148
Biochimica et BiophvsicaActa 825 (1985) 148-153 Elsevier
BBA91466
E x p r e s s i o n of carbamoyl-phosphate synthetase I m R N A in Reuber hepatoma H - 3 5 cells. Regulation by glucocorticoid and insulin Y a s u o K i t a g a w a *, J e r e m i a h Ryall, M a i N g u y e n a n d G o r d o n C. S h o r e Department of Biochemistry, Mclntyre Medical Sciences Building, McGill University, Montreal, H3G 1 Y6 (Canada) (Received January 14th, 1985)
Key words: Carbamoyl-phosphate synthetase I; Glucocorticoid; Insulin; mRNA; Hormone regulation; (Reuber hepatoma H-35 cell)
Reuber hepatoma H-35 cells actively synthesize the urea cycle enzyme, carbamoyl-phosphate synthetase 1. Treatment of H-35 cells with dexamethasone (0.14 #M), however, enhanced synthesis of the enzyme (as measured by incorporation of [3SS]methionine) by 4-5-fold. Insulin (0.18 pM) completely inhibited dexamethasone-dependent stimulation of enzyme synthesis. In vitro translation and cDNA hybridization assays were employed to measure effects of dexamethasone plus or minus insulin on levels of mRNA encoding the biosynthetic precursor of carbamoyl-phosphate synthetase i (pCPS) in Reuher H-35 cells. Both measurements yielded similar results: dexamethasone increased pCPS mRNA levels by 4-5-fold and insulin suppressed this response, but only by 50%. Specific cDNA hybridization assays also demonstrated that Reuber H-35 cells, even after hormone treatments, contain only very low levels of albumin mRNA, and no detectable ornithine carbamoyl-transferase mRNA.
Introduction Carbamoyl-phosphate synthetase I (EC 6.3.4.16; monomeric M r 160000) is a mitochondrial matrix enzyme catalyzing the first step of the urea cycle [1,2]. The enzyme is most abundant in hepatocytes, where it constitutes about 4% of total liver protein [3]; intestinal mucosal epithelial cells contain less than 10% of the level found in liver, whereas in other tissues the enzyme is essentially undetectable (Ref. 4 and unpublished results). Carbamoyl-phosphate synthetase I is encoded by a nuclear gene and, therefore, its primary translation product is transported into mitochondria follow* On leave from permanent address: Institute for Biochemical Regulation, School of Agriculture, Nagoya University, Nagoya 464, Japan. Abbreviations: pCPS, precursor to carbamoyl-phosphate synthetase; pOCT, precursor to ornithine carbamyl transferase; SDS, sodium dodecyl sulfate.
ing synthesis by 80 S cytoplasmic ribosomes (reviewed in Ref. 5). The overall process involves synthesis of a higher molecular weight precursor polypeptide (pCPS, M r 165000) containing an amino-terminal extension which functions to target pCPS to the surface of the organelle. This is immediately followed by transmembrane import of the precursor into the mitochondrial matrix where proteolytic processing takes place to yield mature enzyme [5-7]; targeting, import and processing in vivo takes about 2 min to complete. Besides being expressed in a tissue-specific manner, the nuclear gene encoding pCPS is also subject to a variety of developmental, hormonal and dietary controls. Regarding regulation by hormones, the evidence indicates that control resides with glucagon and/or glucocorticoid [9-16]. Hormonal studies have been somewhat hampered, however, because of inherent difficulties encountered with the various experimental systems: whole
0167-4781/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)
149
or adrenalectomized animals, or primary hepatocyte cultures whose growth characteristics may not accurately reflect the in situ condition. Hepatoma cell lines offer the advantage of an in vitro system but, of course, they may also express a variety of neoplastic abnormalities. Morris hepatoma 5123D cells, for example, express high levels of pCPS mRNA [8] but do so in a constitutive manner which is unresponsive to hormone [16]. Both expression and regulation of carbamoyl-phosphate synthetase I appear to have been retained, however, by Reuber hepatoma H-35 cells; in these cells, synthesis of the enzyme is stimulated by glucocorticoid [13]. In the present study, we have employed a pCPS clone to further investigate hormonal regulation of pCPS mRNA levels in Reuber hepatoma H-35 cells. We show that H-35 ceils express pCPS mRNA at levels which are comparable to those found in the liver of intact animals, and they do so without exogenous addition of hormone. In the presence of glucocorticoid (dexamethasone), however, the level of pCPS mRNA is rapidly stimulated by 4-5-fold; insulin, at physiological concentrations, inhibits this dexamethasone effect by 50%, even though it completely blocks dexamethasone-dependent stimulation of enzyme biosynthesis. Finally, we have also investigated the expression in Reuber H-35 cells of mRNAs encoding two other proteins made in liver, serum albumin and a mitochondrial enzyme catalyzing the second step of the urea cycle, ornithine carbamoyl-transferase. In contrast to pCPS mRNA, both albumin and pOCT mRNAs are expressed at very low levels in these cells.
