- Email: [email protected]
DNA Sequences Essential for Transcription of Human Phospholipid Transfer Protein Gene in HepG2 Cells
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS ARTICLE NO.
232, 574–577 (1997)
RC976330
DNA Sequences Essential for Transcription of Human Phospholipid Transfer Protein Gene in HepG2 Cells An-Yue Tu, Gertrud Wolfbauer, Hongbo Chen, and John J. Albers1 Northwest Lipid Research Laboratories, Department of Medicine, University of Washington, 2121 N. 35th Street, Seattle, Washington 98103
Received January 28, 1997
The present study was conducted to determine the essential DNA sequences required for the transcription of the human phospholipid transfer protein gene. Truncation studies revealed that DNA sequences between 0230 and 0159, particularly those at the upstream region, were responsible for the full promoter activity. This region was able to compete with AP-2 and GRE oligonucloetides for the binding to HepG2 cell nuclear extract as shown by gel mobility shift assay. Further analysis, using site-directed mutagenesis, indicated that DNA sequences identical to Sp1 and highly homologous to GRE and Ap-2 consensus sequences were essential for the transcription. These findings support the concept that several elements, spread over the entire functional promoter, synergistically drive the basal transcription. q 1997 Academic Press
One of the important functions of plasma phospholipid transfer protein (PLTP) is the transfer of phospholipids among circulating lipoproteins (1-3). The removal of excess phospholipids during lipolysis and postprandial lipemia is an initial and essential process for the catabolism of triglyceride-rich lipoproteins (4). Delayed clearance of these lipoproteins is believed to be related to the development of atherosclerosis (5); therefore, PLTP may play a key role in the prevention of the development of premature cardiovascular disease. Recent in vitro studies suggest that PLTP has multiple functional roles. For example, PLTP has been shown to modulate high density lipoprotein (HDL) size and composition (6-8), facilitate generation of pre-beta HDL (9), catalyze the exchange of alpha-tocopherol between 1
To whom correspondence should be addressed. Fax: (206) 6853279. E-mail: [email protected]. Abbreviations: CETP, cholesteryl ester transfer protein; FBS, fetal bovine serum; GRE, glucocorticoid responsive element; GR, glucocorticoid receptor; HDL, high density lipoprotein; HepG2, human hepatoma G2; LBP, lipopolysaccharide-binding protein; PLTP, phospholipid transfer protein. 0006-291X/97 $25.00
lipoprotein particles (10), and neutralize and transfer lipopolysaccharide (11). Evidence supports the concept that pre-beta HDL plays an essential role in cholesterol efflux from cell membranes and serves as the earliest acceptor of cellular cholesterol in reverse cholesterol transport (12). The conversion of HDL3 to larger HDL2like particles, mediated by PLTP, and the regeneration of initial acceptors of cellular cholesterol is likely not only to promote the efflux of cholesterol from peripheral cells and target cholesterol to the liver, but also retard atherogenesis and relate to the protective effect of HDL. The cDNA of human PLTP has been cloned (13), and its gene structure and functional promoter have been characterized (14, 15). However, the regulation of this gene at the transcriptional level has not been studied, because of the lack of information about basal transcription. The present study has identified the essential sequences responsible for the transcription of the human PLTP gene and has indicated several potential binding sites for the transcription factors glucocorticoid receptor (GR), AP-2, and Sp1. This new information should facilitate future studies on the physiological regulation of human PLTP and its impact on lipid transfer and lipoprotein metabolism. MATERIAL AND METHODS Construction of truncated and mutated PLTP promoter/luciferase fusion plasmids. The PLTP functional promoter between 0230 and /15 relative to the transcription starting site was previously cloned into the pGL2-Basic vector (Promega Corp., Madison, WI) to create the fused PLTP promoter/luciferase plasmid for promoter activity studies (15). By taking advantage of this construct, a series of sequential deletions were carried out to generate plasmids consisting of various lengths of the PLTP 5*-flanking regions and the luciferase reporter gene. Erase-a-Base kit of Promega Corp. (Madison, WI) was used according to the protocol of the manufacturer for sequential deletion. All the constructs possessed a fixed 3 * end at position / 15. These plasmids were then transformed into JM 109 competent cells and subjected to white/blue selection on LB plates containing 100 mg/ml ampicillin, 0.5 mM IPTG and 40 mg/ml X-Gal. Maps of restriction enzyme sites and DNA sequences, obtained by chain-ter-
574
Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.
