- Email: [email protected]
Enhancer Binding Protein Activates an Enhancer in the Glutamine Synthetase Distal 5′-Flanking Sequence
Archives of Biochemistry and Biophysics Vol. 397, No. 2, January 15, pp. 258 –261, 2002 doi:10.1006/abbi.2001.2666, available online at http://www.idealibrary.com on
CAAT/Enhancer Binding Protein Activates an Enhancer in the Glutamine Synthetase Distal 5⬘-Flanking Sequence Timothy J. Hadden,* ,† ,1,2 Chongsuk Ryou,* ,‡ ,1,3 Liping Zhu,* ,† and Richard E. Miller* ,† ,‡ *Research Service, The John D. Dingell VA Medical Center, 4646 John R., Detroit, Michigan 48201; and †Department of Internal Medicine and ‡Center for Molecular Medicine and Genetics, School of Medicine, Wayne State University, 540 E. Canfield, Detroit, Michigan 48201 Received September 6, 2001, and in revised form October 15, 2001; published online December 13, 2001
The glutamine synthetase (GS) gene is expressed at high levels in several cell types, including astrocytes, pericentral hepatocytes, and adipocytes. During hormone-mediated adipocyte differentiation of 3T3-L1 cells, GS gene expression increases several hundred fold. We previously reported that elements in the distal 5ⴕ-flanking sequence and intron-1 participate in establishing the temporal pattern of GS transcription during adipocyte differentiation. To examine the role of the distal 5ⴕ-flanking region in regulating adipocytespecific GS expression, GS–CAT fusion genes were constructed and analyzed in transiently transfected 3T3-L1 cells. In this way, adipocyte differentiationresponsive enhancer activity was localized to a 422-bp sequence that occurs about 3.5 kb upstream from the transcription start site. This sequence includes several putative C/EBP binding sites and is activated by ectopic expression of C/EBP␣ in NIH-3T3 cells. Thus, our data indicate that C/EBP␣ has the capacity to activate functional C/EBP sites in the GS gene distal 5ⴕ-flanking region. © 2001 Elsevier Science Key Words: glutamine synthetase; gene expression; adipocytes; CAAT/enhancer binding proteins; diabetes mellitus; obesity.
Adipose tissue is the primary site of energy storage in mammals. Excessive adiposity is a major concern in the developed world and is associated with elevated risk of several widespread medical conditions, including insulin resistance, diabetes mellitus, and cardiovascular disease. Adipose tissue mass can expand as a result of increased size of preexisting adipocytes or by de novo differentiation of preadipocytes. An under1
T.J.H. and C.R. contributed equally to the work reported here. To whom correspondence should be addressed. Fax: (313) 5761112. E-mail: [email protected]. 3 Present address: Laboratory of Neurodegenerative Diseases, University of California at San Francisco, San Francisco, CA 94143. 2
258
standing of the mechanisms that regulate adipocyte differentiation may reveal novel targets for therapeutic intervention for diabetes and obesity. Adipocyte differentiation of cultured Swiss mouse 3T3-L1 cells can be triggered by incubation of confluent cultures with dexamethasone, methylisobutylxanthine, and insulin (DMI) 4 in medium containing fetal calf serum. Differentiation of 3T3-L1 cells recapitulates the corresponding in vivo process with a high degree of fidelity. The key roles of several transcription factors, especially CAAT/enhancer binding proteins (C/ EBPs) and peroxisome proliferator-activated receptor ␥2 (PPAR␥2), in promoting adipogenesis have been described (reviewed in Refs. 1 and 2). Consistent with their role as master regulators of adipogenesis, coexpression of C/EBP␣ and PPAR␥2 in nonadipogenic NIH-3T3 cells induces adipose conversion. The glutamine synthetase (GS) gene is expressed at a high level in adipose tissue in vivo (3) and its expression increases dramatically during adipocyte differentiation of 3T3-L1 cells. In addition to adipose tissue, GS is expressed at a high level in astrocytes, pericentral hepatocytes, and proximal tubule cells of the kidney (4). Several recent studies have provided new insight into the control of GS expression, especially in rat liver. Fahrner et al. (5) detected DNase I hypersensitive sites in intron-1 of the rat GS gene and reported that sequences from these regions directed high-level reporter gene expression in transiently transfected HepG2 cells. These observations were extended by Gaunitz et al. (6) who showed that sequences from the rat GS intron-1 were capable of directing reporter gene expression in cells derived from the pericentral and periportal regions of the liver. Since GS in adult liver is expressed 4
Abbreviations used: GS, glutamine synthetase; C/EBP, CAAT/ enhancer binding protein; PPAR␥2, peroxisome proliferator activated receptor ␥2; DMI, dexamethasone, methyl isobutylxanthine, insulin; HSV-tk, herpes simplex virus thymidine kinase core promoter; CAT, chloramphenicol acetyltransferase; LUC, luciferase. 0003-9861/01 $35.00 © 2001 Elsevier Science All rights reserved.