Experimentalprocedures General. Routine procedures were performed as previously described (Refs. 5 and 8 and references cited therein). They include measurements of protein and RNA, isolation of total and poly(A) ÷RNA, in vitro translation in a messenger-dependent rabbit reticulocyte cell-free system, immunoprecipitation of specific polypeptides made in vivo and in vitro, and SDS-polyacrylamide gel electrophoresis and fluorography of dried gels. Whole livers were obtained from adult male Sprague-Dawley rats (Canadian Breeding Farms, Montreal).
Cell cultures. Reuber hepatoma H-35 cells were maintained in Eagle's minimal essential medium containing 6% fetal bovine serum (GIBCO) at 37°C in a humidified atmosphere of 5% CO 2 and 95% air until they reached about 80% confluency, The culture medium was then replaced with serum-free Eagle's minimal essential medium without additives (control) or containing dexamethasone (0.14 #M), either alone or in combination with insulin (0.18/~M). Cultures were maintained for 13 h prior to labeling with [35S]methionine or harvesting for RNA extraction. In oioo labeling of Reuber H-35 cells. H-35 cells were maintained in microtiter wells and were labeled with 1.42 m C i / m l [35S]methionine (Amersham) for 30 min at 37°C as described previously [13,17]. Labeling was terminated by the addition of 950 gl of phosphate-buffered saline containing 1% Triton and 20 mM unlabeled methionine. Cell lysates were obtained and clarified by centrifugation at 45 000 rpm for 45 rain in a Beckman Ti 75 rotor. Supernatants were subjected to immunoprecipitation with the appropriate monospecific antiserum. Antibodies. Antisera monospecific for rat carbamoyl-phosphate synthetase I, ornithine carbamyl transferase, and serum albumin were from the same preparations which were employed and characterized in earlier studies (refs. 18, 19 and 20, respectively). eDNA clones. Cloned pCPS and albumin cDNAs were characterized in Ref. 8. For use as probes, a 1.3 kb (5')HindlII/PstI(3') fragment from the pCPS eDNA insert of clone pKB4 and a 1.0 kb PstI-excised insert from the albumin eDNA clone were employed. Rat liver pOCT eDNA was cloned (Nguyen, M. and Shore, G.C., unpublished results) in pBR322 employing the DNA polymerase I/RNAase H / D N A ligase procedure for second-strand eDNA synthesis [21]. One such clone contained a 1.3 kb insert in the PstI site of pBR322 and was identified (unpublished data) as encoding pOCT sequences by hybrid-selected translation and by Sanger dideoxy sequencing which identified published regions of rat [22] and human [23] liver pOCT nucleotide sequences. For use as a probe, a 0.6 kb (5')PstI/HindlII(Y) fragment was obtained from the 5'-end of the 1.3 kb insert. All eDNA probes were labeled by nick translation in the
