, suggesting that hormones might have an important role in regulating cholesterogenesis in vivo.
Introduction
REGULATION OF GENES encoding enzymes of cholesterol biosynthesis and uptake is mediated generally by the negative cholesterol feedback loop and sterol regulatory element binding protein (SREBP) transcription factors. SREBPs need partners for performing their task of maintaining the cellular cholesterol level. One of the coregulatory signaling pathways is the cAMP-dependent pathway, which is characterized by the cAMP-response element binding protein (CREB)/cAMP-response element modulator (CREM)/activating transcription factor 1 family of transcription factors and protein kinase A. Induction of the cAMP pathway leads to phosphorylation of cAMP-dependent transcription factors, which bind to cAMP-responsive elements (CRE) in the promoter/regulatory regions of the cAMP-responsive genes and in this way enhance their transcription (1). The induction of transcription can be immediate (a few minutes to hours) or delayed (chronic, several hours). Genes that are rapidly and transiently induced in response to the intracellular signaling cascades are described as immediate early response genes (IEG). The induction of IEG occurs in the absence of de novo protein synthesis and thus could not be blocked by protein synthesis inhibitors. IEG encode secreted proteins, cytoplasmic enzymes, and transcription factors (2). Most of our knowledge is about IEG encoding transcription factors c-jun and c-fos in the nervous and immune systems (3). The transcription of genes that respond to the induction by cAMP-dependent signaling is later repressed. The mediator of the repression is the cAMP-inducible transcription factor inducible cAMP early repressor (ICER), the synthesis of which is mediated by cAMP (4).
The SREBP regulatory pathway is characterized by sterol regulatory element binding proteins (reviewed in Refs.5, 6, 7, 8). SREBPs are synthesized as membrane-bound inactive precursors whose processing is regulated by a lipid-sensor mechanism. Alteration in cholesterol content of membranes promotes changes in physical properties of the membrane in cholesterol (lipid)-rich conditions, which prevents the sterol cleavage activating protein (SCAP)-SREBP complex from reaching the Golgi apparatus (9). Insig was proposed to anchor the SCAP-SREBP complex in the endoplasmic reticulum in the presence of cholesterol (10). Consequently, SREBPs remain membrane-bound precursors of the endoplasmic reticulum in lipid-rich conditions. The central dogma of cholesterol homeostasis says that when cholesterol, oxysterol, and fatty acid levels are limiting (lipid-poor conditions), precursor SREBPs are transported to the Golgi apparatus and are proteolytically cleaved. The N-terminal soluble basic-helix-loop-helix-Zip portions of mature SREBPs enter the nuclei and transactivate target genes by binding to sterol regulatory elements (SRE) of the target promoters. The exception is SREBP-2gc, which is synthesized as an already soluble protein in male germ cells and can thus transactivate SREBP-dependent genes irrespective of the cell cholesterol level (11). Together with other data (12, 13), this has suggested that under certain physiological or pathophysiological conditions, the genes encoding cholesterogenic enzymes might more generally be up-regulated irrespective of the cellular cholesterol level.
Lanosterol 14-demethylase (CYP51) is an enzyme of the late portion of cholesterol biosynthesis. It represents the most evolutionarily conserved member of the cytochrome P450 (CYP) gene superfamily (14, 15, 16). The direct product of the CYP51 enzymatic reaction is follicular fluid meiosis activating sterol, one of the short-lived intermediates of cholesterol biosynthesis that regularly accumulates only in testis and ovary (17).
Herein we demonstrate for the first time that forskolin, a mediator of the cAMP-dependent signaling pathway, activates the cholesterogenic CYP51 with an immediate early time response also in the absence of mature SREBP. A single CRE is sufficient to mediate this transactivation, and ICER is involved in attenuation of transcription. The immediate early transactivation of the CYP51 gene by forskolin is the first clear demonstration that cholesterogenic genes might respond to cAMP signaling also independently of the cholesterol feedback regulation. This might influence the de novo sterol synthesis when SREBP are limiting. We used the model system of JEG-3 cells that are derived from an endocrine organ, the human placenta. These cells harbor the components of cholesterol feedback regulation as well as of the cAMP-signaling pathway. Forskolin as a naturally occurring diterpene has been used to simulate the effects of physiologically occurring hormones (18), such as epinephrine, norepinephrine, FSH, LH, ACTH, and glucagon, that are known to activate the cAMP-signaling pathway.
Materials and Methods
Construction of reporter gene plasmids
Preparation of the wild-type human CYP51 D7 chloramphenicol acetyl transferase (CAT) (–334/+316) has been described previously (12), as well as preparation of CYP51 D7 luc (–334/+316), CYP51 D7 CAT (–121/+316), and CAT reporter mutant constructs CYP51-CRE2a-mut (TGACGCGA changed to TGATTCGA), CRE2b-mut (TGACGCGA was changed to AATGGCGA), CYP51-SRE1-mut (AATCACCTCAG was changed into TTTTTTTTTT), and CYP51-GC-mut (GGGGGCGG was changed into GGGTTTTTG) (13). Both CYP51-CRE-like mutants (CRE2a-mut and CRE2b-mut) have similar attributes in transfection studies (13). The substitution mutation CAT reporter constructs CYP51-CRE3-mut and CRE2-mut were prepared using QuikChange Site-Directed Mutagenesis Kit (Stratagene, La Jolla, CA). CYP51 D7 CAT served as a template for mutagenesis and preparation of CYP51-CRE3-mut, whereas for generation of a double mutant, CRE2a-mut was used as a template.
Human CYP51 C10 CAT (–471/+316) was amplified by the cloned Pfu polymerase (Stratagene) using sense 5'-TTTAG TTGGA GTGGG ACGGG TGACC-3' and antisense 5'-ACCTC GAACT GTGGC ACCTC ACCCT TCTCC-3' primers from the cosmid 121G12 containing the entire human CYP51 gene (19). The fragment was cloned into the SmaI-digested pCATbasic plasmid (Promega, Madison, WI).
Mouse Cyp51 CAT construct (–527/+104) was amplified by the cloned Pfu polymerase (Stratagene) using sense 5'-CCTTG AACCG ACGCG TGCCC AGAGG TGACG TCTCC-3' and antisense 5'-CCTTG GCATC AAGAT CTCAT GGCCT GTCCG AGCAC C-3' primers from the BAC 51921 clone, containing the entire mouse Cyp51 gene. This region of the Cyp51 promoter was cloned into the pCR-TOPO vector (Invitrogen, Carlsbad, CA) digested with MluI and BglII. The –527/+124 region of the mouse Cyp51 promoter was then cloned into the gel-purified pCAT basic from the MluI- and BglII-digested human CYP51 D7 CAT. The –527/+124 region of the mouse Cyp51 promoter is analogous to the –334/+316 region of the human CYP51 promoter. All human and mouse CYP51 CAT and luciferase reporter constructs were verified by sequencing.
