Abstract
Phorone, a glutathione (GSH) depletor, induces the expression of mRNAs of heme oxygenase-1 (HO-1) and c-jun by mediating the activation of activated protein-1 (AP-1) in rat livers. We have shown that phorone activates c-Jun N-terminal kinase (JNK), thus leading to c-Jun phosphorylation, and transactivation of AP-1 and HO-1 gene expression in the rat liver in response to oxidative stress. The in-gel kinase assay showed that phorone activated JNK1 predominantly in the rat liver nuclear extract. The JNK activation by phorone was slightly observed at 1 hr after administration and gradually increased with time. Ser73-phosphorylation of c-Jun catalyzed by JNK was significantly altered by changing hepatic GSH levels based on the results observed by the combined injection of buthionine sulfoximine (BSO) or GSH isopropyl ester (GIP) with phorone. Namely, BSO, an inhibitor of GSH biosynthesis, enhanced phorone-mediated c-Jun phosphorylation as well as AP-1 binding activity. However, GSH isopropyl ester prevented GSH depletion and abolished both c-Jun phosphorylation and the activation of AP-1 binding evoked by phorone. GSH isopropyl ester also suppressed phorone-produced HO-1 and c-jun gene expressions to 25 and 30% of the induced level. Perfluorodecanoic acid (PFDA) reduced GSH S-transferase activity, prevented phorone-mediated GSH depletion and abolished either HO-1 or c-jun mRNA induction by phorone. These results indicated that oxidative stress under GSH depletion produced by phorone could activate preferentially JNK and lead to the transcriptional activation of AP-1 and consequently to HO-1 gene expression. This study suggests that JNK activation could be one of the major signaling pathways to transmit intracellular events to the nuclei during oxidative stress via GSH depletion by phorone in rat livers.
Evidence has been accumulating that HO-1, which is the rate-limiting enzyme of heme degradation, is a stress responsive protein; it responds to many kinds of chemically or physiologically produced oxidative stress in various cells and tissues (Applegate et al., 1991;Rizzardini et al., 1994; Rossi and Santoro, 1995; Choi and Alam, 1996). We have reported that GSH depletors with divergent chemical structures are able to induce HO-1 in the rat liver (Yoshidaet al., 1987; Oguro et al., 1990, 1996a). We have recently shown that phorone and stilbene oxides, which are GSH depletors, increase c-jun, but not c-fos mRNA, in parallel with HO-1 mRNA induction (Oguro et al., 1996b, 1997). We have also revealed that phorone produces HO-1 gene expression via AP-1 activation in the rat liver (Oguro et al., 1996b). Many reports have shown that AP-1 binding elements of the HO-1 gene play an important role in the expression process of this gene in response to several kinds of inducers (Alam, 1994; Alam et al., 1995; Camhiet al., 1995; Choi and Alam, 1996). However, a signaling pathway leading to HO-1 gene expression from each HO-1 inducer is not yet definite.
The products of the jun and fos family genes are components of the transcription factor AP-1 (Abate et al., 1990). The transcriptional activity of AP-1 is regulated by the phosphorylation status of c-Jun (Pulverer et al., 1991). Phosphorylation of the c-Jun N-terminal transcription activation domain at Ser63 and Ser73 enhances transcriptional activity (Smeal et al., 1991). Phosphorylation of the amino-terminal domains is catalyzed by the JNK, which is also called stress-activated protein kinase (Dérijard et al., 1994; Kyriakis et al., 1994). JNKs are a superfamily of the MAP kinase family of serine/threonine kinases. JNKs are activated in response to the inflammatory cytokines, UV irradiation and other stresses such as heat shock, osmotic shock and ischemia/reperfusion (Hibi et al., 1993; Pombo et al., 1994; Adleret al., 1995; Shrode et al., 1997). Recently,Mendelson et al. (1996) showed that carbon tetrachloride stimulated JNK activation in the mouse liver. However, there are a fewin vivo studies that show the intracellular signaling pathways responded to chemicals of which any specific receptors are not known.
We demonstrated that GSH depletion evoked by phorone caused c-Jun phosphorylation leading to HO-1 gene expression in the rat liver. Additionally, we have revealed that phorone produced the activation of JNK in response to oxidative stress in the rat liver.
Materials and Methods
Materials.