AID
BBRC 6330
/
6922$$1341
02-27-97 11:03:22
bbrcg
AP: BBRC
Vol. 232, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
mination method (16) using a Sequenase kit (US Biochemical Corp., Cleveland, OH), were employed to verify the selected clones. To construct plasmids carrying various mutation sites, oligonucleotide-directed mutagenesis was performed using Muta-Gene phagemid mutagenesis kit (Rio-Rad Laboratories, Hercules, CA). The kit is based on a method previously described by Kunkel et al. (17), which provides a strong selection against the non-mutated strand of a double stranded DNA. The PLTP functional promoter (0230//15) was first cloned into pTZ19 vector and mutagenesis was performed according to the protocol of the manufacturer. The generated mutant DNA sequences were verified by sequencing (16). These PLTP promoter mutants were then cloned back to pGL2-Basic vector for further analysis of the promoter activity in human hepatoma (HepG2) cells. Cell culture, DNA transfection, and luciferase assay. HepG2 cells were maintained in Dulbeccco’s modified Eagle medium supplemented with 10% fetal bovine serum (FBS) and passaged at confluency 1-2 times per week. One day prior to transfections cells were seeded at a density of 1.21105/35mm cell culture dish. Transient transfections were initiated by exposing the cells to plasmid/Lipofectamine (BRL Life Technologies, Inc., Grand Island, NY) complexes, containing 1 mg of DNA and 4 ml lipid emulsion per dish, following the manufacturers suggested procedure. After a 6 hour incubation at 377C growing medium containing 20% FBS was added to each dish, and cells were harvested at 24 hour post transfection. Luciferase activity in cell lysates was measured in a luciferase assay system (Promega Corp., Madison, WI) by following the manufacturers protocol, using a liquid scintillation counter (Packard Instr. Co., Downers Grove, IL) to measure chemiluminescense. In each experiment luciferase activities expressed for promoterless pGL2-Basic, and pGL2/ PLTP(0230//15) with full functional PLTP promoter were used as negative and positive controls, respectively. Gel mobility shift assay. Nuclear extract was prepared from cultured HepG2 cells following the procedures described by Dignam et al. (18). The concentration of the extracted proteins was measured by Lowry method (19). Gel mobility shift assays were conducted using the Gel Shift Assay System of Promega Corp. (Madison, WI): HepG2 nuclear extract (5-10 mg) was preincubated for 10 min at 47C in a reaction mixture (20 ml) containing 20 mM Tris, pH 8, 60 mM KCl, 0.2 mM EDTA, 0.5 mM DTT, 0.25 mM phenylmethylsulfonyl fluoride, 1.3 mM MgCl2 , 10% glycerol, 3% Ficoll, and 3 mg of double-stranded poly (dA-dT) or poly (dI-dC) in the presence or absence of 50-fold excess of competitor. Radiolabeled consensus oligonucleotide for transcription factor was added and incubated for 20 min at room temperature. The incubation mixture was loaded on a 4% nondenaturing polyacrylamide gel in TBE buffer and subjected to electrophoresis at 100 V for 1 hr. The gel was then fixed, dried, and visualized by autoradiography.
RESULTS AND DISCUSSION We have previously determined that the region responsible for the functional transcription of the human PLTP gene was located between 0230 and 072 relative to the first transcriptional initiation site (15). In the present study, we selected HepG2 cells because of the previous finding that they exhibited the highest luciferase activity among the tested cells (15), suggesting that they may contain all transcription factors necessary for the full function of the PLTP promoter. DNA sequences within the functional promoter (0230 to 072) were sequentially deleted and fused with the coding region of the luciferase gene in the promoterless plasmid pGL2-Basic. These new constructs con-
FIG. 1. Transcriptional activity of truncated PLTP promoter constructs. Left, diagram indicates the positions of the 5* and 3* ends of the wild-type (0230//15) and truncated constructs. Right, 1 mg of DNA of each construct was transfected into HepG2 cells and luciferase activity was measured. Activity expressed by promoterless pGL2Basic is included as a negative control. At least three independent experiments were carried out for each construct and data shown are the mean values. In all cases the standard deviation of the mean value was less than 10%.
tained PLTP promoter fragments with different 5* ends residing at 0230, 0205, 0180, 0159, 0136, 0102, and 072, respectively, and a common 3* end at position /15. Each of these recombinant DNAs was then used for transient gene expression studies by transfection into HepG2 cells. Luciferase activity was evaluated by measuring the production of chemiluminescense after adding the cell extracts to a luciferase substrate. The constructs with DNA sequences spanning from 0159, 0136, 0102, or 072 to /15 each exhibited relatively limited activities specifically, less than 20% of the full promoter activity as compared to the construct spanning from 0230 to /15 (Figure 1). However, approximately 25% of the full acivity was observed for the construct 0180//15, and 56% of the full activity was observed for the construct 0205//15. These findings suggest that the DNA sequences between 0230 and 0159, especially those at the upstream areas, are essential for the full promoter activity. Competitive gel mobility shift assays were conducted to evaluate the potential transcription factors responsible for the binding of the DNA fragment with full promoter activity. We used radiolabeled consensus oligonucleotides of several transcription factors for the binding to HepG2 cell nuclear extract and 50 times more cold DNA fragments spanning from 0230 to 0148, generated by polymerase chain reaction, to compete. This promoter region was able to compete with AP-2 and glucocorticoid responsive element (GRE) oligonucloetides for the binding to HepG2 cell extract (Figure 2). Although no sequences in this DNA region are completely identical to the consensus sequences of AP-2 and GRE, several potential sites with high sequence homology were identified. For example, as shown in Figure 3 Panel A, DNA sequences from 0228 to 0220 (designated as site A) and from 0206 to 0198 (site B)
575
AID
BBRC 6330
/
6922$$1341
02-27-97 11:03:22
bbrcg
AP: BBRC
Vol. 232, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. Competitive gel shift studies. The radiolabeled consensus oligonucleotides of AP-2 and GRE were incubated as indicated with HepG2 nuclear extract (5-10 mg) for binding in the absence (Line 2) or presence (Line 3) of 50-fold excess of unlabeled PLTP promoter fragment (0230 to 0148). The oligonucleotides incubated without nuclear extract were included as a control (Line 1). Autoradiography of 4% nondenaturing polyacrylamide gels was used to identify the DNA/protein complexes.
