[CANCER RESEARCH 58. .1950-3956, September 1, 1998]

Regulation of Expression of the DNA Repair Gene O6-Methylguanine-DNAMethyltransferase via Protein Kinase C-mediated Signaling1

Istvan Boldogh, Chilakamarti V. Ramana,2 Zhenping Chen, Tapan Biswas, Tapas K. Hazra, Sabine Grösch,Thomas Grombacher, Sankar Mitra,3 and Bernd Kaina

Sealy Center for Molecular Science and Department of Hitman Biological Chemistry and Genetics ¡C.V. R., T. B., T. K. H.. S. M.j and Department of Microbiolog\ andImmunologv [I. B., Z. C.I. Unhersin of Texas Medical Branch, Galveston, Texas 77555-1079: and Division of Applied Toxicology; University of Mainz, D-55I3I, Mainz,

Germany IS. C.. T. C.. B. K.I

ABSTRACT

O6-Alkylguanine is the major mutagenic and cytotoxic DNA lesion

induced by alkylating agents, including 2-chloroethyl-jV-nitrosourea-based antitumor drugs. This lesion is repaired by O'-methylguanine-DNA

methyltransferase (MGMT), the expression of which is highly variable inboth normal tissues and in tumor cells. The promoter of the humanMGMT gene was found to contain two putative activator protein (AP)-l

sites. Here, we show that the level of MGMT mRNA in HeLa S3 cells wasincreased 3-5-fold by phorbol-12-myristate-13-acetate (TPA) and 1,2-diacyl-sn-glycerol (DAG), which are activators of protein kinase C (PKC),

as well as by okadaic acid, an inhibitor of protein phosphatases. The PKCinhibitor l-(5-isoquinoline sulfonyD-2-mcthylpiperazine-HCI eliminatedMGMT activation by TPA and DAG but not by OA. Prior down-regula

tion of PKC abolished subsequent effects of TPA or DAG. The resultsindicate AP-1 to be involved in regulation of MGMT expression. Thishypothesis was supported by showing AP-1 binding to two target se

quences of the MGMT promoter and transactivation of the MGMTpromoter upon cotransfection with c-fos and c-jun in F9 cells. ThatTPA-mediated induction of MGMT caused increased cellular resistance to2-chloroethyl-/V-nitrosourea suggests a therapeutic significance for PKC-

mediated MGMT modulation.

INTRODUCTION

O6-Alkylguanine, the major mutagenic base adduci induced in

DNA by a variety of alkylating agents, including environmentalcarcinogens, is repaired by MGMT4 (EC 2.1.1.63; Refs. 1 and 2). This

ubiquitous protein is irreversibly inactivated as a result of the acceptance of the alkyl group in a single-step, stoichiometric reaction.Antitumor alkylating drugs of the CNU type induce O6-chloroethyl-

guanine in DNA, ultimately leading to the formation of cytotoxicDNA cross-links (3, 4). Repair of the Oh-alkylguanine adduci by

MGMT, thus, prevents the cytotoxicity of these drugs (5, 6).MGMT expression is highly variable in normal tissues as well as in

tumor cells (7. 8), and the cellular sensitivity of these cells to CNUvaries accordingly (9-11 ). Tumor cells that are resistant to drugs of

the CNU type often overexpress MGMT (2, 12). The regulation of

Received 2/23/98; accepted 7/2/98.The costs of publication of this article were defrayed in pan by the payment of page

charges. This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

' This research was supported by United States Public Health Sen-ice Grants ES07572

and CA31721; National Institute of Environmental Health Sciences Center GrantES06676 (to S. Ml: Deutsche Forschungsgemeinschaft DFG KA 724/4-3. SFB 509 and

B4: and Stiftung Rhein land Pfalz (lo B. K.).1 Presenl address: The Cleveland Clinic Foundation, 9500 Euclid Avenue. Cleveland.

OH 44195.' To whom requests for reprints should be addressed, at Sealy Center for Molecular

Science. University of Texas Medical Branch. Galveston. TX 77555-1079. Phone:(409)772-1780; Fax: (409)747-8608.

4 The abbreviations used are: MGMT, (/"-methylguanine-DNA methyllransferase;

CNU. 2-chloroethyl-/V-nitrosourea; hMGMT. human MGMT; AP, activator protein; PKC.protein kinase C; cAPK. cyclic AMP-dependenl protein kinase A: TPA. phorbol-12-myristale-13-acetate; FBS. felal bovine serum; DAG, 1.2-diacyl-sn-glycerol: db-cAMP, dibu-lyryl cyclic AMP: OA. okadaic acid; H-7. l-(5-isoquinoline sulfonyl)-2-methylpipera/.ine-HC1; H-89, N-|2-(memylamino)ernyl]-5-isoquinolincsu1fonamide-HCl: BCNU. l.3-bis-(2-chloroelhyl I-1-nitrosourea: RSV, Rous sarcoma virus; ß-gal.ß-galactosidase: EMSA.electrophoretic mobility-shift assay.

MGMT expression, thus, is of significant clinical importance. Earlierstudies showed that MGMT expression can be activated by a varietyof genotoxic agents in rodent and human cell lines (13-15). However,

the molecular mechanisms of MGMT regulation are not well understood. Recent studies have implicated altered chromatin organizationand nucleosome positioning and aberrant silencing in the CpG islandin the promoter region to altered expression of the gene (16-18), but

the relevant cis elements in the promoter have not been explored. Toelucidate the signaling pathways that contribute to MGMT overex-

pression, we cloned and characterized the hMGMT promoter andidentified several candidate cis elements (19). An enhancer-like se

quence was subsequently identified (20), but its role in MGMTregulation remains to be analyzed. Obvious sequences in the MGMTpromoter that are potential targets for involvement in in vivo tran-scriptional activation of the gene include two each of the glucocorti-coid response elements, AP-1 and AP-2. Indeed, the MGMT gene was

found to be inducible by the glucocorticoid dexamethasone (21), andthe potential function of the glucocorticoid response elements ininduced expression of MGMT was explored in detail (22). Becausemany treatments, including genotoxic exposure, activate AP-1 as an

immediate early response (23), it is also important to elucidate thefunctional role of the AP-1 sequences within the hMGMT promoter.

PKC and cAPK activate the AP-1 and AP-2 transcription factors,

respectively, and this activation is required for their binding to thecognate DNA sequences (24). PKC levels and/or activity are alteredin many tumor cells, and a change in the expression of PKC isotypesis often associated with a change in the regulation of cell growth,immortalization, and/or malignant transformation (25-27). Changing

levels and the functional divergence of PKC family members and theirmolecular heterogeneity make them attractive targets for anticancerdrug development (28-30). We show in this study that: (a) PKC butnot cAPK up-regulates hMGMT promoter activity in tumor cells; (b)AP-1 is involved in enhanced MGMT expression; and (c) a TPA-

mediated increase in MGMT level enhances cellular resistance againstalkylating agents.