C/EBP REGULATES GLUTAMINE SYNTHETASE GENE TRANSCRIPTION
only in pericentral hepatocytes, these sequences, although they appear to have the capacity to support liver-specific gene expression, are incapable of independently establishing the cell-type-restricted pattern of GS expression in the liver. Lie-Venema et al. (7–9) reported that a major determinant of rat GS gene expression lies between nucleotides ⫺3.1 and ⫺0.5 kb with respect to the transcription start site. Chandrasekhar et al. (10) reported that glucocorticoid-responsive sequences may participate in regulating transcription of the rat GS gene in lung and muscle tissue. Previously we reported that sequences from the distal 5⬘-flanking region and the first intron function cooperatively to regulate transcription of the mouse GS gene during adipocyte differentiation of 3T3-L1 cells (11). Using reporter gene analysis, intron-1 regulatory activity was localized to a 310-bp sequence. Under control of this sequence, reporter gene expression increased transiently and preceded the dramatic increase in GS activity observed during differentiation. In contrast, reporter activity from a composite fusion gene consisting of a 3-kb sequence from the GS distal 5⬘-flanking region and the intron-1 sequence fused upstream of HSVtk-CAT had both early and late components. The late component corresponded well with GS activity in differentiating adipocytes. We concluded that the 5⬘-flanking sequence has no intrinsic enhancer activity but interacts cooperatively with the intron-1 regulatory element to drive expression of the native GS gene (11). However, data reported here indicate that elements in the distal 5⬘-flanking sequence alone are sufficient to drive GS-CAT fusion gene expression with the pattern exhibited by the native gene. Here we show that a 422-bp sequence derived from the GS distal 5⬘-flanking sequence has the capacity to drive reporter gene expression with a temporal pattern that mimics that of native GS activity during adipocyte differentiation. This sequence includes four putative C/EBP binding sites. Moreover, cotransfection of NIH3T3 cells with a C/EBP␣ expression construct and a reporter construct that includes this sequence yields high levels of expression, while ectopic expression of PPAR␥2 had little effect. Thus, it appears that C/EBP␣ has the capacity to activate the distal 5⬘-flanking regulatory region. However, since the native GS gene was not expressed in NIH-3T3 cells expressing C/EBP␣ it is clear that factors other than C/EBP␣ must also be involved in regulating GS expression. MATERIALS AND METHODS Cell culture and transfections. 3T3-L1 preadipocytes were grown and adipocyte differentiation was induced as previously described (12). Cells were harvested for GS, CAT, and protein assays in 1 ml of TEN buffer (10 mM Tris, pH 7.5, 1 mM EDTA, 150 mM NaCl) and stored at ⫺70°C prior to analysis. Transient transfection of 3T3-L1 preadipocytes and NIH-3T3 cells was performed at 80% confluence using SuperFect Reagent (Qiagen). After reaching confluence, 3T3-L1 cells were induced to differentiate. On the days indicated
259
FIG. 1. An adipocyte differentiation-responsive enhancer resides in the GS distal 5⬘-flanking sequence. (A) Diagram of the 5⬘ end of the mouse glutamine synthetase gene from ⫺5 kb upstream to more than 3 kb downstream from the transcription start site (TSS). [Adapted from Kuo and Darnell (4).] Exons 1 and 2, interrupted by the 3-kb intron-1, are indicated by open boxes. Also shown is a representation of fusion gene 3ab, which consists of the 422-bp GS sequence, indicated by a solid rectangle, fused to HSV-tk-CAT. (B) Preadipocytes transiently transfected with fusion gene 3ab were maintained for 3 days after confluence in the absence or presence of DMI (days ⫺3 to 0). For the subsequent 5 days, cultures were maintained in standard medium without additions. Cells were harvested on each day for measurement of CAT and GS activities. Open circles (E) CAT activity in control (-DMI) cells; closed circles (●) CAT activity in DMI-treated (⫹DMI) cells; open triangles (⌬) GS activity in control cells; closed triangles (Œ) GS activity in DMI-treated cells.
(Fig. 1B), control cells and cells triggered to differentiate were harvested. NIH-3T3 cells were cotransfected with a GS–CAT fusion gene, aP2–LUC, and a vector expressing either C/EBP␣ or PPAR␥2. The GS–CAT fusion gene was the 422-bp GS sequence from ⫺3.5 to ⫺3.1 kb fused upstream of HSVtk-CAT (Fig. 1A). The aP2–LUC construct includes 5.4 kb of aP2 5⬘-flanking sequence fused to the luciferase gene. Control cells were cotransfected with the reporter gene constructs only, i.e., C/EBP␣ and PPAR␥2 expression vectors were omitted. NIH-3T3 cells were harvested for reporter analysis 48 h after transfection. For determination of luciferase activities cells were harvested into Reporter Lysis Buffer (Promega). Assays. Chloramphenicol acetyltransferase assays were performed using a modification of the method of Seed and Sheen (13) as
260
HADDEN ET AL.
FIG. 2. Nucleotide sequence of the functional GS distal 5⬘-flanking regulatory region. The 422-bp 3ab sequence that displays functional adipocyte differentiation-responsive enhancer activity is shown. Four C/EBP consensus binding sites in the sequence are underlined.
previously described (11). GS activity in cell lysates was measured by the ␥-glutamyl transfer assay (12, 14). Luciferase assay substrate was obtained from Promega and assays were carried out according to the manufacturer’s recommendations using a Berthold LB953 luminometer. LUC activity is expressed in relative light units. Total protein was estimated using the method of Lowry et al. (15). Sequences analyzed. For functional studies of transcriptional regulatory activity several fusion genes were constructed. Sequences tested were cloned into pBLCAT2.0, which contains the minimal HSVtk promoter upstream of the CAT gene (11). The first included GS sequences from ⫺5 to ⫺2 kb in the distal 5⬘-flanking region. The second (Gene 3ab) included a 422-bp sequence derived from the distal 5⬘-flanking region and extending from ⫺3.5 to ⫺3.1 kb. Sequences for construction of fusion genes were obtained by restriction endonuclease digestion or by PCR. These constructs were transiently transfected into 3T3-L1 or NIH-3T3 cells.