Cell cultures
Human choriocarcinoma JEG-3 and human hepatocarcinoma HepG2 cells were cultured in DMEM (Sigma, Taufkirchen, Germany) containing 5% bovine calf serum and 1% L-glutamine in a 5% CO2 incubator at 37 C. In some experiments, lipid-rich medium (COPUFA) has been used (DMEM with 1% BSA, 10 μg/ml cholesterol, 1 μg/ml 25-hydroxycholesterol, and 0.15 mM linoleic acid) or 10% delipidated serum (Sigma). In experiments with delipidated serum, the 10% bovine calf serum medium was used as control (12). Cells were grown to 90% confluency in T150 tissue culture flasks and were split 1:4 into 60-cm2 dishes 24 h before transfections into normal or lipid-rich medium (with COPUFA).
Transfections and determination of reporter gene activity
For transfection studies, CAT and firefly luciferase reporter systems have been used. All transfections using CAT reporters were performed in human choriocarcinoma JEG-3 cells, whereas with human CYP51 D7 luciferase reporter, human hepatoma HepG2 and JEG-3 cells were transfected. Transfections and analysis of CAT activity have been performed as described previously (12, 13). In experiments with deletion CAT constructs, 3 μg pSV CREM has been used. In experiments with mouse Cyp51 CAT construct, where interactions between SREBP and forskolin have been studied, 3 μg pCMV SREBP-1a has been applied. In all other cotransfection studies, 200 ng pRSV CREB, pSV CREM, pCMV-SREBP1a, and pSV-specificity protein 1 (pSV-Sp1) has been applied. -Galactosidase has been used for normalization of the transfection efficiency with CAT reporters. pCAT basic was used as a DNA carrier to 20 μg.
Transfections of HepG2 or JEG-3 cells were performed with the firefly luciferase human CYP51 D7 reporter construct. Cells were transfected with 2 μg human CYP51 D7 luciferase reporter, 500 ng pSV -galactosidase plasmid, and 200 ng pSV CREM, pRSV CREB, or pSV ICERII, pSV Sp1, and pCMV SREBP-1a expression vectors or as indicated in the figure. pCAT basic was used as the carrier DNA to 3 μg. -Galactosidase was used for normalization of the transfection efficiency also in this case. The assay for determining the -galactosidase activity was performed as described (20). Firefly luciferase activity was analyzed with the commercial kit for luciferase (Promega) and Turner TD-20/20 luminometer. All transfection experiments were performed at least three times with two petri dishes for each experimental condition. In all transfection experiments, reporter activity has been calculated by the following formula: reporter activity = (normalized reporter gene activity)/(reporter activity in untreated cells). Reporter activity in untreated cells represents unit 1 and is shown as a black bar in column 1 of each diagram. The average value, SEM, and P values from two-tailed t test were calculated with the Excel program (Microsoft Corp., Redmond, WA). pSV CREM and pSV ICERII were a gift of Dr. P. Sassone-Corsi (Centre National de la Recherche Scientifique, Strasbourg, France). pRSV CREB and pSV Sp1 originate from the laboratory of Dr. M. R. Waterman (Vanderbilt University, Nashville, TN). It was confirmed previously that the results of transfections of human CYP51 CAT and firefly luciferase reporter genes are comparable (Aimovi, J., M. Fink, and D. Rozman, unpublished).
Nuclear extract preparation, expression, and purification of recombinant CREM protein and gel shift analysis
Nuclear extracts were isolated from JEG-3 cells that were grown in normal and COPUFA media as described above. Medium was changed every 24 h. Nuclear extracts were isolated from four petri dishes after 72 h, as described previously (12).
Purification of CREM expressed in bacteria has been described previously (21). Oligonucleotides used to generate double-stranded fragments containing human, mouse, and rat CYP51-CRE and human CYP51-SRE1 are presented in Table 1. Gel shift experiments were performed as described previously (12, 13).
RNA isolation and quantification of CYP51 mRNA by real-time PCR and Northern analysis
Total cellular RNA was isolated from JEG-3 cells. Forty-eight hours before the beginning of the experiment, cells were split 1:5 and were subjected to normal or lipid-rich medium (with COPUFA). Media were replaced every 24 h. After 15 min and 0.5, 1.5, 2, 4, 8, 12, 16, 24, 36, 48, 50, 52, 56, and 60 h, the medium was aspirated. Cells were washed twice with RNase-free PBS buffer and harvested with a rubber policeman, and total RNA was isolated with TRI reagent (Sigma). At 48 h, the medium was changed into fresh COPUFA medium, and cycloheximide (CHX) was added (10 μg/ml final concentration). Fifteen minutes after the addition of CHX, forskolin (25 μM final concentration) was added. Cells were harvested after 15 and 30 min and 1, 1.5, 2, 4, 8, and 12 h, and RNA was isolated as described above.
Total RNA from two 55-mm petri dishes was combined in each experiment. RNA concentration and quality were determined by RNA 6000 Nano Assay with Agilent 2100 Bioanalyzer (Agilent Technologies, Waldbronn, Germany). At least two different samples of each RNA sample were isolated and investigated for each time point.
One microgram of total RNA was converted into cDNA in a 20-μl reaction mixture using a SuperScript II reverse transcriptase (Invitrogen) with random primers (Promega). The reaction mixture was treated with DNase I (Sigma) to remove the contaminating DNA. The quality of each cDNA was tested in PCR with -actin primers.
Real-time PCR was performed with the ABI PRISM 7900 HT sequence detection system (Applied Biosystems, Foster City, CA) applying TaqMan technologies with the assay-by-design human CYP51 primers, sense 5'-CAG GTT GGC TGC CTT TGC-3' and antisense 5'-CTT GAT TTC CCG ATG AGC TCT GT-3', and probe CCC TGC GTC TGA AAC T labeled with FAM. The expression of the CYP51 was normalized to 18S rRNA content using a TaqMan rRNA control reagent kit. Each 20-μl reaction mixture contained 1 μl of 100x diluted cDNA template, 10 μl TaqMan Universal PCR Master Mix, 8 μl water and 1 μl of 20x control 18S rRNA or investigated CYP51 primers and probe and was amplified as follows: after incubation at 50 C for 2 min and denaturating at 95 C for 10 min, 50 cycles were performed at 95 C for 15 sec and 60 C for 1 min. The level of each cDNA was determined by the comparative CT method as described in User Bulletin 2, 1997. All procedures followed the Applied Biosystems protocols. For each time point, real-time PCR was conducted in triplicate from at least two independently isolated RNA samples. All calculations (average CT values, CT, average values, SEM, and P values from two-tailed t test) were calculated with the Excel program (Microsoft).