Phorone, antipain, aprotinin, chymostatin, leupeptin, pepstatin A, sodium molybdate, sodium orthovanadate (v) and APMSF were obtained from Wako Pure Chemical Co. Ltd. (Tokyo, Japan). Bestatin and BSO were purchased from Sigma Chemical Co. (St. Louis, MO). Deoxycytidine-5′-[α-32P] triphosphate (3000 Ci/mmol) and [γ-32P] adenosine-5′-triphosphate (5000 Ci/mmol) were from Japan Isotope Assoc. (Tokyo, Japan). Anti-JNK1 (raised against full length human JNK1) polyclonal antibodies were purchased from Santa Cruz Biotechnology Inc. (Santa Cruz, CA). Anti-phospho c-Jun (Ser73) polyclonal antibodies (raised against a synthetic phospho-Ser73 peptide corresponding to residues 68 to 77 of human c-Jun) were from New England BioLabs, Inc. (Beverly, MA). PFDA and GIP were kindly donated by Dr. T. Ikeda (Sankyo Co., Tokyo, Japan) and Dr. M. Shibata (Yamanouchi Pharmaceutical Co., Tokyo, Japan), respectively. All other reagents used were of the highest grade commercially available.
Animals and treatment.
Male Wistar rats (weighing 160–200 g), obtained from Nippon Seibutsu Zairyo Ctr. (Tokyo, Japan), received injections i.p. with phorone dissolved in corn oil at 1 or 2 mmol/kg. In some experiments, rats were pretreated with PFDA (40 mg/kg, dissolved in corn oil), GIP (4 mmol/kg, dissolved in Na-phosphate buffer, pH 8.0) or BSO (4 mmol/kg, dissolved in Na-phosphate buffer, pH 8.0) 3 wk, 1 or 4 hr before phorone administration, respectively. Rats were starved for 24 hr before tissue collection and livers were collected at the times indicated in the figures. GSH content was determined by the method of Ellman (1959) as described by Costa and Murphy (1986).
RNA extraction and Northern blot.
Total RNA was isolated from each rat liver, using the acid guanidinium thiocyanate-phenol-chloroform extraction method (Chomczynski and Sacchi, 1987). Total RNA (20 μg) was fractionated by electrophoresis on a 1% agarose-formaldehyde gel followed by transfer onto a nylon membrane. The RNA blot was hybridized with 32P-labeled cDNA for HO-1 (Shibahara et al., 1985), c-jun (Kitabayashiet al., 1990) and glyceraldehyde 3-phosphate dehydrogenase (gift from Dr. K. Nose, Department of Microbiology, Showa University, Tokyo, Japan), as reported previously. The hybridization levels were quantitated by a bio-imaging analyzer (BAS3000, Fuji Photo Film Co., Tokyo, Japan).
Nuclear extraction.
Nuclear extracts from livers were prepared as described by Frain et al. (1989), except that all buffers contained protease inhibitors and phosphatase inhibitors, such as 0.5 mM APMSF, 14 μg/ml aprotinin, 2 μg/ml leupeptin, 2 μg/ml bestatin, 2 μg/ml antipain, 2 μg/ml chymostatin, 0.7 μg/ml pepstain A, 20 mM β-glycerophosphate, 10 mMp-nitrophenylphosphate, 50 μM sodium vanadiate and 10 mM sodium molybdate. Protein concentrations of nuclear extracts were determined by the method of Lowry et al. (1951).
Gel mobility shift assay.
The gel mobility shift assay was carried out using the GelShift Assay Kit (Stratagene, La Jolla, CA) as described previously (Oguro et al., 1996b). AP-1 double-strand oligonucleotide (5′-CTAGTGATGAGTCAGCCGGATC-3′) was labeled by reaction with T4 polynucleotide kinase (New England BioLabs) and [γ-32P]ATP; free probe was removed through a NAP5 column (Pharmacia Biotech AB, Uppsala, Sweden). Binding reactions were performed as previously reported (Oguro et al., 1996b). The gel was vacuum-dried and exposed to RX film (Fuji film).
Electrophoresis and immunoblot analysis.
The nuclear extract (40 μg protein) was solubilized in 2% SDS and the proteins were separated by polyacrylamide gel electrophoresis (3% stacking gel, 10% separating gel) according to the method of Laemmli (30). The relative mass was confirmed by comparison with protein molecular weight standards. After electrophoresis, the proteins were transferred to polyvinylidene difluoride membranes (Japan Genetics Co., Tokyo, Japan) at 80 mA for 50 min. Western blots were performed using polyclonal antibodies against Ser73 phosphorylated c-Jun. Signal was detected using chemiluminescence (CDP-Star reagent, New England BioLabs). Molecular weight was calculated with prestained SDS-PAGE standards (Bio-Rad Lab., Hercules, CA) that was applied the same gel run samples.