are similar to the binding motif of AP-2 (GCCNNNGGC); those from 0186 to 0181 (site C) and from 0165 to 0160 (site D) are similar to one half of the binding motif of GRE (TCTTGT). Additionally, the consensus sequences for transcription factors Sp1 at 0114 to 0108 (site E) and Ap-2 at 094 to 085 (site F) were previously identified in the gene promoters of both the human and mouse PLTP (15, 20). The sequences at sites E anf F are slightly downstream of the primary functional region (0230 to 0159); however, these sites may coordinate with other sequences for promoter function. We have mutated two to three base pairs of each of these proposed targeted sites, either identical to the consensus sequences of Sp1 or with high homology to those of Ap-2 and GRE, in order to determine whether they are essential for promoter activity. In Figure 3 Panel A, the mutation at either site C (mutant C) or D (mutant D) caused loss of promoter activity by 59 and 51%, respectivily. Mutant E lost promoter activity by 37%. Mutations at sites A, B, or F resulted in the loss of less than 28% of the activity. These findings indicate that no single region is fully responsible for the promoter function. To further verify that the transcription of the human PLTP gene may require several essential sequences, constructs with various combinations of mutations at two to three different sites were generated and their luciferase activities analyzed. All mutants consisted of additional mutation sites (sites A through F) to either mutants C or D, each of which was shown to be responsible for a significant portion of the promoter activity. Among the mutants assayed for promoter activity, the triple mutation at sites C, D, and E resulted in the maximal loss of promoter function, with less than 10%
of full activity as shown in Figure 3 Panel B. Three constructs, with mutations at any two sites of C, D, and E, also exhibited a substantial loss of promoter activity (67 to 73%). Our studies on multiple mutations indicate 46 to 79% loss of the activity for mutants with two mutation sites, and 73 to 91% loss of the activity for those with three mutation sites. These findings suggest that at least three distinct DNA sequences are crucial for the basal transcription of the human PLTP gene. In many genes the major regulatory elements are present within 1 kilobase of the 5*-upstream region of the coding sequence; however, for some genes these can be located at a greater distance, for example, the regulatory elements of apolipoprotein E gene have been found at least 2 kilobases 3* downstream to the structural gene (21). However, some elements could possibly reside inside this gene. For example, the first intron of apolipoprotein A-II gene was found to possess regulatory elements necessary for its transcription (22). Therefore, we can not rule out the possibility that DNA sequences other than those we have identified are also required for transcription. Little is known about the transcriptional regulation of the PLTP gene as compared to other members of the same gene family, such as cholesteryl ester transfer protein (CETP) and lipopolysaccharide-binding protein (LBP). In the present study, it is particularly noteworthy that the GRE motif could be one of the essential sequences for the PLTP transcription. The presence of GRE in promoters can serve as binding site for GR
FIG. 3. Transcriptional activity of mutated PLTP promoter constructs. The activities of mutants with a single mutation site (A) and with multiple mutation sites (B) are expressed as a percentage of the wild-type construct. Solid lines represent the promoter region from 0230 to 072 and boxes indicate the mutated sites. The positions of the 5* and 3* ends of the sites are indicated. The substituted nucleotides of the mutants are highlighted. One mg of DNA of each construct was transfected into HepG2 cells for luciferase activity assay. Data shown are the means of at least three independent experiments.
576
AID
BBRC 6330
/
6922$$1341
02-27-97 11:03:22
bbrcg
AP: BBRC
Vol. 232, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
to regulate the transcription of glucocorticoid-inducible genes involved in the inflammatory response. It has been found that GRE correlates with the induction of LBP mRNA in HepG2 cells by pro-inflammatory cytokines like IL-1 and IL-6 (23, 24). In addition to LBP, CETP is regulated by endotoxin in both the mouse and hamster model (25, 26). In the hamster model the combination of tumor necrosis factor and interleukin-1 was shown to mimic the effects of the endotoxin-induced decrease in the expression of CETP, suggesting that these cytokines may mediate, in part, the effects of endotoxin on CETP expression (26). PLTP also has been found to be regulated by lipopolysaccharide in a mouse model (27); however, whether the response is on the transcriptional level remains to be investigated. The overall analysis of our mutagenesis data indicates that of seveal DNA sequences examined, three appear to be essential for the basal transcription of the human PLTP gene. Furthermore, the transcriptional process likely is driven by the coordination of several cis-acting elements and trans-acting transcription factors. This information should facilitate elucidating the regulatory mechanisms and the physiological modulators of PLTP expression at the transcriptional level. ACKNOWLEDGMENTS We express our appreciation to Hal Kennedy for data analysis and illustration preparation. This work was supported by Program Project Grant HL 30086 from the National Institutes of Health to John J. Albers.