150 presence of a-[32p]dCTP and a-[32p]dGTP [24]. R N A extraction. Cells were grown in Falcon T175 flasks and then maintained in Eagle's minimal essential medium for 13 h either alone or with dexamethasone (0,14 /~M) plus or minus insulin (0.18 /~M). The culture medium was then replaced with ice-cold phosphate-buffered saline containing 100 /~g/ml cycloheximide, cells were scraped with a rubber policeman, and recovered by centrifugation at 1200 × g for 5 rain. Three flasks were employed per treatment. Total R N A was isolated by the guanidium-isothiocyanate/ CsC1 procedure [27]. R N A dot-blot hybridizations. Denaturation of total R N A and subsequent application to nitrocellulose paper was carried out as described [8]. Prehybridization was performed overnight at 65°C in 3 × SSC, 1 0 x Denhardt's solution [25], 0.1% SDS, 50 # g / m l denatured herring sperm DNA, and 1 0 / ~ g / m l poly(A). Hybridization and washing was performed according to Andrews et al. [26]. Results and Discussion Fig. 1 shows that exposure of Reuber hepatoma H-35 cells to the glucocorticoid analogue, dexamethasone, resulted in a significant stimulation of synthesis of carbamoyl-phosphate synthetase I (about 4-5-fold, based on incorporation of [35S]methionine). The duration of hormone treatment in this experiment was 13 h; maximal stimulation of synthesis is normally attained, however, by 6 - 7 h after the addition of hormone, and continues for at least 24 h (the longest time-period tested) (data not shown). Maximal stimulation was obtained at a relatively low [28,29] concentration of dexamethasone, 0.14/tM. Over the course of the hormone treatments described in Fig. 1, cell proliferation was relatively unaffected; total incorporation of [ 35S]methionine, however, increased by about 2-fold (not shown), so that the absolute stimulation of carbamoyl-phosphate synthetase biosynthesis (per cell) was about 8-10-fold. Interestingly, insulin (0.18/~M), which had no measurable effect on total incorporation of [35S]methionine into protein (not shown), largely suppressed the dexamethasone-dependent stimulation of enzyme synthesis (Fig. 1). When added alone, insulin was without effect on carbamoyl-
a
b
c
d
e
f
CPS'-
Fig. 1. Effect of dexamethasone and insulin on synthesis of carbamoyl-phosphate synthetase I in Reuber hepatoma H-35 cells. Cells were grown in microtiter wells, treated with or without hormones for 13 h, and labeled with [35S]rnethionine as described in Experimental procedures. Aliquots of total cell lysates, containing 0.5.10 6 cpm of [35S]methionine incorporated into protein, were subjected to immunoprecipitationwith anti-carbamoyl-phosphatesynthetase. Immunoprecipitateswere resolved by SDS-polyacrylamidegel electrophoresisand fluorographed. Duplicate cultures of cells were analyzed. Lanes a and b, untreated H-35 cells; lanes c and d, H-35 ceils treated for 13 h with dexamethasone (0.14 #M); lanes e and f. H-35 cells treated for 13 h with dexamethasone (0.14 /~M) and insulin (0.18/tM). CPS, carbamyl phosphate synthetase.