MATERIALS AND METHODS

Cell Lines and Reagents. HeLa S3 (ATCC CCL 2.2) cells were maintained in DMEM/F-12 (Lite Technologies, Inc.) supplemented with 10% FBS

(Life Technologies, Inc.), glutamine (0.292 g/liter). penicillin (100 units/ml),and streptomycin (100 /xg/ml). Mouse teratocarcinoma cells (F9) were culturedin DMEM/F10 medium supplemented with 10% FCS. 2 mM L-glutamine, and0.001% ß-mercaptoethanol and antibiotics. All cultures were routinely tested

for mycoplasma contamination using Hoechst stain 33258. TPA was obtainedfrom Sigma Chemical Co. DAG, db-cAMP, OA, H-7, and H-89 were pur

chased from Calbiochem and Immunochem. BCNU was obtained from BristolLaboratories.

Northern Blot Analysis. Total cellular RNA was isolated from HeLa S3cells using RNA/ol (Tel-test. Inc.) and quantitated spectrophotometrically.

Northern blots were performed according to standard protocols (31) andprobed for hMGMT or 18S rRNA. The MGMT cDNA probe was an £r«RIfragment from pKTIOO (32), and the 18S rRNA probe was a 1.15-kb BamHl-

EcoRl fragment from p5B (33). The radioactivity on the membranes were

3950

on March 14, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

AP-1 ACTIVATION OF HUMAN MOMT

subjected to Phosphorlmager analysis, and the signals were quantified usingImage Quant (Molecular Dynamics) software.

Western Blot Analysis. For Western blot analysis, protein extracts fromHeLa S3 cells were resolved by 10% SDS/PAGE and electroblotted to nitrocellulose membranes (Protran, Schleicher & Schuell). Primary antibody (poly-

clonal rabbit antisera against bacterially expressed hMGMT) was used at a1:500 dilution. Immune complexes were visualized using horseradish peroxi-dase-conjugated secondary antibody ( 1:3000 dilution) and the enhanced

chemiluminescence system (Amersham Life Sciences) using suitable exposuretimes so that the signals were in the linear range.

Construction of Expression Vectors. A 978-bp BamHl/Xbti\ fragmentcontaining the AP-1 sites in the promoter and the transcription start site wasfirst subcloned in pUCI9. The resultant recombinant was called pl911. Site-directed mutagenesis was carried out using Stratagene's double-stranded mu-

tagenesis kit based on a modification of the unique site elimination mutagenesis procedure (34), to mutate one or both of these AP-1 sites. The AP-K1)sequence CGAGTCA was changed to CGAGTTG and the AP-1 (2) sequence

TGAGTCA was changed to TGAGTTG. The mutated plasmid sequences wereconfirmed by dideoxy sequencing. The Kpn\IXba\ fragment from pl911 containing wild-type or mutated AP-1 sites were cloned into Kftnl/Nhel site of

pGL2 basic to generate four different expression plasmids with mutations innone or either one or both of the AP-1 sites and were named p-954/+24 ML.P-954/+24 MLAPM1, p-954/+24 MLAPM2. and p-954/+24 MLAPMI2,

respectively.Assay for Transiently Expressed Reporter Gene. F-9 cells were seeded

in 10-cm dishes (IO5 cells/dish) and transfected by calcium phosphate copre-

cipitation as described previously (11). For each transfection. we used 10 jigof one of the hMGMT luciferase constructs together with 1 /ng of each ofRSV-ß-gal. RSV-c-y««,RSV-c-/(«, or RSV-0 as a control, without RSV-c-junfRSV-c-fos. The amount of DNA was adjusted to 20 j¿gwith sonicated

salmon sperm DNA. Cells were harvested for determination of luciferaseactivity in cell extracts 17 h after transfection. Luciferase activity was measured by chemiluminescence luciferase assay (Promega) according to the manufacturer's protocol. Luciferase activities were normalized for transfection

efficiency by measuring in parallel the ß-galactivity in the same cell extractusing a commercial ß-galassay kit (Tropixl.

EMSA. EMSA was performed as described previously (35). Oligonucleo-tides used in this assay were MGMT AP-K1) (aggacacgagtcatactgga). MGMTAP-1 (2) (agcggctgagtcaggctctg). mutated MGMT AP-1(1) (aggacacgagttgtact-gga), MGMT AP-1 (2) (agcggctgagttgggctctg), and the mouse collagenaseAP-1 oligonucleotide (agtggtgactcatcact). We used a random 42-mer as a

nonspecific oligonucleotide. Oligonucleotides were synthesized commercially

(MWG Biotech, Ebersberg, Germany). In brief. 1-4 /^g of nuclear extract from

exponentially growing cells were incubated in bandshift buffer [ 10% glycerol.50 mM KC1, 10 mM HEPES-OH (pH 7.9). 4 min Tris-CI. 0.5 mM DTT, 0.5 mMEDTA, 10 ng BSA, and 1 jig poly(dl-dC)] for 30 min on ice together with 25

fmol of radiolabeled oligonucleotide as indicated. To test the effect of c-Jun

antibody (Santa Cru/. Biotechnology) on the mobility of the complex, 500 ngof the antibody were added to the nuclear extract 30 min prior to the oligonucleotide binding reaction and incubated at room temperature.

Survival Assay. Exponentially growing HeLa S3 cells were serum-starvedfor 18 h and then seeded at a density of 200 cells per 60-mm Petri dish in

medium containing 0.5% FBS and. 3 h later, were treated with activators/inhibitors of the protein kinase pathway or inhibitors of cellular phosphatasesfor 1 h. After another 6 h of incubation in serum-tree medium, the cells were

washed and treated with increasing doses of BCNU for 1 h. by addition ofstock solutions directly to the medium. The cells were washed with PBS onceand then grown in medium containing 10% FBS. Colonies were allowed togrow to I mm in diameter (—10-12 days) and were then fixed and stained.

Relative cell survival was calculated from the number of colonies per dish.

