- Email: [email protected]
Promoter activity of carbonic anhydrase II regulatory regions in cultured renal proximal tubular cells
PII SO0243205(98)00247-l
ELSEVIER
PROMOTER
Life Sciences,Vol. 63, No. 2, pp. 121-126,1998 copyright0 1998 rGevicrsciem Inc. Printedin the USA. All tightsrerctvd 0024-3205/98 $19.00 + .oo
ACTIVITY OF CARBONIC ANHYDRASE II REGULATORY IN CULTURED RENAL PROXIMAL TUBULAR CELLS
REGIONS
Li-Wen Lai*, Robert P. Erickson’, Patrick J. Venta§, Richard E. Tashiann, and Yeong-Hau H. Lien* Department of *Medicine and #Pediatrics, University of Arizona Health Sciences Center, Tucson, Arizona 85724; “Department of Small Animal Clinical Sciences, Michigan State University, East Lansing, Michigan 48224; ‘Department of Human Genetics, University of Michigan Medical School, Ann Arbor, Michigan 48 109
(Received in fmal form April 17, 1998)
Summary Carbonic anhydrase II (CAII) plays an important role in the acid-base homeostasis of the body and its deficiency results in renal tubular acidosis. In order to identify the regulatory regions in the CAII gene for the future development of kidneytargeted gene therapy, we investigated the 5’ region of the gene for its promoter activity. Deletion constructs with various lengths of the 5’ flanking region of the human CAII promoter were ligated to the CAT reporter gene and lipofected in primary cultures of mouse proximal renal tubular cells and in cells of the established porcine proximal tubular cell line, LLC-PK,. The CAT activity was measured 48 hours after gene transfection. The -12000lCAT and -1300/CAT constructs expressed the highest CAT activity in both types of renal tubular cells (143- and 1X0-fold increase, respectively, in mouse proximal tubular cells; 50- and 70-fold increase, respectively, in LLC-PK, cells) but not the -420/CAT, -270/CAT, or -18O/CAT constructs (9, 12, and 9% of that of -1300/CAT construct, respectively, in mouse proximal tubular cells and, 23, 9, and 8%, respectively, in LLC-PK, cells, all p co.01 vs. -1300/CAT construct). No cytotoxicity was detected in the transfected cells. A computer search identified multiple putative transcription factor binding elements including Apl and Ap2 binding elements, which are present in the -13001CAT construct but not in the shorter constructs. In conclusion, we demonstrate that the human CAR 5’ sequence of proximal 1.3 kb contains strong promoter sequence(s) for renal tubular cells. Key Words: carbonic anhydrase, promoter activity, renal cells
Carbonic
anhydrase
(CA) catalyzes
the reversible
hydration
of carbon dioxide
and plays an
Corresponding Author: Yeong-Hau H. Lien, M.D., Ph.D. Section of Nephrology, Department of Medicine, University of Arizona Health Science Center, Tucson, 1501 N. Campbell Ave., Tucson, AZ 85724. Phone: 520-626-6370; Fax: 520-626-2024; E-mail: wna.edu
122
Carbonic Anhydrase II Regulatory Regions
Vol. 63, No. 2, 1998
role in acid-base homeostasis. CA activity in the mammalian kidney was first described in 1941 by Davenport and Wilelmi (1). CA11 is a cytoplasmic enzyme present in the proximal convoluted tubule, thick limb of Henle, and intercalated cells, but not principal cells, of collecting ducts (2. 3, 4). This pattern of expression suggests that regulation of CA11 expression is likely involved a variety of transcription factors. In human, CA11 deficiency is associated with a syndrome of renal tubular acidosis, osteopetrosis, and cerebral calcification (5). Lewis et al. produced a strain of CAII null mice by an induced mutagenesis that resulted in mice with growth retardation and renal tubular acidosis (6). Recently. we have used CA11 null mice as an animal model for testing gene therapy targeting to renal tubular cells (7, 8, 9). When CMV promoter was ligated to either b-galactosidase (7) or CA11 cDNA (8,9), compound with liposome, and injected into the renal pelvis of normal and CA11 deficient mice, respectively, the expression of the reporter genes last for 4-6 weeks. In order to further prolong the duration of the transgene expression, one strategy is to use the mammalian promoters. Mammalian promoters such as murine albumin promoter and the preproenkephalin promoter have been shown to yield region-specific and longterm expression in liver (10) and brain (11) respectively, after in vivo gene transfer. Whether mammalian promoters have a similar effect in the kidney has not been tested. The first step will be identification of the regulatory regions in the CA11 gene that promote transcriptional activity in renal tubular cells. In this study, we used deletion constructs with various lengths of CA11 5’ regions ligated to CAT reporter gene to evaluate the promoter activity. This information is important for the future construction of a mammalian promoter-based expression vector in the development of kidney-directed gene therapy. important
Methods Cell cultures. Renal proximal tubules were isolated using a protocol described by Vinay et al (12) and suspended in a serum-free medium containing 1: 1 mixture of Dulbecco’s Modified Eagle’s Medium (DMEM) and Ham’s F12 Medium supplemented with sodium bicarbonate (20 mM), Na$eO,SH,O (10 nM), insulin (5 pgiml), PGEl (25 ng/ml), triiodothyronine (5 PM), hydrocortisone (50 nM) and transferrin (5 pg/ml), and grown in a 6-well tissue culture dish precoated with collagen. The hormone-supplemented serum-free medium was used to suppress the growth of fibroblasts, but enhance the growth of renal tubular cells (13). The LLC-PK, cells, a well characterized pig renal proximal tubular cell line (14), were obtained from American Type Culture Collection and maintained in MEM supplemented with 9% newborn calf serum. Both mouse proximal tubular cells in primary culture and LLC-PK, cells express CAII (data not shown).
