EXPERIMENTALCELL

RESEARCH

186,385-389

(19%))

SHORT NOT Expression of Transferrin and Vitamin D-Bindin in an Osteogenic Sarcoma Cell Line GWENDOLYN S. ADRIAN,* FUNMEI YANG,* DANA T. GRAVES,? JAMESM. BUCHANAN,* AND BARBARAII. BOWMAN**~ *Department

of Cellular

and Structural Biology, The University of Texas Health Science Center at San Antonio, San Antonio, Texas 78284; and TBoston University School of Graduate Dentistry, Boston, Massachusetts 02118

Expression of genes encoding transferrin and the vitamin D-binding protein is described in a cell line, U2 OS, derived from a human osteogenic sarcoma. The mRNA transcripts of transferrin and vitamin D-binding protein were shown to be the lengths of those found in normal human liver. The cells synthesize and secrete the transferrin and vitamin D-binding proteins, in addition to human albumin and ceruloplasmin. The U-2 OS cells were successfully transfected with chimeric genes carrying 670 bp of the 5’ regulatory sequence of the human transferrin gene fused to a reporter chloramphenicol acetyltransferase gene. These data indicate that the appropriate transcriptional factors required for expression of four plasma proteins are produced by U-2 OS nuclei and that the U-2 OS cell line will be useful for studies analyzing regulation of these genes. 0 1990

Academic

Press,

Inc.

INTRODUCTION Genes encoding human plasma proteins serve as useful models for analyzing expression modulated by multiregulatory circuits such as inflammatory and malignant factors, hormones, heavy metals, and developmental signals. These modulatory events appear to mediate tissuespecific expression by activating transcriptional factors from cell nuclei to interact with conserved DNA sequences in the regulatory regions of appropriate genes [l]. Plasma proteins are thought to act for the most part systemically. Other systemically acting factors such as interleukin-1 [2] and insulin-like growth factor-l [3], however, have also been shown to act locally, having paracrine or autocrine functions. Thus the extrahepatic production of plasma proteins, particularly acute-phase

1To whom dressed.

reprint

requests

and correspondence should be ad385

proteins, may be significant. Transferrin an min D-binding protein are po and negative acutephase reactants, respectively. sferrin (TF)’ is sy of the testis [6], an thesized by brain [4,5], sertol lymphocytes [7]. Extrahepatic expression of the vitamin D-binding protein (DBP) in rat and mouse has recently been analyzed (Yang et al., in ~re~arat~~n~. Ceruloplasmin (CP) is expressed by liver and macrophages [S]. In order to further investigate the ~ot~~tia~ for extrahepatic synthesis of plasma proteins we have examined the production of transferrin and ~~~a~~~ D-binding protein in the U-2 OS cell line derived from differentiated osteogenic sarcoma [9]. IJ-2 been studied as a model for tbe expressio let-derived growth factor gene in bode-behaved cells [lo, 111. Northern hybridization analysis of mRNA from U2 OS cells has demonstrated tra~scr~~tio~ of transferrin and the vitamin D-binding mRNAs. ~~~~~oprecipitation of metabolically label ns from cells and media demonstrate that U-2 synthesize the transferrin and vitamin D-binding proteins, in addition to ceruloplasmin and albumin. Furthermore, to demonstrate that the cell line can be applie model to investigate gene regulation, transfection expression of a chimerit human TF gene fused to a successfully analyzed. EXP~R~~~NTAL

P

Isolation and blot hybridization of RNA front U-2 OS cells. U-2 OS cells were obtained from American Type Culture Collection (Rockville, MD 20852). Cellular RNA was extracted by bomogenization in guanidinium thiocyanate as described previously [12]. Poly(A)*RNA was obtained by fractionating total RNA using ohgo cellulose

’ Abbreviations used: Alb, albumin; CAT, chloramphenicol acetyltransferase; CP, ceruloplasmin; DBP, vitamin D-binding protein; kb, kilobase; kDa, kilodalton; M,, molecular weight; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel e~ectr~~bores~s~ TF, transferrin. 0014.#27/90 $3.QO Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.

