- Email: [email protected]
EBPβ TRANSCRIPTION FACTOR IN THE ACUTE-PHASE REGULATION OF THE RAT HAPTOGLOBIN GENE
Cell Biology International 1998, Vol. 22, No. 9/10, 685–693 Article No. cb980307, available online at http://www.idealibrary.com on
PARTICIPATION OF TWO ISOFORMS OF C/EBPâ TRANSCRIPTION FACTOR IN THE ACUTE-PHASE REGULATION OF THE RAT HAPTOGLOBIN GENE ILIJANA GRIGOROV*, TANJA MILOSAVLJEVIC u , IVANA CVETKOVIC u and MIODRAG PETROVIC u Molecular Biology Laboratory, Institute for Biological Research, 29 November 142, 11060, Belgrade, Yugoslav Federal Republic Received 1 April 1998; accepted 25 September 1998
Previous analyses of the mechanism of the transcriptional induction of the rat haptoglobin (Hp) gene during acute-phase (AP)-reaction have revealed the involvement of several trans-acting nucleoproteins (NPs) in controlling this process. In this study, by using antibodies against C/EBPâ factor in Western immunoblot assay, we found that rat liver trans-acting NPs p35 and p20 are two characteristic C/EBPâ isoforms whose expression is induced under AP-conditions. DNA-binding assays identified the binding sites for these two C/EBPâ proteins in the functionally defined elements A and C of the rat Hp gene and also revealed that they have specific binding affinity towards these elements. Under non-induced conditions, p35 was the only C/EBPâ binding factor; however, upon AP-conditions both, 35 kDa- and 20 kDa-C/EBPâ binding activities were significantly induced suggesting that these interactions are necessary for the activation of the Hp gene. By in vitro phosphorylation assay and selective proteolysis, we also present evidence that p35 requires phosphorylation for its DNA binding ability. Thus, we conclude that increase in binding of C/EBPâ isoforms during AP-reaction occurs through their 1998 Academic Press upregulation and structural modification. K: haptoglobin gene; trans-acting proteins; C/EBPâ; phosphorylation; acute-phase reaction
INTRODUCTION The systemic response to acute inflammation, infection and tissue injury is characterized by changes in the concentration of a wide variety of plasma proteins collectively called the AP-proteins (Koj, 1986; Fey and Fuller, 1987; Baumann and Gauldie, 1994). Expression of these genes is primarily regulated by the inflammatory cytokines interleukin-1 (IL-1), IL-6, and tumour necrosis factor (Chen et al., 1991; Gauldie and Baumann, 1991) and by steroid hormone glucocorticoids (Baumann et al., 1987; Chen et al., 1991). The elevated expression of AP-genes is regulated at the transcriptional level as a result of increased interaction of trans-acting proteins induced by inflammatory cytokines with their cognate regulatory cis-acting DNA elements (Li and Liao, 1991; Xanthopoulos et al., 1991; Alam and *Author to whom correspondence should be addressed. 1065–6995/98/090685+09 $30.00/0
Papaconstantinou, 1992; Wegenka et al., 1993). Analyses of many AP-promoters have revealed three general types of regulatory elements in the transcriptional induction by cytokines—the binding sites for members of the C/EBP transcription factors, the binding sites for the NFêB/Rel family of proteins and the binding sites for AP-response factor. The C/EBP proteins constitute the most abundant transcription factor family in the liver. They appear as three principal isoforms á, â and ä. All three C/EBPs have similar sequence specificities in their DNA binding, each having the modified consensus sequence T(T/G)NNG(A/C/T)AA(T/G) (Akira et al., 1990; Chen and Liao, 1993). Within this protein family, the C/EBPâ isoform is essential in mediating lipopolysaccharide, turpentine and cytokine-dependent transcriptional activation of AP-genes (Akira et al., 1990; Alam et al., 1993; Ramji et al., 1993). The binding sites for C/EBPâ have been identified within the IL-1/IL-6 1998 Academic Press
686
responsive elements of SAA (Li and Liao, 1991; Ray and Ray, 1993), AGP (Alam and Papaconstantinou, 1992; Lee et al., 1993) and complement C-3 (Juan et al., 1993) promoters, and in the steroid responsive unit of the á1AGP promoter (Nishio et al., 1993). During AP-reaction, C/EBPâ forms constitutive and inducible complexes with these elements and acts as transactivator involved in signal transduction. Haptoglobin, the haemoglobin binding and transport protein, is one of the major AP-proteins synthesized in hepatocytes during AP-reaction (Kushner and Mackiewicz, 1987). Its elevated synthesis in response to inflammatory stimuli is attributed primarily to the transcriptional induction of its gene (Kushner et al., 1988). In an effort to unravel the molecular mechanism of the Hp gene regulation, the promoters of human (Oliviero and Cortese, 1989) and rat (Marinkovic´ and Baumann, 1990) Hp gene have been studied. It was shown that DNA element of the rat Hp gene promoter (positions 170/56), functions as a transcriptional enhancer in conferring dramatic cytokine response onto a downstream reporter gene. This fragment, termed the hormone response element (HRE) is composed of three cooperatively interacting cis-acting elements A, B and C, as in the Hp gene in humans. Each of these three elements binds to a distinct set of NPs and transition of the Hp gene from basal to an activated transcriptional status is accompanied by structural modifications of several pre-existing trans-acting proteins and enhancement of their DNA binding ability (S {evaljevic´ et al., 1995). The most prominent members of the trans-acting proteins involved in the transcription of the AP-proteins genes have been identified either as isoforms of the regulatory protein C/EBP or as proteins related to C/EBP, with molecular masses ranging from 20 to 45 kDa (Wedel and Loms Ziegler-Heitbrock, 1995). Among the rat liver NPs which display remarkable affinity to interact with the HRE of the Hp gene during AP-reaction are the trans-acting proteins with molecular masses 35(p35) and 20(p20)kDa (S {evaljevic´ et al., 1995). In this study, we confirmed the appearance of C/EBP-like factors among the NPs that interact with HRE of the rat Hp gene and identified p35 and p20 as two C/EBPâ isoforms. We also present evidence that the increase in their DNA binding occurs through their upregulation as well as by their phosphorylative structural modifications. Since binding of C/EBPâ factors to the inducible promoter element of the rat Hp gene is highly specific, we conclude that the induction of the
Cell Biology International, Vol. 22, No. 9/10, 1998
Hp gene under AP-conditions involves their participation. MATERIAL AND METHODS Induction of the AP-reaction The AP-reaction was elicited by two subsequent injections of turpentine oil in the lumbar region of 2-month-old male albino rats (ìl/body weight) and animals were killed 12 h later. Although the injection per se was not observed to cause AP-related changes, the control animals were routinely injected with pyrogen-free saline action. The animals were kept at constant temperature, humidity and regular dark/light intervals. Preparation of nuclear protein extracts Rat liver nuclear extracts were prepared 0 and 12 h after the injection of turpentine oil as suggested by Gorski et al. (1986). Briefly, the livers were perfused with PBS (0.34 NaCl, 0.01 KCl, 0.02 Na2HPO4, 7 m KH2PO4), quickly excised and homogenized in 2 sucrose (10 m HEPES, pH 7.6, 25 m KCl, 5 m MgCl2, 1 m EDTA, 1 m spermidine, 1 m PMSF, 1 m DTT, 10% glycerol). Following filtration through two layers of cheesecloth, the homogenate was centrifuged (24,000 rpm, 30 min, 4C). The obtained precipitate of nuclei was resuspended in the lysis buffer (10 m HEPES, pH 7.6, 100 m KCl, 3.0 m MgCl2, 0.1 m EDTA, 1.0 m DTT, 0.1 m PMSF, 10% glycerol). In order to precipitate chromatin, (NH4)2SO4 pH 7.9 was added slowly with constant stirring to a final concentration of 0.36 . Chromatin was sedimented by centrifugation (35,000 rpm, rotor Ti 50, 60 min, 4C). Nucleoproteins were precipitated from the supernatant by adding crystallized (NH4)2SO4 to final concentration of 2.6 . After ammonium sulphate was dissolved, the supernatant was kept on ice for an additional 15 min. Nucleoproteins were sedimented by centrifugation (35,000 rpm, rotor Ti 50, 30 min, 4C), dissolved in the dialysis buffer (25 m HEPES, pH 7.6, 40 m KCl, 0.1 m EDTA, 1.0 m DTT, 10% glycerol) and dialysed overnight against the same buffer. After 5 min centrifugation, pellet was discarded and supernatant kept at 80C in small aliquots. Isolation and labelling of DNA probes Nucleotide sequences A (165/146), C (97/ 49) and ABC (170/56) of the 5 flanking
Cell Biology International, Vol. 22, No. 9/10, 1998
687
region of the rat Hp gene were used as DNA probes. The fragments of Hp gene subcloned into Hinc II site of pUC13 were obtained from Dr Heinz Baumann from the Department of Molecular Cell Biology, Roswell Park Institute, Buffalo, NY, U.S.A. Preparation of DNA, radioactive labelling and DNA electrophoresis were performed according to the standard procedures described by Sambrook et al. (1989). For the gel retardation assay, fragments of the Hp gene were labelled with [ã-32P]ATP (Amersham Inc. plc) by using T4 polynucleotide kinase. For SouthWestern analysis, elements A and C were labelled by ‘random priming’ technique using [á-32P]dCTP (Amersham).
method of Towbin et al. (1979). Nuclear extracts were denatured, separated by 11% SDS-PAGE and transferred to nitrocellulose filters at a constant current of 135 mA for 1 h. Membranes were blocked by incubation in the blocking buffer containing 59 m Tris-HCl pH 7.6, 150 m NaCl, 0.05% Tween 20 and 2% non-fat condensed milk. Filters were then probed with polyclonal rabbit anti-C/EBPâ antibodies (kindly donated by Dr Heinz Baumann, Roswell Park Cancer Institute, Buffalo, U.S.A.) at a 1:50 dilution for 3 h at room temperature. Following rinsing, the blots were incubated with 125I-labelled anti-rabbit antibodies (dilution 1:1000) for 1 h. After 5 min washes, the filters were exposed to X-ray film.