RESULTS

Increase in hMGMT mRNA Level by Activation of the PKCPathway. TPA and DAG activate their intracellular receptor (PKC)by a similar mechanism (36). To investigate the involvement of PKCpathway in hMGMT mRNA expression, cells maintained in 0.5%serum-containing medium for 24 h (to avoid serum stimulation of

PKC) were treated with various concentrations of TPA (for 1 h), andtotal RNA was harvested at various time points. Northern blot analysiswas carried out with both hMGMT cDNA and an 18S rRNA probe.Our preliminary studies showed that levels of ß-aclinor cyclophylin

RNA, often used as internal normali/.ation standards, were significantly altered as a result of TPA treatment. The signals on theautoradiograms were quantitated, and changes of hMGMT mRNAlevels were normalized to that of 18S rRNA. At doses of >10 ng/ml,TPA increased hMGMT mRNA level in a dose-dependent fashion(Fig. IA). Higher concentrations of TPA (500-1000 ng/ml) appeared

to be cytotoxic (data not shown). Because 100 ng/ml TPA resulted ina consistent induction of hMGMT mRNA without being cytotoxic. weused this dose in later studies. Fig. l B demonstrates the effect of TPAas a function of time after addition of TPA to the medium. A maximal

H-7 TPA DAG H-7 H-7TPA DAG

MGMTISS

«0«•••»,.6SS

51H

4£<

312SS

tà '̄J-l...—n°

- o £ASg;: ooo&

«c <cï1-0. 0.D.1-

Ht-

MGMT

18S I

n0 20 40 60 80 100

TPA (ng/ml)

Fig. 1. Modulation of hMGMT expression by PKC activators and inhibitors. Northern blot analyses were earned out with RNA from serum-starved HcLa S3 cultures treated withPKC activators/inhibitors for the indicated time periods. Data ¡minisand columns, means of two to four independent experiments. The increases in RNA levels were normalized withrespect to the 18S RNA signal. Top, representative blots for each experiment. A, TPA dose-dependent changes in hMGMT mRNA levels. Cells were exposed to 10. 50, and 100 ng/mlTPA or solvent (DMSO) for 1 h and harvested after 6 h of treatment. B and C time-dependent changes in hMGMT mRNA levels after exposure of cells to TPA and DAG. respectively.hMGMT mRNA levels in cells treated with TPA (B: 100 ng/ml: •¿�)or DAG <C; 100 ng/ml: •¿�)or DMSO (control; O) were analyzed after 0. 3. 6. and 9 h of treatment. D. effect ofTPA pretreatment on TPA-mediated induction of MGMT mRNA. Columns. RNA levels after 6 h in mock-, TPA ( 10 ng/ml)-. TPA (100 ng/ml)-. or TPA (100 ng/ml (-treated cellspretreated with TPA ( 10 ng/ml) for 24 h. E. effect of H-7 on TPA- or DAG-medialed induction of hMGMT. Columns. RNA levels in mock-. H-7 (20 J¿M)-,TPA (100 ng/ml)-. DAG(100 ng/ml)-, H-7 (20 JIM)- plus TPA (100 ng/ml)-. and H-7 (20 JIM)- plus DAG (100 ng/ml »-treatedcells 6 h after addition of the agents.

3951

on March 14, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

AM ACTIVATION OF HUMAN MGMT

TRA H-7 H-7TPA

MGMT

0)S

3.51

31

2.5.E

20>I-

1s|

0.5

0-j-j-'-;•U01i•¿�a-b-

—¿� H

Fig. 2. Effects of TPA and H-7 on hMGMT protein levels. Columns. MGMT levels inextracts from mock-, TPA (100 ng/ml)-. H-7 (20 /IM(-, and H-7 (20 ¡IM)-plus TPA (100ng/ml (-treated HeLa S3 cells. Top. representative Western blot.

5-fold increase in hMGMT mRNA was observed 6 h after exposure to

TPA.Phorbol esters activate PKC but may also exhibit other effects

besides those on PKC isotypes. To determine whether the effectsobserved after TPA treatment can be attributed to activation of thePKC pathway, we evaluated the effect of DAG (generated by thehydrolysis of phosphatidylinositol 4,5-diphosphate), another specific-

activator of PKC (37, 38). In these studies, we used a membrane-permeable synthetic diacyl glycerol. HeLa S3 cells were serum-

starved as above and treated with DAG (100 ng/ml), and the totalRNA was harvested at different time points and subjected to Northernblot analysis. Fig. 1C shows that DAG mediates a significant increase(~4-fold after 6 h of treatment) in hMGMT mRNA levels. The extent

and kinetics of induction were similar to those for TPA. These datasuggest that TPA-induced increases in hMGMT mRNA levels are

most likely mediated via the PKC pathway.Prolonged treatment with low concentrations of TPA resulted in

down-regulation of PKC. This effect was previously used to investigate the specificity of TPA-mediated cellular responses. Treatment ofHeLa S3 cells with 100 ng/ml TPA after a low-dose exposure (10

ng/ml for 24 h ) to the same agent prevented the induction of hMGMTmRNA (Fig. ID). H-7 reversibly inhibits both the holoenzyme and the

catalytic subunit of PKC (39). That the increase in hMGMT mRNAwas abolished in cells exposed to H-7 (20 /MM)before treatment with

100 ng/ml TPA or DAG for l h (Fig. IE) again suggests that PKCplays a critical role in activation of the hMGMT promoter. TPA wasremoved after l h of exposure, but DAG and/or H-7 was present forthe entire duration of the experiment. That addition of H-7 to HeLa S3

cells did not substantially decrease basal levels of hMGMT mRNA

suggested that the promoter elements responsible for basal expressionare not particularly sensitive to PKC-mediated signaling.

Next, Western blot analysis was performed to test whether TPAactivation increases hMGMT protein levels. Parallel cell cultures wereserum-starved as before and then treated with TPA (100 ng/ml), H-7(20 /LIM),or TPA ( 100 ng/ml) plus H-7 (20 /J.M);the amount of MGMT

protein was quantitated at various times after treatment. Fig. 2 showsan ~4-fold increase of the MGMT protein level after TPA treatment.In agreement with results of Northern blot analysis, H-7 did notsubstantially affect basal levels; however, it eliminated the TPA-

induced increase in hMGMT protein levels. Similar results wereobtained after DAG and/or H-7 exposure of cells (data not shown).

These results also indicate a concordance of MGMT mRNA andprotein levels during regulation of the gene.

Effect of OA on hMGMT mRNA level. In view of the involvement of PKC in regulation of the hMGMT promoter, we askedwhether inhibiting the dephosphorylation of transcriptional activatorswould affect expression of the hMGMT gene. OA, a potent inhibitorof protein phosphatase-1 and protein phosphatase-2A in vitro and in

vivo (24), was used to obtain insights into the potential role ofphosphorylation. Exposure of HeLa S3 cells to OA (20 nM) resulted ina 3-4-fold increase in hMGMT mRNA level (Fig. 3). Parallel cultures

of HeLa S3 cells were also treated with TPA (100 ng/ml) or DAG(100 ng/ml) together with OA (20 ng/ml). TPA was removed after al h exposure, whereas DAG and OA were present for the entireexperimental period. No potentiation of TPA- or DAG-mediated

activation of hMGMT mRNA was observed with OA.cAPK Did Not Significantly Affect MGMT mRNA Level in

HeLa Cells. db-cAMP (10 /MM)and/or theophylline (100 /MM;Ref.