Northern blot analysis was performed by standard procedures (20). 32P-labeled 449-bp probe from the coding region of the human CYP51 gene (exons 1–3) was prepared with Prime-It II random primer labeling kit (Stratagene). Filters were hybridized for 2 h with QuikHyb hybridization solution (Stratagene), washed twice for 15 min at room temperature (0.1% SDS, 0.3 M NaCl, 0.03 M sodium citrate), once for 30 min at 60 C (0.1% SDS, 0.3 M NaCl, 0.03 M sodium citrate), and exposed for 16–48 h.
Immunoblot analysis
Western blot analysis was performed with 20 μg of nuclear proteins isolated from JEG-3 cells after forskolin treatment for different times and analyzed using an anti-CREM polyclonal antibody (21) and visualized by the SuperSignal West Pico chemiluminescent substrate (Pierce, Rockford, Il).
Metabolic labeling and sterol analysis
JEG-3 cells were split 1:2 the day before the beginning of the experiment. Cells were split 1:6 into 75-cm2 cell culture flasks at d 0, and 10 ml of normal medium was added. On d 1, medium was changed to normal medium or lipid-rich medium (with COPUFA). After 48 h (d 3), ketoconazole (1 μM final concentration) and/or forskolin was added (25 μM final concentration) together with the mixture of 80 μCi [3H]sodium acetate (NEN Life Science Products, Boston, MA) (10 mCi/ml, 3.1 Ci/mmol) and 0.2 μmol cold sodium acetate per 1 ml medium.
Cells were returned to the cell culture incubator for an additional 8 h. After trypsinization and homogenization, the concentration of total cellular proteins was determined in the homogenate and sterols extracted from 0.8 ml of the homogenate in glass vials. Ten microliters of ergosterol in ethanol (0.5 mg/ml), 100 μl of 0.3 M NaH2PO4 (pH 1.0), and 3 ml of extraction solution (75% n-heptane, 25% isopropanol) were added. The vials were incubated for at least 2 h on a shaker in the dark at room temperature and centrifuged (2000 x g for 10 min at room temperature). The organic phase was transferred and dried under vacuum, washed with 2 ml n-heptane, and centrifuged. Supernatants were transferred into fresh tubes and dried again. Samples were dissolved in the weight quantity of HPLC solvent. HPLC analysis was performed as described previously (22). The cholesterol fraction contains up to 20% lathosterol. Postlanosterol intermediates 7-dehydrocholesterol, zymosterol, and desmosterol eluted at identical retention time under applied separation conditions. Sterol standards (cholesterol, lanosterol, 7-dehydrocholesterol, zymosterol, and desmosterol) were from Steraloids (Newport, RI).
The average values, SEM, and P values from two-tailed t test were calculated with the Excel program (Microsoft).
Results
CYP51 mRNA expression pattern in lipid-repressed and cAMP-induced conditions follows the immediate early response
The cAMP-signaling pathway is one of the most investigated and in biological systems one of the most frequently used pathways mediating cellular responses to different physiological stimuli, but its involvement in cholesterol homeostasis has been suggested only recently. The aim of our work was to investigate the molecular mechanisms of cAMP-mediated transcription of cholesterogenic CYP51 that participates in late phases of cholesterol biosynthesis. CYP51 is transactivated by transcription factors of the SREBP family (12, 13) in a similar manner as other cholesterogenic genes. In accordance with the cholesterol feedback regulation, the amount of nuclear SREBP is highest when COPUFA are at a minimum (7, 23) and is decreased in lipid-rich media (23). Figure 1A shows that lipid-rich medium with COPUFA, containing 25-hydroxycholesterol, cholesterol, and linoleic acid, prevents the formation of a specific SREBP-dependent transactivation complex at the CYP51 promoter. A 22-bp CYP51-SRE1 element of the human CYP51 promoter binds SREBP specifically (Fig. 1A). Addition of SREBP-1a antibodies prevented binding of nuclear proteins to the human CYP51-SRE1 (compare lanes 1 and 2). No binding to the CYP51-SRE1 was detected in nuclear extracts from the COPUFA medium (lane 3). Results indicate that lipid-rich medium abolishes SREBP binding to CYP51-SRE1. Figure 1B shows a gradual decrease in CYP51 mRNA in lipid-rich medium (0–12 h), which was studied in detail by real-time PCR (Fig. 1C). The level of CYP51 mRNA is reduced by 60% in 2 h and 90% in 12–16 h and is later stabilized at about 30% for the entire time of the experiment. The data of Fig. 1 thus show that no SREBP are bound to the CYP51-SRE in COPUFA medium and that this results in a marked decrease of CYP51 mRNA. If under such conditions, cells are treated with forskolin, a mediator of cAMP response, the CYP51 mRNA level raises again, in a transient fashion. Forty-eight hours of COPUFA treatment represents also time 0 of forskolin addition. Forskolin, in the presence of COPUFA, increased the CYP51 mRNA in 1.5–2 h (49.5–50 h in Fig. 2A). After this time, the CYP51 mRNA progressively decreased to the basal level in 4–6 h (52–54 h in Fig. 2A). The addition of CHX, an inhibitor of protein synthesis, not only abolishes the decrease in CYP51 mRNA accumulation with time but allows the increase to continue (Fig. 2A). These results show that CYP51 transcription is induced by cAMP-dependent agents in an immediate-early fashion and that for such activation, de novo protein synthesis is not needed. However, de novo protein synthesis is required for attenuation of transcription.
To monitor whether the forskolin-induced CYP51 transcription affects also sterol synthesis, JEG-3 cells have been metabolically labeled with [3H]acetate, a C-2 precursor of cholesterol. Addition of forskolin to JEG-3 cells leads to a statistically significant decrease in the CYP51 enzyme substrate lanosterol after addition of forskolin, whereas the quantity of postlanosterol intermediates and cholesterol did not change significantly (Fig. 2B), which could be because of the well-known forskolin-induced consumption of cholesterol in the synthesis of steroid hormones (24, 25). Data indicate that the forskolin-mediated activation of CYP51 transcription leads to an increased activity of the CYP51 protein followed by an increased expenditure of lanosterol. Addition of COPUFA medium reduced the cholesterol quantity 10-fold, and the lanosterol quantity almost 6-fold, reaching the detection limit of the method (data not shown).