Electrophoretic mobility shift assay
Phosphorylation of nuclear extracts
For the gel retardation experiments (Fried and Crothers, 1981), 10 ìg aliquots of the nuclear extract were preincubated with poly dI-dC (at the 1:0.8 ratio) and 200–400 ng of salmon sperm DNA in the buffer containing 25 m Hepes pH 7.6, 60 m KCl, 7.5% glycerol, 0.1 m EDTA, 0.75 m DTT, 5 m MgCl2. After 10 min at room temperature, 5 ng of the corresponding [ã-32P]-labelled DNA fragment was added and the binding reaction was continued for another 20 min. Free samples were then applied onto 5% polyacrylamide gels with 0.5Tris-borate EDTA (45 m Tris-borate, 2 m EDTA) as the electrophoresis buffer. Gels were dried and autoradiographed for 3–5 days.
In vitro phosphorylation of isolated nuclear extracts was performed by a modified phosphorylation assay of Piccoletti et al. (1993). Nuclear extracts were incubated in a reaction mixture containing 20 m Hepes pH 7.6, 60 m KCl, 10 m MgCl2, 20 ì ATP and 10 ìCi [ã-32P] ATP (Amersham) for 5 min at 37C. The phosphorylation reaction was stopped by the addition of the electrophoresis buffer according to Laemmli (1970). The samples were kept for 3 min in a boiling water bath and immediately subjected to electrophoresis. For the South-Western analysis, nuclear extracts from the contol rat livers were phosphorylated in the same buffer containing 50 ì unlabelled ATP.
South-Western blot analysis of nuclear proteins After SDS-polyacrylamide gel electrophoresis (SDS-PAGE) in the first dimension (O’Farrell et al., 1977), the proteins were transferred to Hybond TM-C nitrocellulose filters (45 ìm; Amersham) by electroblotting. Following transfer, the filters were soaked in the binding buffer containing 1 m Na-EDTA, 10 m Tris-HCl pH 7.0, 0.02% bovine serum albumin (BSA), 0.02% Ficoll, 0.02% polyvinylpyrrolidone, 50 m NaCl for 1 h as described by Bowen et al. (1980). The filters were probed with 106 cpm/lane of 32P-labelled DNA fragments in the binding buffer for 1 h at room temperature. After binding, the filters were washed twice for 15 min, with the binding buffer containing 200 m NaCl at room temperature and exposed to X-ray film for 1–4 days. Western immunoblot assay Western immunoblot analysis of the rat liver nuclear extracts was performed according to the
Peptide mapping of nucleoproteins by staphylococcal aureus V8 protease Nucleoproteins from the control and the AP rat livers were submitted to selective proteolysis by staphylococcal aureus V8 protease following the procedure based on the one introduced by Cleveland et al. (1977). The NPs were separated by SDS-PAGE and after Coomassie blue staining, band corresponding to the p35 was cut out from the gel and the protein was eluted according to Berezney (1991). Eluted protein was dissolved in the sample buffer which contained 125 m TrisHCl pH 6.8, 0.5% SDS and 10% glycerol. The samples were then heated at 100C for 2 min. Proteolytic digestion was carried out at 37C for 1 h by addition of 2 ìg of staphylococcal aureus V8 protease per 20 ìg of protein. Following addition of â-mercaptoethanol and SDS to the final concentrations of 10% and 2% respectively, proteolysis was stopped by boiling the samples for 2 min.
688
Fig. 1. Nucleotide sequence of the regions A, B and C from the HRE of the rat Hp gene. C/EBP consensus sequences are underlined.
Peptide fragments were separated on the 15% SDS-PAGE and visualized by silver staining. RESULTS To identify the liver trans-acting nucleoproteins that interact with HRE of the rat Hp gene, also known as ABC sequence, it was compared with binding sequences of several well-known transcription factors. This comparative analysis (Fig. 1) revealed that elements A (165/146) and C (97/49) contain a sequence that is highly homologous to C/EBPâ binding sequence in the human IL-6 promoter and matches the modified C/EBP-binding concensus sequence T(T/ G)NNG(A/C/T)AA(T/G) (Akira et al., 1990). To determine whether the matched sequences indeed correspond to C/EBPâ transcription binding sites and participate in DNA-protein interactions, we radioactively labelled ABC element and its subfragments A and C and used them as probes in gel mobility shift assays. As seen in Figure 2, the entire ABC fragment formed two triplets of complexes with nuclear extracts from the control rat livers indicated with numbers 1–3 and 4–6 and one fast moving complex F (lane 1). AP-reaction (lane 2) was followed by an appearance of one slow migrating complex designated as S and an increase in the abundance of pre-existing complexes 6, 5, 3 and F in contrast to 1 and 2. With trimer of fragment A and nuclear extracts from control (lane 5) and induced (lane 6) livers, a preferential formation of equal amounts of complexes 1, 2 and 3 was observed. The protein-binding patterns of the dimer of fragment C (lanes 9 and 10) resembled that of fragment ABC exept for that the complex S moved faster than the same produced with ABC. From these quantitative differences between the patterns of control and AP-nuclear extracts it can be concluded that NPs from both livers interact
Cell Biology International, Vol. 22, No. 9/10, 1998
Fig. 2. Gel mobility shift assay with a probe corresponding to the C/EBP binding site in the Hp gene promoter. Nucleoproteins from the control (lanes 1, 3, 5, 7, 9 and 11) and the acute-phase (lanes 2, 4, 6, 8, 10 and 12) rat livers were incubated with [ã-32P]ATP labelled sequences ABC (170/ 56), trimer of A (165/146) and dimer of C (97/49) from the rat Hp promoter. Lanes 3, 4, 7, 8, 11 and 12 represent binding of heated at 85C for 10 min, control and acute-phase nucleoproteins prior to use in binding assay. The resultant DNA–protein complexes were resolved on a native 5% polyacrylamide gel. Numerals indicate the different bands formed.
with the same DNA-binding sites of the Hp gene, and that they belong to the same family(s) of transcription factors. Since heat stability is a characteristic property of C/EBPs (Johnson et al., 1987), we heated control and AP-liver nuclear extracts at 85C for 10 min and tested their DNA binding ability in the gel mobility shift assys. Heated nuclear extracts from both, control and induced livers still retained their capacity of forming complexes 1, 2 and 3 with the entire ABC element (Fig. 2, lanes 3 and 4), and dimer of fragment C (lanes 11 and 12) separately. At the same conditions, trimer of fragment A (lanes 7 and 8) represented a preferential binding site for proteins forming complexes 2 and 3. With heated nuclear extract from AP-liver, this element bound protein(s) which formed complex 1. Based on the fact that elements A and C contain sequences homologous to the C/EBP binding concensuse sequence, these results indicate that some of the NPs involved in formation of mentioned complexes are members of the C/EBP protein family. Since element C has a different protein-binding pattern and higher binding affinity for C/EBPs than does the element A, these result suggest involvement of different constitutive and inducible C/EBP isoforms in the regulation of the Hp gene expression. Although our gel mobility data demonstrate that C/EBP family proteins bind to the elements A and C in the HRE of Hp gene, it was not clear which members of this family actually contributed to these interactions. To identify C/EBPâ factors in the rat liver nuclear extract, we performed Western
Cell Biology International, Vol. 22, No. 9/10, 1998
Fig. 3. Western immunoblot analysis. Nucleoproteins from the control (lane 1) and the acute-phase (lane 2) rat livers were separated by SDS-PAGE, blotted and probed with polyclonal anti-C/EBPâ antibodies. The corresponding immune precipitates were detected by 125I-labelled secondary antibody. Those formed with the 35- and 20-kDa proteins are indicated by arrows.
immunoblot assay by using polyclonal rabbit antiC/EBPâ antibody. Figure 3 shows that anti-C/ EBPâ antisera recognized two C/EBPâ isoforms in the control (C) and induced (T) nuclear extracts at the 35- and 20-kDa positions. The amount of these isoforms in the nucleus was significantly increased after treatment with turpentine suggesting that its formation appears to be induced upon AP-conditions. In order to establish whether the p35 and p20 among the other liver NPs of control and turpentine-treated rat livers have the binding ability towards the elements A and C, we performed South-Westen blot analysis. Figure 4 shows that the NP at 35 kDa, bound both sequences with higher affinity during AP-reaction than in the control. The 20 kDa NP has no binding activity in the control nuclear extract, but after turpentine treatment an increase in its binding activity towards both sequences was detected (lanes 2 and 4). Studies described above suggest that turpentine treatment results in the quantitative changes in 35 kDa- and 20 kDa-C/EBPâ nuclear concentrations and that these changes could be due to differential bindings of this trans-acting factors to the elements A, C or an entire HRE of the Hp gene. However, elevated C/EBPâ-binding activity
689
Fig. 4. South-Western blot analysis of nucleoproteins that bind to the elements A and C of the rat Hp gene. Nucleoproteins from the control (lanes 1 and 3) and the acute-phase (lanes 2 and 4) livers were subjected to 11% SDS-PAGE, blotted and probed with [á-32P]dCTP labelled trimer of sequence A and dimer of sequence C. The radioactive spots appearing at the positions of 35 and 20 kDa are indicated by an arrow.