40) neither altered hMGMT mRNA expression nor modified theactivating effect of TPA on hMGMT expression. Similarly, H-89, aninhibitor of cAPK (41), did not alter the TPA-mediated increase in

MGMT RNA levels (data not shown). These results suggest thatcAPK, the target of which can be AP-2, is not involved in signaling

that modulates hMGMT expression.AP-1 Binds to MGMT Promoter Sequences. Given the role of

PKC in activating MAP kinases and Fos/Jun as downstream targets,our results on the PKC-mediated up-regulation of hMGMT supportthe view that the observed response was elicited via AP-1-mediated

activation of the hMGMT promoter. To confirm this possibility, wemeasured AP-1 binding to hMGMT promoter sequences, as well asAP-1-mediated promoter activation. Proteins in nuclear extracts ofHeLa S3 cells bind to the AP-1 sequences located in the hMGMT

promoter as detected by EMS A (Fig. 4A). The binding sites were

Fig. 3. Effect of OA on hMGMT mRNA levels. Northern blot analysis of total RNAfrom serum-starved HeLa S3 cells treated with OA (20 nM) alone or in conjunction withTPA (A: 100 ng/ml) or DAG (Si 100 ng/ml) at 0, 3. 6. and 9 h. O. mock-treated cells: •¿�.TPA- or DAG-trealed cells; D, OA-treated cells; •¿�.OA- plus TPA- or DAG-treated cells.

3952

on March 14, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

AP-1 ACTIVATION OF HUMAN MGMT

A B

? S E E ^ ^

S S £S o oCL 0. 0_ O. CL Q.

competingoligonucleotide

-fold molar excess

AP-1 Coli.

o

?^-s^DÌ.<^*—*^CL<'SE^—*^_GL^_,is*^CM¿<OOtí.oEiM3

OOO

O o OOOOOOO- - ~ ~

OOO

D £"5

ü

. * = =^ E E =S £S o

O- O. CL O. O.< < < < <

oC8C(CoCqC3•|—1i4->

C8C3'—COHcompeting

oligonucleotide(5-fold molar excess)

AP-1 (1)wt

AP-1 (2) wt

Fig. 4. Binding of AP-1 to hMGMT promoter sequences. A. binding of nuclear factors to two putative AP-1-binding sites in the hMGMT promoter. Labeled oligonucleotidescontaining the putative AP-1 binding sites of the hMGMT-promoter [AP-I(It u7 and AP-1 (2) wt] and or mutated sequences [AP-HÃŒt"im and AP-l(2) mut) were each incubated with

4 ¿igprotein of mouse fibroblasl nuclear extracts. Binding of nuclear factors was monitored by EMSA. As a control, the same extract was tested with an oligonucleotide correspondingto the AP-1 site sequence present in the collagenase promoter. B, competition for binding to the collagenase AP-1 site by the hMGMT AP-1 oligonucleotides. A radioactive!)' labeledcollagenase AP-1-binding site oligonucleotide was incubated with nuclear extract (1 fig protein) in the absence or presence of competing unlabeled oligonucleotides in different molarexcesses as indicated. Competition for binding was shown by EMSA. As a control, a 100-fold molar excess of a nonspecific oligonucleotide (40-mer) without the AP-1-binding sitewas used for competition. C. competition for binding to the AP-1 sequence in the hMGMT promoter. Labeled AP-lf I) wt and AP-112) wt oligonucleotides were incubated as describedin A in the absence or presence of a 5-fold molar excess of competing unlabeled oligonucleotides. Competition for binding was determined by EMSA. D, binding of c-Jun protein toAP-1 sites in the hMGMT promoter. Labeled AP-¡ID wt and AP-1 (2) wt oligonucleotides were incubated with 4 /ig of nuclear extract, which were preincubated with anti-c-Jun antibody(antijun) or were mock-treated (contrai)- Binding of nuclear factors produced a shift (j) that was further supershifted after incubation with an anti-Jun antibody (ss; supershift).

designated here AP-l(l) and AP-1 (2), respectively (for sequences, seeFig. 5A). There was significantly less binding to the AP-l(l) sequencethan to the AP-1 (2) sequence. In both cases, however, the observedbinding was lower than to the AP-1 site in the context of the colla

genase promoter sequence (Fig. 4A). Introduction of mutations intoeither one of the binding sites completely abolished AP-1 binding(Fig. 4A). The factors that bind to the AP-1 site of the collagenasepromoter also bind to the AP-1 site in the context of the hMGMT

promoter, as shown in competition studies (Fig. 45). Addition ofunlabeled AP-l(l) or AP-1 (2) oligodeoxynucleotides of the hMGMTpromoter abolished bandshift of the collagenase AP-1. In contrast, themutated AP-1 sequences of the hMGMT promoter were ineffective insuch competition experiments. Interestingly. AP-1 (2) of the wild-type

hMGMT promoter had a higher ability to compete with the collagenase AP-1 site for protein binding than did the AP-1(1) sequence of the

hMGMT promoter (Fig. 45). This result is in agreement with theAP-1-binding activity of both sequences, shown in Fig. 4A. Fig. 4C

shows that when competition was performed with either radiolabeledwild-type AP-l(l) or AP-1 (2) sequences of the hMGMT promoter,the AP-1 sequence of the collagenase promoter prevented bandshiftformation completely, AP-l(l) and AP-1 (2) of MGMT were less

effective, and the corresponding mutated oligonucleotides were completely ineffective. The complex formed with hMGMT AP-1 promoter sequences contained c-Jun, as revealed by the supershift in thepresence of antibodies specific for c-Jun (Fig. 4D).

