Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Requirement of Tyrosine- and Serine/Threonine Kinases in the Transcriptional Activation of the Mammalian grp78/BiP Promoter by Thapsigargin

Requirement of Tyrosine- and Serine/Threonine Kinases in the Transcriptional Activation of the... THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 270, No.1, Issue of January 6, pp. 494-502, 1995 © 1995 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. Requirement of Tyrosine- and SerinelThreonine Kinases in the Transcriptional Activation of the Mammalian grp78/BiP Promoter by Thapsigargin* (Received for publication, August 8, 1994, and in revised form, October 21, 1994) Xianjun Cao, Yanhong Zhou, and Amy S. Leez From the Department ofBiochemistry and Molecular Biology and the Norris Cancer Center, University of Southern California School of Medicine, Los Angeles, California 90033-0800 Depletion of endoplasmic reticulum (ER) Ca + store by conditions, particularly those which perturb ER function (Lee, thapsigargin (Tg) in mammalian cells induces a set of ER 1987; Watowich et al., 1988; Kozutsumi et al., 1988; Lee, 1992; protein genes known as the glucose-regulated proteins. Liu et al., 1992; Price et al., 1992). In mammalian cells, one Recently, IRElp, a transmembrane protein postulated to of the most potent inducers for grp78 is chronic depletion of have a serine/threonine kinase activity, has been identi­ ER calcium stores by Tg or the calcium ionophore A23187 fied as required for the induction ofER resident proteins (Resendez et al., 1986; Drummond et al., 1987; Li et al., 1993). genes in Saccharomyces cerevisiae. To investigate The induction process is independent of a rise in cytoplasmic whether IRElp can stimulate mammalian grp transcrip­ 2+; Ca furthermore, it requires a prolonged period of (2 h) of tion, a stable Chinese hamster ovary cell line containing treatment and is sensitive to cycloheximide. Thus, the signal­ amplified copies of IRElp has been created. The IRElp ing mechanism for the activation of grp78 transcription in­ expressing transfectants exhibited a modest (2-fold) en­ 2+ volves more components than merely the efflux of Ca from hancement ofboth the basal and Tg induced level ofgrp78 the ER. and grp94, two coordinately regulated grp genes. Using In examining the regulatory elements of the rat grp78 pro­ okadaic acid as a specific inhibitor for the endogenous moter which mediate the Tg and A23187 induction response, serine/threonine protein phosphatase activities, a mild (2·fold) stimulative effect was observed for Tg induction two functional domains have been identified (Wooden et al., of grp78 transcription. The okadaic acid potentiating ef­ 1991; Li et al., 1993). One region spans the promoter sequence fect requires a 50-base pair region in the vicinity of the from -159 to -110 and contains the grp core element which is grp78 TATA element. In contrast, the transcriptional ac­ conserved from yeast to human (Resendez et al., 1988) and a C3 tivation of grp78 by Tg is almost totally eliminated by element (Li et al., 1993). The other region spans -109 to -74 genistein, a tyrosine kinase inhibitor. The grp core, the C3 and contains a CCAAT motif (Cl) most proximal to the TATA and Cl elements which are major Tg response elements of element (Wooden et al., 1991). Both regions are able to confer the rat grp78 promoter, are also major targets of the in­ Tg and A23187 inducibility to a heterologous promoter upon hibitive effects of genistein. transient transfection into mammalian cells (Li et al., 1993). Using ligation-mediated PCR, we found that specific changes in factor occupancy occurred after stress induction and that the Thapsigargin (Tg)1 is a tumor promoter whose intracellular major changes occur within the grp core element (Li et al., target has been identified as the endoplasmic reticulum (ER) 1994). The factor which exhibits specific binding to this region 2+-ATPase Ca (Jackson et al., 1988; Thastrup et al., 1990). appears to be constitutive, suggesting that during stress, the Through its inhibition of the enzyme, Tg specifically discharges factor either undergoes a conformational change or has disso­ the ER calcium store. Within the ER resides a set of proteins ciated from inhibiting elements, resulting in the changes in the known as the glucose-regulated proteins (GRPs) (Lee, 1992). in vivo footprinting pattern. GRP78, a 78-kDa protein also referred to as the immunoglob­ 2+ The discovery that chronic depletion of Ca from the ER ulin heavy chain protein (BiP), is one of the best characterized store could lead to grp78 transactivation correlated with factor GRPs with protein chaperoning and calcium binding properties interaction changes at the grp78 promoter, raises the impor­ (Pelham, 1986; Little et al., 1994). Although GRP78 is consti­ tant question of intracellular compartment signaling. What are tutively expressed at a low basal level in many different tis­ the mediators of this response? The transcription of grp78 in sues, its transcription is highly inducible by a variety of stress fibroblast cells is partially inhibited by W-7 which inhibits 2+ Ca /calmodulin-regulated enzyme activities (Resendez et al., * The work was supported by United States Public Health Service 1986). Other studies suggest an involvement of tyrosine and Grant R37 CA27607 from the National Cancer Institute (to A. S. L.). serine/threonine kinases and phosphatases. In the mammalian The costs of publication of this article were defrayed in part by the system, pretreatment ofNIH3T3 cells with genistein, an isofla­ payment of page charges. This article must therefore be hereby marked vone which specifically inhibits a number of tyrosine kinases, "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. reduces Tg induction of grp78 mRNA levels, whereas okadaic :j:To whom correspondence should be addressed: Tel.: 213-764-0507; acid (OA), a specific inhibitor of serine/threonine protein phos­ Fax: 213-764-0094. phatases type 1 (Plt-L) and type 2A (PP-2A) (Schonthal, 1992), 1 The abbreviations used are: Tg, thapsigargin; ER, endoplasmic re­ enhances Tg induction ofgrp78 (Price et al., 1992). In the yeast ticulum; GRP, glucose-regulated proteins; PCR, polymerase chain re­ action; OA,okadaic acid; PP-l and -2, protein phosphatases type 1 and system, a IRE1 /Ern1 gene encoding for a transmembrane ser­ type 2; EGF, epidermal growth factor; EGFR, epidermal growth factor ine/threonine kinase has recently been identified as required receptor: L-A,lavendustin-A; DHFR, dihydrofolate reductase; kb, kilo­ for the transcriptional induction of KAR2, the yeast homologue base pairts); CAT, chloramphenicol acetyltransferase; bp, base pairts); of grp78, in response to the accumulation of malfolded protein MMTV, murine mammary tumor virus; PAGE, polyacrylamide gel elec­ trophoresis; MTX, methotrexate; RSV, Rous sarcoma virus. in the ER (Cox et al., 1993; Mori et al., 1993). It has been This is an Open Access article under the CC BY license. Thapsigargin Induction ofgrp78 Requires Protein Kinases 495 Northern Blotting-After treatment with protein kinase inhibitors postulated that IRE1p is the proximal sensor of events in the and thapsigargin for 7 h, the cells were washed twice in phosphate­ ER and that binding of ligands causes transduction of informa­ buffer saline. Total cellular RNA was extracted following isohigh tion across the ER membrane, leading to activation of a specific method as described previously (Zhong et al., 1994). 10 JLg of RNA was set of transcription factors. IREp1 is of very low abundance in loaded to 1% agarose, 2.2 M formaldehyde gels. The probe, a grp78 yeast cells. Its ligand and the involvement of IRE1plErn1p in cDNA fragment (1.5 kb) from p3C5 digested with EcoRI and PstI, was Ca + signaling of grp78 have not yet been determined. labeled with the hexamer method to specific activities of 2 X 10 Since IRE1p is the first and the only serine/threonine kinase cpm/ug, p3A10, the entire plasmid, was labeled with the same method and used to assess the amount of loaded RNA (Lin and Lee, 1984). identified so far which can affect KAR2/grp78 transcription, we Conditions ofthe RNA blot and hybridization were described previously created a mammalian expression vector for the yeast IRE 1p (Lee et al., 1983). Autoradiograms were quantitated by the densitome­ and selected for stable transfectants containing amplified cop­ try to obtain relative levels of the grp78 and grp94 mRNA. ies of the yeast gene. We report here that expression ofIRE1p Plasmids-Plasmids (-456)CAT and (-355)CAT, previously re­ in hamster ovary cells is correlated with a 2-fold increase in the ferred to as pE43 and p3K, respectively (Changet al., 1987), contain the basal and Tg-induced levels of the endogenousgrp78 mRNA. In rat grp78 promoter at deletion end points of -456 and -355. The mammalian cells, grp78 is coordinately regulated at the tran­ construction of the plasmids (-375/-88)MCAT, (-159/-MCAT, (-109/ scriptional level with grp94, which encodes a 94-kDa GRP, -74)MCAT has been described (Li et al., 1993). Plasmid MCAT con­ tains the minimal MMTV promoter fused to the chloramphenicol acetyl­ through common cis-regulatory elements and trans-acting fac­ transferase (CAT) gene (Kim and Lee, 1989). pRSV-IRE was tors (Chang et aZ., 1987; Liu and Lee, 1991). As in the yeast constructed by inserting the full-length of IREI gene (3.3 kb) into Sail system where IREl also induces other ER resident protein and BamHI sites of pRSV-1 (gift of Dr. C. Gorman)(Li et al., 1992). The genes (Cox et aZ., 1993), the mammalian IRE1p transfectants full-length of IREI gene was generated by the PCR using the plasmid also exhibit 2-fold increase in the basal and Tg-induced levels of pCS110 (Cox et al., 1993; gift of Dr. P. Walter, University of California, grp94 mRNA. San Francisco) as template. pBluescripts KS (pBS) was purchased from Because of the modest effect observed with overexpression of the Stratagene. pBS-IRE contains the IREI gene derived from pRSV­ IRE digested with Saii and BamHI. the yeast IRE1p protein in mammalian cells and the inherent In Vitro Transcription and Translation-The transcription of IREI difficulties associated with expression of a heterologous pro­ gene from pBS-IRE was carried out in 20 JLl of reaction mixture follow­ tein, we investigated the respective contributions of serine/ ing the T3 transcription kit protocol (Amersham Corp.). The RNA was threonine protein phosphatases and tyrosine kinases toward precipitated, and further translated in the 25 JLl of reaction mixture to the Tg induction of grp78 in two different mammalian cell 35S1 synthesize [ methionine-Iabeled IRE1 protein (IRE1p) using a rabbit lines. Our results indicate that genistein eliminates the re­ reticulocyte translation system (Promega). 5 JLl of translation reaction sponse almost completely. We further demonstrate that the mixture was separated on 8% SDS-PAGE. inhibitive effect of genistein is at the level of transcription. We Transient Transfection and CAT Assay-A total of 2 X 10 NIH3T3 cells/100-mm dish was plated 20 h before transfection. 10 JLg of test found that the grp core, the C3 and C1 elements which are the plasmid was co-transfected with 5 fJ-g of PCHllO, an expression vector major Tg response elements ofthe ratgrp78 promoter (Li et aZ., for J3-galactosidase (Hall et al., 1983), with the calcium phosphate 1993), are also major targets of the inhibitive effects of genis­ precipitation method in the presence of 3 JLg of carrier HeLa genomic tein. The tyrosine kinase activity essential for the Tg induction DNA. After 24 h the cells were treated with either thapsigargin (300 ofgrp78 is likely to be distinct from EGFR, since lavendustin-A nM), genistein (140 fJ-M), okadaic acid (100 nxr), or lavendustin-A (50 nM (L-A), a potent inhibitor for EGF-associated tyrosine kinase or 1 fJ-M) alone or in combination of two reagents for 16 h. For the cells activity (Onoda et al., 1989), is without effect. that were treated with two reagents, okadaic acid, genistein, or laven­ dustin-A was added 15 min prior to the addition of thapsigargin. The Using OA as a specific inhibitor for serine/threonine protein cells were further incubated in the presence of both reagents for 16 h phosphatase activities, a mild (about 2-fold) stimulative effect before protein extraction. The protein concentration of the cell extract was observed for Tg induction of grp78 mRNA levels. Nonethe­ was determined by Bio-Rad protein assay. The CAT activities were less, the OA stimulative effect is not mediated through the Tg measured at the linear range after normalization to the J3-galactosidase response elements; rather, it requires sequences in the vicinity activities. The developed TLC plates were quantified on an Ambis of the grp78 TATA element unrelated to the known OA re­ Radioanalytic Imaging System (Arnbis System, San Diego, CAl and sponse elements. Thus, in the mammalian system, although a exposed to films for autoradiography. Most plasmids were transfected serine/threonine kinase activity may potentiate slightly the two to five times independently. Generation of Stable Transfectants-lO fJ-g of pRSV-1 or pRSV-IRE grp78 Tg response, a genistein-sensitive activity acting was transfected into DG44 cells by using the calcium phosphate pre­ through the Tg response elements is the critical component of cipitation method. Twenty-four h after transfection, the cells were the signal transduction system that leads to transcription ac­ changed into the a-minus medium and grown for 14 days. The surviving tivation of grp78. colonies were exposed to 0.005,0.01,0.02,0.05,0.1,0.2,0.5, 1,2, and 5 MATERIALS AND METHODS JLM MTX for the further selection as described previously (Li and Lee, 1991). The resistant colonies were pooled together and maintained in Cell Cultures and Inhibitor Treatments-K12 cells, which are ham­ the a-minus medium with the different concentrations of MTX. ster fibroblast temperature-sensitive mutants derived from WglA as Genomic DNA Hybridization-The stable transformants were grown to described previously (Lee, 1981), were maintained in Dulbecco's modi­ 90% confluence and washed twice in phosphate-buffer saline. The fied Eagle's medium containing 10% of cadet calf serum. NIH3T3 cells genomic DNA was extracted by using the previously described method (Li were maintained in Dulbecco's modified Eagle's medium containing et al., 1989). 