phosphate synthetase biosynthesis in both Reuber hepatoma cells (not shown) and normal hepatocytes [16]. To further elucidate the nature of the dexamethasone and insulin effects on the biogenesis of carbamoyl-phosphate synthetase I in Reuber H-35 cells, we have employed translational and c D N A hybridization assays to measure changes in m R N A levels. In vitro translation The guanidium isothiocyanate/CsC1 procedure was employed to extract total R N A from Reuber H-35 cells maintained for 13 h in the presence or absence of either dexamethasone (0.14 #M) or
151
dexamethasone (0.14/~M) plus insulin (0.18/~M). In vitro translation was performed in the rabbit reticulocyte cell-free system, with each RNA preparation yielding similar incorporation of [35S]methionine into total polypeptide product. Equal quantities of radioactive product were subjected to immunoprecipitation with monospecific antibody against carbamoyl-phosphate synthetase I or rat serum albumin, and the primary translation products were visualized following SDS-poly-
a
b
cd
pCPS -
acrylamide gel electrophoresis and fluorography (Fig. 2). For comparisons, total RNA from whole livers of adult Sprague-Dawley rats was also employed in these assays. At least based on translational activity in vitro, the results show that the relative pCPS mRNA content of Reuber H-35 ceils is very similar to that of normal rat liver (Fig. 1, lanes a and b). It is stimulated about 4-fold by dexamethasone (lane c) and this stimulation is suppressed, though not entirely, by insulin (lane d). Fig. 2 also demonstrates that, whereas Reuber cells apparently retain normal expression of caramoyl-phosphate synthetase I, they contain drastically reduced amounts of mRNAs encoding other hepatic proteins, including serum albumin (Fig. 2, lower panel) and an enzyme catalyzing the second step of the urea cycle, ornithine carbamyl transferase (see below). cDNA hybridization assays The estimates obtained from in vitro translational assays for relative levels of mRNAs encoding carbamoyl-phosphate synthetase I, serum albumin and ornithine carbamyl transferase in normal liver and Reuber H-35 cells were confirmed by RNA dot-blot hybridization employing specific cDNA probes (Figs. 3 and 4). In particular, direct
ALB" L IVER H-35 Fig. 2. Effect of dexamethasone and insulin treatment on levels of translatable pCPS and albumin mRNA in Renber hepatoma H-35 cells. Total cellular RNA was translated in a rabbit reticulocyte cell-free system in the presence of 1.0 mCi/ml [35S]methionine ( ~ 1 0 0 0 Ci/mmol). Sources of RNA were: normal adult rat liver (lane a); untreated Reuber hepatoma H-35 cells (lane b); dexamethasone-treated H-35 cells (lane c); and H-35 cells treated with dexamethasone and insulin (lane d). Aliquots from each translation containing 2.10 .6 cpm incorporated into protein were immunoprecipitated with either anticarbamoyl-phosphate synthetase (upper panel) or anti-albumin (lower panel). Immunoprecipitates were resolved by electrophoresis in a 10% SDS-polyacrylamide gel and visualized by fluorography. The bands appearing below the main pCPS band disappeared when immunoprecipitation was performed in the presence of excess enzyme and, therefore, they probably represent premature translation products of pCPS mRNA. pCPS, precursor to carbamoyl-phosphate syntlietase; ALB, preproalbumin.
H-35 dex
H-35 dex/ins 16
8
4
2
1
IJg RNA Fig. 3. Relative levels of pCPS mRNA in normal adult fiver and hormone-treated Renber hepatoma H-35 cells. Serial dilu. tions of denatured total cellular RNA from the sources indicated were spotted onto nitrocellulose and hybridized to a a2P-labeled pCPS eDNA probe (5.106 cpm/ml; 3.10 a cpm//~g DNA), as described in Experimental procedures. H-35, untreated Reuber hepatoma H-35 cells; H-35 (dex), dexamethasone-treated H-35 cells; H-35 (dex/ins), dexamethasone/insulin-treated H-35 cells.
152 ALB
pOCT
LIVER H-35 H-35 dex H-35 dex/ins 16
8
4
2
16
8
4
2
tjg RNA
Fig. 4. Relativelevelsof albumin and pOCT mRNA in normal adult liver and hormone-treatedReuber hepatoma H-35 cells. RNA dot-blot hybridizations were carried out as outlined in Fig. 3, except that 32p-labeled pOCT and albumin cDNA probes were employed(see Experimentalprocedures).Abbreviations are given in the legend to Fig. 3. quantitation of dot-blots (Fig. 3) indicated that dexamethasone treatment of Reuber H-35 cells resulted in a 4.5-fold increase of pCPS mRNA, relative to total mRNA; insulin blocked this increase by about 50% (Fig. 3). Northern hybridization analysis gave the same quantitative results as those obtained by RNA dot-blots, and also showed that Reuber and normal liver pCPS mRNA are the same size (6.0 kb, data not shown; see also ref. 8). Reuber cells, with or without hormone treatment, were largely inactive with respect to expressing albumin and pOCT mRNA (Fig. 4).