REFERENCES 1. Tollefson, J. H., Ravnik, S., and Albers, J. J. (1988) J. Lipid Res. 29, 1593–1602. 2. Tall, A. R. (1993) J. Lipid Res. 34, 1255–1274. 3. Cheung, M. C., Wolfbauer, G., and Albers, J. J. (1996) Biochim. Biophys. Acta 1303, 103–110. 4. Tall, A. R. (1986) Methods Enzymol. 129, 469–483. 5. Zilversmit, D. B. (1979) Circulation 60, 473–485. 6. Tu, A-Y., Nishida, H. I., and Nishida, T. (1993) J. Biol. Chem. 268, 23098–23105.
7. Jauhiainen, M., Metso, J., Pahlman, R., Blomqvist, S., van Tol, A., and Ehnholm, C. (1993) J. Biol. Chem. 268, 4032–4036. 8. Albers, J. J., Wolfbauer, G., Cheung, M. C., Day, J. R., Ching, A. F. T., Lok, S., and Tu, A-Y. (1995) Biochim. Biophys. Acta 1258, 27–34. 9. von Eckardstein, A., Jauhiainen, M., Huang, Y., Metso, J., Langer, C., Pussinen, P., Wu, S., Ehnholm, C., and Assmann, G. (1996) Biochim. Biophys. Acta 1301, 255–262. 10. Kostner, G. M., Oettl, K., Jauhiainen, M., Ehnholm, C., Esterbauer, H., and Dieplinger, H. (1995) Biochem. J. 305, 659– 667. 11. Hailman, E., Albers, J. J., Wolfbauer, G., Tu, A-Y., and Wright, S. D. (1996) J. Biol. Chem. 271, 12172–12178. 12. Fielding, C. J., and Fielding, P. E. (1995) J. Lipid Res. 36, 211– 228. 13. Day, J. R., Albers, J. J., Lofton-Day, C. E., Gilbert, T. L., Ching, A. F. T., Grant, F. J., O’Hara, P. J., Marcovina, S. M., and Adolphson, J. L. (1994) J. Biol. Chem. 269, 9388–9391. 14. Tu, A-Y., Deeb, S. S., Iwasaki, L., Day, J. R., and Albers, J. J. (1995) Biochem. Biophys. Res. Commun. 207, 552–558. 15. Tu, A-Y., Wolfbauer, G., and Albers, J. J. (1995) Biochem. Biophys. Res. Commun. 217, 705–711. 16. Sanger, F., Nicklen, S., and Coulson, A. R. (1977) Proc. Natl. Acad. Sci. (USA) 74, 5463–5467. 17. Kunkel, T. A., Roberts, J. D., and Zakour, R. A. (1987) Methods in Enzymol. 154, 367–382. 18. Dignam, J. D., Lebovitz, R. M., and Roeder, R. G. (1983) Nucleic Acids Res. 11, 1475–1489. 19. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193, 265–275. 20. Tu, A-Y., Chen, H., Johnson, K. A., Paigen, B., and Albers, J. J. (1997) Gene, in press. 21. Lauer, S. J., Walker, D., Elshourbagy, N. A., Reardon, C. A., Levy-Wilson, B., and Taylor, J. M. (1988) J. Biol. Chem. 263, 7277–7286. 22. Bossu, J.-P., Chartier, F. L., Fruchart, J.-C., Auwerx, J., Staels, B., and Laine, B. (1996) Biochem. J. 318, 547–553. 23. Schumann, R. R. (1995) Prog. Clin. Biol. Res. 392, 297–304. 24. Schumann, R. R., Kirschning, C. J., Unbehaun, A., Aberle, H., Knopf, H-P., Lamping, N., Ulevitch, R. J., and Herrmann, F. (1996) Mol. Cell. Biol. 16, 3490–3503. 25. Masucci-Magoulas, L., Moulin, P., Jiang, X. C., Richardson, H., Walsh, A., Breslow, J. L., and Tall, A. R. (1995) J. Clin. Invest. 95, 1587–1594. 26. Hardard’ottir, I., Moser, A. H., Fuller, J., Fielding, C., Feingold, K., and Grunfeld, C. (1996) J. Clin. Invest. 97, 2585–2592. 27. Jiang, X. C., and Bruce, C. (1995) J. Biol. Chem. 270, 17133– 17138.
