[CANCER RESEARCH55, 5203-5206, November 15, 19951 Advances in Brief

The Differential Expression of 18 in Cisplatin-sensitive and -resistant Human Ovarian Adenocarcinoma Cells and Its Association with Drug Sensitivity

Hemant K. Parekh and Henry Simpkins'

Department of Pathology, Temple University School of Medicine and Fels Institute of Molecular Biology and Cancer Research, Philadelphia, Pennsylvania 19140

Abstract interaction to form an 8—10-nmfilament (12). Intermediate filaments, along with microtubular pmteins, play a key role in the maintenance of DNA is the primary target ofcis-diamminedichloroplatinum (cisplatin), cell shape and in the spatial organization of the cellular organelles. They but the drug also interacts with the cellular composed of are also thought to be involved in the modulation of membrane transport microtabules and intermediate filaments. It was found that the cisplatin (13). Thus, changes in content/structure could p0- resistant 2008/C13* cell line contained markedly lower levels (6-fold) of cytokeratin 18, when compared to the cisplatin-sensitlve 2008 cell line. tentially alter the cellular cytoskeleton and affect a multitude of cellular Northern blot analysis revealed a markedly decreased level of cytokeratin events, including drug transport. 18 mRNA in the resistant cell line. Southern blot analysis of the DNA Christen et al. (14) have shown that the ultrastructure of the microtu extracted from the two cell lines and then digested with Hpall and its bules was altered in a cisplatin-resistant human ovarian adenocarcinoma methylatlon-sensitive isoschizomer, MspI, revealed no detectable differ cell line (2008/C13*) as compared to the parental cisplatin-sensitive 2008 ences in the methylation status of the cytokeratin . Neither 5-azacy cell line, and decreased expression of a membrane-associated, (3- tidine (5 pM)nor retinoic acid (1 pM)treatment enhanced the expression fraction was observed in the cisplatin-resistant cells. Interestingly, using of cytokeratin 18 in the resistant cell line. However, transfection of full a radiolabeled analogue of cisplatin, this resistant cell line (2008/C13*) length cytokeratin 18 cDNA into the cisplatin-resistant 2008/C13* cells was shown to have a defect in the initial uptake rate, but over a period of resulted in clones with increased levels of cytokeratin 18, which was accompanied in the majority of clones by a marked increase in their 24-48 h, the cisplatin aceumulation was found to be similar in 2008 and sensitivity to cisplatin. These results demonstrate that modulating the 2008/C13* cells (2). Also, cisplatin accumulation studies performed in expression of an intermediate filament results in sensitization of a our laboratory using atomic absorption spectroscopy after treatment of drug-resistant human ovarian cell line to cisplatin. 