Activation of MGMT Promoter by Fos and Jun. To demonstratea functional role for the AP-1 sites in the promoter of the repair gene,hMGMT promoter-luciferase expression plasmids were constructed inwhich both AP-1 sites were either intact or one or both of these sites

were mutated (Fig. 5/4); the mutations introduced were the same asthose used in EMSA. Because the mouse F9 teratocarcinoma cellswere found to express very low levels of c-Fos and c-Jun, these cells

provided a sensitive system for directly testing the contribution ofAP-1 to activation of hMGMT. Cotransfection of the intact hMGMTpromoter constructs with RSV-c-/o.v and RSV-c-jun in F9 cells gaverise to a dramatic increase in promoter activity (~30-fold) relative to

the control. This induction was significantly reduced although notcompletely prevented by the hMGMT promoters harboring mutationsin either one or both AP-1 sites (Fig. 55). These data clearly establishthat AP-1 is the downstream target of PKC-mediated signaling in

volved in hMGMT promoter activation.PKC-induced Increase in hMGMT Level Is Associated with

Alkylating Drug Resistance. If stimulation of the PKC pathwayincreases hMGMT expression in HeLa cells, there should be a resulting increase in cellular resistance to CNU. We tested this possibilityby comparing the sensitivity of HeLa S3 cells to toxic doses of BCNUwith or without PKC activation, using a clonogenic survival assay.Fig. 6 shows the cell survival determined in cells treated with TPA.H-7, or H-7 plus TPA. TPA mediated a significant increase in cellularresistance to toxic doses of BCNU. H-7, an inhibitor of PKC, elimi-

3953

on March 14, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

AP-1 ACTIVATION OF HUMAN MOMT

BamHlLr-tAP-l(l)

AP-1(2)Xba\M

luciferasc1 SV40 Dolv(A) 1t—y

Fig. 5. Effect of compression of c-fos/c-jun on the activity of varioushMGMT promoter constructs. A. schematic diagram of the expressionplasmid p-954/+24 ML. The 5' and 3' located AP-1 sites within the

promoter [with respect to the transcription start site] were designatedAP-Iil ) andAP-I{2). respectively. The wild-type (u7) and mutated (muí)AP-1 sequences are shown with the mutated bases (boldface letters). B,P-954/ + 24 ML (Control) and the corresponding derivatives (see "Materials and Methods") harboring mutations in either one of the AP-1 sites

or in both sites were cotransfected with RSV-0-gal, RSV-c-/os. RSV-c-jun, or, as a control, RSV-0 into F9 cells. The luciferase and ß-gal

expression levels were determined, and the luciferase activity was normalized to ß-galactivity in each assay. There was no effect of Fos/Junexpression on RSV-ß-galactivity. The extent of induction was calculated from relative luciferase activity upon cotransfection of hMGMTpromoter constructs with RSV-fos and RSV-jun on the one hand andRSV-0 on the other (background control). Columns, means of threeindependent transfection experiments; bars, SE. The differences in expression between wild-type and the mutated MGMT promoter constructs

are statistically significant.

AP-l(l) wt: AGGACACGAGTCATACTGGA AP-1(2) wt: AGCGGCTGAGTCAGGCTCTG

AP-l(l) mut: AGGACACGAGTTGTACTGGA AP-1(2) mut: AGCGGCTGAGTTGGGCTCTG

B

I 25

nated the increase in cellular resistance to the alkylating agent. Incontrol experiments, exposure of cells to H-7 alone did not significantly increase the cells' sensitivity to BCNU. These results clearly

indicate that activation of PKC influences the sensitivity of tumorcells to CNU-type drugs.

100

40 60

BCNU (UM)Fig. 6. Survival curves (determined by clonogenic assay) showing TPA-induced

protection of HeLa S3 cells from BCNU cytotoxicity. O. mock-treated cells; •¿�,TPA-trealed cells; 0, H-7-ireated cells; •¿�.H-7- plus TPA-treated cells. Data poinis, percentagesurvival, expressed as the mean of three independent experiments; bars, SE. The concentrations of TP A (100 ng/ml) and H-7 (20 /IM) used in combination with BCNU were

not cytotoxic.

DISCUSSION

DNA repair proteins are generally expressed constitutively, but forsome other proteins, a significant variation in its expression betweentissues and even within tissues has been observed (1, 2, 42-44).

MGMT typifies such cases because a large variation in its level invarious normal and tumor tissues is well documented. An understanding of the mechanism of overexpression of MGMT, particularly intumor cells, has clinical significance because reversal of such over-

expression may lead to sensitization of the tumor to methylating andCNU-type alkylating antineoplastic drugs.

Regulation of MGMT expression in both normal and tumor cellshas been widely investigated, but the molecular mechanism(s) havenot been elucidated in detail. The levels of hMGMT mRNA andprotein are dependent on the cell type, cell cycle stage, and differentiation status (45), as well as on the activity of the signaling networkthat regulates the promoter. The presence of putative AP-1 and AP-2

response elements in the hMGMT promoter, thus, suggested theirinvolvement in MGMT activation. Here, we have investigated theroles of multiple AP-1 sites in the hMGMT promoter region in HeLa

S3 cells, which express various isoforms of PKC. FBS, a regularcomponent of culture medium, often masks effects mediated by acti-

vator(s)/inhibitor(s) of PKCs, and we did not observe substantialresponses to either TPA or H-7 in terms of mRNA levels for c-fos,c-jun (46), or hMGMT in the presence of 5-10% FBS. However,

significant overexpression of hMGMT was observed when cells at60-70% confluence were conditioned to low serum (0.5%)-contain-ing medium for 24 h prior to TPA treatment. Thus, serum-starved cell

cultures were routinely used in this study. There is strong biologicalrationale for the common procedure of studying gene regulation incells under starved conditions to down-regulate various signal trans-

3954

on March 14, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

AP.I ACTIVATION OF HUMAN MGMT

auction pathways. In addition to the obvious benefit of inducing alarger signal in gene regulation, the starved cells may be closer tothose in vivo in regard to their growth rate and suboptimal conditionsof no growth requirements.

Our results suggest that AP-1 sequences within the MGMT promoter are involved in MGMT activation. One of the two AP-1

elements in hMGMT promoter matches the consensus sequence(CGAGTCA) exactly, whereas the other is slightly divergent(TGAGTCA). That TPA, an activator of the PKC family, caused anincrease in the steady-state levels of hMGMT mRNA, as well ashMGMT protein, in a time- and dose-dependent fashion argues for animportant role of AP-1 factors in regulating hMGMT promoter activ

ity. Moreover, EMSAs and transient expression assays strengthenedthe idea that the AP-1 sites are involved in affecting transcription.

Similar activation of hMGMT by DAG, another specific activator ofPKC, provided further evidence of PKC involvement. Although theextent of activation was somewhat less than that observed with TPA,the kinetics of increase in hMGMT mRNA were comparable in thetwo cases. The effect of H-7, which prevented the increase in MGMT

mRNA, also provided evidence for the involvement of PKC inMGMT activation. However, that the basal level of the RNA was notaffected suggests that AP-1 contributes only to inducible expression

of the gene. Such a possibility is consistent with our earlier observation that the basal level of MGMT is comparable in the wild-type andc-Fos-negative cells (47). An earlier report showed that in vivo di

methyl sulfate footprinting of the hMGMT promoter in a cell line didnot detect AP-1 binding to the AP-1 sites in the promoter nor did it

provide any evidence of its inaccessibility to transcription factors (18).There are multiple reasons to explain the inconsistency of our studiesversus the footprinting data. It is not clear whether the cell lineexpressed AP-1 to a significant level. Furthermore, the growth conditions of the cells may affect the promoter binding of trans-actingfactors. Our results showing AP-1 expression-dependent activation ofMGMT in AP-1-deficient F9 cells provide a strong argument for thein vivo function of AP-1 in the MGMT promoter.