577
AID
BBRC 6330
/
6922$$1341
02-27-97 11:03:22
bbrcg
AP: BBRC
232, 574–577 (1997)
RC976330
DNA Sequences Essential for Transcription of Human Phospholipid Transfer Protein Gene in HepG2 Cells An-Yue Tu, Gertrud Wolfbauer, Hongbo Chen, and John J. Albers1 Northwest Lipid Research Laboratories, Department of Medicine, University of Washington, 2121 N. 35th Street, Seattle, Washington 98103
Received January 28, 1997
The present study was conducted to determine the essential DNA sequences required for the transcription of the human phospholipid transfer protein gene. Truncation studies revealed that DNA sequences between 0230 and 0159, particularly those at the upstream region, were responsible for the full promoter activity. This region was able to compete with AP-2 and GRE oligonucloetides for the binding to HepG2 cell nuclear extract as shown by gel mobility shift assay. Further analysis, using site-directed mutagenesis, indicated that DNA sequences identical to Sp1 and highly homologous to GRE and Ap-2 consensus sequences were essential for the transcription. These findings support the concept that several elements, spread over the entire functional promoter, synergistically drive the basal transcription. q 1997 Academic Press
One of the important functions of plasma phospholipid transfer protein (PLTP) is the transfer of phospholipids among circulating lipoproteins (1-3). The removal of excess phospholipids during lipolysis and postprandial lipemia is an initial and essential process for the catabolism of triglyceride-rich lipoproteins (4). Delayed clearance of these lipoproteins is believed to be related to the development of atherosclerosis (5); therefore, PLTP may play a key role in the prevention of the development of premature cardiovascular disease. Recent in vitro studies suggest that PLTP has multiple functional roles. For example, PLTP has been shown to modulate high density lipoprotein (HDL) size and composition (6-8), facilitate generation of pre-beta HDL (9), catalyze the exchange of alpha-tocopherol between 1
To whom correspondence should be addressed. Fax: (206) 6853279. E-mail: [email protected]. Abbreviations: CETP, cholesteryl ester transfer protein; FBS, fetal bovine serum; GRE, glucocorticoid responsive element; GR, glucocorticoid receptor; HDL, high density lipoprotein; HepG2, human hepatoma G2; LBP, lipopolysaccharide-binding protein; PLTP, phospholipid transfer protein. 0006-291X/97 $25.00
lipoprotein particles (10), and neutralize and transfer lipopolysaccharide (11). Evidence supports the concept that pre-beta HDL plays an essential role in cholesterol efflux from cell membranes and serves as the earliest acceptor of cellular cholesterol in reverse cholesterol transport (12). The conversion of HDL3 to larger HDL2like particles, mediated by PLTP, and the regeneration of initial acceptors of cellular cholesterol is likely not only to promote the efflux of cholesterol from peripheral cells and target cholesterol to the liver, but also retard atherogenesis and relate to the protective effect of HDL. The cDNA of human PLTP has been cloned (13), and its gene structure and functional promoter have been characterized (14, 15). However, the regulation of this gene at the transcriptional level has not been studied, because of the lack of information about basal transcription. The present study has identified the essential sequences responsible for the transcription of the human PLTP gene and has indicated several potential binding sites for the transcription factors glucocorticoid receptor (GR), AP-2, and Sp1. This new information should facilitate future studies on the physiological regulation of human PLTP and its impact on lipid transfer and lipoprotein metabolism. MATERIAL AND METHODS Construction of truncated and mutated PLTP promoter/luciferase fusion plasmids. The PLTP functional promoter between 0230 and /15 relative to the transcription starting site was previously cloned into the pGL2-Basic vector (Promega Corp., Madison, WI) to create the fused PLTP promoter/luciferase plasmid for promoter activity studies (15). By taking advantage of this construct, a series of sequential deletions were carried out to generate plasmids consisting of various lengths of the PLTP 5*-flanking regions and the luciferase reporter gene. Erase-a-Base kit of Promega Corp. (Madison, WI) was used according to the protocol of the manufacturer for sequential deletion. All the constructs possessed a fixed 3 * end at position / 15. These plasmids were then transformed into JM 109 competent cells and subjected to white/blue selection on LB plates containing 100 mg/ml ampicillin, 0.5 mM IPTG and 40 mg/ml X-Gal. Maps of restriction enzyme sites and DNA sequences, obtained by chain-ter-
574
Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.