2008 and 2008/C13* cells with 100 @Mcisplatmfor 4 h revealed no significant differences between these two cell lines.3 Since intermediate Introduction filaments exist in close juxtaposition with the , an investi gation as to the role of intermediate filaments in the development of Cisplatin2 is an anticancer drug used widely for the treatment of cisplatin-resistance in human ovarian carcinoma cells (2008 and 2780) various cancers. However, the development of resistance to its cyto and their cisplatin-resistant variants (2008/C13* and C70, respectively) toxic effect is a major problem in its use. Experiments using tumor was undertaken. cell lines sensitive and resistant to cisplatin indicate that resistance acquired to the drug is due to one or more factors acting alone or in Materials and Methods conjunction with each other. These factors include decreased intra cellular drug accumulation, increased drug metabolization or macti Materials. Cisplatin was purchased from Aldrich (Milwaukee, WI). vation via conjugation to glutathione or metallothionein, increased a-[32P]dCl'P (3000 Ci/mol) was purchased from NEN-Dupont Research Prod DNA repair, and/or the activation of certain oncogenes (1—3). ucts (Boston, MA). The multiprime DNA labeling system was obtained from Cisplatin binds to the N7 position of guanosine and forms intrastrand the Amersham Corp. (Arlington Heights, IL). The Cam 5.2 and the CK4 mouse and interstrand cross-links. In addition to its direct binding to DNA, the monoclonal antibodies were purchased from Becton Dickinson (San Jose, CA) and Oncogene Science (Manhasset, NY), respectively. A mouse monoclonal other cellular effects of cisplatin that have been reported include disrup antibody to was purchased from Boehringer Mannheim (Indianapolis, tion of the mitochondrial membrane potential, depolymerization of the IN5).The alkaline phosphatase conjugated antimouse rabbit monoclonal anti microtubules, and collapse of the intermediate filament network (4—7). body was obtained from Jackson Immunoresearch Laboratories, Inc. (West Recently, decreased expression of cytokeratin 14 was reported in a Grove, PA). The full-length cytokeratin 18 cDNA (15) was kindly provided by cisplatin-resistant variant of a human lung squamous cell carcinoma line Dr. R. Oshima (La Jolla Cancer Center, La Jolla, CA). The eukaryotic (8). Since intermediate filament bind to cisplatin-damaged DNA expression vector pCDNA3 was purchased from Invitrogen (San Diego, CA). (9—11),the possibility that they may play a role in the cytotoxic action of Cell Culture Conditions. The parental(2008 and 2780) humanovarian cisplatin cannot be ignored. carcinoma cells and their cisplatin-resistant variants (2008/C13 and C70, are intermediate filament proteins that are subdivided into respectively) used in this study were maintained in RPM! medium as described type I (acidic) and type II (basic) (12). One member of type I previously (16). preferentially binds with one member of type II via a hydrophobic Cytotoxicity Studies. The sensitivity of 2008 and 2008/C13* cells to the cytotoxic effect of cisplatin was assessed using the tetrazolium dye 3-[4,5- dimethylthiazol-2-yl]-2,5 diphenyltetrazolium bromide (16). Received 6/19/95; accepted 10/5/95. Protein Determinations. The protein content of the cytoskeletal fractions The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with was determined by the Coomassie Brilliant Blue dye-binding assay (17) using 18 U.S.C. Section 1734 solely to indicate this fact. the commercially available Bio-Rad Protein Assay reagent and BSA as the , To whom requests for reprints should be addressed, at Department of Pathology, standard. Temple University School of Medicine, Room 106 OMS, 3400 North Broad Street, Philadelphia, PA 19140. Phone: (215) 707-4353; Fax: (215) 707-2781.