5 JLg of each genomic DNA and 0.1 JLg of pRSV-IRE were 10% of fetal calf serum. For the inhibitor studies, cells were grown to digested with EcoRI. The probe, consisting of the full-length IRE] gene 90% of confluence in the above media containing 1% antibiotics (50 from Saii and BamHI double digestion of pRSV-IRE, was labeled by the JLg/ml penicillin, 50 JLg/ml streptomycin, and 100 JLg/ml of neomycin) hexamer method with specific activities of 2 x 10 cpm/ILg. Conditions of and then preincubated with the inhibitors (genistein, lavendustin-A, the DNA blot and hybridization were described previously (Li et al., 1989). and okadaic acid) (Life Technologies, Inc.) for 15 min in the fresh Western Blot-The stable transformants in 100-mm dishes were medium. Afterwards, the thapsigargin (Sigma) was added to incubate grown to 90% confluence in the presence of 5.0 ILM MTX and washed for another 7 or 16 h. DG44 cells, a double mutant of Chinese hamster ovary cell lines, lack the hamster dihydrofolate reductase (DHFR) gene twice in phosphate-buffer saline. Cells were resuspended in 100 fJ-l of (gift of Dr. Lawrence Chasin, Columbia University). These cells were lysis buffer (8 M urea, 10% Nonidet P-40 and 5% J3-mercaptoethanol). maintained in the minimal essential a medium containing Lvglutamine, The supernatants were separated on 8% SDS-PAGE. The IREp was ribonucleotides, and deoxyribonucleosides (Life Technologies, Inc.) sup­ detected by using a rabbit polyclonal antisera (gift of C. Shamu and P. plemented with 10% of fetal calf serum (Urlaub et al., 1986). The Walter, University of California, San Francisco) directed against the N DHFR+ transformants can be selected in the same medium without the terminus of IRElp 0:2000 dilution) following Immun-Life Chemilumi­ nucleosides (a-minus). nescent Assay Kits protocol (Bio-Rad), 496 Thapsigargin Induction ofgrp78 Requires Protein Kinases RESULTS Expression of the Yeast IREp in Mammalian Cells-Re­ cently, a 1115-amino acid transmembrane protein encoded by IREl /ERNl in Saccharomyces cerevisiae has been isolated from genetic screens based on its ability to activate KAR2/ grp78 transcription (Cox et al., 1993; Mori et al., 1993). The cytoplasmic C terminus of this protein carries an essential kinase activity related to a cdc2+/CDC28 kinase family and is postulated to be a serine/threonine kinase (Mori et al., 1993). Although it is not known whether a homologue of IREl exists K::J in mammalian cells, with its discovery in the yeast system, it is ! peR possible to test whether IRElp is functional in mammalian cells in activating grp78 transcription. Our approach is to ob­ TAA ATG tain stable mammalian cell lines where the yeast IREl gene is ITJ DJ amplified to produce high levels of IRElp and to test for its ability to enhance grp78 expression over the endogenous level. For this purpose, the yeast IRElp coding sequence was ob­ tained by engineering PCR primers directly flanking its ATG s B translation initiation codon and the termination codon (Fig. 1). After PCR amplification using the yeast genomic clone pCS110 as a template, the 3.3-kb fragment containing the IRElp coding sequence was subcloned into the multiple cloning site located between the Rous sarcoma virus (RSV) long terminal repeat SV40 and an SV40 polyrA) site of the DHFR expression vector pRSV-l (Li et al., 1992). This mammalian expression vector for IRElp is referred to as pRSV-IRE. For purpose of in vitro transcription, the same IRElp coding sequence was subcloned into a T3 vector and is referred to as pBS-IRE. Prior to transfection into mammalian cells, we established sit!- ~ that the IRElp coding sequence obtained from PCR was capa­ ble of translating into a protein close to the 127-kDa size predicted for IRElp. In vitro transcribed IRElp RNA was translated in a rabbit reticulocyte system in the presence of 3 5 [ SJmethionine. As shown in Fig. 2A, only the sample contain­ ing the IRElp mRNA was capable of synthesizing a protein band of about 130 kDa. Thus, the PCR DNA sequence is devoid of cryptic termination codons. Next, pRSV-IRE and as a con­ trol, the vector pRSV-l, were transfected into DG44 cells, which is a CHO cell line devoid ofDHFR activity. Stable trans­ fectants were selected, and the transfected plasmids were am­ FIG. 1. Construction of the expression plasmids. Plasmid plified by subjecting the transfected cells to increasing concen­ pCSllO containing the yeast IREl gene was used to PCR-amplify a 3.3-kb full-length of IREl gene with Sall and BamHI sites at two ends. trations of MTX. Southern blot analysis of genomic DNA The IREl gene was then subcloned into pRSV-l immediately after the prepared from these cells revealed that at MTX concentrations LTR promoter to generate the pRSV-IRE plasmid, in which there is a of 0.2 P.M and higher, there was notable amplification oflRElp selection marker, DHFR gene under the control of SV40 promoter. This coding sequence in the pRSV-IRE transfectants (Fig. 2B). By plasmid was used to generate the stable cell line for the expression of IREl gene. The IREl gene derived from pRSV-IRE was further sub­ comparing the band intensities with the control plasmid ran in cloned into pBS to generate the pBS-IRE plasmid. The pBS-IRE was parallel in the same DNA blot, we estimated the copy number used as a template in in vitro transcription under the direction of a T3 of the IRElp coding sequence in the stable transfectants was promoter. about 50. To confirm that pRSV-IRE stable transfectants were expressing IRElp, total protein extract was prepared from mRNA was measured by Northern blot analysis (Fig. 3A), and stable transfectants selected by 5 P.M MTX, as well as from pRSV-l transfectants subjected to the same MTX selection. To the fold induction was determined after normalization with a detect IRElp, a Western blot was performed using antisera nonvariant control transcript 3AI0 (Lin and Lee, 1984; Li et al., 1993) (Fig. 3B). We observed that in the IRElp-expressing raised against the N terminus oflRElp. In the protein sample from the pRSV-IRE transfectants, an immunoreactive band of cells, the basal and Tg-induced levels of grp78 mRNA were 130 kDa was detected. This band was not observed in the approximately 2-fold higher than the control cells. An interest­ ing characteristic of irel negative mutants in yeast is that not control sample from pRSV-l transfectants (Fig. 2C). These only is KAR2 /grp78 transcription downregulated, transcrip­ results establish that in the pRSV-IRE stable cell line where the IRElp coding sequence was highly amplified, sufficient tion activation of other ER resident protein genes such as amounts of IRElp was produced to allow detection by immu­ protein disulfide isomerase is also negatively affected (Cox et al., 1993). In mammalian cells, there is tight coordinate regu­ noreactivity with antibody specific for IREp. Endogenous grp78 and grp94 mRNA Levels in IRElp-ex­ lation of the grp 78 gene with thegrp94 gene (Lee, 1987). To test pressing Transfectants-To test whether expression of IRElp whether expression of IRElp in mammalian cells would also affectgrp94 expression, the level ofgrp94 mRNA in the pRSV­ has any effect on grp expression, total cytoplasmic RNA was isolated from both the pRSV-IRE and pRSV-l transfectants IRE and pRSV-l stable transfectants was also examined. As in under nontreated and Tg-treated conditions. The level ofgrp78 the case of grp78 mRNA, the level of grp94 mRNA was in- 497 Tha pniga rg in Induction of grp78 Requ ires Protein Kinases A c cr:: ::;: :> c. (]) (/) (/) w c cr:: cr:: ~ g c. c. kDa 21 1- 211- 10 7­ 10 7­ 69 - 69 - 47- 47- 2 3 FIG. 2. IRElp expression ill vit ro and ill vivo. A , 1""S lmethionine-labeled lucifernse tlane 1) and IHEI p tlane 2 ) were synt hes ized in the ill vitro ra bbit reticulocyte system and analyzed by 8';f, SDS·PAGE and autora diogra phy. As a negative control ilane ,'J), a translation reaction mixture was performed without any mRNA. The arrow indicates the location ofl''''S lmethionine-labeled IREl p. 13, pHSV-l and pHSV-IHE were stably trans fected into DG44 cell lines. Genomic DNA were isolated from DG44 ilane 1), pRSV-l stable tran sfectants selected by 5 J.L ~I MTX ilane 2) and pHSV-IHE transfecta nts selected by increasi ng concen trations <0.005-5.0 J.L ~ I) of MTX tlanes .'1- 6). The DNA was digested with EcoHIa nd separa ted on the ugarose gel, blott ed onto the filter , and hybrid ized with the !R EI gene. In lan e 7, the pRSV-mE plasmid was digest ed with EcoHI. C, th e IHElp express ion in pHSV-l (lane 1) and pHSV-mE tlane 2)stably tra nsfected cells selected by 5 J.L ~ I l\ITX was ana lyzed by Western blot usin g antisera aga inst yeas t m E 1p, The arrow indicates the location of lREp. creased by about 2-fold in nonstressed and Tg-stressed cells due to cha nges in the tran scriptiona l rate of the gl1J78 gene or (Fig. 3, A a nd B). Overall, our results indicate that amplifica­ due to changes in the stability of the gl1J78 mRNA . To addres s tion of the yeast IRE l p gene in mamm alian cells and overex­ this issue, we exa mined whether geniste in and OA could exert pression of its protein only showed a modest sti mulative effect the sa me effects on the express ion of a marker gene linked to a grp78 promoter . Previously, we have shown (- 456)CAT, which on the grp78 and grp94 mRNA levels. Se ns itivity of Tg In du ction ofgrp78 to Genistein and Ohadaic contains a 4 18-bp fragment spa nning - 456 to - 38 of the rat Aci d- We next investigated the respective cont ributions of en­ grp78 promoter link ed to a CAT gene, is able to confer a dogen ous se rine/threonine pr otein phosphatases a nd ty rosine 5-6-fold induction of CAT activ ity following Tg induction of K12 transfect ant s harborin g this plasmid (Li et al., 1993). In kinases toward the 'I'g induction of grp78 through the use of th eir specific inh ibitors. Two mammal ian cell lines (NIH3T3 this study, using NIH3T3 cells as recip ien t cells in transient and K12 ha mster fibr oblast cells ) were tested for the effect of tra ns fection assays, a similar result was observed such that the genistein and okadaic acid on the induction of grp78 by Tg. In CAT activity was enhanced ab out 6-fold when the cells were induced by Tg (Fig. 5, A a nd B ). Compa red wit h th e nontrea ted the in hibi tor studies , the cells wer e pre treated with the inhib ­ itors, and subsequently Tg was add ed for 7 h, since the maxi­ cells wit h basal level set as 1, OA an d genistein tr eatment mum level of grp78 mRNA accumu lation occurred afte r 6-h of alone resul ted in 1.4- a nd 0.8-fold, respectively. When OA· addition of 300 nxt ofTg for K12 cells (Li et al., 1993). RNA blot treated cells were ind uced by Tg, the indu ction was increased ana lysis, as shown in Fig. 4A , was performed to quantit ate the to 12-fold. In contrast, geniste in treatment elimina ted almost grp78 mRNA level which was norm a lized agains t 3A10. The completely the Tg response. Since the CAT activities are com­ relative RNA levels for both cell lines are su mma rized in Fig. parab le with the endogenous grp78 mRNA level upon addition 4B . Consiste nt with the previous study (Li et al., 1993), there of OA or G in the Tg induction (compare Fig. 4B and 5B ), we conclude that th e major effect of genistein and OA on grp78 Tg was a 12-15-fold increase in grp78 mRNA cells a fte r Tg treat­ men t. OA or ge niste in alone only showe d min or effect on grp78 ind uction is at the transcriptional level and is mediated mRNA basal level express ion. The most striking resul t is that through the glp 78 promoter eleme nts . when genistein, an inhibitor for ty rosine kin ase acti vity, was