through phosphorylation of pre-existing protein has also been reported (Ramji et al., 1993). On the basis of these data, we tested whether the DNA binding activity of p35 and p20 for the Hp gene HRE is regulated by a phosphorylation event. First, the same amount of nuclear extracts from control and induced livers were in vitro phosphorylated in the presence of endogenous kinases and ã[32P]ATP and samples were analysed by 11% SDS-PAGE. Autoradiogram in Figure 5 revealed that p35, in contrast to p20, is a phosphoprotein displaying a higher capacity to incorporate radioactive phosphorus during the AP-conditions. This suggests that p35 from the two livers differ in the number of phosphate groups or residues accesible to phosphorylation, probably as a consequence of differential endogenous kinase activity in basal and AP-conditions. Considering that modular structure of transacting proteins enables them to become phosphorylated at multiple sites (Choy et al., 1993), it seems that enhancement of p35 affinity to bind HRE of the Hp gene during AP-reaction comes from an AP-induced phosphorylation event. To confirm this, we performed South-Western blot analysis of in vitro phosphorylated control NPs with unlabelled ATP. The quality of the phosphorylated NPs was monitored by SDS-PAGE analysis and
690
Fig. 5. Autoradiography of in vitro phosphorylated nucleoproteins. Nucleoproteins from the control (lane C) and the acute-phase (lane T) livers were phosphorylated by endogenous kinases in the presence of [ã-32P]ATP, and analysed by 11% SDS-PAGE.
Fig. 6. South-Western analysis of p35 binding properties to the HRE of the rat Hp gene after in vitro phosphorylation. Lanes 1 and 3, p35 from the control and the aute-phase livers; lane 2, p35 from the control livers pretreated with unlabelled ATP.
Coomassie blue staining which revealed no evidence of non-specific protein degradation due to the ATP treatment (not shown). Figure 6 shows that phosphorylated modifications of p35 from control liver (lane 2) increased its DNA-binding affinity to the level similar to that observed in AP-liver (lane 3). This result clearly indicates that p35 as well as 35 kDa-C/EBPâ requires phosphorylation for its DNA-binding ability. Recent studies have shown that several mechanisms may account for the effects of phospohorylation on protein DNA binding (Hunter and Karin, 1992). In many of the cases, it is likely that
Cell Biology International, Vol. 22, No. 9/10, 1998
Fig. 7. Partial proteolysis of p35 from the control (lane 1), the acute-phase (lane 2) nuclear extracts and ATP-treated control nuclear extracts (lane 3) by staphyloccocal aureus V8 protease. The arrow indicates the position of new peptide fragment formed under acute-phase and phosphorylative conditions. E: position of staphyloccocal V8 protease.
phosphorylation affects the conformation of the protein in such a way that its binding activity is altered. Thus, we assume that AP-dependent increase in the binding affinity of p35 relies on the conformational changes of its native form through phosphorylation events which result in providing surfaces that can better adapt to the protein binding sites in the HRE of rat Hp gene. To confirm our assumption we performed comparative analysis of digestion patterns of p35 from the control, in vitro phosphorylated and the AP-nuclear extracts by staphylococcal aureus V8 protease. Limited proteolysis of p35 from the control and the AP-liver resulted in different peptide profiles (Fig. 7, lanes 1 and 2). The proteolysis of p35 from the AP liver (lane 2) resulted in the appearance of one new peptide fragment in the 16 kDa region of gel. Mapping of the p35 from the in vitro phosphorylated control nuclear extracts, resulted in peptide map which was identical to that observed for p35 from the AP-liver (lane 3). The observed changes in the accessibility of p35 to V8 protease following induction of AP-reaction and phosphorylation argued for the AP-dependent structural alteration which coincided with the increase in the p35 afinity to bind the HRE of the Hp gene. Thus, these experiments clearly show that both, increased synthesis of the p35-C/EBPâ and its activation via a protein phosphorylation
Cell Biology International, Vol. 22, No. 9/10, 1998
mechanism are required for the induction of the Hp gene expression. DISCUSSION Studies on the mechanism of the transcriptional activation of the rat Hp gene revealed the presence of several liver trans-acting NPs for which binding affinity towards the HRE of Hp gene is altered during AP response (S {evaljevic´ et al., 1995). In order to characterize some of these nuclear factors, in this study we examined the HRE sequence that revealed the existence of two consensus sites for C/EBPâ-like transcription factors in functionally defined A and C regions. By using Western blotting technique, the characteristic 35-kDa, termed LAP or liver activatory protein and 20 kDa isoforms of C/EBPâ termed LIP or liver inhibitory protein (Descombes and Schibler, 1991; Ossipow et al., 1993), were detected in control and AP-liver nuclear extracts. Their relative amounts increased after turpentine treatment suggesting that APreaction mediated an increase in the expression of C/EBPâ protein in liver. Such an interpretation is consistent with the observation that C/EBPâ mRNA is normally expressed in low levels whereas during the AP-respose its level increases significantly (Alam et al., 1992, 1993; An et al., 1996). Considering that high expression of C/EBPâ protein in vitro (Baumman et al., 1992, 1993) is regulated by the cytokines, we propose the influence of the same factors on the amount of nuclear C/EBPâ in vivo. The existence of 35 kDa- and 20 kDa-C/EBPâ isoforms has been detected in nuclear extracts from IL-1/IL-6/dexamethasone-stimulated H-35, HTC (Baumann et al., 1992, 1993) and HepG2 cells (Ossipow et al., 1993). Similar isoforms have been observed in experimentally induced AP-reaction in the mouse and rabbit liver (An et al., 1996; Ray and Ray, 1994) and evidence of their involvement in regulation of certain AP-genes have been provided. Increased amounts of 35 kDa- and 20 kDaC/EBPâ isoforms when the transcription of Hp gene is highly induced suggests that they could play a general role in the regulation of the rat Hp gene expression. Studies by several laboratories have shown that C/EBPâ isoforms have specific transcriptional regulatory properties. The 35 kDaisoform acts as a powerful transactivator in contrast to 20 kDa form that mostly acts as an attenuator of transcription stimulation. The 20 kDa-C/EBPâ isoform has DNA binding and dimerization domain but it lacks part of the trans-
691
activation domain, retaining at least one of the attenuate regions. This truncated form has been shown to bind to DNA, but does not transactivate (Wedel and Loms Ziegler-Heitbrock, 1995). Although the molecular mechanism of this is still unclear, it is conceivable that this form dimerizes with other C/EBP members (Sears and Sealy, 1994), and thereby blocks transactivation. In contrast, Alam et al. (1992) have shown that 20 kDa C/EBPâ is a potential activator of the AGP1 gene and this was the first indication that this protein may function as a transactivator. Our results of South-Western analysis indicate that 20 kDa C/EBPâ does not have DNA-binding activity in control liver, whereas turpentine treatment induced its binding with high affinity for the elements A and C. Since Western blotting analysis revealed increased nuclear concentration of both proteins in the AP-nuclear extract, we propose that these changes could be due to their differential binding activities. However, there is growing evidence that the rate of transcription is determined by functional modifications of trans-acting proteins (Hahn, 1993). Recent studies have shown that various growth factors and cytokines control expression of many genes by inducing phosphorylation events mediated by a cascade of protein kinases (Sadowski et al., 1993) that in turn phosphorylate many transcription factors. Phosphorylation activates their DNA-binding activity which may account for the enhanced transcription of specific genes in reaction to growth factors and cytokines (Kishimoto et al., 1994). Phosphorylation of C/EBPâ at several functional domains has been reported. Various protein kinases, including protein kinase A (Metz and Ziff, 1991), calmodulin kinase II (Wegner et al., 1992), and MAP kinase (Nakajima et al., 1993) indirectly or directly have been implicated in the phosphorylation of the C/EBPâ and are requisite for its full activation. Herein, we show that only 35 kDa-C/EBPâ isoform is phosphorylated and the selective proteolysis confirmed that such a post-translational event enhanced their DNA-binding activity to the level obtained during AP-conditions. Since 20 kDa-C/ EBPâ is not affected by phosphorylation events, we conclude that increase in its nuclear concentration is responsible for its enhanced DNA binding activity. In spite of its known attenuating function in stimulation of transcription, it is not yet understood in what way it contributes to the Hp gene expression. Thus, we propose the possible mode of p35- and p20-C/EBPâ-isoforms action during APconditions. In the absence of inflammation both C/EBPâ protein levels are relatively low, although
692