386

SHORT

chromatography [13]. Northern blots were prepared as described by Lehrach et al. [14]. Immunoprecipitation and identification of plasma proteins synthesized by U-2 OS cells. U-2 OS cells (1.0 X 107) were cultured at 37°C in 4 ml of cysteine-free medium (MEM Select-Amine Kit, GIBCO, Grand Island, NY) for 1 h in the presence of 0.200 mCi cysteine, L[?S]-, 1013 Ci/mmol (NEN Research Products, Boston, MA). Cells were scraped from the culture flasks after two washes with Hank’s balanced salt solution. Cellular lysates of U-2 OS were obtained by suspension in 50 m&f potassium phosphate buffer, pH 7.5, and 1 mit4 2-mercaptoethanol, followed by sonication. The lysates were used for immunoprecipitation experiments. Radiolabeled medium in which the U-2 OS cells were cultured was lyophilized and resuspended in 250 ~1 of the immunoprecipitation buffer containing 1 mMphenylmethylsulfonyl fluoride as described by Silverstone et al. [ 151. Immunoprecipitations were performed as described by Adrian and Hutton [16]. The human plasma protein antibody preparations were obtained as follows: transferrin antiserum from U.S. Biochemical Corp. (Cleveland, OH); vitamin D-binding protein antiserum from Calbiochem (San Diego, CA), called anti-Gc globulin; ceruloplasmin antiserum from Calbiochem; albumin antiserum from BoehringerMannheim (Indianapolis, IN). Analysis of immunoprecipitated, radiolabeled plasma proteins was carried out on an 8% SDS-PAGE [17]. The radiolabeled electrophoretic bands were visualized by fluorography. Gels were treated with En3Hance (Du Pont, NEN, Boston, MA) and exposed to X-ray film. Visualization required l-4 days of exposure depending on the amount of [35S]cysteine incorporated into the synthesized proteins. Construction of a chimeric gene and analysis of CAT. The TF-CAT fusion gene used to transfect U-2 OS cells was constructed using a 5’ fragment of the human TF gene fused to the chloramphenicol acetyltransferase (CAT) gene. The details and background of the TF cDNA, genomic TF constructs, and TF-CAT plasmids are given elsewhere [l&20]. U-2 OS cells were cultured in minimal essential media: Waymouth’s 87/2 (GIBCO) 3:l v:v plus 10% fetal calf serum. The cells were transfected with 10 mg/plate of plasmid DNA by the Cas(PO& DNA coprecipitation technique [21]. Controls for CAT expression were transfected into U-2 OS cells and included pSV2CAT, a positive control that contains the ubiquitously expressing SV4O enhancer and promoter fused to the CAT gene. The negative control was TF(O.B7rev)CAT, a plasmid with the 0.67-kb TF 5’ sequence reversed in orientation (3’ to 5’ orientation instead of 5’ to 3’). Levels of CAT enzyme activity were determined by the procedure of Gorman et al. [22]. Radioactivity in free and acetylated forms of [14C]chloramphenicol was determined by scintillation counting of corresponding regions of thin-layer plates. Protein concentrations of the soluble extracts were determined by the method of Lowry et al. [23].

RESULTS

Transcription

of the TF and DBP Genes in U-2 OS Cells

Northern blot hybridizations of total RNA and poly(A)+RNA demonstrated the transcription of human genes TF and DBP in U-2 OS cells. Figure 1 shows results from hybridization of poly(A)+ prepared from U-2 OS cells and radiolabeled human TF cDNA. The size of U-2 OS TF mRNA transcripts corresponded to that of mRNAs from human and mouse liver and was approximately 2.3 kb. In Fig. 1 the 2.3-kb TF mRNA band appears along with an additional, fainter 4.4-kb band that is nonspecific hybridization with 28 S ribosomal RNA. This band is increased in electrophoresis of total RNA (Fig. 2, lanes 4 and 6). The 28 S band also appears in

XT--lVUlI.!d

123

4.4Kb 2.4 Kb

1.4 Kb

FIG. 1. Detection of transferrin mRNA in U-2 OS cells by Northern hybridization analysis. Ten micrograms of poly(A)+ RNA prepared from U-2 OS cells (lane l), human liver (lane 2), and mouse liver [lane 3) was hybridized with 32P-labeled human transferrin cDNA in this analysis. Molecular weight markers are on the right side of the autoradiograph.

poly(A)) RNA preparations used as negative controls. Figure 2 compares the migration of DBP mRNA from human liver, U-2 OS cells, and a human hepatoma cell line (Hep 3B). The 1.8-kb DBP mRNA was found in U2 OS, the positive controls-human liver and hepatoma cells, but not in the negative control-HeLa cells, a line that does not synthesize the vitamin D-binding protein, The faint bands appearing in Fig. 2 are either degradation products of ribosomal RNA or incompletely processed DBP mRNA precursor. Detection of Transferrin, Vitamin D-binding Ceruloplasmin, and Albumin Synthesis by U-2 OS Cells

Protein,

To ascertain whether or not specific plasma proteins were synthesized by U-2 OS cells, experiments were designed to identify radiolabeled plasma proteins in immunoprecipitates of lysates and supernatants. Radiolabeled cysteine was added to tissue culture medium that contained no other source of this amino acid. Supernates and cell lysates were immunoprecipitated with rabbit antisera to human vitamin D-binding protein, cerulo-

SHORT

123456 1

!