Plasmid construction. The construction of CAB-CAT was previously described by Shapiro et al. (15). Briefly, the human CA11 upstream region was ligated into the m III site of pSV2CAT plasmid. The resultant plasmid contains all SV40 promoter and enhancer sequences in addition to the CA11 upstream region. Deletion plasmids were then constructed by removal of sequences between the & I site and various sites within the 5’-flanking region. The resultant deletion plasmids contain neither SV40 promoter nor enhancer sequences so that the expression of CAT activity is totally driven by the 5’-CAB promoter regions. All the plasmids were purified by Qaigen column (Qaigen) and were RNA free and mostly in the supercoiled form as determined by agarose gel examination. The CMVPgal plasmid contains a promoter from cytomegalovirus and has been shown to express bacterial P-galactosidase in the kidney by liposome-mediated gene transfer (7). Transfections. Cells grown to 60% confluence in a 35 mm culture dish were co-transfected with 1 pg of CAB-CAT plasmid and 1 pg of pCMVPga1 plasmid to correct for the variation in transfection efficiency. DNA-1ipofectAMINE (Gibco) complex formation was optimized
Vol. 63, No. 2, 1998
Carbonic Anbydrase II Regulatory Regions
123
according to the manufacture’s suggestion. The cells were transfected with DNA:liposome complex in 1 ml of OPTI-MEM, serum free medium (Gibco) for 5 h at 37’ C in a CO, incubator, then switched back to their normal growth medium and incubated for a total of 48 h. Each lipofection experiment included a control dish transfected only with pCMVPga1 plasmid and a control dish without DNA. CAT activity was measured by a fluor-diffusion assay (16). Cells grown in a 35 mm culture dish were harvested with 300 pl of 0.1 M Tris-HCI , pH 7.4, 0.1% Triton X-100, frozen once in -70’ C freezer, rapidly thawed in 37’ C water bath, and centrifuged at 12,000 rpm for 5 min. at 4’ C. The supematant was aliquoted into two parts: one for pgalactosidase assay and the other for CAT assay. The unrelated cellular protein was inactivated by incubating the cell lysate at 65’ C for 10 min. After centrifuging at 12,000 rpm for 5 min , 25-50 pl of supematant was added to 150 pl of reaction mixture containing 0.25 pmole of chloramphenicol in 0.1 M Tris-HCl , pH 7.4, 250 nCi of ‘H-acetyl CoA (New England Nuclear), overlaid with 5 ml of ECONOFLUOR-2, and radioactivity is counted immediately and after 16 h incubation at room temperature. Each assay included a standard curve of 2.5 to 50 mu of CAT enzyme (Sigma), the linear range for these reactions, and a control sample without DNA transfection for background subtraction. The CAT activity was normalized to the P-galactosidase activity in order to correct for the variation in transfection efficiency. P-galactosidase activity was measured by Chlorophenol red-B-D-galactopyranoside (CRPG) method: the P-galactosidase reaction mixture contained 88 mM phosphate buffer, pH 7.4, 11 mM KCl, 1 mM MgCI,, 55 mM b-mercaptoethanol, 4.4 mM CRPG, and up to 70 pl of cell lysate in a total volume of 100 ~1. After incubation at 37’ C for 60 min, the reaction was diluted with 300 pl of water and the O.D. Each assay included a standard curve of 0.65 to 5 