the level of p20-C/EBPâ is higher than that of p35-C/EBPâ. Following turpentine treatment, 35 kDa-C/EBPâ DNA binding activity increases as a result of post-translational phosphorylative modification of the pre-existing p35-C/EBPâ, as well as from an increased level of new 35 kDa-C/ EBPâ. This form then probably activates Hp gene transcription. As 20 kDa-C/EBPâ protein level increases, 20 kDa- and 35 kDa-C/EBPâ could form heterodimers, resulting in a relative decline in p35-C/EBPâ dimers capable of binding to the promoter. Therefore, it is the ratio of these two forms which determines whether transcriptional activity will be enhanced or suppressed. Finally, it is also possible that p20-C/EBPâ through its interaction with p35-C/EBPâ could have a stimulatory effect on Hp gene transcription, probably through relief of repression (Kowenz-Leutz et al., 1994). ACKNOWLEDGEMENTS This work was supported by the Research Science Fund of the Serbian Ministry of Science, Contract 03E20. REFERENCES A S, I H, S T, T O, K S, N Y, N T, H T, K T, 1990. A nuclear factor for IL-6 expression (NF-IL6) is a member of a C/EBP family. EMBO J 9: 1897–1906. A T, P J, 1992. Interaction of acutephase-inducible and liver-enriched nuclear factors with the promoter region of the mouse á1-acid glycoprotein gene. Biochemistry 31: 1928–1936. A T, A RM, P J, 1992. Differential expression of three C/EBP isoforms in multiple tissues during the acute phase response. J Biol Chem 267: 5021– 5024. A T, A MR, M RC, H C-C, G X, P J, 1993. Transactivation of the á1-acid glycoprotein gene acute-phase response element by multiple isoforms of C/EBP and glucocorticoid receptor. J Biol Chem 268: 15,681–15,688. A MR, H C-C, R PD, R JP, S SG, K DT, P J, 1996. Evidence for posttranscriptional regulation of C/EBPá and C/EBPâ isoform expression during the lipopolysaccharide-mediated acute-phase response. Mol Cell Biol 16: 2295–2306. B H, G J, 1994. The acute phase response. Immunol Today 15: 74–80. B H, R C, G J, 1987. Interaction among hepatocyte-stimulating factors, interleukin-1 and glucocorticoids for the regulation of acute phase plasma proteins in human hepatoma (HepG2) cells. J Immunol 139: 4122–4128. B H, M KK, C SP, C Z, J GP, 1992. Role of CAAT-enhancer binding protein isoforms in
Cell Biology International, Vol. 22, No. 9/10, 1998
the cytokine regulation of acute-phase plasma protein genes. J Biol Chem 267: 19,744–19,751. B H, M KK, W GHW, 1993. THF-á, IL-1â, hepatocyte growth factor cooperate in stimulating specific acute phase plasma protein genes in rat hepatoma cells. J Immunol 151: 4248–4257. B R, 1991. The nuclear matrix: a heuristic model for investigating genomic organization and function in the cell nucleus. J Cell Biochem 47: 109–123. B B, S J, L UK, W H, 1980. The detection of DNA-binding proteins by protein blotting. Nucleic Acids Res 3: 1–20. C HM, L WS, 1993. Differential acute-phase response of rat kinogen genes involves type I and II interleukin-6 response elements. J Biol Chem 268: 25,311–25,317. C H-M, C KB, L S-L, 1991. Interleukin-6 responsiveness and cell-specific expression of the rat kininogen gene. J Biol Chem 266: 2946–2952. C B, R SGE, G LA, G MR, 1993. How eucaryotic transcription activators increase assembly of preinitiation complexes. Cold Spring Harbor Symp Quant Biol 58: 199–203. C DW, F SG, K MW, L UK, 1977. Peptide mapping by limited proteolysis in sodium dodecil sulfate and analysis by gel electrophoresis. J Biol Chem 252: 1102–1106. D P, S U, 1991. A liver-enriched transcriptional activator protein, LAP, a transcriptional inhibitory protein LIP, are translated from the same mRNA. Cell 67: 569–579. F GH, F GM, 1987. Regulation of acute-phase gene expression by inflammatory mediators. Mol Biol Med 4: 323–328. F M, C DM, 1981. Equilibria and kinetics of lac repressor operator interactions by polyacrylamide gel electrophoresis. Nucl Acids Res 9: 6505–6525. G J, B H, 1991. Cytokines and acute-phase protein expression. In: Kimball ES, ed. Cytokines and Inflammation. Boca Raton, Fl, CRC Press, Inc. 275–305. G K, C M, S U, 1986. Tissue specific in vitro transcription from the mouse albumin promoter. Cell 47: 767–776. H S, 1993. Structure and function of acidic transcription activators. Cell 70: 375–387. H T, K M, 1992. The regulation of transcription by phosphorylation. Cell 70: 375–387. J PE, L WH, G BJ, MK SL, 1987. Identification of a rat liver nuclear protein that binds to the enhancer core element of three animal viruses. Gen Dev 1: 133. J TSC, W DR, W MD, D GJ, 1993. Participation of the transcription factor C/EBP-ä in the acute-phase regulation of the human gene for complement component C3. Proc Natl Acad Sci USA 90: 2584–2588. K T, T T, A T, 1994. Cytokine signal transduction. Cell 76: 253–262. K A, 1986. Biological functions of acute phase proteins and the cytokines involved in their induced synthesis. Falk Symp 43: 331–342. K-L E, T G, A S, L A, 1994. Novel mechanism of C/EBPâ (NF-M) transcriptional control: activation through derepression. Genes Dev 8: 2781–2791. K I, M A, 1987. Acute phase proteins as disease markers. Dis Markers 5: 1–11.
Cell Biology International, Vol. 22, No. 9/10, 1998
K I, G M, S D, 1988. The acute phase response is mediated by heterogeneous mechanisms. Ann NY Acad Sci 557: 19–30. L UK, 1970. Cleavage of structural proteins during the assembly of head of bacteriophage T-4. Nature 227: 680–685. L YM, T WH, L MY, C DS, L SC, 1993. Induction of liver á-1 acid glycoprotein gene expression involves both positive and negative transcription factors. Mol Cell Biol 13: 432–442. L X, L S-L, 1991. Expression of rat serum amyloid A1 gene involves both-C/EBP-like and NF-êB-like transcroption factors. J Biol Chem 266: 15,192–15,201. M´ S, B H, 1990. Structure, hormonal regulation, identification of the interleukin-6 and dexamethasone-responsive element of the rat haptoglobin gene. Mol Cell Biol 10: 1573–1583. M R, Z E, 1991. cAMP stimulate the C/EBP-related transcription factor rNFIL-6 to translocate to the nucleus and induce c-Fos transcription. Genes Dev 5: 1754–1766. N T, K S, S T, S K, N M, K T, A S, 1993. Phosphorylation at threonine-235 by a ras-dependent mitogen-activated protein kinase cascade is essential for transcription factor NF-IL6. Proc Natl Acad Sci USA 90: 2207–2211. N Y, I H, K T, A S, 1993. A nuclear factor for interleukin-6 expression (NF-IL6) and the glucocorticoid receptor synergistically activate transcription of the rat á1-acid glycoprotein gene via direct protein-protein interaction. Mol Cell Biol 13: 1854–1862. O S, C R, 1989. The human haptoglobin gene promoter: interleukin-6 responsive elements interact with a DNA-binding protein by interleukin-6. EMBO J 8: 1145– 1151. O’F PZ, G H, O’F PH, 1977. High resolution two-dimensional electrophoresis of basic as well as acidic proteins. Cell 12: 1133–1142. O V, D P, S U, 1993. CCAAT/ enhancer-binding protein mRNA is translated into multiple proteins with different transcription activation potentials. Proc Natl Acad Sci USA 90: 8219–8223. P R, B P, A D, M P, B-Z A, 1993. Phosphorylation patterns of liver proteins during the early stages of the acute-phase response. Cell Biol Int 17: 425–432. R DP, V A, T F, C R, C G, 1993. The two C/EBP isoforms, IL-6DBP/NF-IL6 and C/EBPä/NF-IL6â, are induced by IL-6 to promote acute phase gene transcription via different mechanisms. Nucleic Acids Res 21: 289–294.
693
R A, R BK, 1993. Analysis of the promoter element of serum amyloid A gene and its interaction with constitutive and inducible nuclear factors from rabbit liver. Gene Expression 3: 151–162. R A, R BK, 1994. Serum amyloid A gene expression under acute-phase conditions involves participation of inducible C/EBP-â and C/EBP-ä and their activation by phosphorylation. Mol Cell Biol 14: 4324–4332. R BK, R A, 1994. Expression of the gene encoding a1-acid glycoprotein in rabbit liver under acute-phase conditions involves induction and activation of â and ä CCAAT-enhancer-binding proteins. Eur J Biochem 222: 891–900. S HB, S K, D JE, G MZ, 1993. A common nuclear signal transduction pathway activated by growth factor and cytokine receptors. Science 261: 1739– 1744. S J, F EF, M T, 1989. Chapter title. In: Irwin N, ed. Molecular Cloning. A Laboratory Manual. Cold Spring Harbor, Cold Spring Harbor Laboratory Press. S RC, S L, 1994. Multiple forms of C/EBPâ bind the EFII enhancer sequence in the Raus sarcoma virus long terminal repeat. Mol Cell Biol 14: 4855–4871. S´ L, M´ -P´ S, G I, B´ D, I´ M´ S, P´ M, 1995. Acute phase-dependent changes in the binding of rat liver nucleoproteins to the cytokine response element of the rat haptoglobin gene. Int J Biochem 27: 185–194. T H, S T, G J, 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci USA 76: 4350–4354. W A, L Z-H HW, 1995. The C/EBP family of transcription factors. Immunobiology 193: 171– 185. W UM, B C, L C, H PC, H F, 1993. Acute-phase response factor, a nuclear factor binding to acute-phase response elements, is rapidly activated by interleukin-6 at the posttranslational level. Mol Cell Biol 13: 276–288. W M, C Z, R MG, 1992. Calcium-regulated phosphorylation within the leucine zipper of C/EBPâ. Science 256: 370–373. X KG, P VR, C WS, S FM, C R, D JE, 1991. The different tissue transcription patterns of genes for HNF-1, C/EBP, HNF-3, HNF-4, protein factors that govern liver-specific transcription. Proc Natl Acad Sci USA 88: 3807–3811.