transfected NIH-3T3 cells with fusion gene 3ab, aP2– LUC, and an expression vector encoding either C/EBP␣ or PPAR␥2. The expression of C/EBP␣ and PPAR␥2 was confirmed by Western blotting (data not shown). As shown in Fig. 3, control cells lacking C/EBP␣ and PPAR␥2 expressed GS–CAT and aP2– LUC at low levels. An approximately 150-fold increase in CAT and a 100-fold increase in LUC activities were observed with forced expression of C/EBP␣. PPAR␥2 expression increased GS–CAT expression by 4-fold, while it increased aP2–LUC expression 25-fold over that in control cells. The effect of C/EBP␣ and PPAR␥2 on aP2–LUC expression is consistent with published reports indicating that these transcription factors participate in activating transcription of the aP2 gene in cultured adipocytes (16, 17). GS activity was not detected in any of the transfected NIH-3T3 cells. DISCUSSION
Adipose tissue is of central importance in energy metabolism and is implicated in the development of obesity, insulin resistance, and type 2 diabetes mellitus (18). The study of adipocyte differentiation and metabolism will produce increased understanding of the pathophysiology of these and other abnormal states. Hormone-mediated adipocyte differentiation of cultured 3T3-L1 cells is accompanied by dramatic changes in the expression of a
RESULTS
Localization of functional differentiation-responsive enhancer activity in the distal 5⬘-flanking sequence. To localize functional enhancer activity in the distal 5⬘-flanking sequence a series of fusion genes was constructed and analyzed in transiently transfected 3T3-L1 cells. In each fusion gene a fragment of the 3-kb distal 5⬘-flanking sequence was fused to HSV-tk-CAT and reporter gene activity was monitored during adipocyte differentiation. A 422-bp sequence derived from this region (gene 3ab) drives reporter gene expression in transiently transfected 3T3-L1 cells (Fig. 1) and the pattern of expression is similar to that of GS activity. Sequences from the distal 5⬘-flanking region other than the 422-bp 3ab sequence did not direct differentiationresponsive reporter gene expression (data not shown). DNA sequence analysis of the 422-bp functional regulatory region. The nucleotide sequence of the 422-bp distal 5⬘-flanking regulatory region was determined and examined for the presence of potential transcription factor binding sites (Fig. 2). The sequence includes four consensus C/EBP binding sites distributed throughout the sequence. Expression of C/EBP␣ in NIH-3T3 cells activates the GS distal 5⬘-flanking regulatory region. To examine the capacity of C/EBP␣ and PPAR␥2 to activate the mouse GS distal 5⬘-flanking region we transiently co-
FIG. 3. C/EBP␣ activates the distal 5⬘-flanking regulatory region in NIH-3T3 cells. CAT and LUC activities in NIH-3T3 cells transfected with GS–CAT fusion gene 3ab, aP2–LUC, and a construct expressing either C/EBP␣ or PPAR␥2. Control indicates transfection with GS–CAT and aP2–LUC but omission of both C/EBP␣ and PPAR␥2 transcription factor expression vectors. Values are the mean ⫾ SE for triplicate cultures. A relative activity of 1.0 in the control was 0.58 ⫾ 0.04 units for GS–CAT and 2702 ⫾ 378 relative light units for aP2–LUC. Closed bars, CAT activity; hatched bars, LUC activity. The P values are ⬍0.001 and ⬍0.008 for CAT in control cells vs cells transfected with the C/EBP␣ and PPAR␥2 expression vectors, respectively. The P values are ⬍0.005 and ⬍0.001 for LUC in control cells vs cells transfected with the C/EBP␣ and PPAR␥2 expression vectors, respectively.
C/EBP REGULATES GLUTAMINE SYNTHETASE GENE TRANSCRIPTION
large number of proteins and culminates in the acquisition of a phenotype closely resembling that of adipocytes in vivo. Among these changes is a dramatic increase in GS gene transcription. Glutamine plays an important role in mammalian nitrogen metabolism (reviewed in Refs. 19 and 20; 21– 26). Therefore, GS, the enzyme responsible for glutamine biosynthesis, is a strategic target for the regulation of nitrogen metabolism because it links the many catabolic processes by which ammonia is produced with numerous biosynthetic pathways that lead ultimately to the formation of proteins, nucleic acids, amino sugars, and complex polysaccharides. The marked rise in GS activity during the course of 3T3-L1 cell differentiation closely follows the removal of inducers (DMI), consistent with regulation by C/EBP␣. We have identified a 422-bp sequence of the mouse GS distal 5⬘-flanking region that has the capacity to confer a pattern of reporter gene expression that mimics that of GS activity in 3T3-L1 cells. This sequence includes four putative C/EBP binding sites. While this fusion gene is expressed at low levels in NIH-3T3 cells, cotransfection with a C/EBP␣ expression vector leads to high-level reporter gene activity. In contrast, expression of PPAR␥2 in these cells does not yield high-level reporter gene expression. Taken together, these results suggest that C/EBP␣ has the capacity to activate GS gene transcription in differentiating adipocytes. However, the fact that C/EBP␣ expression in NIH-3T3 cells does not induce GS expression indicates that C/EBP␣ is not sufficient to induce native GS gene expression. This suggests that GS expression requires factors not expressed in NIH3T3 cells that activate or derepress elements of the GS gene other than those activated by C/EBP␣. We previously reported that in stable transfectants, the 3-kb distal 5⬘-flanking sequence has no intrinsic enhancer activity and that an intron-1 enhancer sequence drives reporter gene expression in a pattern that differs from that of the native GS gene, but that a composite fusion gene composed of 3 kb of distal 5⬘flanking sequence and 2 kb of the intron-1 sequence drives reporter gene expression with a temporal pattern that mimics native GS expression. However, the studies reported here indicate that at least a segment of the distal 5⬘-flanking sequence has the capacity to drive reporter gene expression with a GS-like pattern. Although the cause of this discrepancy is currently unknown, it does serve to highlight the limitations of transfected fusion gene expression as an indicator of native gene transcription. ACKNOWLEDGMENTS We thank Ibrahim Kadura and Mondeep Narewal for outstanding technical assistance. We are grateful to Drs. James Darnell, Jr., Howard Green, C. Frank Kuo, M. Daniel Lane, and Bruce M.