To show that the decrease in lanosterol quantity is a result of an increased consumption of lanosterol and not a decreased production of lanosterol because of a forskolin-mediated inhibition of an enzyme of the prelanosterol cholesterol biosynthesis, a CYP51 inhibitor, ketoconazole, was applied (Fig. 2B). Treatment with ketoconazole increased significantly the quantity of lanosterol (black bars) and 24,25-dehydrolanosterol (not shown), whereas the postlanosterol intermediates have been below the level of detection both in the normal and in the COPUFA medium. The appearance of 24,25-dehydrolanosterol, another possible substrate of CYP51 in cholesterol biosynthesis, indicates that ketoconazole completely blocked CYP51 and that cholesterol biosynthesis did not take place by an alternative pathway. This is in accordance with the belief that a block in any of the cholesterol biosynthesis steps results in the absence of all downstream pathway intermediates. ICER is a cAMP-inducible repressor that is encoded by the last five exons of the CREM gene and is transcribed from the intronic CREM promoter P2 upon cAMP induction (4). ICER lacks the transactivation domain but contains the DNA binding domain, thus representing a cAMP-dependent repressor that binds to identical cAMP-responsive elements as the activator forms of CREB and CREM. The addition of forskolin results in a gradual increase of the cAMP-inducible repressor CREM isoforms ICER II and ICER II, whereas significantly higher molecular weight activator CREM isoforms (CREM, -, -, and -S) do not fluctuate much during the treatment (Fig. 3A), which is consistent with the lack of inducibility of non-ICER CREM transcripts (4). Overexpression of pSV ICER repressed the human CYP51 D7 luciferase reporter gene activity, whereas overexpression of pRSV CREB induced the CYP51 D7 luciferase reporter (Fig. 3). ICER represses the CREB-induced transactivation of CYP51, because lower luciferase activity was determined when ICER and CREB were coexpressed in comparison with the overexpression of CREB alone (P < 0.05). This is consistent with the CYP51 mRNA pattern of Fig. 2A.
Evaluation of CRE in promoters of mammalian CYP51 genes
Mammalian CYP51 promoters are evolutionarily conserved and contain three cAMP-responsive elements, CRE1, CRE2, and CRE3, which all differ from the consensus CRE. Only the human and mouse CYP51-CRE1 are capable of binding the recombinant CREM (Fig. 4A, columns 1 and 2), whereas the rat CYP51-CRE1 failed in binding this cAMP-dependent transcription factor (Fig. 4A, column 3). The sequences of individual CRE-like elements (Fig. 4B) show that the downstream portion of the CYP51-CRE1 palindrome is not conserved. In contrast to CRE1, CYP51-CRE2 and CRE3 elements are conserved and are identical in human/pig and in mouse/rat CYP51 promoters (Fig. 4B).
To explore further the importance of individual CRE, the human CYP51 promoter deletion CAT constructs have been transfected into JEG-3 cells together with overexpressed CREM. Results demonstrate that the shortest construct D7 exhibits a very low basal activity (5% of D7) and does not respond to overexpression of CREM (Fig. 5B, column 1). This indicates that the –121/+316 human CYP51 promoter is not sufficient for basal transcription and is unable to respond to cAMP-dependent stimuli despite the presence of CRE3.
Interestingly, the longest CYP51 promoter construct C10 (Fig. 5B, column 3) also exhibits a lower basal activity (25% of D7) and shows a weaker response to CREM despite the presence of three CRE-like elements in comparison with D7, containing CRE2 and CRE3 only (Fig. 5B, column 2).
The –471/–334 region including CYP51-CRE1 seems not to be important for the cAMP-dependent transactivation of hCYP51, or the repressor element might be present in this region, because no additional activation has been observed in the CYP51 C10 construct that contains CRE1 in addition to CRE2 and CRE3 (26, 27).
To evaluate further the hypotheses based on deletion promoter-reporter construct, promoters with mutant CRE2 and CRE3 regulatory elements have been transfected together with overexpressed CREM or CREB. Although the mutation of CRE2 abolished both basal expression and the cAMP-dependent transactivation, CRE3 mutation retained approximately half of CREM-dependent and over 60% of CREB-dependent transactivation (Fig. 5C). As expected, the double mutation exhibits a similar response as the CRE2-mutant alone. The cumulative results of Fig. 5, B and C, suggest that the human CYP51-CRE2 is the principal CRE-like element responsible for cAMP-dependent transactivation. Interestingly, CRE3 has a strong effect on the basal activity of the CYP51 reporter, suggesting potential interactions with the basal transcriptional machinery.
cAMP-dependent transactivation of human and mouse CYP51 and cross-talk with the sterol regulatory pathway
The –334/+316 human and –527/+104 mouse CYP51 promoters, both containing CRE2 and CRE3 elements, have been applied in functional promoter studies in JEG-3 and HepG2 cells with overexpressed CREM and CREB. CAT activity of the human and mouse CYP51 promoters in normal medium was taken as basal, and all modifications, such as change of the medium or overexpression of transcription factors, were compared with the CYP51 promoter-CAT reporter activities observed in the normal medium.
Generally, both human and mouse CYP51-CAT constructs show a similar pattern of response (transactivation) to overexpression of CREM and CREB in individual cell lines. In normal medium, the overexpression of CREM and CREB results in up to 3-fold activation of hCYP51 in HepG2 cells and 4- to 12-fold activation in JEG-3 cells (Fig. 6A). The mouse Cyp51 promoter is also transactivated by overexpressed CREM and CREB, albeit to a lower extent compared with the human CYP51 promoter. The observed weaker response of the mouse Cyp51 promoter in a human cell line nuclear environment might be explained by the requirement for an optimal species-specific transactivation complex. Several gene-specific transcription factors as well as factors associated with the general transcriptional machinery differ between human and mouse in amino acid sequence and consequently also in the three-dimensional structure. For example, transcription factor SREBP-1a is 80.4% identical between mouse (NP_035610) and human (NP_004167), whereas CREM seems to be more conserved and shows a 94.2% identity between mouse (P27699) and human (Q03060). Numbers in parentheses represent GenBank identification numbers.
The transactivation capacity of CREM and CREB was also monitored in lipid-rich medium COPUFA (Fig. 6B). As shown previously (Fig. 1A), COPUFA prevents binding of SREBP to CYP51-SRE. Figure 6B shows that treatment of cells with COPUFA results in a decrease of human (7-fold in HepG2 and 6-fold in JEG-3 cells) and mouse (20-fold in JEG-3 cells) CYP51-CAT reporter activities. Overexpression of CREM and CREB results in reactivation of CYP51 promoters of both species in lipid-rich medium, 1.5-fold (CREM) or 3.5-fold (CREB) above the basal level for the human promoter.