AID
BBRC 6330
/
6922$$1341
02-27-97 11:03:22
bbrcg
AP: BBRC
Vol. 232, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
mination method (16) using a Sequenase kit (US Biochemical Corp., Cleveland, OH), were employed to verify the selected clones. To construct plasmids carrying various mutation sites, oligonucleotide-directed mutagenesis was performed using Muta-Gene phagemid mutagenesis kit (Rio-Rad Laboratories, Hercules, CA). The kit is based on a method previously described by Kunkel et al. (17), which provides a strong selection against the non-mutated strand of a double stranded DNA. The PLTP functional promoter (0230//15) was first cloned into pTZ19 vector and mutagenesis was performed according to the protocol of the manufacturer. The generated mutant DNA sequences were verified by sequencing (16). These PLTP promoter mutants were then cloned back to pGL2-Basic vector for further analysis of the promoter activity in human hepatoma (HepG2) cells. Cell culture, DNA transfection, and luciferase assay. HepG2 cells were maintained in Dulbeccco’s modified Eagle medium supplemented with 10% fetal bovine serum (FBS) and passaged at confluency 1-2 times per week. One day prior to transfections cells were seeded at a density of 1.21105/35mm cell culture dish. Transient transfections were initiated by exposing the cells to plasmid/Lipofectamine (BRL Life Technologies, Inc., Grand Island, NY) complexes, containing 1 mg of DNA and 4 ml lipid emulsion per dish, following the manufacturers suggested procedure. After a 6 hour incubation at 377C growing medium containing 20% FBS was added to each dish, and cells were harvested at 24 hour post transfection. Luciferase activity in cell lysates was measured in a luciferase assay system (Promega Corp., Madison, WI) by following the manufacturers protocol, using a liquid scintillation counter (Packard Instr. Co., Downers Grove, IL) to measure chemiluminescense. In each experiment luciferase activities expressed for promoterless pGL2-Basic, and pGL2/ PLTP(0230//15) with full functional PLTP promoter were used as negative and positive controls, respectively. Gel mobility shift assay. Nuclear extract was prepared from cultured HepG2 cells following the procedures described by Dignam et al. (18). The concentration of the extracted proteins was measured by Lowry method (19). Gel mobility shift assays were conducted using the Gel Shift Assay System of Promega Corp. (Madison, WI): HepG2 nuclear extract (5-10 mg) was preincubated for 10 min at 47C in a reaction mixture (20 ml) containing 20 mM Tris, pH 8, 60 mM KCl, 0.2 mM EDTA, 0.5 mM DTT, 0.25 mM phenylmethylsulfonyl fluoride, 1.3 mM MgCl2 , 10% glycerol, 3% Ficoll, and 3 mg of double-stranded poly (dA-dT) or poly (dI-dC) in the presence or absence of 50-fold excess of competitor. Radiolabeled consensus oligonucleotide for transcription factor was added and incubated for 20 min at room temperature. The incubation mixture was loaded on a 4% nondenaturing polyacrylamide gel in TBE buffer and subjected to electrophoresis at 100 V for 1 hr. The gel was then fixed, dried, and visualized by autoradiography.
RESULTS AND DISCUSSION We have previously determined that the region responsible for the functional transcription of the human PLTP gene was located between 0230 and 072 relative to the first transcriptional initiation site (15). In the present study, we selected HepG2 cells because of the previous finding that they exhibited the highest luciferase activity among the tested cells (15), suggesting that they may contain all transcription factors necessary for the full function of the PLTP promoter. DNA sequences within the functional promoter (0230 to 072) were sequentially deleted and fused with the coding region of the luciferase gene in the promoterless plasmid pGL2-Basic. These new constructs con-
FIG. 1. Transcriptional activity of truncated PLTP promoter constructs. Left, diagram indicates the positions of the 5* and 3* ends of the wild-type (0230//15) and truncated constructs. Right, 1 mg of DNA of each construct was transfected into HepG2 cells and luciferase activity was measured. Activity expressed by promoterless pGL2Basic is included as a negative control. At least three independent experiments were carried out for each construct and data shown are the mean values. In all cases the standard deviation of the mean value was less than 10%.
tained PLTP promoter fragments with different 5* ends residing at 0230, 0205, 0180, 0159, 0136, 0102, and 072, respectively, and a common 3* end at position /15. Each of these recombinant DNAs was then used for transient gene expression studies by transfection into HepG2 cells. Luciferase activity was evaluated by measuring the production of chemiluminescense after adding the cell extracts to a luciferase substrate. The constructs with DNA sequences spanning from 0159, 0136, 0102, or 072 to /15 each exhibited relatively limited activities specifically, less than 20% of the full promoter activity as compared to the construct spanning from 0230 to /15 (Figure 1). However, approximately 25% of the full acivity was observed for the construct 0180//15, and 56% of the full activity was observed for the construct 0205//15. These findings suggest that the DNA sequences between 0230 and 0159, especially those at the upstream areas, are essential for the full promoter activity. Competitive gel mobility shift assays were conducted to evaluate the potential transcription factors responsible for the binding of the DNA fragment with full promoter activity. We used radiolabeled consensus oligonucleotides of several transcription factors for the binding to HepG2 cell nuclear extract and 50 times more cold DNA fragments spanning from 0230 to 0148, generated by polymerase chain reaction, to compete. This promoter region was able to compete with AP-2 and glucocorticoid responsive element (GRE) oligonucloetides for the binding to HepG2 cell extract (Figure 2). Although no sequences in this DNA region are completely identical to the consensus sequences of AP-2 and GRE, several potential sites with high sequence homology were identified. For example, as shown in Figure 3 Panel A, DNA sequences from 0228 to 0220 (designated as site A) and from 0206 to 0198 (site B)
575
AID
BBRC 6330
/
6922$$1341
02-27-97 11:03:22
bbrcg
AP: BBRC
Vol. 232, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. Competitive gel shift studies. The radiolabeled consensus oligonucleotides of AP-2 and GRE were incubated as indicated with HepG2 nuclear extract (5-10 mg) for binding in the absence (Line 2) or presence (Line 3) of 50-fold excess of unlabeled PLTP promoter fragment (0230 to 0148). The oligonucleotides incubated without nuclear extract were included as a control (Line 1). Autoradiography of 4% nondenaturing polyacrylamide gels was used to identify the DNA/protein complexes.