’

-4.4

387

NOTE

bumin was found in cell extracts and in the tissue culture to approximately 68 media. Its migration correspo in the tissue culture kDa. Ceruloplasmin was seer media and migrated as a protein of greater than 100 k e concentraProteins which were not sized in detec tions in U-2 OS extracts or in tissue c ai-glyco eluded a,-HS-glycoprotein, globulin, and lactoferrin. The Uessiom of vitaparticularly suited for studying min D-binding protein since an (SaOS-2) and normal human o from Dr. Marian F. Young) di (data not shown). Transfection

of U-2 OS Cells with t

The U-2 OS cell line was tr lowing plasmids: pTF(0.67 kb pTF(0.67 rev)CAT. Transient expression of the human transferrin chimeric gene TF(0.6 by CAT enzymatic assay (Fig. 4). demonstrated by the positive control, SVXXT; moder-

FIG. 2. Detection of DBP mRNA in U-2 OS cells by Northern hybridization analysis. Ten micrograms of RNA was loaded in each lane. The RNA samples include total RNA from human liver (lane 1); total RNA (lane 2), and poly(A)+ RNA (lane 3) from U-2 OS cells; total RNA (lane 4) and poly(A)+ RNA (lane 5) from hepatoma 3B cells (lane 6), and total RNA from HeLa cells. Molecular weight markers are placed to the right of the autoradiograph. Longer exposure of the Northern blot revealed a 1%kb band in total RNA from U-2 OS and Hep 3B cells, but not in RNA from HeLa cells.

plasmin, transferrin, and albumin, respectively. The immunoprecipitated, radiolabeled proteins were analyzed by SDS-PAGE under reducing conditions. Specificity of the antisera to each of the four proteins was confirmed by competition experiments in which addition of nonradiolabeled transferrin, vitamin D-binding protein, and albumin were added to each of the respective antisera before immunological reaction with the radiolabeled proteins. A diminution in the radiolabeled band after competition confirmed specificity. The anti-ceruloplasmin antibody preparation was tested by competition with nonlabeled ceruloplasmin and was also shown to be specific (results not shown). As demonstrated in Fig. 3, a radiolabeled band corresponding to transferrin was observed in the cell extracts and tissue culture media of U-2 OS cells. The relative migration of transferrin corresponded to M, of approximately 79,000, as calculated by migration of molecular weight markers. In tissue culture media the vitamin Dbinding protein was secreted into the tissue culture media by U-2 OS cells and appeared as an immunoreactive radiolabeled band of approximately 51 kDa. Human al-

FIG. 3. Synthesis of transferrin (TF), vitamin (DBP), human serum albumin (HSA), and ceruioplasmin (CP) by U2 OS ceils demonstrated by electrophoresis of radiolabeled immunoprecipitates in cell extracts and in tissue culture media. & m glob, & microglobulin; 01~HSGP, u,-HS-glycoprotein; cyl GP, q glycoprotein; LF, lactoferrin. Molecular weight markers are shown to the right of migration of cell extracts. Ni, nonimmune serum; aTF, aDBP, aCP, aHSA, etc., antiserum directed against each of these proteins; TF + aTF; DBP + aDBP; HSA + aHSA, competition of antiserum against each protein with its corresponding ~o~ra~io~abeled antigen. DBP and CP were mainly secreted by cells into the media. No a2HS glycoprotein, or,-glycoprotein, µglobulin, or lactoferrin was detected in cells or media.