PARTICIPATION OF TWO ISOFORMS OF C/EBPâ TRANSCRIPTION FACTOR IN THE ACUTE-PHASE REGULATION OF THE RAT HAPTOGLOBIN GENE ILIJANA GRIGOROV*, TANJA MILOSAVLJEVIC u , IVANA CVETKOVIC u and MIODRAG PETROVIC u Molecular Biology Laboratory, Institute for Biological Research, 29 November 142, 11060, Belgrade, Yugoslav Federal Republic Received 1 April 1998; accepted 25 September 1998
Previous analyses of the mechanism of the transcriptional induction of the rat haptoglobin (Hp) gene during acute-phase (AP)-reaction have revealed the involvement of several trans-acting nucleoproteins (NPs) in controlling this process. In this study, by using antibodies against C/EBPâ factor in Western immunoblot assay, we found that rat liver trans-acting NPs p35 and p20 are two characteristic C/EBPâ isoforms whose expression is induced under AP-conditions. DNA-binding assays identified the binding sites for these two C/EBPâ proteins in the functionally defined elements A and C of the rat Hp gene and also revealed that they have specific binding affinity towards these elements. Under non-induced conditions, p35 was the only C/EBPâ binding factor; however, upon AP-conditions both, 35 kDa- and 20 kDa-C/EBPâ binding activities were significantly induced suggesting that these interactions are necessary for the activation of the Hp gene. By in vitro phosphorylation assay and selective proteolysis, we also present evidence that p35 requires phosphorylation for its DNA binding ability. Thus, we conclude that increase in binding of C/EBPâ isoforms during AP-reaction occurs through their 1998 Academic Press upregulation and structural modification. K: haptoglobin gene; trans-acting proteins; C/EBPâ; phosphorylation; acute-phase reaction
INTRODUCTION The systemic response to acute inflammation, infection and tissue injury is characterized by changes in the concentration of a wide variety of plasma proteins collectively called the AP-proteins (Koj, 1986; Fey and Fuller, 1987; Baumann and Gauldie, 1994). Expression of these genes is primarily regulated by the inflammatory cytokines interleukin-1 (IL-1), IL-6, and tumour necrosis factor (Chen et al., 1991; Gauldie and Baumann, 1991) and by steroid hormone glucocorticoids (Baumann et al., 1987; Chen et al., 1991). The elevated expression of AP-genes is regulated at the transcriptional level as a result of increased interaction of trans-acting proteins induced by inflammatory cytokines with their cognate regulatory cis-acting DNA elements (Li and Liao, 1991; Xanthopoulos et al., 1991; Alam and *Author to whom correspondence should be addressed. 1065–6995/98/090685+09 $30.00/0
Papaconstantinou, 1992; Wegenka et al., 1993). Analyses of many AP-promoters have revealed three general types of regulatory elements in the transcriptional induction by cytokines—the binding sites for members of the C/EBP transcription factors, the binding sites for the NFêB/Rel family of proteins and the binding sites for AP-response factor. The C/EBP proteins constitute the most abundant transcription factor family in the liver. They appear as three principal isoforms á, â and ä. All three C/EBPs have similar sequence specificities in their DNA binding, each having the modified consensus sequence T(T/G)NNG(A/C/T)AA(T/G) (Akira et al., 1990; Chen and Liao, 1993). Within this protein family, the C/EBPâ isoform is essential in mediating lipopolysaccharide, turpentine and cytokine-dependent transcriptional activation of AP-genes (Akira et al., 1990; Alam et al., 1993; Ramji et al., 1993). The binding sites for C/EBPâ have been identified within the IL-1/IL-6 1998 Academic Press
686
responsive elements of SAA (Li and Liao, 1991; Ray and Ray, 1993), AGP (Alam and Papaconstantinou, 1992; Lee et al., 1993) and complement C-3 (Juan et al., 1993) promoters, and in the steroid responsive unit of the á1AGP promoter (Nishio et al., 1993). During AP-reaction, C/EBPâ forms constitutive and inducible complexes with these elements and acts as transactivator involved in signal transduction. Haptoglobin, the haemoglobin binding and transport protein, is one of the major AP-proteins synthesized in hepatocytes during AP-reaction (Kushner and Mackiewicz, 1987). Its elevated synthesis in response to inflammatory stimuli is attributed primarily to the transcriptional induction of its gene (Kushner et al., 1988). In an effort to unravel the molecular mechanism of the Hp gene regulation, the promoters of human (Oliviero and Cortese, 1989) and rat (Marinkovic´ and Baumann, 1990) Hp gene have been studied. It was shown that DNA element of the rat Hp gene promoter (positions 170/56), functions as a transcriptional enhancer in conferring dramatic cytokine response onto a downstream reporter gene. This fragment, termed the hormone response element (HRE) is composed of three cooperatively interacting cis-acting elements A, B and C, as in the Hp gene in humans. Each of these three elements binds to a distinct set of NPs and transition of the Hp gene from basal to an activated transcriptional status is accompanied by structural modifications of several pre-existing trans-acting proteins and enhancement of their DNA binding ability (S {evaljevic´ et al., 1995). The most prominent members of the trans-acting proteins involved in the transcription of the AP-proteins genes have been identified either as isoforms of the regulatory protein C/EBP or as proteins related to C/EBP, with molecular masses ranging from 20 to 45 kDa (Wedel and Loms Ziegler-Heitbrock, 1995). Among the rat liver NPs which display remarkable affinity to interact with the HRE of the Hp gene during AP-reaction are the trans-acting proteins with molecular masses 35(p35) and 20(p20)kDa (S {evaljevic´ et al., 1995). In this study, we confirmed the appearance of C/EBP-like factors among the NPs that interact with HRE of the rat Hp gene and identified p35 and p20 as two C/EBPâ isoforms. We also present evidence that the increase in their DNA binding occurs through their upregulation as well as by their phosphorylative structural modifications. Since binding of C/EBPâ factors to the inducible promoter element of the rat Hp gene is highly specific, we conclude that the induction of the
Cell Biology International, Vol. 22, No. 9/10, 1998
Hp gene under AP-conditions involves their participation. MATERIAL AND METHODS Induction of the AP-reaction The AP-reaction was elicited by two subsequent injections of turpentine oil in the lumbar region of 2-month-old male albino rats (ìl/body weight) and animals were killed 12 h later. Although the injection per se was not observed to cause AP-related changes, the control animals were routinely injected with pyrogen-free saline action. The animals were kept at constant temperature, humidity and regular dark/light intervals. Preparation of nuclear protein extracts Rat liver nuclear extracts were prepared 0 and 12 h after the injection of turpentine oil as suggested by Gorski et al. (1986). Briefly, the livers were perfused with PBS (0.34 NaCl, 0.01 KCl, 0.02 Na2HPO4, 7 m KH2PO4), quickly excised and homogenized in 2 sucrose (10 m HEPES, pH 7.6, 25 m KCl, 5 m MgCl2, 1 m EDTA, 1 m spermidine, 1 m PMSF, 1 m DTT, 10% glycerol). Following filtration through two layers of cheesecloth, the homogenate was centrifuged (24,000 rpm, 30 min, 4C). The obtained precipitate of nuclei was resuspended in the lysis buffer (10 m HEPES, pH 7.6, 100 m KCl, 3.0 m MgCl2, 0.1 m EDTA, 1.0 m DTT, 0.1 m PMSF, 10% glycerol). In order to precipitate chromatin, (NH4)2SO4 pH 7.9 was added slowly with constant stirring to a final concentration of 0.36 . Chromatin was sedimented by centrifugation (35,000 rpm, rotor Ti 50, 60 min, 4C). Nucleoproteins were precipitated from the supernatant by adding crystallized (NH4)2SO4 to final concentration of 2.6 . After ammonium sulphate was dissolved, the supernatant was kept on ice for an additional 15 min. Nucleoproteins were sedimented by centrifugation (35,000 rpm, rotor Ti 50, 30 min, 4C), dissolved in the dialysis buffer (25 m HEPES, pH 7.6, 40 m KCl, 0.1 m EDTA, 1.0 m DTT, 10% glycerol) and dialysed overnight against the same buffer. After 5 min centrifugation, pellet was discarded and supernatant kept at 80C in small aliquots. Isolation and labelling of DNA probes Nucleotide sequences A (165/146), C (97/ 49) and ABC (170/56) of the 5 flanking
Cell Biology International, Vol. 22, No. 9/10, 1998
687
region of the rat Hp gene were used as DNA probes. The fragments of Hp gene subcloned into Hinc II site of pUC13 were obtained from Dr Heinz Baumann from the Department of Molecular Cell Biology, Roswell Park Institute, Buffalo, NY, U.S.A. Preparation of DNA, radioactive labelling and DNA electrophoresis were performed according to the standard procedures described by Sambrook et al. (1989). For the gel retardation assay, fragments of the Hp gene were labelled with [ã-32P]ATP (Amersham Inc. plc) by using T4 polynucleotide kinase. For SouthWestern analysis, elements A and C were labelled by ‘random priming’ technique using [á-32P]dCTP (Amersham).
method of Towbin et al. (1979). Nuclear extracts were denatured, separated by 11% SDS-PAGE and transferred to nitrocellulose filters at a constant current of 135 mA for 1 h. Membranes were blocked by incubation in the blocking buffer containing 59 m Tris-HCl pH 7.6, 150 m NaCl, 0.05% Tween 20 and 2% non-fat condensed milk. Filters were then probed with polyclonal rabbit anti-C/EBPâ antibodies (kindly donated by Dr Heinz Baumann, Roswell Park Cancer Institute, Buffalo, U.S.A.) at a 1:50 dilution for 3 h at room temperature. Following rinsing, the blots were incubated with 125I-labelled anti-rabbit antibodies (dilution 1:1000) for 1 h. After 5 min washes, the filters were exposed to X-ray film.