261
Spiegelman for providing unique materials that were essential for these studies. This work was supported by the VA Medical Research Service in the form of a VA Merit Review award to R.E.M. and an award from the Research Enhancement Award Program to R.E.M., by a fellowship award to C.R. from the Morris Hood, Jr. Comprehensive Diabetes Center, and by the Center for Molecular Medicine & Genetics at Wayne State University.
REFERENCES 1. Lane, M. D., Tang, Q. Q., and Jiang, M. S. (1999) Biochem. Biophys. Res. Commun. 266, 677– 683. 2. Rosen, E. D., Walkey, C. J., Puigserver, P., and Spiegelman, B. M. (2000) Genes Dev. 14, 1293–1307. 3. Miller, R. E., Hackenberg, R. W., and Gershman, H. (1978) Proc. Natl. Acad. Sci. USA 75, 1418 –1422. 4. Kuo, C. F., and Darnell, J. E., Jr. (1989) J. Mol. Biol. 208, 45–56. 5. Fahrner, J., Labruyere, W. T., Gaunitz, C., Moorman, A. F., Gebhardt, R., and Lamers, W. H. (1993) Eur. J. Biochem. 213, 1067–1073. 6. Gaunitz, F., Gaunitz, C., Papke, M., and Gebhardt, R. (1997) Biol. Chem. 378, 11–18. 7. Lie-Venema, H., Labruyere, W. T., van Roon, M. A., de Boer, P. A., Moorman, A. F., Berns, A. J., and Lamers, W. H. (1995) J. Biol. Chem. 270, 28251–28256. 8. Lie-Venema, H., de Boer, P. A., Moorman, A. F., and Lamers, W. H. (1997) Eur. J. Biochem. 248, 644 – 659. 9. Lie-Venema, H., de Boer, P. A., Moorman, A. F., and Lamers, W. H. (1997) Biochem. J. 323, 611– 619. 10. Chandrasekhar, S., Souba, W. W., and Abcouwer, S. F. (1999) Am. J. Physiol. 276, l319 –l331. 11. Hadden, T. J., Ryou, C., and Miller, R. E. (1997) Nucleic Acids Res. 25, 3930 –3936. 12. Miller, R. E., Pope, S. R., DeWille, J. W., and Burns, D. M. (1983) J. Biol. Chem. 258, 5405–5413. 13. Seed, B., and Sheen, J. Y. (1988) Gene 67, 271–277. 14. Bhandari, B., and Miller, R. E. (1987) Mol. Cell. Endocrinol. 51, 7–11. 15. Lowry, O. H., Rosenbrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193, 265–275. 16. Graves, R. A., Tontonoz, P., Platt, K. A., Ross, S. R., and Spiegelman, B. M. (1992) J. Cell. Biochem. 49, 219 –224. 17. Tontonoz, P., Hu, E., Graves, R. A., Budavari, A. I., and Spiegelman, B. M. (1994) Genes Dev. 8, 1224 –1234. 18. Hotamisligil, G. S., Arner, P., Caro, J. F., Atkinson, R. L., and Spiegelman, B. M. (1995) J. Clin. Invest. 95, 2409 –2415. 19. Stadtman, E. R. (1973) in The Enzymes of Glutamine Metabolism (Prusiner, S., and Stadtman, E.R., Eds.) pp. 1– 6, Academic Press, New York. 20. Miller, R. E., Burns, D. M., and Bhandari, B. (1987) in Biology of the Adipocyte: Research Approaches (Hausman, G., and Martin, R., Eds.) pp. 198 –228, Van Nostrand–Reinhold, New York. 21. Felig, P. (1975) Annu. Rev. Biochem. 44, 933–955. 22. Schrock, H., and Goldstein, L. (1981) Am. J. Physiol. 240, E519 – E525. 23. Tischler, M. E., and Goldberg, A. L. (1980) J. Biol. Chem. 255, 8074 – 8081. 24. Windmueller, H. G., and Spaeth, A. E. (1974) J. Biol. Chem. 249, 5070 –5079. 25. Windmueller, H. G., and Spaeth, A. E. (1980) J. Biol. Chem. 255, 107–112. 26. Zielke, H. R., Zielke, C. L., and Ozand, P. T. (1984) Fed. Proc. 43, 121–125.