The effect of forskolin on the SREBP-1a-mediated transactivation is shown in Fig. 6C. Lanes 1–3 show the effect of decreasing amounts of lipids (COPUFA, normal medium, and delipidated serum) on the CYP51-CAT reporter transactivation. Overexpression of nuclear SREBP-1a (lanes 4 and5) amplifies transactivation. No major difference is observed between cells grown in normal serum and delipidated serum. When forskolin is added to the cells with overexpressed SREBP-1a, the transactivation is diminished (compare last two lanes). This suggests interactions and/or competition for binding sites of CRE-2 and SRE-1 factors, because the core elements (CRE-2 octamer and SRE-1 decamer) are separated by only 12 bp (13). The titration assay, where different concentrations of SREBP-1a and CREB expression plasmids were cotransfected to JEG-3 cells together with the human CYP51-luciferase reporter (Fig. 6D), was used to evaluate potential interactions between CRE-2 and SRE-1 binding factors. At low CREB concentrations (0–20 ng CREB expression plasmid), increasing amounts of SREBP-1a increase the transactivation of the CYP51 promoter. The transactivation is highest when 600 ng SREBP-1a plasmid and no CREB are expressed. Similarly, at low SREBP-1a concentrations (0–20 ng SREBP-1a expression plasmid), increasing amounts of CREB increase the transactivation of the CYP51 promoter. The maximal CREB-dependent transactivation of human CYP51 is lower compared with the SREBP-dependent transactivation. Interestingly, increasing amounts of CREB decrease the maximal SREBP-dependent transactivation, suggesting that CRE-dependent transcription factors bind with a higher affinity to the CYP51-CRE2 site. When CREB is not limiting (at least 200 ng of the expression plasmid), even increasing amounts of SREBP-1a cannot transactivate the promoter above the CREB-only maximal level. This suggests that at high concentrations the CYP51-CRE2-bound CREB might be able to displace SREBP-1a from the CYP51-SRE1. Alternatively, CREB protein interaction with SREBP-1a might lead to a less potent transactivation complex, resulting in less efficient transcription of the CYP51 gene. Similar results were obtained for CREM; however, 600 ng of the pRSV CREM expression plasmid did not yet saturate the promoter (not shown). It seems possible that at physiological conditions that result in high concentrations of cAMP-dependent transcription factors, the cholesterogenic CYP51 responds as a cAMP-dependent gene irrespective of the SREBP signaling pathway. Additional studies are required to evaluate this hypothesis and to depict the endogenous concentrations of interacting CRE-binding factors and SREBP.
Relative importance of the CYP51 promoter elements for the cAMP-mediated transactivation
Data above show CRE2 as most important for the cAMP-dependent transactivation of CYP51. This element lies within the –282/–225 promoter region that also contains GC-box and SRE-1 elements (13). In the normal medium, all three mutants (CRE2, SRE, and GC-box) exhibit a lower basal activity compared with the wild-type human CYP51-CAT reporter (Fig. 7, A and B). The highest effect of the mutation is observed with mutated GC-box where only 11% of the basal activity was retained, whereas CRE2b and SRE1 mutations retained about 45% of the basal activity. CRE2b-mut showed the weakest response with only 12% of the wild-type reporter activity after overexpression of CREB. Fifty percent of CAT activity remains after overexpression of CREM with the SRE-1 mutant and 37% after overexpression of CREB. GC-box-mut preserved only 10% of the CREM-dependent transactivation and 30% of the CREB-dependent transactivation.
The data in Fig. 7, A and B, demonstrate that the three DNA regulatory elements (GC-box, CRE2, and SRE1) from the –282/–225 region of the human CYP51 promoter contribute to the basal as well as cAMP-dependent CYP51 gene transactivation in normal medium that contains SREBP from bovine serum. Although an intact GC-box seems to be most important for the basal transcription, CRE2 is essential for the CREM/CREB-dependent transcription, whereas SRE1 seems to have a lower impact. As established previously, SRE1 has a major role in SREBP-dependent transactivation of CYP51 (13).
COPUFA diminished the activity of the wild-type human CYP51-CAT reporter to 16% (Fig. 7C). The activities of CRE2 mut and GC-box-mut were diminished to 50% of the initial value in normal medium. The CYP51 SRE1-mut reporter activity is diminished to 45% of the wild type in the presence of SREBP from the normal, bovine serum-supplied medium. Additionally, our previous results have shown clearly that in the presence of overexpressed SREBP-1a, the CYP51 SRE1-mut activity is diminished to 10% of the wild type (13). The human CYP51 SRE1-mut has a similar response as the wild-type promoter in sterol-rich media, particularly with CREM-dependent transactivation, suggesting activation by CREM in the absence of SREBP-dependent transactivation complex.
Overexpression of Sp1 does not affect the majority the CYP51 transactivation in normal media or in the lipid-rich conditions (Figs. 7, B and C gray bars). This observation is in accordance with previously published work (13); however, it seems to be controversial with dramatic effects observed with mutations of the GC-box. This seeming discrepancy can be explained by the ubiquitous nature of Sp1 that is likely not limiting in JEG-3 cells. Alternatively, Sp1 might serve as a competence factor. For example, overexpression of a competence factor may result in no basal induction of promoter activity, but its binding to response elements may still be required to potentiate transactivation by other factors.
Discussion
Being a part of the cholesterol biosynthesis pathway, CYP51 transcription is regulated primarily by transcription factors SREBP. Cholesterol/SREBP-dependent regulation of CYP51 has been demonstrated previously by several examples (8, 12, 13, 28, 29, 30, 31). Unsaturated fatty acids and 25-hydroxycholesterol work synergistically in lowering the nuclear, transcriptionally active forms of SREBP-1 and SREBP-2 proteins (32) by inhibiting the SREBP precursor cleavage and its translocation to the nucleus. A repressing effect on SREBP-regulated genes has been detected already by 16 h and has been preserved for 48 h at the level of transcription (33, 34, 35, 36, 37). The CYP51 transcription is reduced by over 50% within 4 h in the lipid-rich COPUFA medium. This quick repression could be explained with the short half-life of the nuclear forms of SREBP. Wang et al. (38) reported that reduction at the level of nuclear forms of SREBP is observed already 30 min after addition of cholesterol and 25-hydroxycholesterol. Degradation of SREBP requires protein synthesis and can be blocked by the inhibitor of protein synthesis CHX. Because in the presence of CHX the degradation of SREBP is blocked, the transcription of SREBP target genes is slightly up-regulated (39). A similar observation has been noticed also in our research (Fig. 2A, 52 h).