mu of P-gal enzyme read at 574 mn (17). (Sigma), the linear range for this reaction and a control sample without DNA transfection for background subtraction. Cell viability siua’y. Cell viability was measured in a parallel set of transfection assays by the XTT (2,3-bis[2-Methoxy-4-nitro-5-sulfophenyl]-2H-tetrazolium-5-carboxyanilide inner salt) assay (Sigma), in which XTT, a substrate for mitochondrial dehydrogenase, was metabolized to produce a color reaction measurable at a wavelength of 450 run (18). The XTT assay was performed according to the manufacturer’s instruction and the results were expressed as % of untreated cells. Statistics. The results were expressed as means + standard errors. Differences between groups were assessed with student’s t test. Significance levels were reported at the p ~0.01 and p < 0.05. Results Lipofection efficieency. Renal tubular cells, both primary cultures of mouse proximal tubular cells and LLC-PK, cells, were adequately transfected using the lipofection method. We found that the optimal ratio of DNA:liposome was found to be 1 ug:3 ul in renal tubular cells. The lipofection efficiency was 25-35% in both cell types. Viability of lipofected cells, The cytotoxicity of LipofectAMINE in either cell type was minimal as measured by XTT. The cell viability after transfection with LipofectAMINE was 95f3% for LLC-PK, and 92+5% for PTC cells (n=4). There were no statistical differences in cell viability among these groups. CAT activity induced by human CAIIpromoter regions. Fig. 1 shows the CAT expression by lipofection with various regions of CA II promoter in primary mouse proximal tubular cells (open bars) and the LLC-PK, cells (shaded bars) after correction with the transfection efficiency
Carbonic Anhydrase
124
Vol. 63, No. 2, 1998
11Regulatory Regions
measured from D-galactosidase activity. The -120001CAT and -1300lCAT induced the highest level of CAT expression in both cell types. In the primary mouse proximal tubular cells, the -180/CAT, -270lCAT and -420lCAT induced significant levels of expression (>lO fold increases compared with control), however, the levels were only 9, 12, and 9%, respectively, of that induced by the -1300/CAT (all p
8 _------
i
7 -1
C
12
/
1.3
I
Mouse PTI
0.42 0.27 0.18 kb
Fig. 1. CAT expression in primary mouse renal proximal tubular cells (PTC, open bars) and LLC-PKl cells (shaded bars) by lipofection with various regions of CAII promoter (represented in kilobase (kb) of 5’ human CAII sequence; C (control): results from mock transfected cells). Means and standard errors were calculated based on 3-4 independent experiments. p < 0.01, each group vs. C; p < 0.01, 0.42, 0.27, and 0.18 kb vs. 1.3 kb, and p > 0.05, 12 kb vs. 1.3 kb in both cell types.
Regulatory elements of human CAZZ gene. In order to identify regulatory elements, an extensive computer search of the 1.3 kb human CAII promoter region (gene bank accession #:M77176) was performed. We identified two putative transcription factor binding elements which are present in the -1300/CAT construct but not in the -420/CAT constructs: an Ap2 binding site (-551 through -543), which has a perfect consensus sequence, and an Apl binding site (-992 through -981), which is 89% matched.