CAAT/Enhancer Binding Protein Activates an Enhancer in the Glutamine Synthetase Distal 5⬘-Flanking Sequence Timothy J. Hadden,* ,† ,1,2 Chongsuk Ryou,* ,‡ ,1,3 Liping Zhu,* ,† and Richard E. Miller* ,† ,‡ *Research Service, The John D. Dingell VA Medical Center, 4646 John R., Detroit, Michigan 48201; and †Department of Internal Medicine and ‡Center for Molecular Medicine and Genetics, School of Medicine, Wayne State University, 540 E. Canfield, Detroit, Michigan 48201 Received September 6, 2001, and in revised form October 15, 2001; published online December 13, 2001
The glutamine synthetase (GS) gene is expressed at high levels in several cell types, including astrocytes, pericentral hepatocytes, and adipocytes. During hormone-mediated adipocyte differentiation of 3T3-L1 cells, GS gene expression increases several hundred fold. We previously reported that elements in the distal 5ⴕ-flanking sequence and intron-1 participate in establishing the temporal pattern of GS transcription during adipocyte differentiation. To examine the role of the distal 5ⴕ-flanking region in regulating adipocytespecific GS expression, GS–CAT fusion genes were constructed and analyzed in transiently transfected 3T3-L1 cells. In this way, adipocyte differentiationresponsive enhancer activity was localized to a 422-bp sequence that occurs about 3.5 kb upstream from the transcription start site. This sequence includes several putative C/EBP binding sites and is activated by ectopic expression of C/EBP␣ in NIH-3T3 cells. Thus, our data indicate that C/EBP␣ has the capacity to activate functional C/EBP sites in the GS gene distal 5ⴕ-flanking region. © 2001 Elsevier Science Key Words: glutamine synthetase; gene expression; adipocytes; CAAT/enhancer binding proteins; diabetes mellitus; obesity.
Adipose tissue is the primary site of energy storage in mammals. Excessive adiposity is a major concern in the developed world and is associated with elevated risk of several widespread medical conditions, including insulin resistance, diabetes mellitus, and cardiovascular disease. Adipose tissue mass can expand as a result of increased size of preexisting adipocytes or by de novo differentiation of preadipocytes. An under1
T.J.H. and C.R. contributed equally to the work reported here. To whom correspondence should be addressed. Fax: (313) 5761112. E-mail: [email protected]. 3 Present address: Laboratory of Neurodegenerative Diseases, University of California at San Francisco, San Francisco, CA 94143. 2
258
standing of the mechanisms that regulate adipocyte differentiation may reveal novel targets for therapeutic intervention for diabetes and obesity. Adipocyte differentiation of cultured Swiss mouse 3T3-L1 cells can be triggered by incubation of confluent cultures with dexamethasone, methylisobutylxanthine, and insulin (DMI) 4 in medium containing fetal calf serum. Differentiation of 3T3-L1 cells recapitulates the corresponding in vivo process with a high degree of fidelity. The key roles of several transcription factors, especially CAAT/enhancer binding proteins (C/ EBPs) and peroxisome proliferator-activated receptor ␥2 (PPAR␥2), in promoting adipogenesis have been described (reviewed in Refs. 1 and 2). Consistent with their role as master regulators of adipogenesis, coexpression of C/EBP␣ and PPAR␥2 in nonadipogenic NIH-3T3 cells induces adipose conversion. The glutamine synthetase (GS) gene is expressed at a high level in adipose tissue in vivo (3) and its expression increases dramatically during adipocyte differentiation of 3T3-L1 cells. In addition to adipose tissue, GS is expressed at a high level in astrocytes, pericentral hepatocytes, and proximal tubule cells of the kidney (4). Several recent studies have provided new insight into the control of GS expression, especially in rat liver. Fahrner et al. (5) detected DNase I hypersensitive sites in intron-1 of the rat GS gene and reported that sequences from these regions directed high-level reporter gene expression in transiently transfected HepG2 cells. These observations were extended by Gaunitz et al. (6) who showed that sequences from the rat GS intron-1 were capable of directing reporter gene expression in cells derived from the pericentral and periportal regions of the liver. Since GS in adult liver is expressed 4
Abbreviations used: GS, glutamine synthetase; C/EBP, CAAT/ enhancer binding protein; PPAR␥2, peroxisome proliferator activated receptor ␥2; DMI, dexamethasone, methyl isobutylxanthine, insulin; HSV-tk, herpes simplex virus thymidine kinase core promoter; CAT, chloramphenicol acetyltransferase; LUC, luciferase. 0003-9861/01 $35.00 © 2001 Elsevier Science All rights reserved.