2 The abbreviation used is: cisplatin, cis-diamminedichloroplatinum. 3 Unpublished observations. 5203

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1995 American Association for Cancer Research. CYTOKERATIN EXPRESSION AND CISPLATIN SENS@VITY Intermediate Filament Protein Immunoblots. Cells in the exponential A i 2 3 4 phase of growth were harvestedand washedwith a 10 ms@phosphatebuffer (pH 7.4) containing 120 mMNaa and 2.7 mMKC1.The cell pelletswere suspended in 20 ms@Tris(pH 7.4), 0.6 MKU, 1% (v/v) Triton X-100, and 1 inst MgCI2in the presenceof 0.5 mMphenylmethylsulfonylfluorideand 1000units of DNase I 97 for 5 mm at roomtemperature.The lysatewas thencentrifugedat 13,000X g, and the pellet(cytoskeletalfraction)was solubiized in 8 Murea and 03% (v/v)NP4O. 67 Fifty @gprotein from the solubilized cytoskeletal fraction were mixed with

SDS-PAGEbufferandheatedat100°Cfor15min.Theproteinsampleswerethen — 55 kD* separated on a 10% polyacrylamidegel and electrophoreticallytransferredonto 45 •piF — 45 kDa polyvinylidene difluoride membranes using a Bio-Rad semidry blotter. immuno staining was performed as described earlier (16), using a 1:100 dilution of each of the intermediatefilamentantibodies.Densitometricanalysiswas performedusing a LKB laserscanningdensitometerto assessthe differencein the color intensityof 31 the immunostainedbands. RNA Extraction and Northern Blotting. Total cellular RNA was cx traded from exponentially growing cells and fractionated through a 1% aga 21 rose gel containing 2.2 Mformaldehyde (18). RNA was transferred to Nytran 14 Plus nylon membranes (18), and Northern blot analysis was performed using a full-length cytokeratin 18 cDNA as the probe (18). Equal loading of RNA 4 was ascertained by ethidium bromide staining of the agarose gels. The devel B 1 23 oped autoradiogram was subjected to densitometric analysis on a LKB laser scanning densitometer to assess the differences in the amount of the cyto 18 mRNA between 2008 and 2008/C13* cells. DNA Extraction, Restriction Endonuclease Digestion, and Southern Blotting. High molecular weight genomic DNA was isolated from exponen — 45 kDa tially growing 2008 and 2008/C13 cells using phenol-chloroform extraction and ethanol precipitation. The DNA (10 @&g)wasthen digested overnight at 37°C using 30 units each of HpaII and its methylation-sensitive isoschizomer 31 MspI. The digested DNA was subjected to electrophoresis in a 0.8% (w/v) agarose/1x Tris-Borate-EDTA gel at a constant voltage of 1 V/cm. After depurination and denaturation, the separated DNA was transferred onto Nytran 21 membranes, and Southern blot analysis was performed as described (18) using the full-length cytokeratin 18 cDNA as a probe. 14 Plasmid Construction and Transfection. The full-length cytokeratin 18 cDNA was cloned into the EcoRI site of the eukaryotic expression vector, pCDNA3. Orientation was checked with restriction enzyme digestion. The C 1 2 plasmid carrying the full-length cytokeratin 18 in the right orientation termed 288— pCDK18 was used for the transfection of the 2008/C135 cells. Subconfluently growing cells were transfected with pCDNA3 (10 @.tg/m1)orpCDK18 (10 S p@g/ml)using the calcium phosphate-DNA coprecipitation method. Cells were propagated in a medium containing geneticin (500 @Wml;0418sulfate) for 3 188— weeks. Individual 0418-resistant colonies were picked, grown, and screened for expression of the cytokeratin 18 protein