Vol. 63, No. 2, 1998
Carbonic AnhydraseII Regulatory Regions
125
Discussion Our results indicate that the human CAII 1.3 kb 5’ sequence contains optimal promoter elements for activating transcription in renal tubular cells. Previously, the promoter region of CAB gene t?om different species has been studied in different cell lines. Shapiro et al (15) reported that as little as 200 base pairs of human 5’ flanking sequences allowed expression of CAT reporter gene in cultured murine L cells and human HeLa cells and addition of up to 1.4 kb of the promoter did not result in greater CAT activity. Marion (19) reported that 250 bp of the mouse CAII promoter is an efficient but not cell-specific promoter because it activate the expression of CAT in both HepG2 ( a CAB expressing cell line) and NIH3T3 (a CAB-negative cell line). The variations among reports are likely due to different cell types, which may have different transcription factors and signal transduction pathways. The computer search revealed AP2 and APl binding sites as potential transcription factor binding elements present only in the 1.3 kb sequence but not in the other sequences tested. Ap2 binding element is a CAMP and phobol ester-responsive element, and Ap2 protein is present in the kidney (20). Apl binding site is an element that binds a family of structurally and functionally related proteins encoded by several different proto-oncogenes, including c-Fos and c-Jun. It is now believed that multiple extracellular inputs, including exposure to growth factors, tumor promoters, Ca ionophores, and trans-synaptic signals, can induce various members of the API family and thereby elicit distinct responses at the level of gene expression (21). It is possible that these Apl and Ap2 sequences are required for maximal levels of CA II gene transcription in the renal epithelial cells. Further studies are needed to explore the role of these transcription factor binding elements on transcriptional activity of CAB promoter. In conclusion, transcriptional analyses of mammalian CAII promoter have revealed complex patterns of regulation. Our results indicate that the human CAB 1.3 kb 5’ sequence contains optimal promoter elements for activating transcription in renal tubular cells. This 1.3 kb fkagment will be tested as a promoter in our in vivo gene therapy model. Comparison of the expression patterns of the mammalianvs. viral- driven expression may reveal difference in duration, strength, and location of the gene expression. Moreover, since multiple transcription factor binding sites have been identified in this 1.3 kb promoter, it is possible to regulate the transgene expression through these potential regulatory elements in vivo. Acknowledgments This work was supported by the Southern Arizona Foundation Grant to L. Lai, and NIH Grant ROlDK52358 and a grant from the Dialysis Clinic, Inc., a non-profit organization, to Y.H. Lien. References 1. 2. 3. 4. 5. 6.
H.W. DAVENPORT and A.E. WILELMI, Proc. Sot. Exp. Biol. Med. 48 53-56 (1941). G. LONNERHOLM and Y. RIDDERSTRALE, Kidney Int. 17 162-174 (1980). Y. RIDDERSTRALE, P.J. WISTRAND, and R.E. TASHIAN, J. Histochem. Cytochem. 40 1665-1673 (1992). S. BRETON, S.L. ALPER, S.L. GLUCK, W.S. SLY, J.E. BARKER, and D. BROWN, Am. J. Physiol. 269 F761-774 (1995). W.S. SLY, D. HEWETT-EMMETT, M.P. WHYTE, Y .-S.L. YU, and R.E. TASI-IIAN, Proc. Natl. Acad. Sci.USA 80 2752-2756 (1983). S.E. LEWIS, R.P. ERICKSON, L.B. BARNETT, P.J. VENTA, and R.E. TASHIAN, Proc. Natl. Acad. Sci. USA 85, 1962-1966 (1988).
1%
8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21.
Chonic
Anhydrase 11Regulatory Regions
Vol. 63, No. 2, 1998
L. LAI, G.W. MOECKEL, and Y.H. LIEN, Gene Ther. 4 426-43 1 (1997). Y .H. LIEN and L. LAI, Exp. Nephrol. 5 132-l 36 (1997). Y.H. LIEN, and L. LAI, Kidney Int. 52. Suppl. 61 S85-88 (1997) M.A. KAY, Q. LI, T.J. LIU, F. LELAND, C. TOMAN, M. FINEGOLD, and S.L.C. WOO, Hum. Gene Ther. 3 641-647 (1992). M.G. KAPLITT, A.D. KWONG, S.P. KLEOPOULOS, C.V. MOBBS, S.D. RABKIN, and D.W. PFAFF. Proc. Natl. Acad. Sci. USA 91 8979-8983 (1994). P. VINAY, A. GOUGOUX, and G. LEMIEUX, Am. J. Physiol. 241 F403-F411 (1981) M. TAUB, and G.J. SATO, J. Cell. Physiol. 105 369-378 (1980) R.N. HULL, W.R. CHERRY, and G.W. WEAVER, In Vitro 12 670-677 (1976) L.H. SHAPIRO, P.J. VENTA, and R.E. TASHIAN, Mol. Cell. Biol. 7 4589-4593 (1987) J.R. NEUMANN, C.A. MORENCY, and K.O. RUSSIAN, Bio. Techniques 5 444-448 (1987) T. HOLLON, and F.K. YOSHIMURA, Anal. Biochem. 182 411-418 (1989) N.W. ROEHM, G.H. RODGERS, S.M. HATFIELD, and A.L. GLASBROOK. J. Immunol. Methods 142 257-265 (1991). L.R. MARINO, J. Biol. Chem. 268 7081-7089 (1993) P.J. MITCHELL, P.M. TIMMONS, J.M. HEBERT, P.W.J. RIGBY, and R. TJIAN, Genes Dev. 5 105-l 19 (1991) T. CURRAN and B.R. FRANZA JR, Cell 55 395-397 (1988).