C/EBP REGULATES GLUTAMINE SYNTHETASE GENE TRANSCRIPTION
only in pericentral hepatocytes, these sequences, although they appear to have the capacity to support liver-specific gene expression, are incapable of independently establishing the cell-type-restricted pattern of GS expression in the liver. Lie-Venema et al. (7–9) reported that a major determinant of rat GS gene expression lies between nucleotides ⫺3.1 and ⫺0.5 kb with respect to the transcription start site. Chandrasekhar et al. (10) reported that glucocorticoid-responsive sequences may participate in regulating transcription of the rat GS gene in lung and muscle tissue. Previously we reported that sequences from the distal 5⬘-flanking region and the first intron function cooperatively to regulate transcription of the mouse GS gene during adipocyte differentiation of 3T3-L1 cells (11). Using reporter gene analysis, intron-1 regulatory activity was localized to a 310-bp sequence. Under control of this sequence, reporter gene expression increased transiently and preceded the dramatic increase in GS activity observed during differentiation. In contrast, reporter activity from a composite fusion gene consisting of a 3-kb sequence from the GS distal 5⬘-flanking region and the intron-1 sequence fused upstream of HSVtk-CAT had both early and late components. The late component corresponded well with GS activity in differentiating adipocytes. We concluded that the 5⬘-flanking sequence has no intrinsic enhancer activity but interacts cooperatively with the intron-1 regulatory element to drive expression of the native GS gene (11). However, data reported here indicate that elements in the distal 5⬘-flanking sequence alone are sufficient to drive GS-CAT fusion gene expression with the pattern exhibited by the native gene. Here we show that a 422-bp sequence derived from the GS distal 5⬘-flanking sequence has the capacity to drive reporter gene expression with a temporal pattern that mimics that of native GS activity during adipocyte differentiation. This sequence includes four putative C/EBP binding sites. Moreover, cotransfection of NIH3T3 cells with a C/EBP␣ expression construct and a reporter construct that includes this sequence yields high levels of expression, while ectopic expression of PPAR␥2 had little effect. Thus, it appears that C/EBP␣ has the capacity to activate the distal 5⬘-flanking regulatory region. However, since the native GS gene was not expressed in NIH-3T3 cells expressing C/EBP␣ it is clear that factors other than C/EBP␣ must also be involved in regulating GS expression. MATERIALS AND METHODS Cell culture and transfections. 3T3-L1 preadipocytes were grown and adipocyte differentiation was induced as previously described (12). Cells were harvested for GS, CAT, and protein assays in 1 ml of TEN buffer (10 mM Tris, pH 7.5, 1 mM EDTA, 150 mM NaCl) and stored at ⫺70°C prior to analysis. Transient transfection of 3T3-L1 preadipocytes and NIH-3T3 cells was performed at 80% confluence using SuperFect Reagent (Qiagen). After reaching confluence, 3T3-L1 cells were induced to differentiate. On the days indicated
259
FIG. 1. An adipocyte differentiation-responsive enhancer resides in the GS distal 5⬘-flanking sequence. (A) Diagram of the 5⬘ end of the mouse glutamine synthetase gene from ⫺5 kb upstream to more than 3 kb downstream from the transcription start site (TSS). [Adapted from Kuo and Darnell (4).] Exons 1 and 2, interrupted by the 3-kb intron-1, are indicated by open boxes. Also shown is a representation of fusion gene 3ab, which consists of the 422-bp GS sequence, indicated by a solid rectangle, fused to HSV-tk-CAT. (B) Preadipocytes transiently transfected with fusion gene 3ab were maintained for 3 days after confluence in the absence or presence of DMI (days ⫺3 to 0). For the subsequent 5 days, cultures were maintained in standard medium without additions. Cells were harvested on each day for measurement of CAT and GS activities. Open circles (E) CAT activity in control (-DMI) cells; closed circles (●) CAT activity in DMI-treated (⫹DMI) cells; open triangles (⌬) GS activity in control cells; closed triangles (Œ) GS activity in DMI-treated cells.
(Fig. 1B), control cells and cells triggered to differentiate were harvested. NIH-3T3 cells were cotransfected with a GS–CAT fusion gene, aP2–LUC, and a vector expressing either C/EBP␣ or PPAR␥2. The GS–CAT fusion gene was the 422-bp GS sequence from ⫺3.5 to ⫺3.1 kb fused upstream of HSVtk-CAT (Fig. 1A). The aP2–LUC construct includes 5.4 kb of aP2 5⬘-flanking sequence fused to the luciferase gene. Control cells were cotransfected with the reporter gene constructs only, i.e., C/EBP␣ and PPAR␥2 expression vectors were omitted. NIH-3T3 cells were harvested for reporter analysis 48 h after transfection. For determination of luciferase activities cells were harvested into Reporter Lysis Buffer (Promega). Assays. Chloramphenicol acetyltransferase assays were performed using a modification of the method of Seed and Sheen (13) as
260
HADDEN ET AL.
FIG. 2. Nucleotide sequence of the functional GS distal 5⬘-flanking regulatory region. The 422-bp 3ab sequence that displays functional adipocyte differentiation-responsive enhancer activity is shown. Four C/EBP consensus binding sites in the sequence are underlined.
previously described (11). GS activity in cell lysates was measured by the ␥-glutamyl transfer assay (12, 14). Luciferase assay substrate was obtained from Promega and assays were carried out according to the manufacturer’s recommendations using a Berthold LB953 luminometer. LUC activity is expressed in relative light units. Total protein was estimated using the method of Lowry et al. (15). Sequences analyzed. For functional studies of transcriptional regulatory activity several fusion genes were constructed. Sequences tested were cloned into pBLCAT2.0, which contains the minimal HSVtk promoter upstream of the CAT gene (11). The first included GS sequences from ⫺5 to ⫺2 kb in the distal 5⬘-flanking region. The second (Gene 3ab) included a 422-bp sequence derived from the distal 5⬘-flanking region and extending from ⫺3.5 to ⫺3.1 kb. Sequences for construction of fusion genes were obtained by restriction endonuclease digestion or by PCR. These constructs were transiently transfected into 3T3-L1 or NIH-3T3 cells.