JNK assays.
pGEX-3X expression vectors encoding GST-c-Jun (1-79) that were constructed as described previously (Hibi et al., 1993) were provided by Dr. K. Todokoro (Institute of Physical and Chemical Research, Tsukuba, Japan). The expression vectors were transformed into the BL-21 strain of Escherichia coli. Protein induction and purification were performed as described by Smith and Johnson (1984).
The JNK immunocomplex kinase assay was essentially performed as described by Dérijard et al. (1994), using purified GST-c-Jun as a substrate. Nuclear extract (500 μg protein) was preincubated for 2 hr with normal rabbit serum conjugated with protein A-agarose (Pharmacia Biotech AB) to reduce nonspecific binding to protein A before the incubation with anti-JNK1 polyclonal antibodies conjugated with protein A-agarose (2 hr). Immunocomplexes were washed extensively in 20 mM Tris-HCl buffer (pH 7.4) containing 1 mg/ml bovine serum albumin, 1% NP-40, 100 μg/ml DNase, 50 μg/ml RNase A, 1 mM APMSF and protease inhibitor cocktail (Boehringer Mannheim, Mannheim, Germany), and washed again in phosphate-buffered saline. These complexes were washed once with kinase buffer, and then incubated in kinase buffer (20 mM HEPES, pH 7.6 containing 20 mM MgCl2, 20 mM β-glycerophosphate, 20 mM p-nitrophenyl phosphate, 0.1 mM sodium orthovanadate and 2 mM dithiothreitol) containing GST-c-Jun (6 μg) and 20 μM ATP with 0.1 μCi [γ-32P] ATP for 20 min at 30°C. The reaction was terminated by adding a 10-fold volume of Laemmli loading buffer and the reaction mixture was electrophoresed on 15% polyacrylamide gel containing 0.1% SDS. The gel was vacuum-dried and exposed to RX film.
In-gel kinase assay was examined as described by Hibi et al.(1993). Namely, 80 μg of nuclear extract were separated on a 15% separation gel containing 0.1 mg GST-c-Jun (1–79)/ml and 0.1% SDS. The gel was washed twice with 20% 2-propanol, 50 mM HEPES, at pH 7.6 to remove SDS. The gel was then washed twice with buffer A (50 mM HEPES at pH 7.6, 5 mM β-mercaptoethanol). It was then incubated in 6 M urea in buffer A at 25°C for 1 hr, followed by serial incubations in buffer A containing 0.05% Tween 20 and wither 3, 1.5 or 0.75 M urea. The gel was then washed several times with buffer A containing 0.05% Tween 20 at 40°C. The renatured gel was incubated in kinase buffer containing 25 μM [γ-32P] ATP (5 μCi/ml) for 1 hr at 30°C. The gel was washed with 5% trichloroacetic acid and 1% sodium pyrohposphate at 25°C several times, followed by drying and autoradiography.
Statistical analysis.
The results were analyzed by the Student’s t test.
Results
We have shown that GSH depletion plays a role in triggering HO-1 gene expression evoked by phorone (Yoshida et al., 1987;Oguro et al., 1996b). To confirm further this phenomenon, PFDA, which has been shown to be able to decrease GST proteins and mRNAs (Schramm et al., 1989), was used to inhibit GSH conjugation of phorone and thus reduce the extent of decrease in GSH. Pretreatment of rats with PFDA 3 wk before the administration of phorone (1 mmol/kg) reduced hepatic GST activity to ∼80% of the control or phorone alone (data not shown). As shown in figure1, PFDA pretreatment also resulted in the inhibition of GSH depletion evoked by phorone (18% of the control level) to 45% of the control level. Additionally, PFDA dramatically inhibited either HO-1 or c-jun gene expressions induced by phorone.
Figure 2 shows the pretreatment effect of GIP on phorone-induced HO-1 and c-jun mRNAs in rat livers. Phorone (1 mmol/kg) decreased GSH content to 20% of the control level, and increased HO-1 mRNA (8-fold of the control) and c-jun mRNA (9-fold of the control) at 2 hr after the administration as we have shown previously (Oguro et al., 1996b). GIP is a GSH derivative that is permeable to the plasma membrane (Noguchi et al., 1989). Therefore, the administration of GIP increased hepatic GSH content to 1.5-fold of the control level at 3 hr. Under the present conditions, GIP suppressed GSH depletion evoked by phorone and prevented phorone-mediated HO-1 and c-jun mRNAs induction to 20 and 30% of the induced level, respectively.