Although not involved in PKC activation, OA, a potent inhibitor ofprotein phosphatases 1 and 2A both in vivo and in vitro (24), produceda significant (~3-fold) increase in hMGMT mRNA level in HeLa

cells. This suggests involvement of phosphorylated proteins, possiblyAP-1 factors, in MGMT gene activation. No significant enhancementof TPA- or DAG-mediated increase in MGMT expression was appar

ent upon concomitant OA treatment. These results suggest that bothPKC activation and prolonged protein phosphorylation are involved inup-regulation of the hMGMT promoter.

Combined treatment of cells with theophylline (an inhibitor ofphosphodiesterase), db-cAMP, or H-89 (an inhibitor of cAPK) did notaffect either basal or TPA/DAG-induced hMGMT mRNA/protein

levels. In this context, it is also important to recognize that TPA mayincrease intracellular levels of cAMP sufficiently to mediate a stillunidentified interaction between the PKC and cAPK pathways (48,49). It is interesting to note that, in a different cell type, cAMP did notinduce MGMT mRNA levels; however, H-89, an inhibitor of cAPK,

did. In any event, cAPK does not appear to contribute to MGMTactivation in HeLa S3 cells.

HeLa S3 cells may be ideally suited for these studies involvingAP-1 activation because these cells express various isoforms of PKC.It is known that TPA binds with high affinity to both Ca2+-activated(a, ß,and y; conventional PKC) and Ca2+-insensitive (o, e, 6, TJ,and

fi and novel PKC) PKC isotypes, leading to their activation (24, 25,27, 50). In addition, TPA forms a stable complex with PKC andmediates a persistent activation of PKC that ultimately leads todepletion (down-regulation) of the enzyme (36). That prolonged treat

ment of HeLa S3 cells with TPA (10 ng/ml) eliminated hMGMT

mRNA inducibility by high doses of TPA (100 ng/ml) or DAG (100ng/ml) indicated that up-regulation of hMGMT is dependent on con

ventional and novel PKC isotypes. These data, considered togetherwith the Ca2+-sensitivity of PKC isotypes, may indicate that conven

tional PKCs (a, ß,ßl,and y) are involved in signaling events that leadto AP-1 activation and up-regulation of the MGMT promoter. It is

interesting to note that the a, ß,and y isoenzymes of PKC are oftenfound to be associated with maintenance of various human tumor celltypes (28, 50).

We found a significant increase in cellular resistance to BCNU afterexposure of cells to TPA. This increase in the drug resistance of HeLaS3 cells was reversed by H-7, which inhibits both the holoenzyme and

the catalytic subunit of PKC. These results are consistent with the ideathat AP-1 sequences in the hMGMT promoter, together with PKC-

mediated activation, may enhance the drug resistance of tumor cells.Indeed, biochemical modulation of hMGMT protein activity by use ofspecific inhibitors in Mer+ (methylation repair proficient) cells has

been demonstrated to reverse CNU resistance (51, 52). Thus, pretreatment of Mer^ tumor cells with alkyltransferase substrates (e.g., O6-methyl- or 06-benzylguanine) before exposure to CNUs has demon

strated marked synergism. However, after depletion of hMGMTactivity by acute treatment with these biochemical modulators, thelevel of MGMT activity eventually rebounds, with subsequent reac-

quisition of the resistant phenotype (52, 53). That this restoration ofactivity occurs with no concomitant alterations in hMGMT mRNAsteady-state levels (54) suggests a direct effect on the enzyme activity.

In conclusion, our results establish a functional role for the PKCsignal pathway(s) in regulating hMGMT mRNA levels. Our evidencethat TPA-induced AP-1 activation of MGMT expression is associated

with increased resistance of HeLa S3 cells to the cytotoxicity ofBCNU suggests that MGMT overexpression in certain drug-resistanttumor cells may, in part, reflect transactivation by AP-1. Once this

possibility is established for a specific tumor, it may be possible todesign an adjuvant chemotherapeutic approach for sensitizing thetumor to alkylating drugs with specific inhibitor! s ) of PKC. This is thefirst report showing involvement of PKC/AP-1 in regulation of a

DNA repair gene having impact on cytotoxic drug resistance.

ACKNOWLEDGMENTS

We thank Dr. D. Konkel for critically reading the manuscript and W. Smithfor typing it. We are grateful to Dr. P. Angel (German Cancer Research Center.Heidelberg, Germany) for RSV-c-/o.v and RSV-c-y'n/i expression plasmids.

REFERENCES

1. Mitra, S., and Kaina, B. Regulation of repair of alkylation damage in mammaliangenomes. Prog. Nucleic Acids Res. Mol. Biol., 44: 109-142. 1993.

2. Pegg, A. E. Mammalian O^-alkylguanine-DNA alkyltransferase: regulation and im

portance in response to alkylating carcinogenic and therapeutic agents. Cancer Res.,50:6119-6129, 1990.

3. Erickson. L. C.. Laurent. G.. Sharkey. N. A., and Kohn. K. W. DNA cross-linking andmonoadduct repair in nitrosourea-treated human tumour cells. Nature (Lond.), 288:727-729, 1980.

4. Ludlum. D. B. DNA alkylation by the haloethylnitrosoureas: nature of modificationsproduced and their enzymatic repair or removal. Mutât.Res., 233: 117-126. 1990.

5. Robins, P., Harris, A. L., Goldsmith. 1., and Lindahl. T. Crosslinking of DNA inducedby chloroethylnitrosoureas is prevented by O^-methylguanine-DNA methyltrans-

ferase. Nucleic Acids Res.. //: 7743-7758. 1983.6. Preuss. I., Thust. R., and Kaina. B. Protective effect of O^-melhylguanine-DNA

methyltransferase (MGMT) on the cytotoxic and rccombinogenic activity of differentantineoplastic drugs. Int. J. Cancer, 65: 506-512, 1996.

7. Washington. W. J., Foote. R. S.. Dunn, W. C.. Generoso. W. M.. and Mitra, S.Age-dependent modulation of tissue-specific repair activity for 3-methyladenine and06-methylguanine in DNA in inbred mice. Mech. Aging Dev.. 48: 43-52, 1989.