using the Cam 5.2 mouse mono Fig. 1. Western blot analysis of the expression of cytokeratins 8 and 18 using the Cam clonal antibody as described earlier. For cisplatin chemosensitivity studies, 18 5.2 (A) or the CK4 (B) mouse monoclonal antibody. The cytoskeletal protein fractions (50 p.g) were loaded on a 10% polyacrylamide gel and subjected to electrophoresis. The clones from the 2008/C13* cells transfected with the pCDK18 vector and proteins were then transferred onto a polyvinylidene difluoride membrane and immuno exhibiting cytokeratin 18 expression were analyzed. One clone of 2008/C13* stained with a 1:100 dilution of the antibodies. The Cam 5.2 antibody recognizes the Mr cells transfected with the pCDNA3 vector alone (C13*/pCDNA3) was used as 55,000 (55 Wa) cytokeratin 8 as well as the M, 45,000 (45 kDa) cytokeratin 8, whereas a control in these studies. the CK4 antibody is specific for the Mr 45,000 (45 k0@)cytokeratin 18 protein. Lane 1, 2008; Lane 2. 2008/C13; Lane 3, 2780; Lane 4, C70. C, Northem blot analysis was Statistical Analysis. The regression analysis and paired t test were per performed to detect the presence of cytokeratin 18 mRNA in the 2008 (Lane 1) and the formed using SigmaStat Statistical Analysis System, version 1.01. 2008/C13 (Lane 2) cells using a full-length cytokeratin 18 cDNA as probe. Total RNA was extracted, and 30 i@gwere fractionated in a 1% (w/v) agarose gel under denaturing Results and Discussion conditions. RNA was transferred to the nylon membranes and probed with a randomly primed 32P-labeled full-length cytokeratin 18 cDNA. Equal loading of RNA was ascer tamed by an ethidium bromide-stained RNA gel. The intracellular levels of the cytoskeletal proteins, vimentin and cytokeratin, were determined by Western blot analysis. High, but identical, vimentin levels were observed in the 2780 and C70 cells; low and similar (Fig. 1A). The mRNA level of cytokeratin 18 was however, only very low levels of vimentin were detected in the 2008 assayed by Northern blot analysis using a full-length human cytok and 2008/C13* cells (data not presented). In contrast, expression of eratin 18 cDNA as a probe. A 5-fold lower cytokeratin 18 mRNA the simple epithelial cytokeratins (cytokeratins 8 and 18) was ob level was observed in the 2008/C13* cells as compared to the 2008 served in the 2008 and 2008/C13* cells (Fig. 1). The level of cyto cells (Fig. 1C), results which correlate with the Western blot data. (M@55,000), asjudged by immunostaining with the Cam 5.2 Thus, a relationship apparently exists between decreased expression of antibody, were similar in the 2008 and 2008/C13* cells (Fig. IA), but cytokeratin 18 and resistance to cisplatin. A similar association between those of cytokeratin 18 (Mr 45,000) were 6-fold lower in the cisplatin reduced expression of cytokeratin 14 and cisplatin resistance was re resistant cells (2008/C13*) when compared to the cisplatin-sensitive ported recently in a human lung squamous cell carcinoma cell line (8). cells (2008; Fig. 1A). Similar results were observed with the cyto Interestingly, the cisplatin-resistant lung cells accumulated less intracel -specific antibody, CK4 (Fig. 1B). Interestingly, the levels lular cisplatin then the sensitive cells. Recent studies with the 2008/C13* of cytokeratin 8 and cytokeratin 18 in the 2780 and C70 cells were cells report alterations in the ultrastructural morphology of the microtu 5204