We have previously shown that phorone induces the activation of AP-1 binding activity and BSO, an inhibitor of GSH biosynthesis, enhanced HO-1 and c-jun gene expressions by phorone (Oguro et al., 1996b). In this study, we carried out gel mobility shift assays to assess the contribution of sulfhydryl level to the activation of AP-1 binding to cis-element. According to our previous results (Oguro et al., 1996b), phorone increased AP-1 binding that was composed by Jun/Jun (Oguro et al., 1996b) at 4 hr after the administration (fig. 3). Pretreatment with BSO (4 mmol/kg) 4 hr before phorone (1 mmol/kg) injection resulted in a marked increase of AP-1 binding activity (fig. 3A). BSO produced a more profound depletion of GSH content (BSO and phorone, 0.67 ± 0.01 μmol GSH/g liver; control, 3.96 ± 0.20 μmol GSH/g liver) as compared with that of phorone alone (1.20 ± 0.15 μmol/g liver) that follows our previous reported results (Oguro et al., 1996b). In contrast, GIP significantly suppressed phorone-induced AP-1 binding activity (fig. 3B).
Transcriptional activity of AP-1 is regulated by the phosphorylation state of the transactivation domain of c-Jun N-terminus (Smeal et al., 1991). Because this phosphorylation of c-Jun is catalyzed by JNK, a time course study was done on the effect of phorone on JNK activity in the rat liver nuclear extract (fig.4, A and B). The JNK activity increased 2 to 4 hr after phorone treatment (fig. 4A). The result of in-gel assay was consistent with the result of the immunocomplex kinase assay (fig.4B). In addition, it showed that phorone increased the activity of JNK1 (46 kDa) more than that of JNK2 (55 kDa) up to 4 hr after the injection.
To determine an interrelationship between c-Jun activation and hepatic GSH levels, the effects of pretreatment with BSO or GIP before phorone injection on the Ser73 phosphorylation of c-Jun in the rat liver nuclear extract was examined by using Western blotting (fig.5). In the nuclear extract of the control rat liver, no visible band of Ser73-phosphorylated c-Jun was found. Phorone (1 mmol/kg) induced the phosphorylated c-Jun at 4 hr (fig. 5B). BSO pretreatment resulted in more elevation of Ser73 phosphorylation of c-Jun than phorone alone (fig.5A). In contrast, the combined injection of GIP with phorone caused a less intense band of phosphorylated c-Jun as compared to the compound alone (fig. 5B).
Discussion
This study was undertaken to clarify the molecular mechanisms of HO-1 gene expression induced by phorone. Previously, we have revealed that HO-1 gene expression evoked by phorone is mediated through AP-1 activation; probably it is due to the profound decrease of GSH by the compound (Oguro et al., 1996b). Many studies have shown the importance of AP-1 elements in HO-1 gene expression caused by various HO inducers in vitro and in vivo (Alam, 1994;Alam et al., 1995; Camhi et al., 1995; Choi and Alam, 1996). However, it still remains to be determined how information on cytoplasmic events produced by chemicals, for which specific receptors are not known, are transduced to the nuclear genes. We found that GSH depletion evoked by phorone resulted in the activation of JNK activity, suggesting that JNK is one of the signal transduction pathways for oxidative stress caused by GSH depletor.
Phorone decreased GSSG in parallel with GSH content in the liver (data not shown). And this compound did not change the activities of GSH reductase and GSH peroxidase under our experimental conditions (data not shown). Consequently, GSH depletion by phorone plays an important role in HO-1 gene expression, as we reported previously (Oguro et al., 1990, 1996b). However, it cannot be ruled out that phorone itself is directly involved in this gene expression. Therefore, PFDA was used to reduce metabolism of phorone by GSH conjugation, so as to inhibit GSH depletion by the compound. The experimental results of pretreatment of rats with PFDA before phorone injection clearly indicated that GSH depletion rather than phorone itself plays the key role for either HO-1 or c-jun gene expression by the compound. PFDA almost completely inhibited HO-1 and c-Jun gene expression, but GSH still remained to be 50% of the control level. This result indicates that PFDA could inhibit some gene transcriptions. In this respect, however, further study will be needed.