8. Sagher, D., Karrison, T., Schwartz. J. L.. Larson. R. A., and Strauss. B. Heterogeneityof O6-alkylguanine-DNA alkyltransferase activity in peripheral blood lymphocytes:

relationship between this activity in lymphocytes and in lymphoblastoid lines from

3955

on March 14, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

AP-1 ACTIVATION OF HUMAN MGMT

normal controls and from patients with Hodgkin's disease or non-Hodgkin's lym-

phoma. Cancer Res.. 49: 5339-5344. 1989.

9. Day. R. S.. Ziolkowski. C. H. J.. Seudiero. D. A.. Meyer. S. A.. Lubiniecki, A. S.,Ciardi. A. J.. Galloway. S. M.. and Bynum. G. D. Detective repair of aikylated DNAby human tumor and SV40-transformcd human cell strains. Nature (Lond.), 288:724-727, 1980.

10. Brent, T. P., Houghton. J. A., and Houghton. P. J. O''-Alkylguanine-DNA alkyltrans-

ferase activity correlates with the therapeutic response of human. Proc. Nati. Acad.Sci. USA, 82: 2985-2989, 1985.

11. K,un.i B.. Fritz, G-, Mitra. S., and Coquerelle, T. Transfection and expression ofhuman (/'-methylguanine-DNA melhyltransferasc (MGMT) cDNA in Chinese ham

ster cells: the role of MGMT in protection against the gcnotoxic effects of alkylatingagents. Carcinogenesis (Land.). 12: 1857-1867. 1991.

12. Taño.K., Dunn, W. C., Darroudi. F.. Shiota S.. Preston, R. J.. Natarajan. A. T., andMitra. S. Amplification of the DNA repair gene O^-methylguanine-DNA melhyl-

transferase associated with resistance to aikylated drugs in a mammalian cell line.J. Biol. Chem.. 272.- 13250-13254, 1997.

13. Fritz, G., Taño.K.. Mitra, S., and Kaina. B. Inducibility of the DNA repair geneencoding (/'-methylguanine-DNA methyltransferase in mammalian cells by DNA-

damaging treatments. Mol. Cell. Biol.. //: 4660-4668. 1991.14. Takahashi. S.. Hall. J.. and Montcsano. R. Temporal cell-type-specific mRNA ex

pression of C/'-methylguanine-DNA methyltransferases in liver of rats treated with

dimethylnitrosamine. Am. J. Pathol., 148: 497-507, 1996.15. Fritz, G.. and Kaina. B. Stress factors affecting expression of O6-methylguanine-DNA

methyllransfera.se mRNA in rat hepatoma cells. Biochim. Biophys. Acta. II71:35-40. 1992.

16. Pieper. R. O. Understanding and manipulating i/'-methylguanine-DNA methyltrans

ferase expression. Pharmacol. Ther.. 74: 285-297. 1997.

17. Walts. G. S.. Pieper. R. O.. Costello, J. F.. Peng, Y.. Dalton. W. S.. and Futscher.B. W. Mclhylation of discrete regions of the C/'-methylguanine-DNA methyltrans

ferase (MGMT) gene CpG island is associated with heterochromatinizalion of theMGMT transcription start site and silencing of the gene. Mol. Cell. Biol.. 17:5612-5619, 1997.

18. Palei, S. A., Graunke. D. M., and Pieper. R. O. Aberrant silencing of the CpGisland-containing human gene is associated with the loss of nucleosome like positioning. Mol. Cell. Biol.. 17: 5813-5822. 1997.

19. Harris. L. C.. Potter. P. M., Taño, K.. Shiota. S.. Mitra, S., and Brent. T. P.Characterization of the promoter region of the human O^-methylguanine-DNA meth

yltransferase gene. Nucleic Acids Res.. 19: 6163-6167, 1991.

20. Harris. L. C.. Remack. J. S., and Brent. T. P. Identification of a 59 bp enhancerlocated at the first exon/intron boundary of the human O^-methylguanine DNA

methyltransferase gene. Nucleic Acids Res.. 22: 4614-4619. 1994.

21. Grombacher, T.. Mitra. S., and Kaina. B. Induction of the alkyltransferase (MGMT)gene by DNA damaging agents and the glucocorticoid dexamethasone and comparison with the response of base excision repair genes. Carcinogenesis (Lond.), 17:2329-2336. 1996.

22. Biswas, T., Ramana, C. V.. Srinivasan. G., Boldogh, I., Hazra, T. K.. Chen. Z., Taño,K.. Thompson, E. B., and Mitra. S. Activation of human O^-methylguanine DNA

methyltransferase gene by glucocorticoid hormone. Oncogene, in press. 1998.23. Herrlich, P.. Ponta, H., and Ramsdorf, H. Y. DNA damage-induced gene expression:

signal transduction and relation to growth factor signalling. Rev. Physiol. Biochcm.Pharmacol., 119: 187-212. 1992.

24. Hunter. T. Protein kinascs and phosphatases: the yin and yang of protein phospho-rylation and signalling. Cell. 80: 225-236. 1995.

25. Chen. C. C. Protein kinase C a. S. €¿�and f in C6 glioma cells. TPA inducestranslocation and down-regulation of conventional and new PKC isoform.s but notatypical PKC£.FEBS Lett., 332: 169-173, 1993.

26. Chambers. T. C.. McAvoy. E. M.. Jakobs. J. W.. and Eilon. G. Protein kinase Cphosphorylates P-glycoprotein in multidrug resistant human KB carcinoma cells.J. Biol. Chem.. 265: 7679-7686. 1990.

27. Lee, S. A.. Karaszkiewicz. J. W., and Anderson. W. B. Elevated level of nuclearprotein kinase C in multidrug-resistant MCF-7 human breast carcinoma cells. CancerRes., 52: 3750-3759. 1992.

28. Basu, A. The potential of protein kinase C and A target for anticancer treatment.Pharmacol. Ther., 59: 257-280. 1993.

29. Dphil. G. P. Anticancer drugs acting against signaling pathways. Curr. Opin. Oncol.,7: 554-559, 1995.

30. Chambers, T. C.. Zheng, B., and Kuo. J. F. Regulation by phorbol ester and proteinkinase C inhibitors and by protein phosphatase inhibitor (okadaik acid), of P-glycoprotcin phosphorylation and relationship to drug accumulation in multidrugresistance human KB cells. Mol. Pharmacol.. 41: 1008-1015. 1992.

31. Sambrook. J.. Fritsch. E. F., and Maniatis. T. Molecular Cloning: A LaboratoryManual. Ed. 2. Cold Spring Harbor. NY: Cold Spring Harbor Laboratory. 1989.