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1995 American Association for Cancer Research. CYFOKERATIN ExPRESSION AND CISPLATIN se@smvrr@

Table 1 Effect of treatment with 5-azacytidine and retinoic acid on cisplatin probe may not be specific enough to detect subtle differences in the cytotoxicityThe methylation status of this region, especially since, of the 15—20genes human ovarian carcinoma cells sensitive (2008) and resistant (2008/C13*) to cisplatin were treated with 5-azacytidine (5 @xsi)orretinoic acid (1 SM), either 24 h prior similar to cytokeratin 18 present in the , only one may to exposure to cisplatin or simultaneously. Cell survival was determined by the tetrazo be active (15). 5-Azacytidine inhibits the DNA methyltransferase hum blue dye assay. IC5@swere calculated from a regression analysis of the survival enzyme, and the resultant DNA hypomethylation induces gene cx curves and correspond to the cisplatin concentration that produces 50% inhibition of cell survival. IC50s are expressed ±SD and are the mean of three individual experiments pression (21). Treatment with 5-azacytidine has been shown to induce performed in triplicate.IC50 the expression of cytokeratin 8 in mesenchymal cells and Endo B (the p.MDrugs mouse form of human cytokeratin 18) in myoblast cells (21, 22), although the addition of 5-azacytidine to an embryonic teratocarci 2008/C13Cisplatin 2008 noma cell line did not result in the induction of Endo B (15). Retinoic 35-Azacytidinealone 1.8 ±0.5 22 ± acid treatment of epithelial cells also modulates the expression of 75-Azacytidine(24-h pretreatment) + cisplatin 1.1 ±0.7 16 ± various cytokeratins, including cytokeratins 8 and 18 (23, 24). How 6Retinoic + cisplatin (simultaneously) 1.9 ±0.8 20 ± ever, treatment of the 2008 and 2008/C13* cells with 5-azacytidine (5 3Retinoic acid (24-h pretreatment) + cisplatin 1.2 ±0.8 18 ± p.M) and retinoic acid (1 ,.LM) did not affect the expression of cyto acid + cisplatin (simultaneously) 1.8 ±0.5 18 ±2 keratin 18 (data not presented). Furthermore, a simultaneous exposure with cisplatin or a 24-h pretreatment of 2008 and 2008/C13* cells with 5-azacytidine (5 p@M)andretinoic acid (1 @M)hadno effect on bules, a decreased expression of membrane bound (3-tubulin, and an cisplatin cytotoxicity (Table 1). altered mitochondrial function (4, 14). Since microtubules are closely In an attempt to determine if increasing the expression of cyto associated with the intermediate filaments, it is conceivable that a change keratin 18 in the 2008/C13* cells could lead to increased sensitivity to in one would affect the expression and/or function of the other. Further cisplatin, 2008/C13* cells were transfected with a full-length cyto more, drugs that alter mitochondrial function can also disrupt the struc keratin 18 cDNA cloned in the eukaryotic expression vector, pCDNA3. tore of the intermediate filaments (19). Twenty-six clones were picked to assess the relationship between expres Cytokeratins 8 and 18 are expressed in simple epithelia (12), and sion of cytokeratin 18 and cisplatin sensitivity. Increased cytokeratin their expression is regulated by the differentiation state of the cell and expression was observed in 18 clones (which represent 70% of the total possibly DNA methylation in the promoter region of the gene (15, number of clones). A Western blot of nine of these clones is presented in 20).KuleshandOshima(15)havedemonstratedthatthe5'-flanking Fig. 2. Although the pCDNA3 vector leads to a high level of transcrip region of the human cytokeratin 18 gene is G/C rich; thus, it is tion, the modest increase observed in cytokeratin 18 expression is prob possible that methylation of this region could be involved in its ably due to rapid turnover of this protein. Similar results have been transcriptional regulation. However, when DNA was isolated from the observed in cytokeratin 18-transfected F9 cells and L-cdll fibroblasts 2008 and 2008/C13* cells and subjected to restriction endonuclease (15). Forced expression of one type of cytokeratin has also been shown digestion with HpaII and its methylation-sensitive isoschizomer MspI, to induce expression of the other type (25, 26), but the observed increase followed by Southern blot analysis with a full-length cytokeratin 18 in cytokeratin 18 expression in these human ovarian carcinoma cells did cDNA probe, no gross alterations in the methylation pattern were not result in an increased expression of cytokeratin 8. The cisplatin observed (data not presented). Using a RNA probe corresponding to sensitivity of the transfected clones (C13*/pCK1@C13*/pCK18) was 461 nucleotides from the 5' end of the murine cytokeratin 18 gene, assessed. A significant decrease in the IC@ of cisplatin in 11 of the 18 Kulesh and Oshima have demonstrated that it is possible to study transfectedclones was observed (Fable 2). Of particular importancewas methylation status of the cytokeratin 18 gene (20). However, in this the observation that in 4 of the 18 clones, the IC@ was decreased to study we have used a full-length cytokeratin 18 cDNA as a probe, and 70-80% of the value observed for the 2008/C13* cells, and in 7 more due to the presence of multiple HpaII/MspI sites in the 5' end of the clones, the IC50 was reduced to 40—60% of that observed with the gene (10 sites in the murine cytokeratin 18 gene), a full-length cDNA resistant cell line (Table 2). In the remaining seven transfected clones,