In this work, we describe for the first time an immediate early transactivation of the cholesterogenic CYP51 by a cAMP mediator, forskolin. Reports from the late 1980s have indicated that the low-density lipoprotein receptor (36, 40) and 3-hydroxy-3-methylglutaryl coenzyme A reductase (36, 37) mRNA levels are elevated by cAMP-dependent agents; however, the mechanisms have not been investigated. The immediate early cAMP-dependent response described in this work does not depend on the cellular lipid level, showing that at least some cholesterogenic genes can become transactivated by other stimuli, also irrespectively of the cholesterol feedback regulation. When phosphorylated CREB binds to CRE, it regulates immediate early genes, such as c-fos and junB (41). Together with other immediate early genes, the expression of ICER (42, 43) mRNA is up-regulated after 1.5 h of forskolin treatment, reaching maximal levels after 3.5 h. This rise in expression of ICER correlates well with the decrease in c-fos and junB levels (44) as well as with time and concentration dependency of the repression of CRE-mediated luciferase gene expression after the treatment of CHO cells with forskolin (45). Our results show that the CYP51 gene responds to activation by the cAMP-signaling pathway in the same immediate-early time-dependent manner. The highest level of CYP51 mRNA is detected 1.5–2 h after addition of forskolin to JEG-3 cells grown in lipid-rich medium with COPUFA (Fig. 2A) as well as cells grown in normal medium (data not shown) and returns to basal level after 4 h. The early response does not require protein synthesis in contrast to the delayed cAMP response (46, 47). In our case, the addition of protein synthesis inhibitor CHX did not prevent the induction of CYP51 transcription, suggesting that CYP51 transactivation does not depend on de novo protein synthesis. However, the attenuation of CYP51 transcription likely requires de novo synthesis of ICER. It seems that the declination of CYP51 transcription is a result of the synthesis of ICER, which binds to CRE in the promoters of cAMP-responsive genes and represses their transcription (4, 48). ICER is transcribed from a cAMP-dependent intronic promoter of the CREM gene. It binds identical CRE as the activator CREM and CREB isoforms. However, ICER does not contain a transactivation domain. It displaces the activator transcription factors from CRE in the cAMP-dependent promoters and thus acts as a transcriptional repressor (4). As shown in Fig. 3A, ICER synthesis was induced in a time-dependent manner also in our experiments, in accordance with the attenuated CYP51 mRNA level (Fig. 2A). Overexpression of ICER presumably repressed the CREB-mediated transcription of CYP51 D7 luciferase reporter construct (Fig. 3B), suggesting competitive displacement of CREB by ICER. Even if mammalian CYP51 promoters contain three potential CREs, it seems that only CYP51-CRE2 is responsible for the immediate early response of CYP51, including the cAMP-mediated transactivation as well as repression. This conclusion arises from the following observations: 1) CRE2 is evolutionarily the most conserved CRE-like element in the four mammalian species; 2) mutation of CRE2 abolished the cAMP-dependent response of the human CYP51 gene; 3) the longest promoter-reporter construct, containing CRE1, CRE2, and CRE3, has a weaker cAMP-dependent response compared with the construct containing CRE2 and CRE3 only; and 4) the shortest promoter-reporter construct containing only CRE3 does not respond to cAMP-dependent transactivation. Interestingly, the shortest –121/+316 D7 construct, which contains the downstream promoter element and the initiator element of the four studied mammals (49), shows less than 10% of the basal activity compared with the –334/+316 promoter. This suggests that in the case of mammalian TATA-box lacking CYP51 promoters, the downstream promoter element and initiator element seem not to serve as the RNA pol II complex binding sites.
These results suggest that CYP51 belongs to the group of immediate early response genes. Importantly, forskolin, by modulating the transcription of CYP51, modulates also the precholesterol sterol synthesis in JEG-3 cells ex vivo. Although the quantity of the de novo synthesized lanosterol seems to decrease, the quantity of postlanosterol intermediates of cholesterol biosynthesis did not change or even increased in conditions when cells have been treated with forskolin. This suggests an increased consumption of lanosterol to form postlanosterol intermediates. Reasons for a statistically nonsignificant increase in postlanosterol intermediates might be a result either of the low activity of enzymes of the postlanosterol phases of cholesterol biosynthesis or of an increased expenditure of the synthesized cholesterol. It is well known that forskolin increases the expenditure of cholesterol because genes that encode enzymes involved in production of steroid hormones are highly up-regulated by the cAMP-signaling pathway (24, 25). Identification of all possible steroid metabolites that are derived from cholesterol after the forskolin-mediated immediate early activation in JEG-3 cells is beyond the scope of this work.
Our data show for the first time that a cholesterogenic gene CYP51 can be activated through CYP51-CRE2 in an immediate early fashion after forskolin treatment, avoiding the cholesterol feedback repression. Either CREB or CREM can mediate this transactivation and ICER is responsible for attenuation of transcription. The forskolin stimulus leads to a short-term transactivation of cholesterogenic CYP51 that is reflected also at the level of de novo synthesis of cholesterogenic sterol intermediates.
In accordance with our data, others have indicated that the cholesterol feedback regulation might be bypassed in some physiological or pathophysiological conditions, such as tumor development (50, 51), regeneration of tissues after injury (52, 53, 54), response to cytokines (52, 53, 54, 55, 56), or maturation of male germ cells (12). The high level of cholesterol in the cell presumably protects cells against stress, whereas an increased cholesterol biosynthesis is one of the cellular responses to stress (57, 58, 59). However, the biochemical details that would allow understanding of these phenomena have been lacking so far. Our work shows that the CYP51 gene is transcriptionally activated in sterol-repressed conditions by either CREB or CREM. Although CRE2 and GC-box DNA elements are essential for cAMP-dependent transactivation, the SRE1 site seems to be less important. Transactivation of CYP51 through CRE2 represents the first clear evidence that the cholesterol feedback repression of cholesterogenic genes can be bypassed by transcriptional activation through another signaling pathway. Additionally, our work shows for the first time that a gene participating in cholesterol biosynthesis (CYP51) is regulated in an immediate early fashion, which influences the sterol profile of the cell. This opens new venues of research to better understand cholesterol-independent roles and cross-talk of the cholesterol feedback regulation with other signaling pathways.
Acknowledgments
We thank Dr. P. Sassone-Corsi (Centre National de la Recherche Scientifique, Strasbourg, France), for providing the pSV CREM, pSV ICER expression vectors and anti-CREM antibodies. Thanks also to Dr. M. R. Waterman (Vanderbilt University, Nashville, TN) for providing pRSV CREB and pSV Sp1 expression vectors. Sincere thanks to Dr. S. Svetina and Dr. U. Batista (Institute of Biophysics, Faculty of Medicine, University of Ljubljana) for sharing the cell culture laboratory and to Dr. V. urin erbec, dr. V. Galvani, and Dr. R. Rupreht (Blood Transfusion Center of Slovenia, Ljubljana, Slovenia) for sharing the 7900 HT sequence detection system.
Footnotes
This work was supported by the Ministry of Science of Slovenia, Grants J1-1380, SLO-USA 0002, and P1–0527, by the Fogarty International Research Collaboration Award/National Institutes of Health Grant 1 RO3 TW01174-01, and by International Centre for Genetic Engineering and Biotechnology Grant CRP/SLO0001. T.R. was supported by a fellowship from the Ministry of Science of Slovenia.