Transcriptional activation of AP-1 is regulated by phosphorylation of c-Jun amino-terminal domain catalyzed by JNKs (Smeal et al., 1991; Hibi et al., 1993), which have been shown to be activated by many kinds of oxidative stresses (Hibi et al., 1993; Pombo et al., 1994; Sluss et al., 1994; Loet al., 1996). However, all of these experiments were mainly carried out in vitro using established cell lines. As shown in this study, phorone activates JNKs in a time-dependent manner that coincides with the time-dependent increase of c-Jun phosphorylation (our unpublished data) in the nuclear extract of the rat liver. We did not examine the effects of phorone on the activities of other MAP kinases, ERK and p38 kinase in this experiment. However, the results of time-dependent increases in JNK activity after phorone administration are well coincided with the alterations of AP-1 binding activities, HO-1 and c-jun gene expressions, suggesting that the events, mainly oxidative stress due to GSH depletion produced by phorone, are transduced to genes through JNKs in part. In addition, the combination studies of BSO or GIP with phorone stimulated or diminished phosphorylated c-Jun evoked by phorone, respectively. These results indicated that GSH depletion evoked by phorone was one of the important intracellular events to stimulate transactivation of AP-1 which is partially due to JNK, thus leading to HO-1 gene expression.
Two JNK isoforms (JNK1; Mr 46,000 and JNK2; Mr 55,000) have been identified in HeLa cells; both of them are capable of phosphorylation of the transactivation domain of c-Jun (Hibi et al., 1993). Our results indicate that the JNK1 isoform contributed predominantly to phorone-mediated stimulation of JNK activity in contrast to JNK2; these results are similar to those obtained by UV and cytokines (Hibiet al., 1993; Lo et al., 1996).
JNKs are potently activated by many kinds of stresses such as UV exposure, hyperosmosis, tumor necrosis factor-α, interleukin-1 and heat shock (Hibi et al., 1993; Dérijard et al., 1994; Sluss et al., 1994; Adler et al., 1995; Lo et al., 1996; Shrode et al., 1997). The mechanism by which stress culminates in JNK activation remains largely to be determined. In this respect, Adler et al. (1995)reported that UV-induced, but not heat shock-mediated JNK activation was dependent on membrane-associated components and free oxygen radicals. In addition, it has been recently suggested that reactive oxygen species produced by cytokines and nitric oxide would be a second messenger to stimulate JNK, and some low molecular weight G proteins (such as p21ras, Rac1 and CDC42) might be sensors in JNK activation for reactive oxygen species (Lander et al., 1996; Lo et al., 1996). However, we have evidence that phorone did not produce the increases in interleukin-1, tumor necrosis factor-α and inducible nitric oxide synthetase mRNAs up to 4 hr after administration (Oguro T, Hayashi M, Hausmann EHS and Yoshida T, unpublished results). These facts again indicated that GSH depletion by phorone might stimulate such stress sensor(s) directly to cause JNK activation and the subsequent HO-1 gene expression.
In conclusion, this study has revealed that phorone, a GSH depletor, is able to activate JNKs, thus leading to AP-1 transactivation and consequently to HO-1 gene expression in the rat liver.
Acknowledgments
The authors thank Dr. W. J. Waddell (University of Louisville, KY) for helpful review of this manuscript. We thank Dr. K. Yokoyama [Tsukuba Life Science Center, The Institute of Physical and Chemical Research (Riken), Tsukuba, Ibaraki 305, Japan] and Dr. K. Nose (Department of Microbiology, Showa University, Tokyo, Japan) for supplying the c-jun and GAPDH cDNA probes, respectively. We also thank Dr. T. Ikeda (Sankyo Co., Tokyo, Japan) and Dr. M. Shibata (Yamanouchi Pharmaceutical Co., Tokyo, Japan) for providing the PFDA and GIP, respectively.
Footnotes
-
Send reprint requests to: Dr. Takiko Oguro, Department of Biochemical Toxicology, School of Pharmaceutical Sciences, Showa University, 1-5-8 Hatanodai, Shinagawa-ku, Tokyo 142, Japan.
-
↵1 This work was supported in part by a Grant-in-Aid from the Ministry of Education, Science and Culture, Japan.
- Abbreviations:
- GSH
- glutathione
- HO
- heme oxygenase
- AP-1
- activated protein-1
- JNK
- c-Jun N-terminal kinase
- BSO
- buthionine sulfoximine
- GIP
- GSH isopropyl ester
- PFDA
- perfluorodecanoic acid
- GST
- GSH S-transferase, MAP kinase, mitogen-activated protein kinase
- APMSF
- p-amidinophenylmethanesulfonyl fluoroide
- ERK
- extracellular signal related kinase
- SDS
- sodium dodecyl sulfate
- HEPES
- 2-[4-(2-hydroxyethyl)-1-piperazinyl]ethane sulfonic acid
- Received November 25, 1997.
- Accepted May 26, 1998.
- The American Society for Pharmacology and Experimental Therapeutics