are similar to the binding motif of AP-2 (GCCNNNGGC); those from 0186 to 0181 (site C) and from 0165 to 0160 (site D) are similar to one half of the binding motif of GRE (TCTTGT). Additionally, the consensus sequences for transcription factors Sp1 at 0114 to 0108 (site E) and Ap-2 at 094 to 085 (site F) were previously identified in the gene promoters of both the human and mouse PLTP (15, 20). The sequences at sites E anf F are slightly downstream of the primary functional region (0230 to 0159); however, these sites may coordinate with other sequences for promoter function. We have mutated two to three base pairs of each of these proposed targeted sites, either identical to the consensus sequences of Sp1 or with high homology to those of Ap-2 and GRE, in order to determine whether they are essential for promoter activity. In Figure 3 Panel A, the mutation at either site C (mutant C) or D (mutant D) caused loss of promoter activity by 59 and 51%, respectivily. Mutant E lost promoter activity by 37%. Mutations at sites A, B, or F resulted in the loss of less than 28% of the activity. These findings indicate that no single region is fully responsible for the promoter function. To further verify that the transcription of the human PLTP gene may require several essential sequences, constructs with various combinations of mutations at two to three different sites were generated and their luciferase activities analyzed. All mutants consisted of additional mutation sites (sites A through F) to either mutants C or D, each of which was shown to be responsible for a significant portion of the promoter activity. Among the mutants assayed for promoter activity, the triple mutation at sites C, D, and E resulted in the maximal loss of promoter function, with less than 10%
of full activity as shown in Figure 3 Panel B. Three constructs, with mutations at any two sites of C, D, and E, also exhibited a substantial loss of promoter activity (67 to 73%). Our studies on multiple mutations indicate 46 to 79% loss of the activity for mutants with two mutation sites, and 73 to 91% loss of the activity for those with three mutation sites. These findings suggest that at least three distinct DNA sequences are crucial for the basal transcription of the human PLTP gene. In many genes the major regulatory elements are present within 1 kilobase of the 5*-upstream region of the coding sequence; however, for some genes these can be located at a greater distance, for example, the regulatory elements of apolipoprotein E gene have been found at least 2 kilobases 3* downstream to the structural gene (21). However, some elements could possibly reside inside this gene. For example, the first intron of apolipoprotein A-II gene was found to possess regulatory elements necessary for its transcription (22). Therefore, we can not rule out the possibility that DNA sequences other than those we have identified are also required for transcription. Little is known about the transcriptional regulation of the PLTP gene as compared to other members of the same gene family, such as cholesteryl ester transfer protein (CETP) and lipopolysaccharide-binding protein (LBP). In the present study, it is particularly noteworthy that the GRE motif could be one of the essential sequences for the PLTP transcription. The presence of GRE in promoters can serve as binding site for GR
FIG. 3. Transcriptional activity of mutated PLTP promoter constructs. The activities of mutants with a single mutation site (A) and with multiple mutation sites (B) are expressed as a percentage of the wild-type construct. Solid lines represent the promoter region from 0230 to 072 and boxes indicate the mutated sites. The positions of the 5* and 3* ends of the sites are indicated. The substituted nucleotides of the mutants are highlighted. One mg of DNA of each construct was transfected into HepG2 cells for luciferase activity assay. Data shown are the means of at least three independent experiments.
576
AID
BBRC 6330
/
6922$$1341
02-27-97 11:03:22
bbrcg
AP: BBRC
Vol. 232, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
to regulate the transcription of glucocorticoid-inducible genes involved in the inflammatory response. It has been found that GRE correlates with the induction of LBP mRNA in HepG2 cells by pro-inflammatory cytokines like IL-1 and IL-6 (23, 24). In addition to LBP, CETP is regulated by endotoxin in both the mouse and hamster model (25, 26). In the hamster model the combination of tumor necrosis factor and interleukin-1 was shown to mimic the effects of the endotoxin-induced decrease in the expression of CETP, suggesting that these cytokines may mediate, in part, the effects of endotoxin on CETP expression (26). PLTP also has been found to be regulated by lipopolysaccharide in a mouse model (27); however, whether the response is on the transcriptional level remains to be investigated. The overall analysis of our mutagenesis data indicates that of seveal DNA sequences examined, three appear to be essential for the basal transcription of the human PLTP gene. Furthermore, the transcriptional process likely is driven by the coordination of several cis-acting elements and trans-acting transcription factors. This information should facilitate elucidating the regulatory mechanisms and the physiological modulators of PLTP expression at the transcriptional level. ACKNOWLEDGMENTS We express our appreciation to Hal Kennedy for data analysis and illustration preparation. This work was supported by Program Project Grant HL 30086 from the National Institutes of Health to John J. Albers.