ELSEVIER
PROMOTER
Life Sciences,Vol. 63, No. 2, pp. 121-126,1998 copyright0 1998 rGevicrsciem Inc. Printedin the USA. All tightsrerctvd 0024-3205/98 $19.00 + .oo
ACTIVITY OF CARBONIC ANHYDRASE II REGULATORY IN CULTURED RENAL PROXIMAL TUBULAR CELLS
REGIONS
Li-Wen Lai*, Robert P. Erickson’, Patrick J. Venta§, Richard E. Tashiann, and Yeong-Hau H. Lien* Department of *Medicine and #Pediatrics, University of Arizona Health Sciences Center, Tucson, Arizona 85724; “Department of Small Animal Clinical Sciences, Michigan State University, East Lansing, Michigan 48224; ‘Department of Human Genetics, University of Michigan Medical School, Ann Arbor, Michigan 48 109
(Received in fmal form April 17, 1998)
Summary Carbonic anhydrase II (CAII) plays an important role in the acid-base homeostasis of the body and its deficiency results in renal tubular acidosis. In order to identify the regulatory regions in the CAII gene for the future development of kidneytargeted gene therapy, we investigated the 5’ region of the gene for its promoter activity. Deletion constructs with various lengths of the 5’ flanking region of the human CAII promoter were ligated to the CAT reporter gene and lipofected in primary cultures of mouse proximal renal tubular cells and in cells of the established porcine proximal tubular cell line, LLC-PK,. The CAT activity was measured 48 hours after gene transfection. The -12000lCAT and -1300/CAT constructs expressed the highest CAT activity in both types of renal tubular cells (143- and 1X0-fold increase, respectively, in mouse proximal tubular cells; 50- and 70-fold increase, respectively, in LLC-PK, cells) but not the -420/CAT, -270/CAT, or -18O/CAT constructs (9, 12, and 9% of that of -1300/CAT construct, respectively, in mouse proximal tubular cells and, 23, 9, and 8%, respectively, in LLC-PK, cells, all p co.01 vs. -1300/CAT construct). No cytotoxicity was detected in the transfected cells. A computer search identified multiple putative transcription factor binding elements including Apl and Ap2 binding elements, which are present in the -13001CAT construct but not in the shorter constructs. In conclusion, we demonstrate that the human CAR 5’ sequence of proximal 1.3 kb contains strong promoter sequence(s) for renal tubular cells. Key Words: carbonic anhydrase, promoter activity, renal cells
Carbonic
anhydrase
(CA) catalyzes
the reversible
hydration
of carbon dioxide
and plays an
Corresponding Author: Yeong-Hau H. Lien, M.D., Ph.D. Section of Nephrology, Department of Medicine, University of Arizona Health Science Center, Tucson, 1501 N. Campbell Ave., Tucson, AZ 85724. Phone: 520-626-6370; Fax: 520-626-2024; E-mail: wna.edu
122
Carbonic Anhydrase II Regulatory Regions
Vol. 63, No. 2, 1998
role in acid-base homeostasis. CA activity in the mammalian kidney was first described in 1941 by Davenport and Wilelmi (1). CA11 is a cytoplasmic enzyme present in the proximal convoluted tubule, thick limb of Henle, and intercalated cells, but not principal cells, of collecting ducts (2. 3, 4). This pattern of expression suggests that regulation of CA11 expression is likely involved a variety of transcription factors. In human, CA11 deficiency is associated with a syndrome of renal tubular acidosis, osteopetrosis, and cerebral calcification (5). Lewis et al. produced a strain of CAII null mice by an induced mutagenesis that resulted in mice with growth retardation and renal tubular acidosis (6). Recently. we have used CA11 null mice as an animal model for testing gene therapy targeting to renal tubular cells (7, 8, 9). When CMV promoter was ligated to either b-galactosidase (7) or CA11 cDNA (8,9), compound with liposome, and injected into the renal pelvis of normal and CA11 deficient mice, respectively, the expression of the reporter genes last for 4-6 weeks. In order to further prolong the duration of the transgene expression, one strategy is to use the mammalian promoters. Mammalian promoters such as murine albumin promoter and the preproenkephalin promoter have been shown to yield region-specific and longterm expression in liver (10) and brain (11) respectively, after in vivo gene transfer. Whether mammalian promoters have a similar effect in the kidney has not been tested. The first step will be identification of the regulatory regions in the CA11 gene that promote transcriptional activity in renal tubular cells. In this study, we used deletion constructs with various lengths of CA11 5’ regions ligated to CAT reporter gene to evaluate the promoter activity. This information is important for the future construction of a mammalian promoter-based expression vector in the development of kidney-directed gene therapy. important
Methods Cell cultures. Renal proximal tubules were isolated using a protocol described by Vinay et al (12) and suspended in a serum-free medium containing 1: 1 mixture of Dulbecco’s Modified Eagle’s Medium (DMEM) and Ham’s F12 Medium supplemented with sodium bicarbonate (20 mM), Na$eO,SH,O (10 nM), insulin (5 pgiml), PGEl (25 ng/ml), triiodothyronine (5 PM), hydrocortisone (50 nM) and transferrin (5 pg/ml), and grown in a 6-well tissue culture dish precoated with collagen. The hormone-supplemented serum-free medium was used to suppress the growth of fibroblasts, but enhance the growth of renal tubular cells (13). The LLC-PK, cells, a well characterized pig renal proximal tubular cell line (14), were obtained from American Type Culture Collection and maintained in MEM supplemented with 9% newborn calf serum. Both mouse proximal tubular cells in primary culture and LLC-PK, cells express CAII (data not shown).