Electrophoretic mobility shift assay
Phosphorylation of nuclear extracts
For the gel retardation experiments (Fried and Crothers, 1981), 10 ìg aliquots of the nuclear extract were preincubated with poly dI-dC (at the 1:0.8 ratio) and 200–400 ng of salmon sperm DNA in the buffer containing 25 m Hepes pH 7.6, 60 m KCl, 7.5% glycerol, 0.1 m EDTA, 0.75 m DTT, 5 m MgCl2. After 10 min at room temperature, 5 ng of the corresponding [ã-32P]-labelled DNA fragment was added and the binding reaction was continued for another 20 min. Free samples were then applied onto 5% polyacrylamide gels with 0.5Tris-borate EDTA (45 m Tris-borate, 2 m EDTA) as the electrophoresis buffer. Gels were dried and autoradiographed for 3–5 days.
In vitro phosphorylation of isolated nuclear extracts was performed by a modified phosphorylation assay of Piccoletti et al. (1993). Nuclear extracts were incubated in a reaction mixture containing 20 m Hepes pH 7.6, 60 m KCl, 10 m MgCl2, 20 ì ATP and 10 ìCi [ã-32P] ATP (Amersham) for 5 min at 37C. The phosphorylation reaction was stopped by the addition of the electrophoresis buffer according to Laemmli (1970). The samples were kept for 3 min in a boiling water bath and immediately subjected to electrophoresis. For the South-Western analysis, nuclear extracts from the contol rat livers were phosphorylated in the same buffer containing 50 ì unlabelled ATP.
South-Western blot analysis of nuclear proteins After SDS-polyacrylamide gel electrophoresis (SDS-PAGE) in the first dimension (O’Farrell et al., 1977), the proteins were transferred to Hybond TM-C nitrocellulose filters (45 ìm; Amersham) by electroblotting. Following transfer, the filters were soaked in the binding buffer containing 1 m Na-EDTA, 10 m Tris-HCl pH 7.0, 0.02% bovine serum albumin (BSA), 0.02% Ficoll, 0.02% polyvinylpyrrolidone, 50 m NaCl for 1 h as described by Bowen et al. (1980). The filters were probed with 106 cpm/lane of 32P-labelled DNA fragments in the binding buffer for 1 h at room temperature. After binding, the filters were washed twice for 15 min, with the binding buffer containing 200 m NaCl at room temperature and exposed to X-ray film for 1–4 days. Western immunoblot assay Western immunoblot analysis of the rat liver nuclear extracts was performed according to the
Peptide mapping of nucleoproteins by staphylococcal aureus V8 protease Nucleoproteins from the control and the AP rat livers were submitted to selective proteolysis by staphylococcal aureus V8 protease following the procedure based on the one introduced by Cleveland et al. (1977). The NPs were separated by SDS-PAGE and after Coomassie blue staining, band corresponding to the p35 was cut out from the gel and the protein was eluted according to Berezney (1991). Eluted protein was dissolved in the sample buffer which contained 125 m TrisHCl pH 6.8, 0.5% SDS and 10% glycerol. The samples were then heated at 100C for 2 min. Proteolytic digestion was carried out at 37C for 1 h by addition of 2 ìg of staphylococcal aureus V8 protease per 20 ìg of protein. Following addition of â-mercaptoethanol and SDS to the final concentrations of 10% and 2% respectively, proteolysis was stopped by boiling the samples for 2 min.
688
Fig. 1. Nucleotide sequence of the regions A, B and C from the HRE of the rat Hp gene. C/EBP consensus sequences are underlined.
Peptide fragments were separated on the 15% SDS-PAGE and visualized by silver staining. RESULTS To identify the liver trans-acting nucleoproteins that interact with HRE of the rat Hp gene, also known as ABC sequence, it was compared with binding sequences of several well-known transcription factors. This comparative analysis (Fig. 1) revealed that elements A (165/146) and C (97/49) contain a sequence that is highly homologous to C/EBPâ binding sequence in the human IL-6 promoter and matches the modified C/EBP-binding concensus sequence T(T/ G)NNG(A/C/T)AA(T/G) (Akira et al., 1990). To determine whether the matched sequences indeed correspond to C/EBPâ transcription binding sites and participate in DNA-protein interactions, we radioactively labelled ABC element and its subfragments A and C and used them as probes in gel mobility shift assays. As seen in Figure 2, the entire ABC fragment formed two triplets of complexes with nuclear extracts from the control rat livers indicated with numbers 1–3 and 4–6 and one fast moving complex F (lane 1). AP-reaction (lane 2) was followed by an appearance of one slow migrating complex designated as S and an increase in the abundance of pre-existing complexes 6, 5, 3 and F in contrast to 1 and 2. With trimer of fragment A and nuclear extracts from control (lane 5) and induced (lane 6) livers, a preferential formation of equal amounts of complexes 1, 2 and 3 was observed. The protein-binding patterns of the dimer of fragment C (lanes 9 and 10) resembled that of fragment ABC exept for that the complex S moved faster than the same produced with ABC. From these quantitative differences between the patterns of control and AP-nuclear extracts it can be concluded that NPs from both livers interact
Cell Biology International, Vol. 22, No. 9/10, 1998
Fig. 2. Gel mobility shift assay with a probe corresponding to the C/EBP binding site in the Hp gene promoter. Nucleoproteins from the control (lanes 1, 3, 5, 7, 9 and 11) and the acute-phase (lanes 2, 4, 6, 8, 10 and 12) rat livers were incubated with [ã-32P]ATP labelled sequences ABC (170/ 56), trimer of A (165/146) and dimer of C (97/49) from the rat Hp promoter. Lanes 3, 4, 7, 8, 11 and 12 represent binding of heated at 85C for 10 min, control and acute-phase nucleoproteins prior to use in binding assay. The resultant DNA–protein complexes were resolved on a native 5% polyacrylamide gel. Numerals indicate the different bands formed.
with the same DNA-binding sites of the Hp gene, and that they belong to the same family(s) of transcription factors. Since heat stability is a characteristic property of C/EBPs (Johnson et al., 1987), we heated control and AP-liver nuclear extracts at 85C for 10 min and tested their DNA binding ability in the gel mobility shift assys. Heated nuclear extracts from both, control and induced livers still retained their capacity of forming complexes 1, 2 and 3 with the entire ABC element (Fig. 2, lanes 3 and 4), and dimer of fragment C (lanes 11 and 12) separately. At the same conditions, trimer of fragment A (lanes 7 and 8) represented a preferential binding site for proteins forming complexes 2 and 3. With heated nuclear extract from AP-liver, this element bound protein(s) which formed complex 1. Based on the fact that elements A and C contain sequences homologous to the C/EBP binding concensuse sequence, these results indicate that some of the NPs involved in formation of mentioned complexes are members of the C/EBP protein family. Since element C has a different protein-binding pattern and higher binding affinity for C/EBPs than does the element A, these result suggest involvement of different constitutive and inducible C/EBP isoforms in the regulation of the Hp gene expression. Although our gel mobility data demonstrate that C/EBP family proteins bind to the elements A and C in the HRE of Hp gene, it was not clear which members of this family actually contributed to these interactions. To identify C/EBPâ factors in the rat liver nuclear extract, we performed Western
Cell Biology International, Vol. 22, No. 9/10, 1998
Fig. 3. Western immunoblot analysis. Nucleoproteins from the control (lane 1) and the acute-phase (lane 2) rat livers were separated by SDS-PAGE, blotted and probed with polyclonal anti-C/EBPâ antibodies. The corresponding immune precipitates were detected by 125I-labelled secondary antibody. Those formed with the 35- and 20-kDa proteins are indicated by arrows.
immunoblot assay by using polyclonal rabbit antiC/EBPâ antibody. Figure 3 shows that anti-C/ EBPâ antisera recognized two C/EBPâ isoforms in the control (C) and induced (T) nuclear extracts at the 35- and 20-kDa positions. The amount of these isoforms in the nucleus was significantly increased after treatment with turpentine suggesting that its formation appears to be induced upon AP-conditions. In order to establish whether the p35 and p20 among the other liver NPs of control and turpentine-treated rat livers have the binding ability towards the elements A and C, we performed South-Westen blot analysis. Figure 4 shows that the NP at 35 kDa, bound both sequences with higher affinity during AP-reaction than in the control. The 20 kDa NP has no binding activity in the control nuclear extract, but after turpentine treatment an increase in its binding activity towards both sequences was detected (lanes 2 and 4). Studies described above suggest that turpentine treatment results in the quantitative changes in 35 kDa- and 20 kDa-C/EBPâ nuclear concentrations and that these changes could be due to differential bindings of this trans-acting factors to the elements A, C or an entire HRE of the Hp gene. However, elevated C/EBPâ-binding activity
689
Fig. 4. South-Western blot analysis of nucleoproteins that bind to the elements A and C of the rat Hp gene. Nucleoproteins from the control (lanes 1 and 3) and the acute-phase (lanes 2 and 4) livers were subjected to 11% SDS-PAGE, blotted and probed with [á-32P]dCTP labelled trimer of sequence A and dimer of sequence C. The radioactive spots appearing at the positions of 35 and 20 kDa are indicated by an arrow.