The Inh ibitive Effect of Genistein Acts through the grp78 Tg added in combina tion with Tg, there was a dr astic redu ction in Response Elem ents- Since genistei n treatment elimina ted al­ rast, OA, a serine/ most completely the Tg response, we tested whet her the grp78 the grp 78 mRNA indu ction level. In cont threo nine protein phosp hatase inhibitor, in combination with Tg response eleme nts are target s for the genistein mediated Tg only showe d a slight potenti at ing effect in K12 cells, inhibitive effect. A panel of glp 78 promoter fragmen UCAT fu­ whe reas in NIH3T3 cells, a 2-fold increase was det ected. Based sion genes previously demonstrated to be able to respond to or rvations, we conclude that in mammali an cells, an confer Ca depletion stress was ut ilized (Kim and Lee, 1989; on these obse Li et al., 1993). As shown in Fig. 6, (- 456)CAT cont ain s mul ­ endogenous tyrosine kinase activity sensitive to geniste in is a critical compone nt of the transactivation mechan ism for the Tg tiple arrays of regulatory eleme nts upstream of the TATA element located at - 55. The important elements for Tg induc­ induct ion of grp78. Se rine/t hreonine protein phosphorylation ntribute partially to the overall tion previously iden tified by 5' deletion, linker scanning, and sensitive to OA may also co grp78 ind uction level in NIH3T3 cells. internal deleti on mutants incl ude the core at - 150, the C3 The Genistein and OA Effect A re Med iated thro ugh the grp78 element at - 125, and the C1 eleme nt at - 95 (Wooden et al., 1991; Li et al., 1993). There are additiona l upstream en­ Promoter- T he cha nge in grp78 mR NA levels when cells were pre treated wit h genistein 0 1' OA pr ior to addition of Tg could be han cer clement s such as 12-0-tetradeca noylphorbol-13-acetate 498 Thapsigarg in Induction ofgrp78 Requ ires Protein Kinases OA pRSV- I pRSV-IRE I I Tg C Tg OA Tg + + grp78 ~ _ grp78 ~ -~ --~ 3AIO ~ grp94 ~ DKI Z 3AI O ~ 2 5 ~N I H 3 T3 2 3 4 2 0 DpRSV- l ~ p R S V -I R E c 14 o .~ .§ 1 5 .S a; '0 :§ .!!! 10 Q) .2: 10 a: iii E I Qi a: <I> .2: 6 I lii a; a: c OA G Tg OA G - + - + - + -+ I 1....._ _ ---' + + Tg Tg grp78 grp94 FIG. 4. Effect of ge n istein a n d okadaic acid on the thapsigu r ­ FIG. 3. Rel a ti ve mRNA level s of g rp 78 and grp94 in the sta b ly gin in du ced accu mu latio n of grp78 mRNA. A, total cytoplas mic tr unsfected cells a fter th a psiga r gin treatment. A , tota l RNA were RNA was isolat ed from untreat ed K12 cells (C); cells treat ed with 300 isolat ed from pRSV-l a nd pRSV-IR E stably tran sfected cells selecte d by nxt thapsigargin (Tg ) and cells pretreat ed with geniste in th en Tg added J.l ~ 1 ~I TX . At 16 h prior to RNA extra ction, cells were untreated (- ) or (G + Tg ), with 100 nxt of okada ic acid (OA ), or pret reat ed with OA then treat ed ( +) with 300 nxtof tha psiga rgin. The RNA blots were hybrid ized Tg ad ded (OA + Tg ). RNA blot ana lysis for grp78 and 3AIO mRNA with grp78, grp94 , a nd 3A 1O cDNA probes. The autoradi ogram is levels were performed. The au toradiograms ar c shown. B, the levels of shown. B, levels ofgrp 78 and grp94 mRNA from stably tra nsfected cells grp 78 mRNA from K12 and NIH3T3 cells aft er treatment wit h okadaic were qu antitat ed by densitomet ry and norm alized aga inst the invar i­ acid (OA) , genistein (G) , or thap sigar gin (Tg) or in combina tions (OA ant 3A /O mRNA level. The mRNA levels fill ' grp 78 and grp.94 from + Tg or G + 1'g) were determ ined by den sitomet ry and normalized pRSV-l st ably transfected cells wit hout treatment of tha psiga rgin were against mRNA levels of 3A 10. The levels of grp78 mRNA in untreated set as 1. The fold ind ucti on by Tg is indi cated. RNA levels from cells NIH3T 3 or K12 cells (C) were set as 1. The fold inductio n for each of the treated with (+) or wit hout (-) th apsigargin are plotted. treatment conditions is shown. response element, cAMP respon se element, a nd additional CCAAT elements ; however, th ey do not contribute to grp 78 To demonstrate that th e inhibitive effect of genistein on stress ind ucibili ty (Alexa ndre et al., 199 1). The results of all grp78 transcription is not due to a genera l toxic effect of tyro­ tra nsient tran sfecti ons are summa rized in Table I and some sine kin ase inhibitors and to delim it th e tyrosin e kina se activ­ representative CAT assays are shown in Fig. 7. ity targeted by genistein , we tested th e effect of L-A, a potent We next tested (- 375/-881MCAT, which contai ns 287 bp of inhibitor for EGF receptor-associated tyrosine kinase activities the grp78 promoter fused to a minimal MMTV promoter. The (Onoda et al., 19891on (- 456)CAT a nd (- 3551CAT. Two con­ results showed th at the 4-fold induction of CAT activity by Tg centrations (50 ml and 1 JL~1 1 of L-A were used. The results, as was completely elimina ted when the cells were treated with shown in Table I, demon strated th at L-A, either added alone or genistein in combinat ion with Tg, whereas genistein alone in combination with Tg, did not ha ve any effect. showed no effect (Table I). Thus, genistein response elements are Th e Potent iating Effect of Ohada ic Acid Requires a 50-bp locat ed within thi s 287-bp region. One of the target domains Region in the Vicinity of the grp78 TATA Sequence- In exam­ spans - 159 to - 110, since dupli cat e copies of thi s subfra gment ining th e response of (- 375/- 881MCAT, (- 159/- 1101MCAT when linked to th e minimal MMTV promoter (- 159/- 1101CAT and (- 109/- 741MCAT to OA treatment in combination with was able to confer a 8-fold induction by Tg, However, this induc­ Tg, we discovered that instead of a stimulative effect as ob­ tion was reduced to 1.6-fold when geniste in was added (Table I). served for (- 4561CAT, OA actually reduce d slightly the Tg This region contains the 3'-half of core and C3 element. Another induction. Thu s, the Tg inductio n for the three plasmids listed target region spans - 109 to - 74 and it contains the C1 element. decreased from 4-, 8- and 13-fold to about 2-, 6-, and 8-fold, Duplicate copies of th is subfragment linked to the minimal respectively (Fig. 7 and Tab le I). These results indicate that in l\IMTV promoter (- 109/- 74)CAT were able to confer a 13-fold NIH3T3 cells, the enhancing effect of OA on grp78 tran scrip­ induction by Tg. Addition of genistein and Tg completely elimi­ tion is not mediated by the kn own Tg response elements of the nated the induction (Fig. 7 and Table n grp78 promoter. 499 Thapsigargin Induction ofgrp 78 Requires Protein Kinases strea m region surrounding the grp78 TATA sequence. Previ­ (-4 56)CAT ously, we demonstrated that this same region, when fused to a CAT reporter gene, is almost devoid of any basal promoter .. 3AC H activity and is not inducible by Ca depletion stress (Resendez et al., 1988). DISCUSSION With the recent discovery that in the S. cereoisiae a serine/ threonine kin ase (IRE 1p) plays a critica l role in activating the ER protein gene tra nscript ion in response to malfolded protein accumulation stress, th e question arises as to whether protein + + T9 T9 phosph orylation is also util ized as a mecha nism for th e induc­ tion of the mammalian ER protein genes. In thi s report we focused on grp78 gene expression which is induced to high levels by th ap siga rgin following chronic depleti on of th e ER Ca + store and provide severa l lines of evidence that protein phosph orylation and deph osph orylation are involved in grp tran sactivation. Our first approach is to engineer th e coding sequence of z- IRE1p into a mammalian expression vecto r a nd create a stable : ~ CHO cell line with a mplified copies . Th is was achieved, and a <3 cell line harboring about 50 copies of the yeas t gene was estab­ '" f- <: lished. In these cells, sufficient a mounts of IRE1p were ex­ l1> pressed to allow iden tification with specific anti body aga inst ."' iii th e protein. Nonet heless, under these favo rab le conditio ns, th e Q; cr: enhancement for grp 78 mRNA level in stressed a nd non­ stressed cells was about 2-fold. Could this slight increase in grp78 mRNA levels be due to th e se rine/threonine kin ase ac­ tivity of the tra nsfected gene? Th e fact that the 2-fold induction we obser ved is similar to that report ed for S. cereoisiae when the sa me protein is overproduced over th e endogenous level (Mori et al., 1993) suggests th at this might be the physiological C OA G T9 OA G limit for thi s protein in a norm al cell background. Furthermore, + + the co-induction of other ER protein genes such as grp94 in th e T9 T9 IRE1p expressing cells is consistent wit h th e action of IRE 1p in FIG. 5. Effect of genistein and okadaic acid on the Tg induc­ yeast cells. On th e other hand, expression of a heterologous tion of grp78 promoter activity. A, NIH3T3 cells were transient ly protein has its inherent problems which are difficult to resolve transfected with (- 456)CAT and subjected to trea tment s with OA, in a mammali an system. For example, the modest enhance­ genistein, and Tg alone or in combinations (OA + Tg or G + Tg ). The autora diogra m of the CAT assays is shown. The positions of the chlor­ ment of the grp mRNA levels could have been caused by im­ amphenicol (CM) and its acetylated form s (3Ac and lAc) are indi cated. proper insertion of IRE1p into the design at ed cellul ar compart­ B, the CAT conversion in each sa mple was qua nt itated by an Ambis ment, th us diminishing the capacity of IRE1p to fully induce Radi oanalytic Imagin g Syst em. The level of th e prom oter activity in the grps. Anot her complication is that the portion of IRE1p untreated cells (C) was set as 1. The fold ind uct ion for the various treatments are shown. located in the endoplasmic reticulum might not fold properly. Thi s effect could have elicited a mild grp stress response. Al­ Since OA exerts a potentiating effect on (- 456)CAT bu t not though amino acids essential for th e kinase activity have been on (- 375/-88)MCAT (Fig. 7), th e target for th e stimulation identified within IRE1p (Mori et al. , 1993), to crea te an equiv­ eithe r lies in th e upstream region spanning - 456 and - 375 or alent CHO cell line with the amplified mutant genes at th e in th e downstream region spanning - 88 to - 38 (Fig. 3). Th e sa me level of express ion as the wild type gene to differentiat e upstream region contains basal enha ncer elements and th e between a 2-fold effect is tech nically difficult, if not impossible. downstream region contains th e sequence surrounding th e Th e functional significance of IRE1p in anima l cells is best grp78 TATA element at - 55 bu t excludes transcription initia­ evaluate d when t he fold induction of the tra nsfeeted wild-type tion located at ( + 1) (Li et al., 1993). To differentiate betw een pr otein can be man ipul at ed so that it is sufficiently high to these two possibilities, we first compared th e response between allow mu tational an alysis. Thi s may be possibl e using recipient (- 457)CAT and (- 355)CAT. Our results show th at th ese two cells wit h the homologous mammali an gene deleted. However, plasm ids responded nearly identically to th e treatment of Tg this requires the confirmation th at such a gene exists in mam­ and OA (Table I). Thus, the upstream sequence between - 456 malian cells and its subsequent isolation. Based on the current a nd - 355 is not needed for the potentiating effect as both limitations, we focused our effort int o examining th e contribu­ plasmids showed th at OA enha nced th e Tg induction from tion of endogenous protein kinases and phosph atases in the Tg about 6-12-fold. Next the response between (- 355)CAT and inductio n of grp78. (- 375/-88)MCAT was compared. Whil e both plasmids showed Poten t protein kinase inhibi tors have been used widely as a 4-5-fold induction by Tg, OA in combina tion with Tg stimu­ biochemical tools to study the functi on of kin ase a nd phosph a­ late d the (- 355)CAT activity to ll-fold and redu ced (-375/ tase activities in regul ating gene express ion in mammalian - 88)MCAT activity to 2-fold (Fig. 7 and Tabl e I). The minimal cells. In th is study we use a pan el of specific inhibi tors to MMTV promoter (MCAT) was relatively un affected by Tg or determine their effect on grp78 induction byTg. The majority of the various inhibitors, either added alone or in combinations intracellu lar phosph ata se activ ity has been attributed to PP-l with Tg (Fig. 7 and Tabl e I). Our results show th at th e OA and PP-2A, a nd OA has th e abi lity to inhibit both enzyme stimulative effect of grp 78 Tg induction requ ires a 50-bp down- activities. Genera lly, PP-2A is inhibited more efficiently th an 500 Thapsigargin Induction ofgrp78 Requires Protein Kinases OA G OA G OA ,+ ~.Jl,.--------,,-----, ~~ -150 -125 -95 -55 • • core ~ ~ ;OW -355 (-355)CAT (-375/-88)MCAT ....... ------. -159 -110 ..._----- ---..-.. (-159/-110)MCAT =========,r--- -- -- e I }~~~? ?~ -109 -74 ............................ TATA (-109/-74)MCAT ====:::1",'" r I tl~ TATA MCAT e1111??2@ FIG. 6. Schematic drawings of the grp78/CAT and grp78/MCAT fusion genes used in the transient transfection assays. The 5' and 3' end points of the grp78 promoter fragments used are indicated. The positions of the core, C3, C1, and TATA elements are indicated. The small triangles indicate upstream enhancer elements. The hatched lines in the MCAT constructs represent the promoter sequence of the mouse mammary tumor virus (MMTV). The MCAT plasmid contains the promoter fragment (-105 to +104) of the MMTV fused to the CAT reporter gene. The target regions for the inhibitive effect (-) of genistein (0) and okadaic acid (OA) and stimulative effect (+) of OA during Tg induction are bracketed. TABLE I Response ofgrp78 promoter/ CAT fusion plasmids to kinase and phosphatase inhibitors Fold induction" Construct Control OA G L-A Tg OA + Tg G + Tg L-A + Tg b b (-456)CAT 1 1.5 :!:0.1 0.8:!: 0.1 1.2 6.4 :!:2.4 12:!: 2.3 1.3 :!:0.4 6.5 (-355)CAT 1 1.7 :!: 0.2 ND 0.8 5.2:!: 1.3 11 :!: 1.3 ND 5.5" (-375/-88)MCAT 1 1.0 :!:0.3 ND 4.2:!: 1.0 1.0 :!:0.1 1.0 2.3:!: 0.6 ND (-159/-110)MCAT 1 1.1 :!:0.1 0.8 ND 8.2:!: 1.7 5.6 :!:0.4 1.6 :!:0.2 ND (-109/-74)MCAT 1 1.2 :!:0.2 0.9:!: 0.1 ND 13:!: 0.3 7.7:!: 0.5 1.2 :!:0.1 ND MCAT 1 1.5 :!:0.4 0.9 1.0 1.2 1.2 1.2 1.0 a Plasmids (-159/-110)MCAT and (-109/-74) contain two copies of the synthetic promoter subfragments fused to CAT as described previously (Li et 01., 1993). The plasmids were transiently transfected into NIH3T3 cells. The cells were either nontreated (control) or treated with okadaic acid (OA), genistein (G), lavendustin-A (L-A), or thapsigargin (Tg) alone or in combinations as indicated. For each construct, the fold induction over the control level set as 1 is shown. Most transfections were performed independently from two to five times. The standard deviation is indicated. ND, not determined. b 50 nM L-A. c II-'M L-A. PP-1 (Cohn, 1989). In our experiments, we discovered that OA genes are both stress-inducible genes and share some common only exhibits a mild potentiating effect on grp78 transcription. induction conditions (Lee, 1987; Watowich and Morimoto, Since the effect is more pronounced at 100 TIM of OA than 50 1988), their response to OA during stress is not identical. nM , PP-1, rather than PP-2A, may be the affected phospha­ In examining thegrp78 promoter sequence which is required tase. OA has been reported to stimulate a large number of gene for the OA stimulative effect on Tg induction in NIH3T3 cells, systems, including that of the heat shock gene (Schonthal et al., we discovered that it does not act through the Tg response 1991; Guy et al., 1992; Chang et al., 1993). Since OA strongly elements. Rather, it exerts a mild inhibitive effect on those increases the heat shock hsp70 promoter activity, it was pro­ elements. A 50-bp region in the vicinity of the grp78 TATA posed that the heat-induced transcriptional activation of the sequence is required for the OA stimulative effect of Tg induc­ heat shock gene is associated with the phosphorylation of com­ tion. Interestingly, no sequence resemblance was found with ponentts) of the transcriptional complex and that OA enhances known target sites of OA action (Wera et al., 1993; Kharbanda this phosphorylation (Chang et al., 1993). Based on the optimal et al., 1993; Chauhan et al., 1994; Hyun et al., 1994), and the concentrations of OA, PP-2A, as opposed to PP-1, was impli­ majority of the previously described OA response elements are cated in this induction process. Thus, although grp and hsp upstream regulatory sequences. Thus, the grp78 target may represent a novel site. However, OA does not have to act di­ 2 X. Cao, Y. Zhou, and A. S. Lee, unpublished results. rectly on the transcription factors which bind to the 50 bp 501 Thapsiga rgin. Induction of grp 78 Requires Protein Kinases (-3SS)CAT (- 109/ -74) MCAT ~ 3 A c FI G. 7. Differential effect of genis­ ~ l A c tein a nd okadaic acid on the Tg in­ d uction of grp78/CAT and g rp 78/ MCAT fusion genes. Left pan els, comparison betw een (- 355)CAT and ~ C M (- 375/- 88)i\ICAT. Right pan el, compa ri­ son between (- 109/- MCAT and th e min ­ (-3 7S/ -88) MCAT MCAT imal i\IMTV promoter driven CAT (MCAT). NIH3T3 cells were transiently ~ 3 A c tra ns fecte d with t hese plasm ids an d sub­ jected to the various treatment conditions ~ l A c (C, Tg, OA, G, OA + 7'g, G + Tg ). The autora diogra rns of the CAT as says are shown. The position of chloram pheni col (CM ) and its ace tylated form s (3Ac and ~ C M lAc) are shown. '----' '----' '----' '----' L.--...J l---...J L.--...J l---...J L..------J l---...J C Tg OA OA C Tg G G OA OA + + Tg Tg Tg region. For exa mple, it could inhibit a serine/threonine pr otein treatment of the cells with lavendustin-A, a specific inhibitor for EGF receptor-associated tyrosine kin ase, neither stimu­ phosphatase activity which modifies t he activ ity of any inter­ mediatory regul atory components. Nonetheless , our finding lates nor inhibits the Tg res ponse, sugges ting th at the tyrosin e that OA only exerts a modest effect on t he Tg induction ofgrp78 kinase acti vity mediating the grp response to Tg is likely to be a nd that the response does not involve the Tg response ele­ distinct from that of the EGF receptor type. Based on our findi ngs, we hypothesize th at a tyr osine kin ase activity is es­ ments prompted us to investiga te wheth er kinases other tha n that of th e se rine/threonine typ e are more critical elements for sential for grp indu ction by Tg. It is a lso possibl e that genistein inhibitsgrp induction through a yet unknown mechan ism. The the Tg induction process in mammalian cells. Our experiments indicate that gen istein, a tyrosine kin ase identification of th e genistein se nsitive component s, their sub­ cellular locali zati ons and regulatory mechan ism wiII cont ribute inhibitor, almost totally elimina te d the Tg induction response. Furthe rmore , we have found that the grp core, the C3 and Cl, significantly to th e und erstanding of how ER protein genes which are major targets for Tg induction, are also major target s respond to Ca + depletion stress in ma mma lia n cells. of the inhibitive effect exerted by geniste in. Thus, genistein Aclwowledgmellts- We acknowledge Cory Gorman for the gift of the acts eithe r directly on th e tran script ion factors which inte ract pRSV-l pla smi d, Larry Chas in for the DG44 cells, Ca roline Sha mu and with th ese promoter regul atory elements, or it modifies other Peter Walter for the pCSIIOplasmid and the IRElp a ntise ra. We th ank targets in th e intermedi ary path ways leading to th e activation Axel Schontha l, Wilfred Li, and Meera Ramakri shnan for cri tica l review of the manu script. of th e nu clear tran scription factors. Recent exa mples suggest that genistein ca n exert its effect both ways. In the T cell HEFE ltE NCES antigen receptor signaling system, genistein was used to iden ­ Akiyam a, T.. Ishida, .J., Nak agawa , S., Ognwa rn , II., Wata nahe, S.. ltoh, N., tify a tyrosi ne kinase activity which down-regu lates specifically Shi buya , ~ I. , a nd Fuka mi, Y. (1987 ) J. Hiol. CI" ,m. 2(;2, 5592- 55B5 Alexandre, S., Nakaki, T., Van hnm me, L., a nd Lee, A. S. (l BBI) Mol. End ocrinol. calcium-mediated signaling at a point downstream of the 5, 18(;2- 1872 lc k tyrosine kin ase p56 but up stream of calcium mobili zation Bal da ri, C. T., and Telford, ,J. L. (19 94) Eur.•J. lm m 11I10/. 24, 104(;-1052 Cha ng, S. C., Wooden, S. K., Na kaki, T., Kim, Y. K., Lin, A. Y., Kung, L.. Attenello, (Balda ri and Telford , 1994). Genistein, on th e other hand, .J. W., an d Lee, A. S. (1987) Proc. Not! . Acad. Sci. U. S . A. 84, (;80- (;84 blocks the transcript ional stimulation of interferon-induced Cha ng, N.-T., Hua ng, L. E., a nd Liu, A. Y.-C. (1993 ).1. /lio/. Chem, 2(;8, 14:l6 - 1439 genes by inhibitin g tyrosine phosph orylation of the tra nscrip ­ Cha uha n, D., Kharhanda , S. M.. Uchiya ma , H.. Sukhat me, V. P., Kufe, D. W., and Ander son, K. C. (1994) Can cer Res. 54 , 2234 - 2239 tion facto r ISGF 3 which binds directly to the interferon-stirn­ Cohe n, P. (1989) Allllu. Rev. Biochem . 58, 453- 508 ulated response element found in the promoter of inducible Cox, ./. S., Sh am u, C. E., an d Wal ter, P. (1993 ) Cel/ 7:l, 1197-1 206 Dru mmond, I. A. S., Lee, A. S., Resendez, K , ,/ 1'., a nd St einha rdt. H. A. (1987 ) genes (Gutch et al., 1992). In the grp 78 promoter, th e core, C3, J. Bioi. Chem. 2(;2, 12801-1 2805 a nd Cl act in concert to enhance grp expression (Wooden et al., Gutch , M. •L, Da ly, C.. a nd Reich, N. C. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 1991). Thus, it is possible that upon stress induction, as in the 114 11- 114 15 2 Guy, G. It, Cao, X., Chua , S. P., a nd Tan , Y. H. (W92) .J. Hiol. Chem. 2(;7, case of ER Ca + depleti on, a common tyrosine kinase activ ity is 1846 -1 852 activated which leads to the modification of the binding factors Hall, C. V., J acob, P. E., Ringold, G. ~ I. , a nd Lee, F. ( W831J. .1101. 111'1'1. Genet, 2, 101- 109 or th eir co-act iva tors, resul ting in grp78 induction. Hyun, S. W., Par k, K., Lee, Y. S., Lee, Y. I, Kim , S. •/. (1994) .J. /liol. Chem. 2(;9, Among kin ase inhibitors, gen iste in is a specific inhibitor for 36·1-368 tyrosine kin ases but not se rine and threonine kin ases and J ackson, T., Pat terson, S. I., Thastrup, 0 ., a nd Ha nley, 1\1. It ( 1988 ) Biochcm . •1. 253 , 8 1- 8(; other ATP ana logue-related enzymes (Akiyama et al., 1987). Kha rbandn, S., Rubin, E., Datt a. H., Bass , R., Sukh a tme, V., and Kufe, D. (1993) Th us, it has minimal effect on the enzyme activit ies of cAMP­ Cell Grouith & Differ. 4, 17- 2:1 Kim , Y. K., a nd Lee, A. S. (1989) Gene (Am st,) 77 , 123- 13 1 dependent protein kinases, protein kinase C, phosph orylase Kozutsurni, Y., Sega l, I\I., Nor mington, K., Gethin g, 1\1. J. , and Sarnbrook. vl. (1988) kinase , 5'-nucleotidase and phosph odiestera se. While it is Nat ure :1:12, 462-464 Lee, A. S. (198 1) .J. Cell. Physiol. ros, 119 -1 25 highly potent against tyrosine kinases such as the EGF recep­ v src Lee, A. S. ( HJ87) Trend Biochem. S ci. 12, 20 - 23 tor, pp60 - and ppllOl-:al-:-fc", it is also capable of inhibiting Lee, A. S. (1992) Curl'. Opin. c-utu«. 4, 267-27:l tyrosine kin ases other than these three well cha racterized Lee, A. S., Delegean e, A. 1\1., Bak el', V.. and Chow. P. C. (198:1)d. /Iiol. Chem, 258, 59 7- 603 ones. Since th e ta rgets for geniste in can be broad , we initiated Li, X. A., a nd Lee, A. S. nssn Mol. Cell. Bioi. 11, 3446 - :l45:1 experiments to determine which type of tyrosine kinase activ­ Li, X.. Warren, D. W., Gr egoire, .I., Peder son, H. C.. and Lee, A. S. (1989) .1/01. ity is involved in th e Tg indu ction of grp78. We found tha t Endocrinol. :I, 1944-W52 502 Thapsigargin Induction ofgrp78 Requires Protein Kinases Li, L.-J., Li, x.,Ferrario, A., Rucker, N., Liu, E. S., Wong, S., Gomer, C., and Lee, Physiol.152,545-552 A. S. (1992) J. Cell. Physiol. 153, 575-582 Resendez, E., Jr., Ting, J., Kim, K S., Wooden, S. K, and Lee, A. S. (1986) J. Cell Li, W. W., Alexandre, S., Cao, C., and Lee, A. S. (1993) J. BioI. Chem. 268, Bioi. 103, 2145-2152 12003-12009 Resendez, E. R., Jr., Wooden, S. K, and Lee, A. S. (1988) Mol. Cell. Bioi. 8, Li, W. W., Sistonen, L., Morimoto, R. 1., and Lee, A. S. (1994) Mol. Cell. BioI. 14, 4579-4584 5533-5546 Schonthal, A. (1992) New Bioi. 4, 16-21 Lin, A. Y., and Lee, A. S. (1984) Proc. Nat!. Acad. Sci. U. S. A. 81,988-992 Schonthal, A., Tsuk.itani, Y., and Feramisco, J. R. (1991) Oncogene 6, 423-430 Little, E., Ramakrishnan, M., Roy, B., Gazit, G., and Lee, A. S. (1994) Crit. Rev. Thastrup, 0., Cullen, P. J., Drebak, B. K, Hanley, M. R., and Dawson, A. P. (1990) Eukaryotic Gene Expression 4, 1-18 Proc. Nat!. Acad. Sci. U. S. A. 87, 2466-2470 Liu, E. S., and Lee, A. S. (1991) Nucleic Acids Res. 19, 5425-5431 Urlaub, G., MitchelJ, P. J., Kas, E., Chasin, L. A., Funanage, V. L, Myoda, T. T., Liu, E. S., Ou, J., and Lee, A. S. (1992) J. Bioi. Chem. 267,7128-7133 and Hamlin, J. (1986) Somatic Cell. Mol. Genet. 12,555-566 Mori, K, Ma, W., Gething, M. J., and Sambrook, J. (1993) Cell 74, 743-756 Watowich, S. S., and Morimoto, R. I. (1988) Mol. Cell. BioI. 8,393-405 Nouspikel, T., and Iynedjian, P. B. (1992) Eur. J. Biochem. 210,365-373 Wera, S., Belayew, A., and Martial, J. A. (1993) Mol. Endocrinol. 7,965-971 Onoda, T., Iinuma, H., Sasak.i, Y., Hamada, M., Isshiki, K, Naganawa, H., and Wooden, S. K, Li, L.-J., Navarro, D., Qadri, 1., Pereira, L., and Lee, A. S. (1991) Takeuchi, T. (1989) J. Nat. Prod. 52, 1252-1257 Mol. Cell. Bioi. 11, 5612-5623 Pelham, H. R. B. (1986) Cell 46, 959-961 Price, B. D., Mannheim-Rodman, L. A., and Calderwood, S. K (1992) J. Cell. Zhong, L., Batt, D. B., and Carmichael, G. G. (1994) BioTechniques 16, 56-57 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Biological Chemistry Unpaywall

Requirement of Tyrosine- and Serine/Threonine Kinases in the Transcriptional Activation of the Mammalian grp78/BiP Promoter by Thapsigargin

Cao, Xianjun; Zhou, Yanhong; Lee, Amy S.