Plasmid construction. The construction of CAB-CAT was previously described by Shapiro et al. (15). Briefly, the human CA11 upstream region was ligated into the m III site of pSV2CAT plasmid. The resultant plasmid contains all SV40 promoter and enhancer sequences in addition to the CA11 upstream region. Deletion plasmids were then constructed by removal of sequences between the & I site and various sites within the 5’-flanking region. The resultant deletion plasmids contain neither SV40 promoter nor enhancer sequences so that the expression of CAT activity is totally driven by the 5’-CAB promoter regions. All the plasmids were purified by Qaigen column (Qaigen) and were RNA free and mostly in the supercoiled form as determined by agarose gel examination. The CMVPgal plasmid contains a promoter from cytomegalovirus and has been shown to express bacterial P-galactosidase in the kidney by liposome-mediated gene transfer (7). Transfections. Cells grown to 60% confluence in a 35 mm culture dish were co-transfected with 1 pg of CAB-CAT plasmid and 1 pg of pCMVPga1 plasmid to correct for the variation in transfection efficiency. DNA-1ipofectAMINE (Gibco) complex formation was optimized
Vol. 63, No. 2, 1998
Carbonic Anbydrase II Regulatory Regions
123
according to the manufacture’s suggestion. The cells were transfected with DNA:liposome complex in 1 ml of OPTI-MEM, serum free medium (Gibco) for 5 h at 37’ C in a CO, incubator, then switched back to their normal growth medium and incubated for a total of 48 h. Each lipofection experiment included a control dish transfected only with pCMVPga1 plasmid and a control dish without DNA. CAT activity was measured by a fluor-diffusion assay (16). Cells grown in a 35 mm culture dish were harvested with 300 pl of 0.1 M Tris-HCI , pH 7.4, 0.1% Triton X-100, frozen once in -70’ C freezer, rapidly thawed in 37’ C water bath, and centrifuged at 12,000 rpm for 5 min. at 4’ C. The supematant was aliquoted into two parts: one for pgalactosidase assay and the other for CAT assay. The unrelated cellular protein was inactivated by incubating the cell lysate at 65’ C for 10 min. After centrifuging at 12,000 rpm for 5 min , 25-50 pl of supematant was added to 150 pl of reaction mixture containing 0.25 pmole of chloramphenicol in 0.1 M Tris-HCl , pH 7.4, 250 nCi of ‘H-acetyl CoA (New England Nuclear), overlaid with 5 ml of ECONOFLUOR-2, and radioactivity is counted immediately and after 16 h incubation at room temperature. Each assay included a standard curve of 2.5 to 50 mu of CAT enzyme (Sigma), the linear range for these reactions, and a control sample without DNA transfection for background subtraction. The CAT activity was normalized to the P-galactosidase activity in order to correct for the variation in transfection efficiency. P-galactosidase activity was measured by Chlorophenol red-B-D-galactopyranoside (CRPG) method: the P-galactosidase reaction mixture contained 88 mM phosphate buffer, pH 7.4, 11 mM KCl, 1 mM MgCI,, 55 mM b-mercaptoethanol, 4.4 mM CRPG, and up to 70 pl of cell lysate in a total volume of 100 ~1. After incubation at 37’ C for 60 min, the reaction was diluted with 300 pl of water and the O.D. Each assay included a standard curve of 0.65 to 5 