transfected NIH-3T3 cells with fusion gene 3ab, aP2– LUC, and an expression vector encoding either C/EBP␣ or PPAR␥2. The expression of C/EBP␣ and PPAR␥2 was confirmed by Western blotting (data not shown). As shown in Fig. 3, control cells lacking C/EBP␣ and PPAR␥2 expressed GS–CAT and aP2– LUC at low levels. An approximately 150-fold increase in CAT and a 100-fold increase in LUC activities were observed with forced expression of C/EBP␣. PPAR␥2 expression increased GS–CAT expression by 4-fold, while it increased aP2–LUC expression 25-fold over that in control cells. The effect of C/EBP␣ and PPAR␥2 on aP2–LUC expression is consistent with published reports indicating that these transcription factors participate in activating transcription of the aP2 gene in cultured adipocytes (16, 17). GS activity was not detected in any of the transfected NIH-3T3 cells. DISCUSSION
Adipose tissue is of central importance in energy metabolism and is implicated in the development of obesity, insulin resistance, and type 2 diabetes mellitus (18). The study of adipocyte differentiation and metabolism will produce increased understanding of the pathophysiology of these and other abnormal states. Hormone-mediated adipocyte differentiation of cultured 3T3-L1 cells is accompanied by dramatic changes in the expression of a
RESULTS
Localization of functional differentiation-responsive enhancer activity in the distal 5⬘-flanking sequence. To localize functional enhancer activity in the distal 5⬘-flanking sequence a series of fusion genes was constructed and analyzed in transiently transfected 3T3-L1 cells. In each fusion gene a fragment of the 3-kb distal 5⬘-flanking sequence was fused to HSV-tk-CAT and reporter gene activity was monitored during adipocyte differentiation. A 422-bp sequence derived from this region (gene 3ab) drives reporter gene expression in transiently transfected 3T3-L1 cells (Fig. 1) and the pattern of expression is similar to that of GS activity. Sequences from the distal 5⬘-flanking region other than the 422-bp 3ab sequence did not direct differentiationresponsive reporter gene expression (data not shown). DNA sequence analysis of the 422-bp functional regulatory region. The nucleotide sequence of the 422-bp distal 5⬘-flanking regulatory region was determined and examined for the presence of potential transcription factor binding sites (Fig. 2). The sequence includes four consensus C/EBP binding sites distributed throughout the sequence. Expression of C/EBP␣ in NIH-3T3 cells activates the GS distal 5⬘-flanking regulatory region. To examine the capacity of C/EBP␣ and PPAR␥2 to activate the mouse GS distal 5⬘-flanking region we transiently co-
FIG. 3. C/EBP␣ activates the distal 5⬘-flanking regulatory region in NIH-3T3 cells. CAT and LUC activities in NIH-3T3 cells transfected with GS–CAT fusion gene 3ab, aP2–LUC, and a construct expressing either C/EBP␣ or PPAR␥2. Control indicates transfection with GS–CAT and aP2–LUC but omission of both C/EBP␣ and PPAR␥2 transcription factor expression vectors. Values are the mean ⫾ SE for triplicate cultures. A relative activity of 1.0 in the control was 0.58 ⫾ 0.04 units for GS–CAT and 2702 ⫾ 378 relative light units for aP2–LUC. Closed bars, CAT activity; hatched bars, LUC activity. The P values are ⬍0.001 and ⬍0.008 for CAT in control cells vs cells transfected with the C/EBP␣ and PPAR␥2 expression vectors, respectively. The P values are ⬍0.005 and ⬍0.001 for LUC in control cells vs cells transfected with the C/EBP␣ and PPAR␥2 expression vectors, respectively.
C/EBP REGULATES GLUTAMINE SYNTHETASE GENE TRANSCRIPTION
large number of proteins and culminates in the acquisition of a phenotype closely resembling that of adipocytes in vivo. Among these changes is a dramatic increase in GS gene transcription. Glutamine plays an important role in mammalian nitrogen metabolism (reviewed in Refs. 19 and 20; 21– 26). Therefore, GS, the enzyme responsible for glutamine biosynthesis, is a strategic target for the regulation of nitrogen metabolism because it links the many catabolic processes by which ammonia is produced with numerous biosynthetic pathways that lead ultimately to the formation of proteins, nucleic acids, amino sugars, and complex polysaccharides. The marked rise in GS activity during the course of 3T3-L1 cell differentiation closely follows the removal of inducers (DMI), consistent with regulation by C/EBP␣. We have identified a 422-bp sequence of the mouse GS distal 5⬘-flanking region that has the capacity to confer a pattern of reporter gene expression that mimics that of GS activity in 3T3-L1 cells. This sequence includes four putative C/EBP binding sites. While this fusion gene is expressed at low levels in NIH-3T3 cells, cotransfection with a C/EBP␣ expression vector leads to high-level reporter gene activity. In contrast, expression of PPAR␥2 in these cells does not yield high-level reporter gene expression. Taken together, these results suggest that C/EBP␣ has the capacity to activate GS gene transcription in differentiating adipocytes. However, the fact that C/EBP␣ expression in NIH-3T3 cells does not induce GS expression indicates that C/EBP␣ is not sufficient to induce native GS gene expression. This suggests that GS expression requires factors not expressed in NIH3T3 cells that activate or derepress elements of the GS gene other than those activated by C/EBP␣. We previously reported that in stable transfectants, the 3-kb distal 5⬘-flanking sequence has no intrinsic enhancer activity and that an intron-1 enhancer sequence drives reporter gene expression in a pattern that differs from that of the native GS gene, but that a composite fusion gene composed of 3 kb of distal 5⬘flanking sequence and 2 kb of the intron-1 sequence drives reporter gene expression with a temporal pattern that mimics native GS expression. However, the studies reported here indicate that at least a segment of the distal 5⬘-flanking sequence has the capacity to drive reporter gene expression with a GS-like pattern. Although the cause of this discrepancy is currently unknown, it does serve to highlight the limitations of transfected fusion gene expression as an indicator of native gene transcription. ACKNOWLEDGMENTS We thank Ibrahim Kadura and Mondeep Narewal for outstanding technical assistance. We are grateful to Drs. James Darnell, Jr., Howard Green, C. Frank Kuo, M. Daniel Lane, and Bruce M.