Journal of Biological Chemistry – Jan 1, 1995
Download PDF
9 pages

Loading next page...
 
/lp/unpaywall/requirement-of-tyrosine-and-serine-threonine-kinases-in-the-4R6Qicoh0N

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
0021-9258
DOI
10.1074/jbc.270.1.494
Publisher site
See Article on Publisher Site

Abstract

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 270, No.1, Issue of January 6, pp. 494-502, 1995 © 1995 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. Requirement of Tyrosine- and SerinelThreonine Kinases in the Transcriptional Activation of the Mammalian grp78/BiP Promoter by Thapsigargin* (Received for publication, August 8, 1994, and in revised form, October 21, 1994) Xianjun Cao, Yanhong Zhou, and Amy S. Leez From the Department ofBiochemistry and Molecular Biology and the Norris Cancer Center, University of Southern California School of Medicine, Los Angeles, California 90033-0800 Depletion of endoplasmic reticulum (ER) Ca + store by conditions, particularly those which perturb ER function (Lee, thapsigargin (Tg) in mammalian cells induces a set of ER 1987; Watowich et al., 1988; Kozutsumi et al., 1988; Lee, 1992; protein genes known as the glucose-regulated proteins. Liu et al., 1992; Price et al., 1992). In mammalian cells, one Recently, IRElp, a transmembrane protein postulated to of the most potent inducers for grp78 is chronic depletion of have a serine/threonine kinase activity, has been identi­ ER calcium stores by Tg or the calcium ionophore A23187 fied as required for the induction ofER resident proteins (Resendez et al., 1986; Drummond et al., 1987; Li et al., 1993). genes in Saccharomyces cerevisiae. To investigate The induction process is independent of a rise in cytoplasmic whether IRElp can stimulate mammalian grp transcrip­ 2+; Ca furthermore, it requires a prolonged period of (2 h) of tion, a stable Chinese hamster ovary cell line containing treatment and is sensitive to cycloheximide. Thus, the signal­ amplified copies of IRElp has been created. The IRElp ing mechanism for the activation of grp78 transcription in­ expressing transfectants exhibited a modest (2-fold) en­ 2+ volves more components than merely the efflux of Ca from hancement ofboth the basal and Tg induced level ofgrp78 the ER. and grp94, two coordinately regulated grp genes. Using In examining the regulatory elements of the rat grp78 pro­ okadaic acid as a specific inhibitor for the endogenous moter which mediate the Tg and A23187 induction response, serine/threonine protein phosphatase activities, a mild (2·fold) stimulative effect was observed for Tg induction two functional domains have been identified (Wooden et al., of grp78 transcription. The okadaic acid potentiating ef­ 1991; Li et al., 1993). One region spans the promoter sequence fect requires a 50-base pair region in the vicinity of the from -159 to -110 and contains the grp core element which is grp78 TATA element. In contrast, the transcriptional ac­ conserved from yeast to human (Resendez et al., 1988) and a C3 tivation of grp78 by Tg is almost totally eliminated by element (Li et al., 1993). The other region spans -109 to -74 genistein, a tyrosine kinase inhibitor. The grp core, the C3 and contains a CCAAT motif (Cl) most proximal to the TATA and Cl elements which are major Tg response elements of element (Wooden et al., 1991). Both regions are able to confer the rat grp78 promoter, are also major targets of the in­ Tg and A23187 inducibility to a heterologous promoter upon hibitive effects of genistein. transient transfection into mammalian cells (Li et al., 1993). Using ligation-mediated PCR, we found that specific changes in factor occupancy occurred after stress induction and that the Thapsigargin (Tg)1 is a tumor promoter whose intracellular major changes occur within the grp core element (Li et al., target has been identified as the endoplasmic reticulum (ER) 1994). The factor which exhibits specific binding to this region 2+-ATPase Ca (Jackson et al., 1988; Thastrup et al., 1990). appears to be constitutive, suggesting that during stress, the Through its inhibition of the enzyme, Tg specifically discharges factor either undergoes a conformational change or has disso­ the ER calcium store. Within the ER resides a set of proteins ciated from inhibiting elements, resulting in the changes in the known as the glucose-regulated proteins (GRPs) (Lee, 1992). in vivo footprinting pattern. GRP78, a 78-kDa protein also referred to as the immunoglob­ 2+ The discovery that chronic depletion of Ca from the ER ulin heavy chain protein (BiP), is one of the best characterized store could lead to grp78 transactivation correlated with factor GRPs with protein chaperoning and calcium binding properties interaction changes at the grp78 promoter, raises the impor­ (Pelham, 1986; Little et al., 1994). Although GRP78 is consti­ tant question of intracellular compartment signaling. What are tutively expressed at a low basal level in many different tis­ the mediators of this response? The transcription of grp78 in sues, its transcription is highly inducible by a variety of stress fibroblast cells is partially inhibited by W-7 which inhibits 2+ Ca /calmodulin-regulated enzyme activities (Resendez et al., * The work was supported by United States Public Health Service 1986). Other studies suggest an involvement of tyrosine and Grant R37 CA27607 from the National Cancer Institute (to A. S. L.). serine/threonine kinases and phosphatases. In the mammalian The costs of publication of this article were defrayed in part by the system, pretreatment ofNIH3T3 cells with genistein, an isofla­ payment of page charges. This article must therefore be hereby marked vone which specifically inhibits a number of tyrosine kinases, "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. reduces Tg induction of grp78 mRNA levels, whereas okadaic :j:To whom correspondence should be addressed: Tel.: 213-764-0507; acid (OA), a specific inhibitor of serine/threonine protein phos­ Fax: 213-764-0094. phatases type 1 (Plt-L) and type 2A (PP-2A) (Schonthal, 1992), 1 The abbreviations used are: Tg, thapsigargin; ER, endoplasmic re­ enhances Tg induction ofgrp78 (Price et al., 1992). In the yeast ticulum; GRP, glucose-regulated proteins; PCR, polymerase chain re­ action; OA,okadaic acid; PP-l and -2, protein phosphatases type 1 and system, a IRE1 /Ern1 gene encoding for a transmembrane ser­ type 2; EGF, epidermal growth factor; EGFR, epidermal growth factor ine/threonine kinase has recently been identified as required receptor: L-A,lavendustin-A; DHFR, dihydrofolate reductase; kb, kilo­ for the transcriptional induction of KAR2, the yeast homologue base pairts); CAT, chloramphenicol acetyltransferase; bp, base pairts); of grp78, in response to the accumulation of malfolded protein MMTV, murine mammary tumor virus; PAGE, polyacrylamide gel elec­ trophoresis; MTX, methotrexate; RSV, Rous sarcoma virus. in the ER (Cox et al., 1993; Mori et al., 1993). It has been This is an Open Access article under the CC BY license. Thapsigargin Induction ofgrp78 Requires Protein Kinases 495 Northern Blotting-After treatment with protein kinase inhibitors postulated that IRE1p is the proximal sensor of events in the and thapsigargin for 7 h, the cells were washed twice in phosphate­ ER and that binding of ligands causes transduction of informa­ buffer saline. Total cellular RNA was extracted following isohigh tion across the ER membrane, leading to activation of a specific method as described previously (Zhong et al., 1994). 10 JLg of RNA was set of transcription factors. IREp1 is of very low abundance in loaded to 1% agarose, 2.2 M formaldehyde gels. The probe, a grp78 yeast cells. Its ligand and the involvement of IRE1plErn1p in cDNA fragment (1.5 kb) from p3C5 digested with EcoRI and PstI, was Ca + signaling of grp78 have not yet been determined. labeled with the hexamer method to specific activities of 2 X 10 Since IRE1p is the first and the only serine/threonine kinase cpm/ug, p3A10, the entire plasmid, was labeled with the same method and used to assess the amount of loaded RNA (Lin and Lee, 1984). identified so far which can affect KAR2/grp78 transcription, we Conditions ofthe RNA blot and hybridization were described previously created a mammalian expression vector for the yeast IRE 1p (Lee et al., 1983). Autoradiograms were quantitated by the densitome­ and selected for stable transfectants containing amplified cop­ try to obtain relative levels of the grp78 and grp94 mRNA. ies of the yeast gene. We report here that expression ofIRE1p Plasmids-Plasmids (-456)CAT and (-355)CAT, previously re­ in hamster ovary cells is correlated with a 2-fold increase in the ferred to as pE43 and p3K, respectively (Changet al., 1987), contain the basal and Tg-induced levels of the endogenousgrp78 mRNA. In rat grp78 promoter at deletion end points of -456 and -355. The mammalian cells, grp78 is coordinately regulated at the tran­ construction of the plasmids (-375/-88)MCAT, (-159/-MCAT, (-109/ scriptional level with grp94, which encodes a 94-kDa GRP, -74)MCAT has been described (Li et al., 1993). Plasmid MCAT con­ tains the minimal MMTV promoter fused to the chloramphenicol acetyl­ through common cis-regulatory elements and trans-acting fac­ transferase (CAT) gene (Kim and Lee, 1989). pRSV-IRE was tors (Chang et aZ., 1987; Liu and Lee, 1991). As in the yeast constructed by inserting the full-length of IREI gene (3.3 kb) into Sail system where IREl also induces other ER resident protein and BamHI sites of pRSV-1 (gift of Dr. C. Gorman)(Li et al., 1992). The genes (Cox et aZ., 1993), the mammalian IRE1p transfectants full-length of IREI gene was generated by the PCR using the plasmid also exhibit 2-fold increase in the basal and Tg-induced levels of pCS110 (Cox et al., 1993; gift of Dr. P. Walter, University of California, grp94 mRNA. San Francisco) as template. pBluescripts KS (pBS) was purchased from Because of the modest effect observed with overexpression of the Stratagene. pBS-IRE contains the IREI gene derived from pRSV­ IRE digested with Saii and BamHI. the yeast IRE1p protein in mammalian cells and the inherent In Vitro Transcription and Translation-The transcription of IREI difficulties associated with expression of a heterologous pro­ gene from pBS-IRE was carried out in 20 JLl of reaction mixture follow­ tein, we investigated the respective contributions of serine/ ing the T3 transcription kit protocol (Amersham Corp.). The RNA was threonine protein phosphatases and tyrosine kinases toward precipitated, and further translated in the 25 JLl of reaction mixture to the Tg induction of grp78 in two different mammalian cell 35S1 synthesize [ methionine-Iabeled IRE1 protein (IRE1p) using a rabbit lines. Our results indicate that genistein eliminates the re­ reticulocyte translation system (Promega). 5 JLl of translation reaction sponse almost completely. We further demonstrate that the mixture was separated on 8% SDS-PAGE. inhibitive effect of genistein is at the level of transcription. We Transient Transfection and CAT Assay-A total of 2 X 10 NIH3T3 cells/100-mm dish was plated 20 h before transfection. 10 JLg of test found that the grp core, the C3 and C1 elements which are the plasmid was co-transfected with 5 fJ-g of PCHllO, an expression vector major Tg response elements ofthe ratgrp78 promoter (Li et aZ., for J3-galactosidase (Hall et al., 1983), with the calcium phosphate 1993), are also major targets of the inhibitive effects of genis­ precipitation method in the presence of 3 JLg of carrier HeLa genomic tein. The tyrosine kinase activity essential for the Tg induction DNA. After 24 h the cells were treated with either thapsigargin (300 ofgrp78 is likely to be distinct from EGFR, since lavendustin-A nM), genistein (140 fJ-M), okadaic acid (100 nxr), or lavendustin-A (50 nM (L-A), a potent inhibitor for EGF-associated tyrosine kinase or 1 fJ-M) alone or in combination of two reagents for 16 h. For the cells activity (Onoda et al., 1989), is without effect. that were treated with two reagents, okadaic acid, genistein, or laven­ dustin-A was added 15 min prior to the addition of thapsigargin. The Using OA as a specific inhibitor for serine/threonine protein cells were further incubated in the presence of both reagents for 16 h phosphatase activities, a mild (about 2-fold) stimulative effect before protein extraction. The protein concentration of the cell extract was observed for Tg induction of grp78 mRNA levels. Nonethe­ was determined by Bio-Rad protein assay. The CAT activities were less, the OA stimulative effect is not mediated through the Tg measured at the linear range after normalization to the J3-galactosidase response elements; rather, it requires sequences in the vicinity activities. The developed TLC plates were quantified on an Ambis of the grp78 TATA element unrelated to the known OA re­ Radioanalytic Imaging System (Arnbis System, San Diego, CAl and sponse elements. Thus, in the mammalian system, although a exposed to films for autoradiography. Most plasmids were transfected serine/threonine kinase activity may potentiate slightly the two to five times independently. Generation of Stable Transfectants-lO fJ-g of pRSV-1 or pRSV-IRE grp78 Tg response, a genistein-sensitive activity acting was transfected into DG44 cells by using the calcium phosphate pre­ through the Tg response elements is the critical component of cipitation method. Twenty-four h after transfection, the cells were the signal transduction system that leads to transcription ac­ changed into the a-minus medium and grown for 14 days. The surviving tivation of grp78. colonies were exposed to 0.005,0.01,0.02,0.05,0.1,0.2,0.5, 1,2, and 5 MATERIALS AND METHODS JLM MTX for the further selection as described previously (Li and Lee, 1991). The resistant colonies were pooled together and maintained in Cell Cultures and Inhibitor Treatments-K12 cells, which are ham­ the a-minus medium with the different concentrations of MTX. ster fibroblast temperature-sensitive mutants derived from WglA as Genomic DNA Hybridization-The stable transformants were grown to described previously (Lee, 1981), were maintained in Dulbecco's modi­ 90% confluence and washed twice in phosphate-buffer saline. The fied Eagle's medium containing 10% of cadet calf serum. NIH3T3 cells genomic DNA was extracted by using the previously described method (Li were maintained in Dulbecco's modified Eagle's medium containing et al., 1989). 