through phosphorylation of pre-existing protein has also been reported (Ramji et al., 1993). On the basis of these data, we tested whether the DNA binding activity of p35 and p20 for the Hp gene HRE is regulated by a phosphorylation event. First, the same amount of nuclear extracts from control and induced livers were in vitro phosphorylated in the presence of endogenous kinases and ã[32P]ATP and samples were analysed by 11% SDS-PAGE. Autoradiogram in Figure 5 revealed that p35, in contrast to p20, is a phosphoprotein displaying a higher capacity to incorporate radioactive phosphorus during the AP-conditions. This suggests that p35 from the two livers differ in the number of phosphate groups or residues accesible to phosphorylation, probably as a consequence of differential endogenous kinase activity in basal and AP-conditions. Considering that modular structure of transacting proteins enables them to become phosphorylated at multiple sites (Choy et al., 1993), it seems that enhancement of p35 affinity to bind HRE of the Hp gene during AP-reaction comes from an AP-induced phosphorylation event. To confirm this, we performed South-Western blot analysis of in vitro phosphorylated control NPs with unlabelled ATP. The quality of the phosphorylated NPs was monitored by SDS-PAGE analysis and
690
Fig. 5. Autoradiography of in vitro phosphorylated nucleoproteins. Nucleoproteins from the control (lane C) and the acute-phase (lane T) livers were phosphorylated by endogenous kinases in the presence of [ã-32P]ATP, and analysed by 11% SDS-PAGE.
Fig. 6. South-Western analysis of p35 binding properties to the HRE of the rat Hp gene after in vitro phosphorylation. Lanes 1 and 3, p35 from the control and the aute-phase livers; lane 2, p35 from the control livers pretreated with unlabelled ATP.
Coomassie blue staining which revealed no evidence of non-specific protein degradation due to the ATP treatment (not shown). Figure 6 shows that phosphorylated modifications of p35 from control liver (lane 2) increased its DNA-binding affinity to the level similar to that observed in AP-liver (lane 3). This result clearly indicates that p35 as well as 35 kDa-C/EBPâ requires phosphorylation for its DNA-binding ability. Recent studies have shown that several mechanisms may account for the effects of phospohorylation on protein DNA binding (Hunter and Karin, 1992). In many of the cases, it is likely that
Cell Biology International, Vol. 22, No. 9/10, 1998
Fig. 7. Partial proteolysis of p35 from the control (lane 1), the acute-phase (lane 2) nuclear extracts and ATP-treated control nuclear extracts (lane 3) by staphyloccocal aureus V8 protease. The arrow indicates the position of new peptide fragment formed under acute-phase and phosphorylative conditions. E: position of staphyloccocal V8 protease.
phosphorylation affects the conformation of the protein in such a way that its binding activity is altered. Thus, we assume that AP-dependent increase in the binding affinity of p35 relies on the conformational changes of its native form through phosphorylation events which result in providing surfaces that can better adapt to the protein binding sites in the HRE of rat Hp gene. To confirm our assumption we performed comparative analysis of digestion patterns of p35 from the control, in vitro phosphorylated and the AP-nuclear extracts by staphylococcal aureus V8 protease. Limited proteolysis of p35 from the control and the AP-liver resulted in different peptide profiles (Fig. 7, lanes 1 and 2). The proteolysis of p35 from the AP liver (lane 2) resulted in the appearance of one new peptide fragment in the 16 kDa region of gel. Mapping of the p35 from the in vitro phosphorylated control nuclear extracts, resulted in peptide map which was identical to that observed for p35 from the AP-liver (lane 3). The observed changes in the accessibility of p35 to V8 protease following induction of AP-reaction and phosphorylation argued for the AP-dependent structural alteration which coincided with the increase in the p35 afinity to bind the HRE of the Hp gene. Thus, these experiments clearly show that both, increased synthesis of the p35-C/EBPâ and its activation via a protein phosphorylation
Cell Biology International, Vol. 22, No. 9/10, 1998
mechanism are required for the induction of the Hp gene expression. DISCUSSION Studies on the mechanism of the transcriptional activation of the rat Hp gene revealed the presence of several liver trans-acting NPs for which binding affinity towards the HRE of Hp gene is altered during AP response (S {evaljevic´ et al., 1995). In order to characterize some of these nuclear factors, in this study we examined the HRE sequence that revealed the existence of two consensus sites for C/EBPâ-like transcription factors in functionally defined A and C regions. By using Western blotting technique, the characteristic 35-kDa, termed LAP or liver activatory protein and 20 kDa isoforms of C/EBPâ termed LIP or liver inhibitory protein (Descombes and Schibler, 1991; Ossipow et al., 1993), were detected in control and AP-liver nuclear extracts. Their relative amounts increased after turpentine treatment suggesting that APreaction mediated an increase in the expression of C/EBPâ protein in liver. Such an interpretation is consistent with the observation that C/EBPâ mRNA is normally expressed in low levels whereas during the AP-respose its level increases significantly (Alam et al., 1992, 1993; An et al., 1996). Considering that high expression of C/EBPâ protein in vitro (Baumman et al., 1992, 1993) is regulated by the cytokines, we propose the influence of the same factors on the amount of nuclear C/EBPâ in vivo. The existence of 35 kDa- and 20 kDa-C/EBPâ isoforms has been detected in nuclear extracts from IL-1/IL-6/dexamethasone-stimulated H-35, HTC (Baumann et al., 1992, 1993) and HepG2 cells (Ossipow et al., 1993). Similar isoforms have been observed in experimentally induced AP-reaction in the mouse and rabbit liver (An et al., 1996; Ray and Ray, 1994) and evidence of their involvement in regulation of certain AP-genes have been provided. Increased amounts of 35 kDa- and 20 kDaC/EBPâ isoforms when the transcription of Hp gene is highly induced suggests that they could play a general role in the regulation of the rat Hp gene expression. Studies by several laboratories have shown that C/EBPâ isoforms have specific transcriptional regulatory properties. The 35 kDaisoform acts as a powerful transactivator in contrast to 20 kDa form that mostly acts as an attenuator of transcription stimulation. The 20 kDa-C/EBPâ isoform has DNA binding and dimerization domain but it lacks part of the trans-
691
activation domain, retaining at least one of the attenuate regions. This truncated form has been shown to bind to DNA, but does not transactivate (Wedel and Loms Ziegler-Heitbrock, 1995). Although the molecular mechanism of this is still unclear, it is conceivable that this form dimerizes with other C/EBP members (Sears and Sealy, 1994), and thereby blocks transactivation. In contrast, Alam et al. (1992) have shown that 20 kDa C/EBPâ is a potential activator of the AGP1 gene and this was the first indication that this protein may function as a transactivator. Our results of South-Western analysis indicate that 20 kDa C/EBPâ does not have DNA-binding activity in control liver, whereas turpentine treatment induced its binding with high affinity for the elements A and C. Since Western blotting analysis revealed increased nuclear concentration of both proteins in the AP-nuclear extract, we propose that these changes could be due to their differential binding activities. However, there is growing evidence that the rate of transcription is determined by functional modifications of trans-acting proteins (Hahn, 1993). Recent studies have shown that various growth factors and cytokines control expression of many genes by inducing phosphorylation events mediated by a cascade of protein kinases (Sadowski et al., 1993) that in turn phosphorylate many transcription factors. Phosphorylation activates their DNA-binding activity which may account for the enhanced transcription of specific genes in reaction to growth factors and cytokines (Kishimoto et al., 1994). Phosphorylation of C/EBPâ at several functional domains has been reported. Various protein kinases, including protein kinase A (Metz and Ziff, 1991), calmodulin kinase II (Wegner et al., 1992), and MAP kinase (Nakajima et al., 1993) indirectly or directly have been implicated in the phosphorylation of the C/EBPâ and are requisite for its full activation. Herein, we show that only 35 kDa-C/EBPâ isoform is phosphorylated and the selective proteolysis confirmed that such a post-translational event enhanced their DNA-binding activity to the level obtained during AP-conditions. Since 20 kDa-C/ EBPâ is not affected by phosphorylation events, we conclude that increase in its nuclear concentration is responsible for its enhanced DNA binding activity. In spite of its known attenuating function in stimulation of transcription, it is not yet understood in what way it contributes to the Hp gene expression. Thus, we propose the possible mode of p35- and p20-C/EBPâ-isoforms action during APconditions. In the absence of inflammation both C/EBPâ protein levels are relatively low, although
692