Concluding remarks The present results show that pCPS gene expression in Reuber hepatoma H-35 cells responds to relatively low concentrations of the glucocorticoid analogue, dexamethasone (Figs. 1-3). In this regard, the hepatoma system acts in a manner similar to normal livers from adrenalectomized animals [9], despite the fact that Reuber cells demonstrate a variety of neoplastic abnormalities, including low expression of albumin (Figs. 2 and 4) and no detectable expression of the gene coding for the enzyme catalyzing the second step of the urea cycle, ornithine carbamoyl-transferase (Fig. 4). The increased synthesis of carbamoyl-phosphate synthetase 1 following exposure of H-35 cells to dexamethasone (Fig. 1) correlated closely with the elevated amounts of pCPS mRNA present in these cells (Figs. 2 and 3), rather than
enhanced translatability of pre-existing pCPS mRNA. Of particular interest in this study, however, was the finding that physiological concentrations of insulin completely suppressed the response of carbamoyl-phosphate synthetase biosynthesis to dexamethasone (Fig. 1). Reuber H-35 cells are known to display a growth response to insulin, but this occurs at very low concentrations (30-70 pM) and requires a relatively long lag period (8-9 h) for measurable effects on cell growth to take place [30]. However, under the cell culture conditions employed here (80-90% confluent cells), addition of insulin (0.18/~M) for 13 h did not stimulate cell growth. Moreover, the suppressive effect of insulin on d e x a m e t h a s o n e - s t i m u l a t e d synthesis of carbamoyl-phosphate synthetase I occurred even at early time-points of dexamethasone treatment (1 h, data not shown). Clearly, therefore, the suppressive effect of insulin in the present circumstance occurred by a mechanism distinct from its growth-promoting activities. A role for insulin in controlling urea cycle enzymes has long been postulated [11], but experiments with either whole animals or isolated hepatocytes have generally failed to show a response. This may have been due, however, to experimental limitations in the manipulation of either of these two systems. Here, we have presented evidence that insulin treatment results in almost 100% inhibition of dexamethasone-dependent stimulation of carbamoyl-phosphate synthetase biosynthesis (Fig. 1), a finding which we observe reproducibly. This is in contrast to insulin's effect on the dexamethasone-dependent increase in pCPS mRNA in these cells, where only about 50% inhibition was recorded, both by translational (Fig. 2) and cDNA hybridization (Fig. 3) assays. The basis for this discrepancy (e.g., translational controls in insulin-treated cells) is currently being investigated.