5 JLg of each genomic DNA and 0.1 JLg of pRSV-IRE were 10% of fetal calf serum. For the inhibitor studies, cells were grown to digested with EcoRI. The probe, consisting of the full-length IRE] gene 90% of confluence in the above media containing 1% antibiotics (50 from Saii and BamHI double digestion of pRSV-IRE, was labeled by the JLg/ml penicillin, 50 JLg/ml streptomycin, and 100 JLg/ml of neomycin) hexamer method with specific activities of 2 x 10 cpm/ILg. Conditions of and then preincubated with the inhibitors (genistein, lavendustin-A, the DNA blot and hybridization were described previously (Li et al., 1989). and okadaic acid) (Life Technologies, Inc.) for 15 min in the fresh Western Blot-The stable transformants in 100-mm dishes were medium. Afterwards, the thapsigargin (Sigma) was added to incubate grown to 90% confluence in the presence of 5.0 ILM MTX and washed for another 7 or 16 h. DG44 cells, a double mutant of Chinese hamster ovary cell lines, lack the hamster dihydrofolate reductase (DHFR) gene twice in phosphate-buffer saline. Cells were resuspended in 100 fJ-l of (gift of Dr. Lawrence Chasin, Columbia University). These cells were lysis buffer (8 M urea, 10% Nonidet P-40 and 5% J3-mercaptoethanol). maintained in the minimal essential a medium containing Lvglutamine, The supernatants were separated on 8% SDS-PAGE. The IREp was ribonucleotides, and deoxyribonucleosides (Life Technologies, Inc.) sup­ detected by using a rabbit polyclonal antisera (gift of C. Shamu and P. plemented with 10% of fetal calf serum (Urlaub et al., 1986). The Walter, University of California, San Francisco) directed against the N DHFR+ transformants can be selected in the same medium without the terminus of IRElp 0:2000 dilution) following Immun-Life Chemilumi­ nucleosides (a-minus). nescent Assay Kits protocol (Bio-Rad), 496 Thapsigargin Induction ofgrp78 Requires Protein Kinases RESULTS Expression of the Yeast IREp in Mammalian Cells-Re­ cently, a 1115-amino acid transmembrane protein encoded by IREl /ERNl in Saccharomyces cerevisiae has been isolated from genetic screens based on its ability to activate KAR2/ grp78 transcription (Cox et al., 1993; Mori et al., 1993). The cytoplasmic C terminus of this protein carries an essential kinase activity related to a cdc2+/CDC28 kinase family and is postulated to be a serine/threonine kinase (Mori et al., 1993). Although it is not known whether a homologue of IREl exists K::J in mammalian cells, with its discovery in the yeast system, it is ! peR possible to test whether IRElp is functional in mammalian cells in activating grp78 transcription. Our approach is to ob­ TAA ATG tain stable mammalian cell lines where the yeast IREl gene is ITJ DJ amplified to produce high levels of IRElp and to test for its ability to enhance grp78 expression over the endogenous level. For this purpose, the yeast IRElp coding sequence was ob­ tained by engineering PCR primers directly flanking its ATG s B translation initiation codon and the termination codon (Fig. 1). After PCR amplification using the yeast genomic clone pCS110 as a template, the 3.3-kb fragment containing the IRElp coding sequence was subcloned into the multiple cloning site located between the Rous sarcoma virus (RSV) long terminal repeat SV40 and an SV40 polyrA) site of the DHFR expression vector pRSV-l (Li et al., 1992). This mammalian expression vector for IRElp is referred to as pRSV-IRE. For purpose of in vitro transcription, the same IRElp coding sequence was subcloned into a T3 vector and is referred to as pBS-IRE. Prior to transfection into mammalian cells, we established sit!- ~ that the IRElp coding sequence obtained from PCR was capa­ ble of translating into a protein close to the 127-kDa size predicted for IRElp. In vitro transcribed IRElp RNA was translated in a rabbit reticulocyte system in the presence of 3 5 [ SJmethionine. As shown in Fig. 2A, only the sample contain­ ing the IRElp mRNA was capable of synthesizing a protein band of about 130 kDa. Thus, the PCR DNA sequence is devoid of cryptic termination codons. Next, pRSV-IRE and as a con­ trol, the vector pRSV-l, were transfected into DG44 cells, which is a CHO cell line devoid ofDHFR activity. Stable trans­ fectants were selected, and the transfected plasmids were am­ FIG. 1. Construction of the expression plasmids. Plasmid plified by subjecting the transfected cells to increasing concen­ pCSllO containing the yeast IREl gene was used to PCR-amplify a 3.3-kb full-length of IREl gene with Sall and BamHI sites at two ends. trations of MTX. Southern blot analysis of genomic DNA The IREl gene was then subcloned into pRSV-l immediately after the prepared from these cells revealed that at MTX concentrations LTR promoter to generate the pRSV-IRE plasmid, in which there is a of 0.2 P.M and higher, there was notable amplification oflRElp selection marker, DHFR gene under the control of SV40 promoter. This coding sequence in the pRSV-IRE transfectants (Fig. 2B). By plasmid was used to generate the stable cell line for the expression of IREl gene. The IREl gene derived from pRSV-IRE was further sub­ comparing the band intensities with the control plasmid ran in cloned into pBS to generate the pBS-IRE plasmid. The pBS-IRE was parallel in the same DNA blot, we estimated the copy number used as a template in in vitro transcription under the direction of a T3 of the IRElp coding sequence in the stable transfectants was promoter. about 50. To confirm that pRSV-IRE stable transfectants were expressing IRElp, total protein extract was prepared from mRNA was measured by Northern blot analysis (Fig. 3A), and stable transfectants selected by 5 P.M MTX, as well as from pRSV-l transfectants subjected to the same MTX selection. To the fold induction was determined after normalization with a detect IRElp, a Western blot was performed using antisera nonvariant control transcript 3AI0 (Lin and Lee, 1984; Li et al., 1993) (Fig. 3B). We observed that in the IRElp-expressing raised against the N terminus oflRElp. In the protein sample from the pRSV-IRE transfectants, an immunoreactive band of cells, the basal and Tg-induced levels of grp78 mRNA were 130 kDa was detected. This band was not observed in the approximately 2-fold higher than the control cells. An interest­ ing characteristic of irel negative mutants in yeast is that not control sample from pRSV-l transfectants (Fig. 2C). These only is KAR2 /grp78 transcription downregulated, transcrip­ results establish that in the pRSV-IRE stable cell line where the IRElp coding sequence was highly amplified, sufficient tion activation of other ER resident protein genes such as amounts of IRElp was produced to allow detection by immu­ protein disulfide isomerase is also negatively affected (Cox et al., 1993). In mammalian cells, there is tight coordinate regu­ noreactivity with antibody specific for IREp. Endogenous grp78 and grp94 mRNA Levels in IRElp-ex­ lation of the grp 78 gene with thegrp94 gene (Lee, 1987). To test pressing Transfectants-To test whether expression of IRElp whether expression of IRElp in mammalian cells would also affectgrp94 expression, the level ofgrp94 mRNA in the pRSV­ has any effect on grp expression, total cytoplasmic RNA was isolated from both the pRSV-IRE and pRSV-l transfectants IRE and pRSV-l stable transfectants was also examined. As in under nontreated and Tg-treated conditions. The level ofgrp78 the case of grp78 mRNA, the level of grp94 mRNA was in- 497 Tha pniga rg in Induction of grp78 Requ ires Protein Kinases A c cr:: ::;: :> c. (]) (/) (/) w c cr:: cr:: ~ g c. c. kDa 21 1- 211- 10 7­ 10 7­ 69 - 69 - 47- 47- 2 3 FIG. 2. IRElp expression ill vit ro and ill vivo. A , 1""S lmethionine-labeled lucifernse tlane 1) and IHEI p tlane 2 ) were synt hes ized in the ill vitro ra bbit reticulocyte system and analyzed by 8';f, SDS·PAGE and autora diogra phy. As a negative control ilane ,'J), a translation reaction mixture was performed without any mRNA. The arrow indicates the location ofl''''S lmethionine-labeled IREl p. 13, pHSV-l and pHSV-IHE were stably trans fected into DG44 cell lines. Genomic DNA were isolated from DG44 ilane 1), pRSV-l stable tran sfectants selected by 5 J.L ~I MTX ilane 2) and pHSV-IHE transfecta nts selected by increasi ng concen trations <0.005-5.0 J.L ~ I) of MTX tlanes .'1- 6). The DNA was digested with EcoHIa nd separa ted on the ugarose gel, blott ed onto the filter , and hybrid ized with the !R EI gene. In lan e 7, the pRSV-mE plasmid was digest ed with EcoHI. C, th e IHElp express ion in pHSV-l (lane 1) and pHSV-mE tlane 2)stably tra nsfected cells selected by 5 J.L ~ I l\ITX was ana lyzed by Western blot usin g antisera aga inst yeas t m E 1p, The arrow indicates the location of lREp. creased by about 2-fold in nonstressed and Tg-stressed cells due to cha nges in the tran scriptiona l rate of the gl1J78 gene or (Fig. 3, A a nd B). Overall, our results indicate that amplifica­ due to changes in the stability of the gl1J78 mRNA . To addres s tion of the yeast IRE l p gene in mamm alian cells and overex­ this issue, we exa mined whether geniste in and OA could exert pression of its protein only showed a modest sti mulative effect the sa me effects on the express ion of a marker gene linked to a grp78 promoter . Previously, we have shown (- 456)CAT, which on the grp78 and grp94 mRNA levels. Se ns itivity of Tg In du ction ofgrp78 to Genistein and Ohadaic contains a 4 18-bp fragment spa nning - 456 to - 38 of the rat Aci d- We next investigated the respective cont ributions of en­ grp78 promoter link ed to a CAT gene, is able to confer a dogen ous se rine/threonine pr otein phosphatases a nd ty rosine 5-6-fold induction of CAT activ ity following Tg induction of K12 transfect ant s harborin g this plasmid (Li et al., 1993). In kinases toward the 'I'g induction of grp78 through the use of th eir specific inh ibitors. Two mammal ian cell lines (NIH3T3 this study, using NIH3T3 cells as recip ien t cells in transient and K12 ha mster fibr oblast cells ) were tested for the effect of tra ns fection assays, a similar result was observed such that the genistein and okadaic acid on the induction of grp78 by Tg. In CAT activity was enhanced ab out 6-fold when the cells were induced by Tg (Fig. 5, A a nd B ). Compa red wit h th e nontrea ted the in hibi tor studies , the cells wer e pre treated with the inhib ­ itors, and subsequently Tg was add ed for 7 h, since the maxi­ cells wit h basal level set as 1, OA an d genistein tr eatment mum level of grp78 mRNA accumu lation occurred afte r 6-h of alone resul ted in 1.4- a nd 0.8-fold, respectively. When OA· addition of 300 nxt ofTg for K12 cells (Li et al., 1993). RNA blot treated cells were ind uced by Tg, the indu ction was increased ana lysis, as shown in Fig. 4A , was performed to quantit ate the to 12-fold. In contrast, geniste in treatment elimina ted almost grp78 mRNA level which was norm a lized agains t 3A10. The completely the Tg response. Since the CAT activities are com­ relative RNA levels for both cell lines are su mma rized in Fig. parab le with the endogenous grp78 mRNA level upon addition 4B . Consiste nt with the previous study (Li et al., 1993), there of OA or G in the Tg induction (compare Fig. 4B and 5B ), we conclude that th e major effect of genistein and OA on grp78 Tg was a 12-15-fold increase in grp78 mRNA cells a fte r Tg treat­ men t. OA or ge niste in alone only showe d min or effect on grp78 ind uction is at the transcriptional level and is mediated mRNA basal level express ion. The most striking resul t is that through the glp 78 