REFERENCES 1. Tollefson, J. H., Ravnik, S., and Albers, J. J. (1988) J. Lipid Res. 29, 1593–1602. 2. Tall, A. R. (1993) J. Lipid Res. 34, 1255–1274. 3. Cheung, M. C., Wolfbauer, G., and Albers, J. J. (1996) Biochim. Biophys. Acta 1303, 103–110. 4. Tall, A. R. (1986) Methods Enzymol. 129, 469–483. 5. Zilversmit, D. B. (1979) Circulation 60, 473–485. 6. Tu, A-Y., Nishida, H. I., and Nishida, T. (1993) J. Biol. Chem. 268, 23098–23105.
7. Jauhiainen, M., Metso, J., Pahlman, R., Blomqvist, S., van Tol, A., and Ehnholm, C. (1993) J. Biol. Chem. 268, 4032–4036. 8. Albers, J. J., Wolfbauer, G., Cheung, M. C., Day, J. R., Ching, A. F. T., Lok, S., and Tu, A-Y. (1995) Biochim. Biophys. Acta 1258, 27–34. 9. von Eckardstein, A., Jauhiainen, M., Huang, Y., Metso, J., Langer, C., Pussinen, P., Wu, S., Ehnholm, C., and Assmann, G. (1996) Biochim. Biophys. Acta 1301, 255–262. 10. Kostner, G. M., Oettl, K., Jauhiainen, M., Ehnholm, C., Esterbauer, H., and Dieplinger, H. (1995) Biochem. J. 305, 659– 667. 11. Hailman, E., Albers, J. J., Wolfbauer, G., Tu, A-Y., and Wright, S. D. (1996) J. Biol. Chem. 271, 12172–12178. 12. Fielding, C. J., and Fielding, P. E. (1995) J. Lipid Res. 36, 211– 228. 13. Day, J. R., Albers, J. J., Lofton-Day, C. E., Gilbert, T. L., Ching, A. F. T., Grant, F. J., O’Hara, P. J., Marcovina, S. M., and Adolphson, J. L. (1994) J. Biol. Chem. 269, 9388–9391. 14. Tu, A-Y., Deeb, S. S., Iwasaki, L., Day, J. R., and Albers, J. J. (1995) Biochem. Biophys. Res. Commun. 207, 552–558. 15. Tu, A-Y., Wolfbauer, G., and Albers, J. J. (1995) Biochem. Biophys. Res. Commun. 217, 705–711. 16. Sanger, F., Nicklen, S., and Coulson, A. R. (1977) Proc. Natl. Acad. Sci. (USA) 74, 5463–5467. 17. Kunkel, T. A., Roberts, J. D., and Zakour, R. A. (1987) Methods in Enzymol. 154, 367–382. 18. Dignam, J. D., Lebovitz, R. M., and Roeder, R. G. (1983) Nucleic Acids Res. 11, 1475–1489. 19. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193, 265–275. 20. Tu, A-Y., Chen, H., Johnson, K. A., Paigen, B., and Albers, J. J. (1997) Gene, in press. 21. Lauer, S. J., Walker, D., Elshourbagy, N. A., Reardon, C. A., Levy-Wilson, B., and Taylor, J. M. (1988) J. Biol. Chem. 263, 7277–7286. 22. Bossu, J.-P., Chartier, F. L., Fruchart, J.-C., Auwerx, J., Staels, B., and Laine, B. (1996) Biochem. J. 318, 547–553. 23. Schumann, R. R. (1995) Prog. Clin. Biol. Res. 392, 297–304. 24. Schumann, R. R., Kirschning, C. J., Unbehaun, A., Aberle, H., Knopf, H-P., Lamping, N., Ulevitch, R. J., and Herrmann, F. (1996) Mol. Cell. Biol. 16, 3490–3503. 25. Masucci-Magoulas, L., Moulin, P., Jiang, X. C., Richardson, H., Walsh, A., Breslow, J. L., and Tall, A. R. (1995) J. Clin. Invest. 95, 1587–1594. 26. Hardard’ottir, I., Moser, A. H., Fuller, J., Fielding, C., Feingold, K., and Grunfeld, C. (1996) J. Clin. Invest. 97, 2585–2592. 27. Jiang, X. C., and Bruce, C. (1995) J. Biol. Chem. 270, 17133– 17138.
577
AID
BBRC 6330
/
6922$$1341
02-27-97 11:03:22
bbrcg
AP: BBRC