First Published Online August 25, 2005
Abbreviations: CAT, Chloramphenicol acetyl transferase; CHX, cycloheximide; COPUFA, lipid-rich medium with cholesterol, 25-hydroxycholesterol and linoleic acid; CRE, cAMP-response element; CREB, CRE-binding protein; CREM, CRE modulator; CYP, cytochrome P450; CYP51, lanosterol 14-demethylase; ICER, inducible cAMP early repressor; IEG, immediate early response gene; Sp1; specificity protein 1; SCAP, sterol cleavage activating protein; SRE, sterol regulatory element; SREBP, SRE-binding protein.
Accepted for publication August 19, 2005.
References
Fimia GM, Sassone-Corsi P 2001 Cyclic AMP signalling. J Cell Sci 114:1971–1972
Sng JC, Taniura H, Yoneda Y 2004 A tale of early response genes. Biol Pharm Bull 27:606–612
Herdegen T, Leah JD 1998 Inducible and constitutive transcription factors in the mammalian nervous system: control of gene expression by Jun, Fos and Krox, and CREB/ATF proteins. Brain Res Rev 28:370–490
Molina CA, Foulces NS, Lalli E, Sassone-Corsi P 1993 Inducibility and negative autoregulation of CREM: an alternative promoter directs the expression of ICER, an early repressor. Cell 75:875–886
Brown MS, Goldstein JL 1997 The SREBP pathway: regulation of cholesterol metabolism by proteolysis of a membrane-bound transcription factor. Cell 89:331–340
Goldstein JL, Brown MS 1990 Regulation of the mevalonate pathway. Nature 343:425–430
Horton JD, Goldstein JL, Brown MS 2002 SREBPs: activators of the complete program of cholesterol and fatty acid synthesis in the liver. J Clin Invest 109:1125–1131
Sakakura Y, Shimano H, Sone H, Takahashi A, Inoue K, Toyoshima H, Suzuki S, Yamada N 2001 Sterol regulatory element-binding proteins induce an entire pathway of cholesterol synthesis. Biochem Biophys Res Commun 286:176–183
Espenshade PJ, Li WP, Yabe D 2002 Sterols block binding of COPII proteins to SCAP, thereby controlling SCAP sorting in ER. Proc Natl Acad Sci USA 99:11694–11699
Loewen CJ, Levine TP 2002 Cholesterol homeostasis: not until the SCAP lady INSIGs. Curr Biol 12:R779–R781
Wang H, Liu F, Millette CF, Kilpatrick DL 2002 Expression of a novel, sterol-insensitive form of sterol regulatory element binding protein 2 (SREBP2) in male germ cells suggests important cell- and stage-specific functions for SREBP targets during spermatogenesis. Mol Cell Biol 22:8478–8490
Rozman D, Fink M, Fimia GM, Sassone-Corsi P, Waterman MR 1999 Cyclic adenosine 3',5'-monophosphate (cAMP)/cAMP-responsive element modulator (CREM)-dependent regulation of cholesterogenic lanosterol 14-demethylase (CYP51) in spermatids. Mol Endocrinol 13:1951–1962
Halder S, Fink M, Waterman MR, Rozman D 2002 A cAMP responsive element binding site is essential for sterol regulation of the human lanosterol 14-demethylase gene (CYP51). Mol Endocrinol 16:1853–1863
Aoyama Y, Noshiro M, Gotoh O, Imaoka S, Funae Y, Kurosawa N, Horiuchi T, Yoshida Y 1996 Sterol 14-demethylase P450 (P45014DM) is one of the most ancient and conserved P450 species. J Biochem 119:926–933
Bellamine A, Mangla AT, Nes WD, Waterman MR 1999 Characterization and catalytic properties of the sterol 14-demethylase from Mycobacterium tuberculosis. Proc Natl Acad Sci USA 96:8937–8942
Rezen T, Debeljak N, Kordis D, Rozman D 2004 New aspects on lanosterol 14-demethylase and cytochrome P450 evolution: lanosterol/cycloartenol diversification and lateral transfer. J Mol Evol 59:51–58
Byskov AG, Andersen CY, Nordholm L, Thogersen H, Guoliang X, Wassman O, Guddal JVAE, Roed T 1995 Chemical structure of sterols that activate oocyte meiosis. Nature 374:559–562
Seamon KB, Padgett W, Daly JW 1981 Forskolin: unique diterpene activator of adenylate cyclase in membranes and in intact cells. Proc Natl Acad Sci USA 78:3363–3367
Rozman D, Strmstedt, M., Tsui LC, Scherer SW, Waterman MR 1996 Structure and mapping of the human lanosterol 14-demethylase gene (CYP51) encoding the cytochrome P450 involved in cholesterol biosynthesis: comparison of exon/intron organization with other mammalian and fungal CYP genes. Genomics 38:371–381
Sambrook J, Fritsch EF, Maniatis TM 1989 Molecular cloning: a laboratory manual. New York: Cold Spring Harbor Laboratory Press
Delmas V, van der Horn F, Mellstrm B, Jegou B, Sassone-Corsi P 1993 Induction of CREM activator proteins in spermatids: downstream targets and implications for haploid germ cell differentiation. Mol Endocrinol 7:1502–1514
Fon Tacer K, Haugen TB, Baltsen M, Debeljak N, Rozman D 2002 Tissue-specific transcriptional regulation of the cholesterol biosynthetic pathway leads to accumulation of testis meiosis-activating sterol (T-MAS). J Lipid Res 43:82–89
Worgall TS, Sturley SL, Seo T, Osborne TF, Deckelbaum RJ 1998 Polyunsaturated fatty acids decrease expression of promoters with sterol regulatory elements by decreasing levels of mature sterol regulatory element-binding protein. J Biol Chem 273:25537–25540
Christenson LK, Strauss 3rd JF 2001 Steroidogenic acute regulatory protein: an update on its regulation and mechanism of action. Arch Med Res 32:576–586
Omura T, Morohashi K 1995 Gene regulation of steroidogenesis. J Steroid Biochem Mol Biol 53:19–25
Zhan HC, Liu DP, Liang CC 2001 Insulator: from chromatin domain boundary to gene regulation. Hum Genet 109:471–478
Bell AC, Felsenfeld G 1999 Stopped at the border: boundaries and insulators. Curr Opin Genet Dev 9:191–198
Ness GC, Gertz KR, Holland RC 2001 Regulation of hepatic lanosterol 14-demethylase gene expression by dietary cholesterol and cholesterol-lowering agents. Arch Biochem Biophys 395:233–238