Acknowledgements We are grateful to Dr. Carol Lusty, The Public Health Research Institute of the City of New York, for providing the pCPS cDNA clone employed in this study. Financial support for this work was provided by the Medical Research Council of Canada, the National Cancer Institute
153
of Canada, the Japanese Ministry of Education, Science and Culture, and the Ishida Science Foundation. J.R. is a recipient of the McGill University Alma Mater Fellowship. M.N. is a recipient of a Fellowship from Fonds de la Recherche en Sant6 du Qu6bec (F.R.S.Q.). References 1 Metzenberg, R.L., Hall, L.M., Marshall, M. and Cohen, P.P. (1957) J. Biol. Chem. 229, 1019-1025 2 Gamble, J.G. and Lehninger, A.L. (1973) J. Biol. Chem. 248, 610-618 3 Raymond, Y. and Shore, G.C. (1981) Biochim. Biophys. Acta 656, 111-119 4 Jones, M.E,, Anderson, D., Anderson, C. and Hodis, S. (1961) Arch. Biochem. Biophys. 95, 499-507. 5 Shore, G.C., Rachubinski, R.A., Argan, C., Rozen, R., Pouchelet, M., Lusty, C.J. and Raymond, Y. (1983) Methods Enzymol. 97, 396-408 6 Raymond, Y. and Shore, G.C. (1979) J. Biol. Chem. 254, 9335-9338 7 Mori, M., Morita, T., Ikeda, F., Amaya, Y., Tatibana, M. and Cohen, P.P. (1981) Proc. Natl. Acad. Sci. USA 78, 6055-6060 8 Ryall, J., Rachubinski, R.A., Nguyen, M., Rozen, R., Broglie, K.E. and Shore, G.C. (1984) J. Biol. Chem. 259, 9172-9176 9 Schimke, R.T. (1963) J. Biol. Chem. 238, 1012-1018 10 McLean, P. and Gurney, M.W. (1963) Biochem. J. 87, 96-104 11 McLean, P. and Novello, F. (1965) Biochem. J. 94, 410-422 12 Snodgrass, P.J., Lin, R.C., Muller, W.A. and Aoki, T.T. (1978) J. Biol. Chem. 253, 2748-2753 13 Murakami, A., Kitagawa, Y. and Sugimoto, E. (1983) Biochim. Biophys. Acta 740, 38-45
14 Haggerty, P.F., Spector, E.B., Lynch, M., Kern, R., Frouk, L.B. and Dexterbaum, S.D. (1982) J. Biol. Chem. 257, 2246-2253 15 Gebhardt, R. and Mecke, D. (1979) Eur. J. Biochem. 97, 29-35 16 Rozen, R., Noel, C. and Shore, G.C. (1983) Biochim. Biophys. Acta 741, 47-54 17 Kitagawa, Y., Murakami, A. and Sugimoto, E. (1984) FEBS Lett. 165, 133-137 18 Shore, G.C., Carignan, P. and Raymond, Y. (1979) J. Biol. Chem. 254, 9335-9338 19 Argan, C., Lusty, C.J. and Shore, G.C. (1983) J. Biol. Chem. 258, 6667-6670 20 Shore, G.C. and Harris, R. (1977) J. Cell Biol. 74, 315-321 21 Gubler, U. and Hoffman, B.J. (1983) Gene 25, 263-269 22 Horwich, A.L., Kraus, J.P., Williams, K.R., Kalousek, F., Konigsberg, W. and Rosenberg, L.E. (1983) Proc. Natl. Acad. Sci. USA 80, 4258-4261 23 Horwich, A.L., Fenton, W.A., William, K.R., Kaiousek, F., Kraus, J.P., Doolittle, R.F., Konigsberg, N. and Rosenberg, L.E. (1984) Science 224, 1068-1074 24 Rigby, P.W.J., Diekmann, M., Rhodes, C. and Berg, P. (1977) J. Mol. Biol. 113, 237-251 25 Denhardt, D.T. (1966) Biochem. Biophys. Res. Commun. 23, 641-646 26 Andrews, G.K., Dziadek, M. and Tamaoki, T. (1982) J. Biol. Chem. 257, 5148-5153 27 Chirgwin, J.M., Przybyla, A.E., MacDonald, R.J. and Ratter, W.J. (1979) Biochemistry 18, 5294-5299. 28 Liddle, G.W. and Mellon, K.K. (1974) in Textbook of Endocrinology (Williams, R.H., ¢d.), 5th Edn., p. 257, W.B. Saunders, Philadelphia 29 Haggarty, D.F., Young, P.L., Buess, J.V. and Popjack, G. (1975) J. Biol. Chem. 250, 8428-8437 30 Koontz, J.W. and Iwahashi, M. (1981) Science 211,947-949