388

SHORT

1,3.-di,oAckl

34d4C1

Chl

Chl

l-OAc[‘4Ckh,

k4Cl Chl

FIG. 4. Enzymatic activity of CAT detected in U-2 OS cells transfected with positive control pSV2CAT, pTF(0.67)CAT and negative control pTF(O.67 reversed). Expression of the CAT enzyme was measured as radioactivity in 3-OAc-[r4C]chloramphenicol product/500 pg protein. The radioactivity of the U-2 OS cell extracts transfected with pSV2CAT was 40,932 cpm/500 pg; transfected with pTF(0.67)CAT was 3548 cpm/500 pg; and transfected with pTF(0.6’7 reversed) was 62 cpm/500 pg protein. Abbreviations used are [i4C]Chl, [‘%]chloramphenicol; l-OAc[r%], [14C]chloramphenicol acetylated in the 1 position, 3-OAc[r4C]Chl, [14C]chlorsmphenicol acetylatedin the 3 position and 1,3-dioAc[r4C]Chl, [i4C]chloramphenicol diacetylated in the 1,3 positions.

ate expression could be observed by the transferrin chimeric gene, TF(O.G7)CAT, and no CAT activity could be detected in the negative control TF(O.67reu)CAT, in which the 5’ regulatory sequence of the TF gene was reversed in the plasmid. Expression of SVZCAT by the U2 OS cells indicated an effective transfection. DISCUSSION Tumor cells often constitutively express factors that are only transiently expressed in their normal counterpart. Sporn and Roberts [24] have suggested that the production of such factors is carefully regulated in nontransformed cells, but that their expression becomes deregulated during malignant development. The production of regulatory factors by transformed cells has facilitated the study of several protein hormones including transforming growth factor p [25], transforming growth factor 01 [26], and interleukin-2 [27]. In addition, the constitutive synthesis of regulatory factors by malignant cells has provided the conceptual basis for investigating autocrine stimulation of cell behavior [28]. Thus, tumorderived cells provide a useful model for studying the production of regulatory factors that would be difficult to examine in their normal counterpart.

NOTE

U-2 OS cells have been studied as a model for the expression of the platelet-derived growth factor genes in bone-derived cells [29]. Studies presented here indicate that these cells also serve as a useful model for the expression of the plasma proteins, transferrin, and vitamin D-binding protein. The work described demonstrates expression of transferrin and the vitamin D-bindingprotein at the transcriptional and protein level. The transcripts of the TF and DBP genes were examined by Northern hybridization and found to be the same length in U-2 OS cells as in liver. On the protein level of expression, transferrin and the vitamin D-binding protein were found to be synthesized by U-2 OS cells and secreted into the tissue culture media. Synthesis and secretion of ceruloplasmin and albumin were also detected in U-2 OS cells. To demonstrate the efficacy of U-2 OS cells as a model for studying gene expression, transfection and expression of a TF-CAT chimeric gene were demonstrated. Although speculative, the extrahepatic synthesis and secretion of transferrin and vitamin D-binding protein may have a significant local regulatory function in bone. Since cellular proliferation may be diminished if transferrin concentration is limited [30], a localized, extrahepatic source of transferrin could facilitate growth in developing bone, where the rate of cellular activity is high. As with insulin, virtually every cell type examined has been found to respond to transferrin in serum-free medium [31]. Similarly, the local production of vitamin Dbinding protein could be significant in bone. It is well known that the activity of regulatory factors such as IGF-1 and TGF-P is modulated by binding proteins. Since bone is a target of vitamin D action, the local production of vitamin D-binding protein by bone cells could potentially modulate vitamin D activity and consequently affect bone metabolism. Thus, the production of transferrin and vitamin D-binding protein by bone cells could potentially play an important role in locally regulating bone growth or metabolism. The results of this study suggest that synthesis of transferrin and vitamin D-binding protein may have the potential for paracrine or autocrine function, particularly in bone tissue. The U-2 OS cells have the capacity to express both endogenous plasma proteins as well as a transfected chimeric transferrin gene, indicating that appropriate transcriptional factors are produced by the nuclei of these cells. The U-2 OS cell line is easy to transfect in contrast to some other cell lines and lends itself to gene regulation studies. This provides the opportunity to identify truns-acting factors in U-2 OS cells and to determine if they differ from those which regulate the expression of the same genes in liver cells, studies currently in progress. We thank Dr. Marian F. Young, NIH-NIDR, Bethesda, Maryland, for her generous gift of a culture of primary human osteoblast cells. This work was supported by grants from the American Heart Associa-

SHORT tion-Texas Affiliate, Grant 86R-371 and NIH DE08569, AG06650, AG06872, and AG00165.

Grants

GM33298,

NOTE 14. 15.

REFERENCES 1.

Maniatis, T., Goodbourn, 236,1237-245.

2.

Dinarello, C. A. (1988) PASEB J. 2,108. Nissley, S. P., and Rechler, M. M. (1984) Harm. Proteins

3.