261
Spiegelman for providing unique materials that were essential for these studies. This work was supported by the VA Medical Research Service in the form of a VA Merit Review award to R.E.M. and an award from the Research Enhancement Award Program to R.E.M., by a fellowship award to C.R. from the Morris Hood, Jr. Comprehensive Diabetes Center, and by the Center for Molecular Medicine & Genetics at Wayne State University.
REFERENCES 1. Lane, M. D., Tang, Q. Q., and Jiang, M. S. (1999) Biochem. Biophys. Res. Commun. 266, 677– 683. 2. Rosen, E. D., Walkey, C. J., Puigserver, P., and Spiegelman, B. M. (2000) Genes Dev. 14, 1293–1307. 3. Miller, R. E., Hackenberg, R. W., and Gershman, H. (1978) Proc. Natl. Acad. Sci. USA 75, 1418 –1422. 4. Kuo, C. F., and Darnell, J. E., Jr. (1989) J. Mol. Biol. 208, 45–56. 5. Fahrner, J., Labruyere, W. T., Gaunitz, C., Moorman, A. F., Gebhardt, R., and Lamers, W. H. (1993) Eur. J. Biochem. 213, 1067–1073. 6. Gaunitz, F., Gaunitz, C., Papke, M., and Gebhardt, R. (1997) Biol. Chem. 378, 11–18. 7. Lie-Venema, H., Labruyere, W. T., van Roon, M. A., de Boer, P. A., Moorman, A. F., Berns, A. J., and Lamers, W. H. (1995) J. Biol. Chem. 270, 28251–28256. 8. Lie-Venema, H., de Boer, P. A., Moorman, A. F., and Lamers, W. H. (1997) Eur. J. Biochem. 248, 644 – 659. 9. Lie-Venema, H., de Boer, P. A., Moorman, A. F., and Lamers, W. H. (1997) Biochem. J. 323, 611– 619. 10. Chandrasekhar, S., Souba, W. W., and Abcouwer, S. F. (1999) Am. J. Physiol. 276, l319 –l331. 11. Hadden, T. J., Ryou, C., and Miller, R. E. (1997) Nucleic Acids Res. 25, 3930 –3936. 12. Miller, R. E., Pope, S. R., DeWille, J. W., and Burns, D. M. (1983) J. Biol. Chem. 258, 5405–5413. 13. Seed, B., and Sheen, J. Y. (1988) Gene 67, 271–277. 14. Bhandari, B., and Miller, R. E. (1987) Mol. Cell. Endocrinol. 51, 7–11. 15. Lowry, O. H., Rosenbrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193, 265–275. 16. Graves, R. A., Tontonoz, P., Platt, K. A., Ross, S. R., and Spiegelman, B. M. (1992) J. Cell. Biochem. 49, 219 –224. 17. Tontonoz, P., Hu, E., Graves, R. A., Budavari, A. I., and Spiegelman, B. M. (1994) Genes Dev. 8, 1224 –1234. 18. Hotamisligil, G. S., Arner, P., Caro, J. F., Atkinson, R. L., and Spiegelman, B. M. (1995) J. Clin. Invest. 95, 2409 –2415. 19. Stadtman, E. R. (1973) in The Enzymes of Glutamine Metabolism (Prusiner, S., and Stadtman, E.R., Eds.) pp. 1– 6, Academic Press, New York. 20. Miller, R. E., Burns, D. M., and Bhandari, B. (1987) in Biology of the Adipocyte: Research Approaches (Hausman, G., and Martin, R., Eds.) pp. 198 –228, Van Nostrand–Reinhold, New York. 21. Felig, P. (1975) Annu. Rev. Biochem. 44, 933–955. 22. Schrock, H., and Goldstein, L. (1981) Am. J. Physiol. 240, E519 – E525. 23. Tischler, M. E., and Goldberg, A. L. (1980) J. Biol. Chem. 255, 8074 – 8081. 24. Windmueller, H. G., and Spaeth, A. E. (1974) J. Biol. Chem. 249, 5070 –5079. 25. Windmueller, H. G., and Spaeth, A. E. (1980) J. Biol. Chem. 255, 107–112. 26. Zielke, H. R., Zielke, C. L., and Ozand, P. T. (1984) Fed. Proc. 43, 121–125.