the level of p20-C/EBPâ is higher than that of p35-C/EBPâ. Following turpentine treatment, 35 kDa-C/EBPâ DNA binding activity increases as a result of post-translational phosphorylative modification of the pre-existing p35-C/EBPâ, as well as from an increased level of new 35 kDa-C/ EBPâ. This form then probably activates Hp gene transcription. As 20 kDa-C/EBPâ protein level increases, 20 kDa- and 35 kDa-C/EBPâ could form heterodimers, resulting in a relative decline in p35-C/EBPâ dimers capable of binding to the promoter. Therefore, it is the ratio of these two forms which determines whether transcriptional activity will be enhanced or suppressed. Finally, it is also possible that p20-C/EBPâ through its interaction with p35-C/EBPâ could have a stimulatory effect on Hp gene transcription, probably through relief of repression (Kowenz-Leutz et al., 1994). ACKNOWLEDGEMENTS This work was supported by the Research Science Fund of the Serbian Ministry of Science, Contract 03E20. REFERENCES A S, I H, S T, T O, K S, N Y, N T, H T, K T, 1990. A nuclear factor for IL-6 expression (NF-IL6) is a member of a C/EBP family. EMBO J 9: 1897–1906. A T, P J, 1992. Interaction of acutephase-inducible and liver-enriched nuclear factors with the promoter region of the mouse á1-acid glycoprotein gene. Biochemistry 31: 1928–1936. A T, A RM, P J, 1992. Differential expression of three C/EBP isoforms in multiple tissues during the acute phase response. J Biol Chem 267: 5021– 5024. A T, A MR, M RC, H C-C, G X, P J, 1993. Transactivation of the á1-acid glycoprotein gene acute-phase response element by multiple isoforms of C/EBP and glucocorticoid receptor. J Biol Chem 268: 15,681–15,688. A MR, H C-C, R PD, R JP, S SG, K DT, P J, 1996. Evidence for posttranscriptional regulation of C/EBPá and C/EBPâ isoform expression during the lipopolysaccharide-mediated acute-phase response. Mol Cell Biol 16: 2295–2306. B H, G J, 1994. The acute phase response. Immunol Today 15: 74–80. B H, R C, G J, 1987. Interaction among hepatocyte-stimulating factors, interleukin-1 and glucocorticoids for the regulation of acute phase plasma proteins in human hepatoma (HepG2) cells. J Immunol 139: 4122–4128. B H, M KK, C SP, C Z, J GP, 1992. Role of CAAT-enhancer binding protein isoforms in
Cell Biology International, Vol. 22, No. 9/10, 1998
the cytokine regulation of acute-phase plasma protein genes. J Biol Chem 267: 19,744–19,751. B H, M KK, W GHW, 1993. THF-á, IL-1â, hepatocyte growth factor cooperate in stimulating specific acute phase plasma protein genes in rat hepatoma cells. J Immunol 151: 4248–4257. B R, 1991. The nuclear matrix: a heuristic model for investigating genomic organization and function in the cell nucleus. J Cell Biochem 47: 109–123. B B, S J, L UK, W H, 1980. The detection of DNA-binding proteins by protein blotting. Nucleic Acids Res 3: 1–20. C HM, L WS, 1993. Differential acute-phase response of rat kinogen genes involves type I and II interleukin-6 response elements. J Biol Chem 268: 25,311–25,317. C H-M, C KB, L S-L, 1991. Interleukin-6 responsiveness and cell-specific expression of the rat kininogen gene. J Biol Chem 266: 2946–2952. C B, R SGE, G LA, G MR, 1993. How eucaryotic transcription activators increase assembly of preinitiation complexes. Cold Spring Harbor Symp Quant Biol 58: 199–203. C DW, F SG, K MW, L UK, 1977. Peptide mapping by limited proteolysis in sodium dodecil sulfate and analysis by gel electrophoresis. J Biol Chem 252: 1102–1106. D P, S U, 1991. A liver-enriched transcriptional activator protein, LAP, a transcriptional inhibitory protein LIP, are translated from the same mRNA. Cell 67: 569–579. F GH, F GM, 1987. Regulation of acute-phase gene expression by inflammatory mediators. Mol Biol Med 4: 323–328. F M, C DM, 1981. Equilibria and kinetics of lac repressor operator interactions by polyacrylamide gel electrophoresis. Nucl Acids Res 9: 6505–6525. G J, B H, 1991. Cytokines and acute-phase protein expression. In: Kimball ES, ed. Cytokines and Inflammation. Boca Raton, Fl, CRC Press, Inc. 275–305. G K, C M, S U, 1986. Tissue specific in vitro transcription from the mouse albumin promoter. Cell 47: 767–776. H S, 1993. Structure and function of acidic transcription activators. Cell 70: 375–387. H T, K M, 1992. The regulation of transcription by phosphorylation. Cell 70: 375–387. J PE, L WH, G BJ, MK SL, 1987. Identification of a rat liver nuclear protein that binds to the enhancer core element of three animal viruses. Gen Dev 1: 133. J TSC, W DR, W MD, D GJ, 1993. Participation of the transcription factor C/EBP-ä in the acute-phase regulation of the human gene for complement component C3. Proc Natl Acad Sci USA 90: 2584–2588. K T, T T, A T, 1994. Cytokine signal transduction. Cell 76: 253–262. K A, 1986. Biological functions of acute phase proteins and the cytokines involved in their induced synthesis. Falk Symp 43: 331–342. K-L E, T G, A S, L A, 1994. Novel mechanism of C/EBPâ (NF-M) transcriptional control: activation through derepression. Genes Dev 8: 2781–2791. K I, M A, 1987. Acute phase proteins as disease markers. Dis Markers 5: 1–11.
Cell Biology International, Vol. 22, No. 9/10, 1998
K I, G M, S D, 1988. The acute phase response is mediated by heterogeneous mechanisms. Ann NY Acad Sci 557: 19–30. L UK, 1970. Cleavage of structural proteins during the assembly of head of bacteriophage T-4. Nature 227: 680–685. L YM, T WH, L MY, C DS, L SC, 1993. Induction of liver á-1 acid glycoprotein gene expression involves both positive and negative transcription factors. Mol Cell Biol 13: 432–442. L X, L S-L, 1991. Expression of rat serum amyloid A1 gene involves both-C/EBP-like and NF-êB-like transcroption factors. J Biol Chem 266: 15,192–15,201. M´ S, B H, 1990. Structure, hormonal regulation, identification of the interleukin-6 and dexamethasone-responsive element of the rat haptoglobin gene. Mol Cell Biol 10: 1573–1583. M R, Z E, 1991. cAMP stimulate the C/EBP-related transcription factor rNFIL-6 to translocate to the nucleus and induce c-Fos transcription. Genes Dev 5: 1754–1766. N T, K S, S T, S K, N M, K T, A S, 1993. Phosphorylation at threonine-235 by a ras-dependent mitogen-activated protein kinase cascade is essential for transcription factor NF-IL6. Proc Natl Acad Sci USA 90: 2207–2211. N Y, I H, K T, A S, 1993. A nuclear factor for interleukin-6 expression (NF-IL6) and the glucocorticoid receptor synergistically activate transcription of the rat á1-acid glycoprotein gene via direct protein-protein interaction. Mol Cell Biol 13: 1854–1862. O S, C R, 1989. The human haptoglobin gene promoter: interleukin-6 responsive elements interact with a DNA-binding protein by interleukin-6. EMBO J 8: 1145– 1151. O’F PZ, G H, O’F PH, 1977. High resolution two-dimensional electrophoresis of basic as well as acidic proteins. Cell 12: 1133–1142. O V, D P, S U, 1993. CCAAT/ enhancer-binding protein mRNA is translated into multiple proteins with different transcription activation potentials. Proc Natl Acad Sci USA 90: 8219–8223. P R, B P, A D, M P, B-Z A, 1993. Phosphorylation patterns of liver proteins during the early stages of the acute-phase response. Cell Biol Int 17: 425–432. R DP, V A, T F, C R, C G, 1993. The two C/EBP isoforms, IL-6DBP/NF-IL6 and C/EBPä/NF-IL6â, are induced by IL-6 to promote acute phase gene transcription via different mechanisms. Nucleic Acids Res 21: 289–294.
693
R A, R BK, 1993. Analysis of the promoter element of serum amyloid A gene and its interaction with constitutive and inducible nuclear factors from rabbit liver. Gene Expression 3: 151–162. R A, R BK, 1994. Serum amyloid A gene expression under acute-phase conditions involves participation of inducible C/EBP-â and C/EBP-ä and their activation by phosphorylation. Mol Cell Biol 14: 4324–4332. R BK, R A, 1994. Expression of the gene encoding a1-acid glycoprotein in rabbit liver under acute-phase conditions involves induction and activation of â and ä CCAAT-enhancer-binding proteins. Eur J Biochem 222: 891–900. S HB, S K, D JE, G MZ, 1993. A common nuclear signal transduction pathway activated by growth factor and cytokine receptors. Science 261: 1739– 1744. S J, F EF, M T, 1989. Chapter title. In: Irwin N, ed. Molecular Cloning. A Laboratory Manual. Cold Spring Harbor, Cold Spring Harbor Laboratory Press. S RC, S L, 1994. Multiple forms of C/EBPâ bind the EFII enhancer sequence in the Raus sarcoma virus long terminal repeat. Mol Cell Biol 14: 4855–4871. S´ L, M´ -P´ S, G I, B´ D, I´ M´ S, P´ M, 1995. Acute phase-dependent changes in the binding of rat liver nucleoproteins to the cytokine response element of the rat haptoglobin gene. Int J Biochem 27: 185–194. T H, S T, G J, 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci USA 76: 4350–4354. W A, L Z-H HW, 1995. The C/EBP family of transcription factors. Immunobiology 193: 171– 185. W UM, B C, L C, H PC, H F, 1993. Acute-phase response factor, a nuclear factor binding to acute-phase response elements, is rapidly activated by interleukin-6 at the posttranslational level. Mol Cell Biol 13: 276–288. W M, C Z, R MG, 1992. Calcium-regulated phosphorylation within the leucine zipper of C/EBPâ. Science 256: 370–373. X KG, P VR, C WS, S FM, C R, D JE, 1991. The different tissue transcription patterns of genes for HNF-1, C/EBP, HNF-3, HNF-4, protein factors that govern liver-specific transcription. Proc Natl Acad Sci USA 88: 3807–3811.