32. Taño.K., Shiota, S., Collier. J., Foote, R. S., and Mitra. S. Isolation and structuralcharacterization of a cDNA clone encoding human DNA repair protein for O6-

alkylguanine. Proc. Nati. Acad. Sci. USA. 87: 686-690, 1990.

33. Bowman. L. H.. Rabin, B., and Schlessinger. D. Multiple ribosomal RNA cleavagepathways in mammalian cells. Nucleic Acids Res., 9: 4951-4966. 1981.

34. Deng. W. P.. and Nicholoff. J. A. Site-directed mutagenesis of virtually any plasmidby eliminating a unique site. Anal. Biochem.. 200: 81-88, 1992.

35. Dosch. J., and Kaina. B. Induction of c-fos. c-jun. junB and junD mRNA and AP-1

by alkylaling mutagens in cells deficient and proficient for the DNA repair proteinO''-methylguanine-DNA methyltransferase (MGMT) and its relationship to cell

death, mutation induction and chromosomal instability. Oncogene, 13: 1927-1935,

1996.36. Castagna, M.. Takai, Y., Kaibuchi, K., Sano, K., Kikkawa, U., and Nishizuka, Y.

Direct activation of calcium-activated phospholipid-dependent protein kinase bytumor-promoting phorbol esters. J. Biol. Chem.. 257: 7847-7851. 1982.

37. Nishizuka, Y. The role of protein kinase C in cell surface signal transduction andtumor promotion. Nature (Lond.). 308: 693-698, 1984.

38. Kishimoto. A., Takai, Y., Mori, T.. Kikkawa. U., and Nishizuka, Y. Activation ofcalcium and phospholipid-dependent protein kinase by diacylglycerol, its possiblerelation to phosphatidylinositol turnover. J. Biol. Chem.. 255: 2273-2276, 1980.

39. Hidaka. H.. Inagaki. M.. Kawamoto. S.. and Sasaki. Y. Isoquinolinesulfonamides,novel and potent inhibitors of cyclic nucleolide dependent protein kinase and proteinkinase C. Biochemistry, 2.?: 5036-5041, 1984.

40. Suwanichkul. A., DePaolis. L. A., Lee, P. D., and Powell. D. R. Identification of apromoter element which participates in cAMP-stimulated expression of human insulin-like growth factor-binding protein-1. J. Biol. Chem., 268: 9730-9736, 1993.

41. Findik. D.. Song. O.. Hidaka. H., and Lavin, M. Protein kinase A inhibitors enhanceradiation-induced apoptosis. J. Cell. Biochem.. 57: 12-21, 1995.

42. Harrison, L., Galanopoulos, T.. Ascione, A. G., Antoniades. H. N., and Demple. B.Regulated expression of APE apurinic endonuclease mRNA during wound healing inporcine epidermis. Carcinogenesis (Lond.). 17: 377-381, 1996.

43. Wilson. T. M.. Rivkees, S. A., Deutsch. W. A., and Kelley. M. R. Differentialexpression of the apurinic/apyrimidinic endonuclease (APE/refI ) multifunctionalDNA base excision repair gene during fetal development and in adult rat brain andtestis. Mutât.Res., 362: 237-248. 1996.

44. Xu. Y., Moore. D. H.. Broshears, J., Liu, L. F., Wilson, T. M., and Kelley, M. R. Theapurinic/apyrimidinic endonuclease (APE-1/refl) DNA repair enzyme is elevated inpremalignant and malignant cervical cancer. Anticancer Res.. 17: 3713-3719. 1997.

45. Grombacher. T.. and Kaina. B. Constitutive expression and inducibility of 0ft-

methylguanine-DNA methyltransferase and /V-melhylpurine-DNA glycosylase in rat

liver cells exhibiting different status of differentiation. Biochim. Biophys. Acta. 7270:63-72, 1995.

46. Boldogh, I., Abubakar. S.. and Albrecht, T. Activation of proto-oncogenes: an

immediate early event in human cytomegalovirus infection. Science (WashingtonDC), 247: 561-564. 1990.

47. Kaina. B.. Haas, S., and Kappes, H. A general role for c-Fos in cellular protectionagainst DNA-damaging carcinogens and cytostatic drugs. Cancer Res.. 57: 2721-

2731, 1997.48. Ziegler. A.. Knesel, J., Fabbro. D.. and Nagamine. Y. Protein kinase C down-

regulation enhances cAMP-mediated induction of urokinase-type plasminogen activator mRNA in LLC-PK1 cells. J. Biol. Chem.. 266: 21067-21074. 1991.

49. Huang. D.. Hubbard. C. J.. and Jungmann. R. A. Láclatedehydrogenase A subunitmessenger stability is synergistically regulated via protein kinase A and C signaltransduction pathways. Mol. Endocrinol.. 9: 994-1004. 1995.

50. Levitzki. A. Signal-transduction therapy: a novel approach to disease management.Eur. J. Biochem.. 226: 1-13, 1994.

51. Dolan. M. E.. Mitchell, R. B.. Mummert. C.. Moschel. R. C.. and Pegg. A. E. Effectof Oh-benzylguanine analogs on sensitivity of human tumor cells to the cytotoxic

effects of alkylating agents. Cancer Res.. 5/: 3367-3372, 1991.52. Pieper. R. O.. Futscher. B. W., Dong, Q.. and Erickson. L. C. Effects of streptozo-

tocin/bis-chloroethylnotrosourea combination therapy on Oh-methylguanine DNA

methyltransferase activity and mRNA levels in HT-29 cells in vitro. Cancer Res.. 51 :1581-1585. 1991.

53. Erickson. L. C.. Micetich, K. C., and Fisher. R. I. Preclinical and clinical experienceswith drug combinations designed to inhibit DNA repair enzymes. In: P. Wooley andK. Tew (eds.). Mechanisms of Drug Resistance in Neoplastic Cells, pp. 173-183.

New York: Academic Press, 1988.54. Marathi. U. K. Kroes. R. A.. Dolan. M. E., and Erickson. L. C. Prolonged depletion

of O^-methylguanine-DNA methyltransferase activity following exposure to O6-

benzylguanine with or without streptozotocin enhances 1.3-bis(2-chloroethyl)-l-ni-trosourea sensitivity in viirn. Cancer Res.. 53: 4281-4286. 1993.

3956

on March 14, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

1998;58:3950-3956. Cancer Res   Istvan Boldogh, Chilakamarti V. Ramana, Zhenping Chen, et al.   C-mediated Signaling-Methylguanine-DNA Methyltransferase via Protein Kinase

6ORegulation of Expression of the DNA Repair Gene

  Updated version

  http://cancerres.aacrjournals.org/content/58/17/3950

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/58/17/3950To request permission to re-use all or part of this article, use this link

on March 14, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from