mu of P-gal enzyme read at 574 mn (17). (Sigma), the linear range for this reaction and a control sample without DNA transfection for background subtraction. Cell viability siua’y. Cell viability was measured in a parallel set of transfection assays by the XTT (2,3-bis[2-Methoxy-4-nitro-5-sulfophenyl]-2H-tetrazolium-5-carboxyanilide inner salt) assay (Sigma), in which XTT, a substrate for mitochondrial dehydrogenase, was metabolized to produce a color reaction measurable at a wavelength of 450 run (18). The XTT assay was performed according to the manufacturer’s instruction and the results were expressed as % of untreated cells. Statistics. The results were expressed as means + standard errors. Differences between groups were assessed with student’s t test. Significance levels were reported at the p ~0.01 and p < 0.05. Results Lipofection efficieency. Renal tubular cells, both primary cultures of mouse proximal tubular cells and LLC-PK, cells, were adequately transfected using the lipofection method. We found that the optimal ratio of DNA:liposome was found to be 1 ug:3 ul in renal tubular cells. The lipofection efficiency was 25-35% in both cell types. Viability of lipofected cells, The cytotoxicity of LipofectAMINE in either cell type was minimal as measured by XTT. The cell viability after transfection with LipofectAMINE was 95f3% for LLC-PK, and 92+5% for PTC cells (n=4). There were no statistical differences in cell viability among these groups. CAT activity induced by human CAIIpromoter regions. Fig. 1 shows the CAT expression by lipofection with various regions of CA II promoter in primary mouse proximal tubular cells (open bars) and the LLC-PK, cells (shaded bars) after correction with the transfection efficiency
Carbonic Anhydrase
124
Vol. 63, No. 2, 1998
11Regulatory Regions
measured from D-galactosidase activity. The -120001CAT and -1300lCAT induced the highest level of CAT expression in both cell types. In the primary mouse proximal tubular cells, the -180/CAT, -270lCAT and -420lCAT induced significant levels of expression (>lO fold increases compared with control), however, the levels were only 9, 12, and 9%, respectively, of that induced by the -1300/CAT (all p
8 _------
i
7 -1
C
12
/
1.3
I
Mouse PTI
0.42 0.27 0.18 kb
Fig. 1. CAT expression in primary mouse renal proximal tubular cells (PTC, open bars) and LLC-PKl cells (shaded bars) by lipofection with various regions of CAII promoter (represented in kilobase (kb) of 5’ human CAII sequence; C (control): results from mock transfected cells). Means and standard errors were calculated based on 3-4 independent experiments. p < 0.01, each group vs. C; p < 0.01, 0.42, 0.27, and 0.18 kb vs. 1.3 kb, and p > 0.05, 12 kb vs. 1.3 kb in both cell types.
Regulatory elements of human CAZZ gene. In order to identify regulatory elements, an extensive computer search of the 1.3 kb human CAII promoter region (gene bank accession #:M77176) was performed. We identified two putative transcription factor binding elements which are present in the -1300/CAT construct but not in the -420/CAT constructs: an Ap2 binding site (-551 through -543), which has a perfect consensus sequence, and an Apl binding site (-992 through -981), which is 89% matched.