promoter eleme nts . when genistein, an inhibitor for ty rosine kin ase acti vity, was The Inh ibitive Effect of Genistein Acts through the grp78 Tg added in combina tion with Tg, there was a dr astic redu ction in Response Elem ents- Since genistei n treatment elimina ted al­ rast, OA, a serine/ most completely the Tg response, we tested whet her the grp78 the grp 78 mRNA indu ction level. In cont threo nine protein phosp hatase inhibitor, in combination with Tg response eleme nts are target s for the genistein mediated Tg only showe d a slight potenti at ing effect in K12 cells, inhibitive effect. A panel of glp 78 promoter fragmen UCAT fu­ whe reas in NIH3T3 cells, a 2-fold increase was det ected. Based sion genes previously demonstrated to be able to respond to or rvations, we conclude that in mammali an cells, an confer Ca depletion stress was ut ilized (Kim and Lee, 1989; on these obse Li et al., 1993). As shown in Fig. 6, (- 456)CAT cont ain s mul ­ endogenous tyrosine kinase activity sensitive to geniste in is a critical compone nt of the transactivation mechan ism for the Tg tiple arrays of regulatory eleme nts upstream of the TATA element located at - 55. The important elements for Tg induc­ induct ion of grp78. Se rine/t hreonine protein phosphorylation ntribute partially to the overall tion previously iden tified by 5' deletion, linker scanning, and sensitive to OA may also co grp78 ind uction level in NIH3T3 cells. internal deleti on mutants incl ude the core at - 150, the C3 The Genistein and OA Effect A re Med iated thro ugh the grp78 element at - 125, and the C1 eleme nt at - 95 (Wooden et al., 1991; Li et al., 1993). There are additiona l upstream en­ Promoter- T he cha nge in grp78 mR NA levels when cells were pre treated wit h genistein 0 1' OA pr ior to addition of Tg could be han cer clement s such as 12-0-tetradeca noylphorbol-13-acetate 498 Thapsigarg in Induction ofgrp78 Requ ires Protein Kinases OA pRSV- I pRSV-IRE I I Tg C Tg OA Tg + + grp78 ~ _ grp78 ~ -~ --~ 3AIO ~ grp94 ~ DKI Z 3AI O ~ 2 5 ~N I H 3 T3 2 3 4 2 0 DpRSV- l ~ p R S V -I R E c 14 o .~ .§ 1 5 .S a; '0 :§ .!!! 10 Q) .2: 10 a: iii E I Qi a: <I> .2: 6 I lii a; a: c OA G Tg OA G - + - + - + -+ I 1....._ _ ---' + + Tg Tg grp78 grp94 FIG. 4. Effect of ge n istein a n d okadaic acid on the thapsigu r ­ FIG. 3. Rel a ti ve mRNA level s of g rp 78 and grp94 in the sta b ly gin in du ced accu mu latio n of grp78 mRNA. A, total cytoplas mic tr unsfected cells a fter th a psiga r gin treatment. A , tota l RNA were RNA was isolat ed from untreat ed K12 cells (C); cells treat ed with 300 isolat ed from pRSV-l a nd pRSV-IR E stably tran sfected cells selecte d by nxt thapsigargin (Tg ) and cells pretreat ed with geniste in th en Tg added J.l ~ 1 ~I TX . At 16 h prior to RNA extra ction, cells were untreated (- ) or (G + Tg ), with 100 nxt of okada ic acid (OA ), or pret reat ed with OA then treat ed ( +) with 300 nxtof tha psiga rgin. The RNA blots were hybrid ized Tg ad ded (OA + Tg ). RNA blot ana lysis for grp78 and 3AIO mRNA with grp78, grp94 , a nd 3A 1O cDNA probes. The autoradi ogram is levels were performed. The au toradiograms ar c shown. B, the levels of shown. B, levels ofgrp 78 and grp94 mRNA from stably tra nsfected cells grp 78 mRNA from K12 and NIH3T3 cells aft er treatment wit h okadaic were qu antitat ed by densitomet ry and norm alized aga inst the invar i­ acid (OA) , genistein (G) , or thap sigar gin (Tg) or in combina tions (OA ant 3A /O mRNA level. The mRNA levels fill ' grp 78 and grp.94 from + Tg or G + 1'g) were determ ined by den sitomet ry and normalized pRSV-l st ably transfected cells wit hout treatment of tha psiga rgin were against mRNA levels of 3A 10. The levels of grp78 mRNA in untreated set as 1. The fold ind ucti on by Tg is indi cated. RNA levels from cells NIH3T 3 or K12 cells (C) were set as 1. The fold inductio n for each of the treated with (+) or wit hout (-) th apsigargin are plotted. treatment conditions is shown. response element, cAMP respon se element, a nd additional CCAAT elements ; however, th ey do not contribute to grp 78 To demonstrate that th e inhibitive effect of genistein on stress ind ucibili ty (Alexa ndre et al., 199 1). The results of all grp78 transcription is not due to a genera l toxic effect of tyro­ tra nsient tran sfecti ons are summa rized in Table I and some sine kin ase inhibitors and to delim it th e tyrosin e kina se activ­ representative CAT assays are shown in Fig. 7. ity targeted by genistein , we tested th e effect of L-A, a potent We next tested (- 375/-881MCAT, which contai ns 287 bp of inhibitor for EGF receptor-associated tyrosine kinase activities the grp78 promoter fused to a minimal MMTV promoter. The (Onoda et al., 19891on (- 456)CAT a nd (- 3551CAT. Two con­ results showed th at the 4-fold induction of CAT activity by Tg centrations (50 ml and 1 JL~1 1 of L-A were used. The results, as was completely elimina ted when the cells were treated with shown in Table I, demon strated th at L-A, either added alone or genistein in combinat ion with Tg, whereas genistein alone in combination with Tg, did not ha ve any effect. showed no effect (Table I). Thus, genistein response elements are Th e Potent iating Effect of Ohada ic Acid Requires a 50-bp locat ed within thi s 287-bp region. One of the target domains Region in the Vicinity of the grp78 TATA Sequence- In exam­ spans - 159 to - 110, since dupli cat e copies of thi s subfra gment ining th e response of (- 375/- 881MCAT, (- 159/- 1101MCAT when linked to th e minimal MMTV promoter (- 159/- 1101CAT and (- 109/- 741MCAT to OA treatment in combination with was able to confer a 8-fold induction by Tg, However, this induc­ Tg, we discovered that instead of a stimulative effect as ob­ tion was reduced to 1.6-fold when geniste in was added (Table I). served for (- 4561CAT, OA actually reduce d slightly the Tg This region contains the 3'-half of core and C3 element. Another induction. Thu s, the Tg inductio n for the three plasmids listed target region spans - 109 to - 74 and it contains the C1 element. decreased from 4-, 8- and 13-fold to about 2-, 6-, and 8-fold, Duplicate copies of th is subfragment linked to the minimal respectively (Fig. 7 and Tab le I). These results indicate that in l\IMTV promoter (- 109/- 74)CAT were able to confer a 13-fold NIH3T3 cells, the enhancing effect of OA on grp78 tran scrip­ induction by Tg. Addition of genistein and Tg completely elimi­ tion is not mediated by the kn own Tg response elements of the nated the induction (Fig. 7 and Table n grp78 promoter. 499 Thapsigargin Induction ofgrp 78 Requires Protein Kinases strea m region surrounding the grp78 TATA sequence. Previ­ (-4 56)CAT ously, we demonstrated that this same region, when fused to a CAT reporter gene, is almost devoid of any basal promoter .. 3AC H activity and is not inducible by Ca depletion stress (Resendez et al., 1988). DISCUSSION With the recent discovery that in the S. cereoisiae a serine/ threonine kin ase (IRE 1p) plays a critica l role in activating the ER protein gene tra nscript ion in response to malfolded protein accumulation stress, th e question arises as to whether protein + + T9 T9 phosph orylation is also util ized as a mecha nism for th e induc­ tion of the mammalian ER protein genes. In thi s report we focused on grp78 gene expression which is induced to high levels by th ap siga rgin following chronic depleti on of th e ER Ca + store and provide severa l lines of evidence that protein phosph orylation and deph osph orylation are involved in grp tran sactivation. Our first approach is to engineer th e coding sequence of z- IRE1p into a mammalian expression vecto r a nd create a stable : ~ CHO cell line with a mplified copies . Th is was achieved, and a <3 cell line harboring about 50 copies of the yeas t gene was estab­ '" f- <: lished. In these cells, sufficient a mounts of IRE1p were ex­ l1> pressed to allow iden tification with specific anti body aga inst ."' iii th e protein. Nonet heless, under these favo rab le conditio ns, th e Q; cr: enhancement for grp 78 mRNA level in stressed a nd non­ stressed cells was about 2-fold. Could this slight increase in grp78 mRNA levels be due to th e se rine/threonine kin ase ac­ tivity of the tra nsfected gene? Th e fact that the 2-fold induction we obser ved is similar to that report ed for S. cereoisiae when the sa me protein is overproduced over th e endogenous level (Mori et al., 1993) suggests th at this might be the physiological C OA G T9 OA G limit for thi s protein in a norm al cell background. Furthermore, + + the co-induction of other ER protein genes such as grp94 in th e T9 T9 IRE1p expressing cells is consistent wit h th e action of IRE 1p in FIG. 5. Effect of genistein and okadaic acid on the Tg induc­ yeast cells. On th e other hand, expression of a heterologous tion of grp78 promoter activity. A, NIH3T3 cells were transient ly protein has its inherent problems which are difficult to resolve transfected with (- 456)CAT and subjected to trea tment s with OA, in a mammali an system. For example, the modest enhance­ genistein, and Tg alone or in combinations (OA + Tg or G + Tg ). The autora diogra m of the CAT assays is shown. The positions of the chlor­ ment of the grp mRNA levels could have been caused by im­ amphenicol (CM) and its acetylated form s (3Ac and lAc) are indi cated. proper insertion of IRE1p into the design at ed cellul ar compart­ B, the CAT conversion in each sa mple was qua nt itated by an Ambis ment, th us diminishing the capacity of IRE1p to fully induce Radi oanalytic Imagin g Syst em. The level of th e prom oter activity in the grps. Anot her complication is that the portion of IRE1p untreated cells (C) was set as 1. The fold ind uct ion for the various treatments are shown. located in the endoplasmic reticulum might not fold properly. Thi s effect could have elicited a mild grp stress response. Al­ Since OA exerts a potentiating effect on (- 456)CAT bu t not though amino acids essential for th e kinase activity have been on (- 375/-88)MCAT (Fig. 7), th e target for th e stimulation identified within IRE1p (Mori et al. , 1993), to crea te an equiv­ eithe r lies in th e upstream region spanning - 456 and - 375 or alent CHO cell line with the amplified mutant genes at th e in th e downstream region spanning - 88 to - 38 (Fig. 3). Th e sa me level of express ion as the wild type gene to differentiat e upstream region contains basal enha ncer elements and th e between a 2-fold effect is tech nically difficult, if not impossible. downstream region contains th e sequence surrounding th e Th e functional significance of IRE1p in anima l cells is best grp78 TATA element at - 55 bu t excludes transcription initia­ evaluate d when t he fold induction of the tra nsfeeted wild-type tion located at ( + 1) (Li et al., 1993). To differentiate betw een pr otein can be man ipul at ed so that it is sufficiently high to these two possibilities, we first compared th e response between allow mu tational an alysis. Thi s may be possibl e using recipient (- 457)CAT and (- 355)CAT. Our results show th at th ese two cells wit h the homologous mammali an gene deleted. However, plasm ids responded nearly identically to th e treatment of Tg this requires the confirmation th at such a gene exists in mam­ and OA (Table I). Thus, the upstream sequence between - 456 malian cells and its subsequent isolation. Based on the current a nd - 355 is not needed for the potentiating effect as both limitations, we focused our effort int o examining th e contribu­ plasmids showed th at OA enha nced th e Tg induction from tion of endogenous protein kinases and phosph atases in the Tg about 6-12-fold. Next the response between (- 355)CAT and inductio n of grp78. (- 375/-88)MCAT was compared. Whil e both plasmids showed Poten t protein kinase inhibi tors have been used widely as a 4-5-fold induction by Tg, OA in combina tion with Tg stimu­ biochemical tools to study the functi on of kin ase a nd phosph a­ late d the (- 355)CAT activity to ll-fold and redu ced (-375/ tase activities in regul ating gene express ion in mammalian - 88)MCAT activity to 2-fold (Fig. 7 and Tabl e I). The minimal cells. In th is study we use a pan el of specific inhibi tors to MMTV promoter (MCAT) was relatively un affected by Tg or determine their effect on grp78 induction byTg. The majority of the various inhibitors, either added alone or in combinations intracellu lar phosph ata se activ ity has been attributed to PP-l with Tg (Fig. 7 and Tabl e I). Our results show th at th e OA and PP-2A, a nd OA has th e abi lity to inhibit both enzyme stimulative effect of grp 78 Tg induction requ ires a 50-bp down- activities. Genera lly, PP-2A is inhibited more efficiently th an 500 Thapsigargin Induction ofgrp78 Requires Protein Kinases OA G OA G OA ,+ ~.Jl,.--------,,-----, ~~ -150 -125 -95 -55 • • core ~ ~ ;OW -355 (-355)CAT (-375/-88)MCAT ....... ------. -159 -110 ..._----- ---..-.. (-159/-110)MCAT =========,r--- -- -- e I }~~~? ?