S., and Fisher,

16. J. A. (1987) Science

Pep. 19.

X2,128-203. 4.

Bloch, B., Popovici, T., Levin, M. J., Tuil, D., and Kahn, (1985) Proc. Natl. Acad. Sci. USA 82,6706-6710.

A.

5.

Dickson, P. W., Aldred, A. R., Marley, P. D., Tu, G. F., Howlett, F. J., and Schreiber, G. (1985) Biochem. Biophys. Res. Commun.

6.

Skinner, N. K., and Griswold, 9523-9525.

7.

Lum, 9. B., Infante, A. J., Makker, D. M., Yang, F., and Bowman, B. H. (1986) J. Clin. Inuest. 77,841-849.

8.

Yang, F., Naylor, S. L., Lum, J. B., Cutshaw, S., McCombs, J. S., Naberhaus, K. H., McGill, J. R., Adrian, G. S., Moore, C. M., Barnett, D. R., and Bowman, B. H. (1986) Proc. N&Z. Acad. Sci. USA 83,3257-3261.

127,890~895.

9.

17. 18.

M. D. (1980) J. Biol. Chem. 255,

Ponten, J., and Saksela, E. (1967) Int. J. Cancer 2,434-447.

20.

21. 22. 23. 24. 25.

10.

Graves, D. T., Owen, A. J., Barth, R. K., Tempst, P., Winoto, A., Fors, L., Hood, L. E., and Antoniades, H. N. (1984) Science 226, 972-974.

26. 21.

Il.

Betsholtz, C., Johnsson, A., Heldin, 6. H., Westermark, B., Lind, P., Urdea, M. S., Eddy, IX., Shows, T. B., Philpott, K., Mellor, A. L., Knott, T. J., and Scott, J. (1986) Nature (London) 320, 695-699.

28.

12.

Chirgwin, J. ha., Przybyla, A. E., MacDonald, W. J. (1979) Biochemistry X8,5294-5299.

13.

Aviv, H., and Leder, P. (1972) Proc. Natl. Acad. Sci. USA 69, 408-1412.

Received July 4,1989 Revised version received September

13,1989

29.

R. J., and Rutter, 30. 31.

38 Lehrach, H., Diamond, D., Wozney, J. M., and Boedtder, H. (197’7) Biochemistry l&4143-4151. SiIverstone, A., Sun, L., Witte, 0. N., end Baltimore, D. (1930) J. Biol. Chem. 255,791-796. Adrian, 6. S., and Hutton, J. 9. (1983) J. C2in. Invest. 7’6,16491660. Laemmli, U. K. (1970) Nature (Londan,J 229,680-685. Yang, F., Lum, J. B., McGill, J. R., Moore, C. VanBragt, P. H., Baldwin, W. D., and Bowman, B. Ii. (1984) Proc. Natl. Acad. Sci. USA 81,7875-7819. Adrian, 6. S., Korinek, B. W., Bowman, B. H., and Yang, F. (1986) Gene 49,167-175. Adrian, 6. S., Riebl, R., Herbert, D. C., Weaker, F. J., Adrian, E. K., Robinson, L. K., Walter, C. A., Eddy, C. A., Pauerstein, H. (1989) in Molecular Biology C. J., Yang, F.; and Bowman, of Aging: UCLA Symposia, A. R. Liss, New York. Graham, F. L., and van der Eb, A. J. (1973) firobogy 52,456467. Gorman, C. M., Moffat, L. F., and Howard, 6. H. (1982) MOE. Cell. Biol. 2, 1010-1051. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. 9. (1951) J. BioL Chem. 193,265-2%. Sporn, M. B., and Roberts, A. B. (1986) J. Ciin. Pnuest. 78,329332. Sporn, M. B., Roberts, A. B., Wakefield, L. M., and de Crombrugghe, B. (1987) J. Ceil Biol. X05,1039-1045. Derynck, R. (1988) Cell 54,593-595. Robb, R. J., Kutny, R. M., Panico, M., Morris, H. R., and Chowdhry, V. (1984) Proc. Natl. Acad. Sci. USA 3 1 p 6486-6490. Sporn, M. B., and Todaro, G. J. (1986) N. ET&. J. Med. 303, 878-880. Graves, D. T., Valentin-Opran, A., Del Mundy, G. R., and Picbe, J. E. (1989) press. Tormey, D. C., Imrie, R. C., and Mueller, G. C. (1972) Exp. Cell Res. 44,163-169. ) 649-655. Barnes, D., and Sato, G. (1980) CelE