Vol. 63, No. 2, 1998
Carbonic AnhydraseII Regulatory Regions
125
Discussion Our results indicate that the human CAII 1.3 kb 5’ sequence contains optimal promoter elements for activating transcription in renal tubular cells. Previously, the promoter region of CAB gene t?om different species has been studied in different cell lines. Shapiro et al (15) reported that as little as 200 base pairs of human 5’ flanking sequences allowed expression of CAT reporter gene in cultured murine L cells and human HeLa cells and addition of up to 1.4 kb of the promoter did not result in greater CAT activity. Marion (19) reported that 250 bp of the mouse CAII promoter is an efficient but not cell-specific promoter because it activate the expression of CAT in both HepG2 ( a CAB expressing cell line) and NIH3T3 (a CAB-negative cell line). The variations among reports are likely due to different cell types, which may have different transcription factors and signal transduction pathways. The computer search revealed AP2 and APl binding sites as potential transcription factor binding elements present only in the 1.3 kb sequence but not in the other sequences tested. Ap2 binding element is a CAMP and phobol ester-responsive element, and Ap2 protein is present in the kidney (20). Apl binding site is an element that binds a family of structurally and functionally related proteins encoded by several different proto-oncogenes, including c-Fos and c-Jun. It is now believed that multiple extracellular inputs, including exposure to growth factors, tumor promoters, Ca ionophores, and trans-synaptic signals, can induce various members of the API family and thereby elicit distinct responses at the level of gene expression (21). It is possible that these Apl and Ap2 sequences are required for maximal levels of CA II gene transcription in the renal epithelial cells. Further studies are needed to explore the role of these transcription factor binding elements on transcriptional activity of CAB promoter. In conclusion, transcriptional analyses of mammalian CAII promoter have revealed complex patterns of regulation. Our results indicate that the human CAB 1.3 kb 5’ sequence contains optimal promoter elements for activating transcription in renal tubular cells. This 1.3 kb fkagment will be tested as a promoter in our in vivo gene therapy model. Comparison of the expression patterns of the mammalianvs. viral- driven expression may reveal difference in duration, strength, and location of the gene expression. Moreover, since multiple transcription factor binding sites have been identified in this 1.3 kb promoter, it is possible to regulate the transgene expression through these potential regulatory elements in vivo. Acknowledgments This work was supported by the Southern Arizona Foundation Grant to L. Lai, and NIH Grant ROlDK52358 and a grant from the Dialysis Clinic, Inc., a non-profit organization, to Y.H. Lien. References 1. 2. 3. 4. 5. 6.
H.W. DAVENPORT and A.E. WILELMI, Proc. Sot. Exp. Biol. Med. 48 53-56 (1941). G. LONNERHOLM and Y. RIDDERSTRALE, Kidney Int. 17 162-174 (1980). Y. RIDDERSTRALE, P.J. WISTRAND, and R.E. TASHIAN, J. Histochem. Cytochem. 40 1665-1673 (1992). S. BRETON, S.L. ALPER, S.L. GLUCK, W.S. SLY, J.E. BARKER, and D. BROWN, Am. J. Physiol. 269 F761-774 (1995). W.S. SLY, D. HEWETT-EMMETT, M.P. WHYTE, Y .-S.L. YU, and R.E. TASI-IIAN, Proc. Natl. Acad. Sci.USA 80 2752-2756 (1983). S.E. LEWIS, R.P. ERICKSON, L.B. BARNETT, P.J. VENTA, and R.E. TASHIAN, Proc. Natl. Acad. Sci. USA 85, 1962-1966 (1988).
1%
8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21.
Chonic
Anhydrase 11Regulatory Regions
Vol. 63, No. 2, 1998
L. LAI, G.W. MOECKEL, and Y.H. LIEN, Gene Ther. 4 426-43 1 (1997). Y .H. LIEN and L. LAI, Exp. Nephrol. 5 132-l 36 (1997). Y.H. LIEN, and L. LAI, Kidney Int. 52. Suppl. 61 S85-88 (1997) M.A. KAY, Q. LI, T.J. LIU, F. LELAND, C. TOMAN, M. FINEGOLD, and S.L.C. WOO, Hum. Gene Ther. 3 641-647 (1992). M.G. KAPLITT, A.D. KWONG, S.P. KLEOPOULOS, C.V. MOBBS, S.D. RABKIN, and D.W. PFAFF. Proc. Natl. Acad. Sci. USA 91 8979-8983 (1994). P. VINAY, A. GOUGOUX, and G. LEMIEUX, Am. J. Physiol. 241 F403-F411 (1981) M. TAUB, and G.J. SATO, J. Cell. Physiol. 105 369-378 (1980) R.N. HULL, W.R. CHERRY, and G.W. WEAVER, In Vitro 12 670-677 (1976) L.H. SHAPIRO, P.J. VENTA, and R.E. TASHIAN, Mol. Cell. Biol. 7 4589-4593 (1987) J.R. NEUMANN, C.A. MORENCY, and K.O. RUSSIAN, Bio. Techniques 5 444-448 (1987) T. HOLLON, and F.K. YOSHIMURA, Anal. Biochem. 182 411-418 (1989) N.W. ROEHM, G.H. RODGERS, S.M. HATFIELD, and A.L. GLASBROOK. J. Immunol. Methods 142 257-265 (1991). L.R. MARINO, J. Biol. Chem. 268 7081-7089 (1993) P.J. MITCHELL, P.M. TIMMONS, J.M. HEBERT, P.W.J. RIGBY, and R. TJIAN, Genes Dev. 5 105-l 19 (1991) T. CURRAN and B.R. FRANZA JR, Cell 55 395-397 (1988).