~ -109 -74 ............................ TATA (-109/-74)MCAT ====:::1",'" r I tl~ TATA MCAT e1111??2@ FIG. 6. Schematic drawings of the grp78/CAT and grp78/MCAT fusion genes used in the transient transfection assays. The 5' and 3' end points of the grp78 promoter fragments used are indicated. The positions of the core, C3, C1, and TATA elements are indicated. The small triangles indicate upstream enhancer elements. The hatched lines in the MCAT constructs represent the promoter sequence of the mouse mammary tumor virus (MMTV). The MCAT plasmid contains the promoter fragment (-105 to +104) of the MMTV fused to the CAT reporter gene. The target regions for the inhibitive effect (-) of genistein (0) and okadaic acid (OA) and stimulative effect (+) of OA during Tg induction are bracketed. TABLE I Response ofgrp78 promoter/ CAT fusion plasmids to kinase and phosphatase inhibitors Fold induction" Construct Control OA G L-A Tg OA + Tg G + Tg L-A + Tg b b (-456)CAT 1 1.5 :!:0.1 0.8:!: 0.1 1.2 6.4 :!:2.4 12:!: 2.3 1.3 :!:0.4 6.5 (-355)CAT 1 1.7 :!: 0.2 ND 0.8 5.2:!: 1.3 11 :!: 1.3 ND 5.5" (-375/-88)MCAT 1 1.0 :!:0.3 ND 4.2:!: 1.0 1.0 :!:0.1 1.0 2.3:!: 0.6 ND (-159/-110)MCAT 1 1.1 :!:0.1 0.8 ND 8.2:!: 1.7 5.6 :!:0.4 1.6 :!:0.2 ND (-109/-74)MCAT 1 1.2 :!:0.2 0.9:!: 0.1 ND 13:!: 0.3 7.7:!: 0.5 1.2 :!:0.1 ND MCAT 1 1.5 :!:0.4 0.9 1.0 1.2 1.2 1.2 1.0 a Plasmids (-159/-110)MCAT and (-109/-74) contain two copies of the synthetic promoter subfragments fused to CAT as described previously (Li et 01., 1993). The plasmids were transiently transfected into NIH3T3 cells. The cells were either nontreated (control) or treated with okadaic acid (OA), genistein (G), lavendustin-A (L-A), or thapsigargin (Tg) alone or in combinations as indicated. For each construct, the fold induction over the control level set as 1 is shown. Most transfections were performed independently from two to five times. The standard deviation is indicated. ND, not determined. b 50 nM L-A. c II-'M L-A. PP-1 (Cohn, 1989). In our experiments, we discovered that OA genes are both stress-inducible genes and share some common only exhibits a mild potentiating effect on grp78 transcription. induction conditions (Lee, 1987; Watowich and Morimoto, Since the effect is more pronounced at 100 TIM of OA than 50 1988), their response to OA during stress is not identical. nM , PP-1, rather than PP-2A, may be the affected phospha­ In examining thegrp78 promoter sequence which is required tase. OA has been reported to stimulate a large number of gene for the OA stimulative effect on Tg induction in NIH3T3 cells, systems, including that of the heat shock gene (Schonthal et al., we discovered that it does not act through the Tg response 1991; Guy et al., 1992; Chang et al., 1993). Since OA strongly elements. Rather, it exerts a mild inhibitive effect on those increases the heat shock hsp70 promoter activity, it was pro­ elements. A 50-bp region in the vicinity of the grp78 TATA posed that the heat-induced transcriptional activation of the sequence is required for the OA stimulative effect of Tg induc­ heat shock gene is associated with the phosphorylation of com­ tion. Interestingly, no sequence resemblance was found with ponentts) of the transcriptional complex and that OA enhances known target sites of OA action (Wera et al., 1993; Kharbanda this phosphorylation (Chang et al., 1993). Based on the optimal et al., 1993; Chauhan et al., 1994; Hyun et al., 1994), and the concentrations of OA, PP-2A, as opposed to PP-1, was impli­ majority of the previously described OA response elements are cated in this induction process. Thus, although grp and hsp upstream regulatory sequences. Thus, the grp78 target may represent a novel site. However, OA does not have to act di­ 2 X. Cao, Y. Zhou, and A. S. Lee, unpublished results. rectly on the transcription factors which bind to the 50 bp 501 Thapsiga rgin. Induction of grp 78 Requires Protein Kinases (-3SS)CAT (- 109/ -74) MCAT ~ 3 A c FI G. 7. Differential effect of genis­ ~ l A c tein a nd okadaic acid on the Tg in­ d uction of grp78/CAT and g rp 78/ MCAT fusion genes. Left pan els, comparison betw een (- 355)CAT and ~ C M (- 375/- 88)i\ICAT. Right pan el, compa ri­ son between (- 109/- MCAT and th e min ­ (-3 7S/ -88) MCAT MCAT imal i\IMTV promoter driven CAT (MCAT). NIH3T3 cells were transiently ~ 3 A c tra ns fecte d with t hese plasm ids an d sub­ jected to the various treatment conditions ~ l A c (C, Tg, OA, G, OA + 7'g, G + Tg ). The autora diogra rns of the CAT as says are shown. The position of chloram pheni col (CM ) and its ace tylated form s (3Ac and ~ C M lAc) are shown. '----' '----' '----' '----' L.--...J l---...J L.--...J l---...J L..------J l---...J C Tg OA OA C Tg G G OA OA + + Tg Tg Tg region. For exa mple, it could inhibit a serine/threonine pr otein treatment of the cells with lavendustin-A, a specific inhibitor for EGF receptor-associated tyrosine kin ase, neither stimu­ phosphatase activity which modifies t he activ ity of any inter­ mediatory regul atory components. Nonetheless , our finding lates nor inhibits the Tg res ponse, sugges ting th at the tyrosin e that OA only exerts a modest effect on t he Tg induction ofgrp78 kinase acti vity mediating the grp response to Tg is likely to be a nd that the response does not involve the Tg response ele­ distinct from that of the EGF receptor type. Based on our findi ngs, we hypothesize th at a tyr osine kin ase activity is es­ ments prompted us to investiga te wheth er kinases other tha n that of th e se rine/threonine typ e are more critical elements for sential for grp indu ction by Tg. It is a lso possibl e that genistein inhibitsgrp induction through a yet unknown mechan ism. The the Tg induction process in mammalian cells. Our experiments indicate that gen istein, a tyrosine kin ase identification of th e genistein se nsitive component s, their sub­ cellular locali zati ons and regulatory mechan ism wiII cont ribute inhibitor, almost totally elimina te d the Tg induction response. Furthe rmore , we have found that the grp core, the C3 and Cl, significantly to th e und erstanding of how ER protein genes which are major targets for Tg induction, are also major target s respond to Ca + depletion stress in ma mma lia n cells. of the inhibitive effect exerted by geniste in. Thus, genistein Aclwowledgmellts- We acknowledge Cory Gorman for the gift of the acts eithe r directly on th e tran script ion factors which inte ract pRSV-l pla smi d, Larry Chas in for the DG44 cells, Ca roline Sha mu and with th ese promoter regul atory elements, or it modifies other Peter Walter for the pCSIIOplasmid and the IRElp a ntise ra. We th ank targets in th e intermedi ary path ways leading to th e activation Axel Schontha l, Wilfred Li, and Meera Ramakri shnan for cri tica l review of the manu script. of th e nu clear tran scription factors. Recent exa mples suggest that genistein ca n exert its effect both ways. In the T cell HEFE ltE NCES antigen receptor signaling system, genistein was used to iden ­ Akiyam a, T.. Ishida, .J., Nak agawa , S., Ognwa rn , II., Wata nahe, S.. ltoh, N., tify a tyrosi ne kinase activity which down-regu lates specifically Shi buya , ~ I. , a nd Fuka mi, Y. (1987 ) J. Hiol. CI" ,m. 2(;2, 5592- 55B5 Alexandre, S., Nakaki, T., Van hnm me, L., a nd Lee, A. S. (l BBI) Mol. End ocrinol. calcium-mediated signaling at a point downstream of the 5, 18(;2- 1872 lc k tyrosine kin ase p56 but up stream of calcium mobili zation Bal da ri, C. T., and Telford, ,J. L. (19 94) Eur.•J. lm m 11I10/. 24, 104(;-1052 Cha ng, S. C., Wooden, S. K., Na kaki, T., Kim, Y. K., Lin, A. Y., Kung, L.. Attenello, (Balda ri and Telford , 1994). Genistein, on th e other hand, .J. W., an d Lee, A. S. (1987) Proc. Not! . Acad. Sci. U. S . A. 84, (;80- (;84 blocks the transcript ional stimulation of interferon-induced Cha ng, N.-T., Hua ng, L. E., a nd Liu, A. Y.-C. (1993 ).1. /lio/. Chem, 2(;8, 14:l6 - 1439 genes by inhibitin g tyrosine phosph orylation of the tra nscrip ­ Cha uha n, D., Kharhanda , S. M.. Uchiya ma , H.. Sukhat me, V. P., Kufe, D. W., and Ander son, K. C. (1994) Can cer Res. 54 , 2234 - 2239 tion facto r ISGF 3 which binds directly to the interferon-stirn­ Cohe n, P. (1989) Allllu. Rev. Biochem . 58, 453- 508 ulated response element found in the promoter of inducible Cox, ./. S., Sh am u, C. E., an d Wal ter, P. (1993 ) Cel/ 7:l, 1197-1 206 Dru mmond, I. A. S., Lee, A. S., Resendez, K , ,/ 1'., a nd St einha rdt. H. A. (1987 ) genes (Gutch et al., 1992). In the grp 78 promoter, th e core, C3, J. Bioi. Chem. 2(;2, 12801-1 2805 a nd Cl act in concert to enhance grp expression (Wooden et al., Gutch , M. •L, Da ly, C.. a nd Reich, N. C. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 1991). Thus, it is possible that upon stress induction, as in the 114 11- 114 15 2 Guy, G. It, Cao, X., Chua , S. P., a nd Tan , Y. H. (W92) .J. Hiol. Chem. 2(;7, case of ER Ca + depleti on, a common tyrosine kinase activ ity is 1846 -1 852 activated which leads to the modification of the binding factors Hall, C. V., J acob, P. E., Ringold, G. ~ I. , a nd Lee, F. ( W831J. .1101. 111'1'1. Genet, 2, 101- 109 or th eir co-act iva tors, resul ting in grp78 induction. Hyun, S. W., Par k, K., Lee, Y. S., Lee, Y. I, Kim , S. •/. (1994) .J. /liol. Chem. 2(;9, Among kin ase inhibitors, gen iste in is a specific inhibitor for 36·1-368 tyrosine kin ases but not se rine and threonine kin ases and J ackson, T., Pat terson, S. I., Thastrup, 0 ., a nd Ha nley, 1\1. It ( 1988 ) Biochcm . •1. 253 , 8 1- 8(; other ATP ana logue-related enzymes (Akiyama et al., 1987). Kha rbandn, S., Rubin, E., Datt a. H., Bass , R., Sukh a tme, V., and Kufe, D. (1993) Th us, it has minimal effect on the enzyme activit ies of cAMP­ Cell Grouith & Differ. 4, 17- 2:1 Kim , Y. K., a nd Lee, A. S. (1989) Gene (Am st,) 77 , 123- 13 1 dependent protein kinases, protein kinase C, phosph orylase Kozutsurni, Y., Sega l, I\I., Nor mington, K., Gethin g, 1\1. J. , and Sarnbrook. vl. (1988) kinase , 5'-nucleotidase and phosph odiestera se. While it is Nat ure :1:12, 462-464 Lee, A. S. (198 1) .J. Cell. Physiol. ros, 119 -1 25 highly potent against tyrosine kinases such as the EGF recep­ v src Lee, A. S. ( HJ87) Trend Biochem. S ci. 12, 20 - 23 tor, pp60 - and ppllOl-:al-:-fc", it is also capable of inhibiting Lee, A. S. (1992) Curl'. Opin. c-utu«. 4, 267-27:l tyrosine kin ases other than these three well cha racterized Lee, A. S., Delegean e, A. 1\1., Bak el', V.. and Chow. P. C. (198:1)d. /Iiol. Chem, 258, 59 7- 603 ones. Since th e ta rgets for geniste in can be broad , we initiated Li, X. A., a nd Lee, A. S. nssn Mol. Cell. Bioi. 11, 3446 - :l45:1 experiments to determine which type of tyrosine kinase activ­ Li, X.. Warren, D. W., Gr egoire, .I., Peder son, H. C.. and Lee, A. S. (1989) .1/01. ity is involved in th e Tg indu ction of grp78. We found tha t Endocrinol. :I, 1944-W52 502 Thapsigargin Induction ofgrp78 Requires Protein Kinases Li, L.-J., Li, x.,Ferrario, A., Rucker, N., Liu, E. S., Wong, S., Gomer, C., and Lee, Physiol.152,545-552 A. S. (1992) J. Cell. Physiol. 153, 575-582 Resendez, E., Jr., Ting, J., Kim, K S., Wooden, S. K, and Lee, A. S. (1986) J. Cell Li, W. W., Alexandre, S., Cao, C., and Lee, A. S. (1993) J. BioI. Chem. 268, Bioi. 103, 2145-2152 12003-12009 Resendez, E. R., Jr., Wooden, S. K, and Lee, A. S. (1988) Mol. Cell. Bioi. 8, Li, W. W., Sistonen, L., Morimoto, R. 1., and Lee, A. S. (1994) Mol. Cell. BioI. 14, 4579-4584 5533-5546 Schonthal, A. (1992) New Bioi. 4, 16-21 Lin, A. Y., and Lee, A. S. (1984) Proc. Nat!. Acad. Sci. U. S. A. 81,988-992 Schonthal, A., Tsuk.itani, Y., and Feramisco, J. R. (1991) Oncogene 6, 423-430 Little, E., Ramakrishnan, M., Roy, B., Gazit, G., and Lee, A. S. (1994) Crit. Rev. Thastrup, 0., Cullen, P. J., Drebak, B. K, Hanley, M. R., and Dawson, A. P. (1990) Eukaryotic Gene Expression 4, 1-18 Proc. Nat!. Acad. Sci. U. S. A. 87, 2466-2470 Liu, E. S., and Lee, A. S. (1991) Nucleic Acids Res. 19, 5425-5431 Urlaub, G., MitchelJ, P. J., Kas, E., Chasin, L. A., Funanage, V. L, Myoda, T. T., Liu, E. S., Ou, J., and Lee, A. S. (1992) J. Bioi. Chem. 267,7128-7133 and Hamlin, J. (1986) Somatic Cell. Mol. Genet. 12,555-566 Mori, K, Ma, W., Gething, M. J., and Sambrook, J. (1993) Cell 74, 743-756 Watowich, S. S., and Morimoto, R. I. (1988) Mol. Cell. BioI. 8,393-405 Nouspikel, T., and Iynedjian, P. B. (1992) Eur. J. Biochem. 210,365-373 Wera, S., Belayew, A., and Martial, J. A. (1993) Mol. Endocrinol. 7,965-971 Onoda, T., Iinuma, H., Sasak.i, Y., Hamada, M., Isshiki, K, Naganawa, H., and Wooden, S. K, Li, L.-J., Navarro, D., Qadri, 1., Pereira, L., and Lee, A. S. (1991) Takeuchi, T. (1989) J. Nat. Prod. 52, 1252-1257 Mol. Cell. Bioi. 11, 5612-5623 Pelham, H. R. B. (1986) Cell 46, 959-961 Price, B. D., Mannheim-Rodman, L. A., and Calderwood, S. K (1992) J. Cell. Zhong, L., Batt, D. B., and Carmichael, G. G. (1994) BioTechniques 16, 56-57

Journal

Journal of Biological Chemistry – Unpaywall

Published: Jan 1, 1995

There are no references for this article.