Rodriguez C, Martinez-Gonzalez J, Sanchez-Gomez S, Badimon L 2001 LDL downregulates CYP51 in porcine vascular endothelial cells and in the arterial wall through a sterol regulatory element binding protein-2-dependent mechanism. Circ Res 88:268–274
Sonoda Y, Amano C, Endo M, Sato Y, Sekigawa Y, Fukuhara M 1995 Lanosterol 14-demethylase (cytochrome P-45014DM): modulation of its enzyme activity by cholesterol feeding. Biol Pharm Bull 18:1009–1011
Trzaskos JM, Bowen WD, Shafiee A, Fischer RT, Gaylor JL 1984 Cytochrome P-450-dependent oxidation of lanosterol in cholesterol biosynthesis. J Biol Chem 259:13402–13412
Thewke DP, Panini SR, Sinensky M 1998 Oleate potentiates oxysterol inhibition of transcription from sterol regulatory element-1-regulated promoters and maturation of sterol regulatory element-binding proteins. J Biol Chem 273:21402–21407
Hannah VC, Ou J, Luong A, Goldstein JL, Brown MS 2001 Unsaturated fatty acids down-regulate SREBP isoforms 1a and 1c by two mechanisms in HEK-293 cells. J Biol Chem 276:4365–4372
Ellsworth JL, Lloyd DB, Carlstrom AJ, Thompson JF 1995 Protein binding to the low density lipoprotein receptor promoter in vivo is differentially affected by gene activation in primary human cells. J Lipid Res 36:383–392
Mascaro C, Nadal A, Hegardt FG, Marrero PF, Haro D 2000 Contribution of steroidogenic factor 1 to the regulation of cholesterol synthesis. Biochem J 350:785–790
Takagi K, Strauss 3rd JF 1989 Control of low density lipoprotein receptor gene expression in steroidogenic cells. Can J Physiol Pharmacol 67:968–973
Golos TG, Strauss 3rd JF 1988 8-Bromoadenosine cyclic 3',5'-phosphate rapidly increases 3-hydroxy-3-methylglutaryl coenzyme A reductase mRNA in human granulosa cells: role of cellular sterol balance in controlling the response to tropic stimulation. Biochemistry 27:3503–3606
Wang X, Sato R, Brown MS, Hua X, Goldstein JL 1994 SREBP-1, a membrane-bound transcription factor released by sterol-regulated proteolysis. Cell 77:53–62
Xu J, Teran-Garcia M, Park JH, Nakamura MT, Clarke SD 2001 Polyunsaturated fatty acids suppress hepatic sterol regulatory element-binding protein-1 expression by accelerating transcript decay. J Biol Chem 276:9800–9807
Golos TG, August AM, Strauss 3rd JF 1986 Expression of low density lipoprotein receptor in cultured human granulosa cells: regulation by human chorionic gonadotropin, cyclic AMP, and sterol. J Lipid Res 27:1089–1096
Ma W, Hatzis C, Eisenach JC 2003 Intrathecal injection of cAMP response element binding protein (CREB) antisense oligonucleotide attenuates tactile allodynia caused by partial sciatic nerve ligation. Brain Res 988:97–104
Sassone-Corsi P 1998 Molecular clocks: mastering time by gene regulation. Nature 392:871–874
Sassone-Corsi P 1998 Coupling gene expression to cAMP signalling: role of CREB and CREM. Int J Biochem Cell Biol 30:27–38
Mao D, Warner EA, Gurwitch SA, Dowd DR 1998 Differential regulation and transcriptional control of immediate early gene expression in forskolin-treated WEHI7.2 thymoma cells. Mol Endocrinol 12:492–503
Kemp DM, George SE, Kent TC, Bungay PJ, Naylor LH 2002 The effect of ICER on screening methods involving CRE-mediated reporter gene expression. J Biomol Screen 7:141–148
Simpson ER, Waterman MR 1988 Regulation of the synthesis of steroidogenic enzymes in adrenal cortical cells by ACTH. Annu Rev Physiol 50:427–440
Waterman MR, Simpson ER 1989 Regulation of steroid hydroxylase gene expression is multifactorial in nature. Recent Prog Horm Res 45:533–563
Thommesen L, Norsett K, Sandvik AK, Hofsli E, Laegreid A 2000 Regulation of inducible cAMP early repressor expression by gastrin and cholecystokinin in the pancreatic cell line AR42J. J Biol Chem 275:4244–4250
Debeljak N, Fink M, Rozman D 2003 Many facets of mammalian lanosterol 14-demethylase from the evolutionarily conserved cytochrome P450 family CYP51. Arch Biochem Biophys 409:159–171
Rao KN 1995 The significance of the cholesterol biosynthetic pathway in cell growth and carcinogenesis (review). Anticancer Res 15:309–314
Hentosh P, Yuh SH, Elson CE, Peffley DM 2001 Sterol-independent regulation of 3-hydroxy-3-methylglutaryl coenzyme A reductase in tumor cells. Mol Carcinog 32:154–166
Bocchetta M, Bruscalupi G, Castellano F, Trentalance A, Komaromy M, Fong LG, Cooper AD 1993 Early induction of LDL receptor gene during rat liver regeneration. J Cell Physiol 156:601–609
Feingold KR, Pollock AS, Moser AH, Shigenaga JK, Grunfeld C 1995 Discordant regulation of proteins of cholesterol metabolism during the acute phase response. J Lipid Res 36:1474–1482
Zager RA, Shah VO, Shah HV, Zager PG, Johnson AC, Hanson S 2002 The mevalonate pathway during acute tubular injury: selected determinants and consequences. Am J Pathol 161:681–692
Ruan XZ, Varghese Z, Fernando R, Moorhead JF 1998 Cytokine regulation of low-density lipoprotein receptor gene transcription in human mesangial cells. Nephrol Dial Transplant 13:1391–1397
Ruan XZ, Varghese Z, Powis SH, Moorhead JF 2001 Dysregulation of LDL receptor under the influence of inflammatory cytokines: a new pathway for foam cell formation. Kidney Int 60:1716–1725
Cutler RG, Kelly J, Storie K, Pedersen WA, Tammara A, Hatanpaa K, Troncoso JC, Mattson MP 2004 Involvement of oxidative stress-induced abnormalities in ceramide and cholesterol metabolism in brain aging and Alzheimer’s disease. Proc Natl Acad Sci USA 101:2070–2075
Gesquiere L, Loreau N, Minnich A, Davignon J, Blache D 1999 Oxidative stress leads to cholesterol accumulation in vascular smooth muscle cells. Free Radic Biol Med 27:134–145
Shack S, Gorospe M, Fawcett TW, Hudgins WR, Holbrook NJ 1999 Activation of the cholesterol pathway and Ras maturation in response to stress. Oncogene 18:6021–6028(Martina Fink, Jure Aimovi, Tadeja Reen, )