1 2 3 4 5 6 7 8 9 10 11 12 kDa

97

67

@ . _ —@ — .@ — cytokeratin 8 45 @@@@@ .@. ,@lø•I• “ --—-- — cytokeratin 18

31

21

Fig. 2. A representative Western blot of the expression of cytokeratin 18 in 2008, 2008/C13, 2008/C13 mock-transfected with pCDNA3 alone, and full-length cytokeratin 18 cDNA transfected 2008/C13 cells. Cytoskeletal fractions were isolated as before and subjected to immunostaining after electrophoresis and transfer, using the Cam 5.2 antibody. Lane 1, 2008; Lane 2, 2008/C13*; Lane 3, C13@/pCDNA3; Lane 4, C13/pCK1; Lane 5, C13@/pCK3; Lane 6, C13/pCK5; Lane 7, C13/pCK7; Lane 8, C13/pCK9; Lane 9. C13*/pCK11; Lane 10, C13@/pCK13; Lane 11. C13*/pCK15; Lane 12, C13/pCK17. 5205

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1995 American Association for Cancer Research. CYTOKERATIN EXPRESSION AND CISPLATIN sENsmvrI-Y

Table 2 Cisplatinproteinafter sensitivity of 2008/C13* clones expressing cytokeratin 18 Hamilton (Fox Chase Cancer Center, Philadelphia, PA) for providing us with cDNADrug transfection with full.length cytokeratin 18 the human cisplatin-sensitive and -resistant ovarian carcinoma cells. the“Materialssensitivity was determined by the tetrazolium blue dye assay as described in ofthe and Methods―section.IC@swere calculated from linear regression analysis cell survival data. P < 0.05 anddifferent.IC50 lower was defined as being significantly References p.M % decrease 1. Scanlon, K. J., Kashani-Sabet, M., Tone, T., and Funato, T. Cisplatin resistance in Cells cisplatin P― in IC50― human cancers. Pharmacol. & Ther., 52: 385—406,1991. 2008 1.8±0.5 2. Gately, D. P., and Howell, S. B. Cellular accumulation of the anticancer agent 2008/C13 22±3 cisplatin (Review). Br. J. Cancer, 67: 1171—1176,1993. C13@/pCDNA3 21±2 NSC 5 3. Kelland, L. R. The molecular basis of cisplatin sensitivity/resistance. Eur. J. Cancer, Cytokeratin 18 cDNA transfected clones displaying significant reduction in cisplatin 30.4: 725-727, 1994. ICso 4. Andrews, P. A., and Albright, K. D. Mitochondrial defects in cis-diamminedichlo C13*/pCK15 4.7 ±1.3 <0.001 79 roplatinum (11)-resistant human ovarian carcinoma cells. Cancer Res., 52: 1895—1901, C13/pCKlO 5.6 ±0.6 <0.001 75 1992. C13/pCK9 6.6 ±0.7 <0.001 70 5. Boekelheide, K., Arcila, M. E., and Eveleth, J. cis-Diamminedichloroplatinum (ll) C13/pCK16 6.4 ±1.8 0.002 71 (cisplatin) alters assembly dynamics. Toxicol. AppI. Pharmacol., 116: Cytokeratin 18 cDNA transfected clones displaying intermediate reduction in cisplatin 146—151,1992. IC50 6. Kopf-Maier, P., and Muhlhausen, S. K. Changes in the cytoskeleton pattern of tumor C13/pCK14 9.2 ±0.9 0.002 58 cells by cisplatin in vitro. Chem.-Biol. Interact., 82: 295—316,1992. C13*/pCK13 9.7 ±1.5 0.003 56 7. Evans, R. M., and Simpkins, H. Cisplatin induced intermediate filament reorganiza C13*/pCKll 10.7±0.8 0.003 51 tion and altered mitochondrial function in 3D cells and drug sensitive and resistant C13*/pCK3 11.1 ±2 0.006 50 Walker 256 cells. Exp. Cell Res., in press, 1995. C13*/pCK7 11.6 ±1.4 0.006 47 8. Katabami, M., Fujita, H., Honke, K., Makita, A., Akita, H., Miyamoto, H., C13/pCK2 12.1 ±1.3 0.006 45 Kawakami, Y., and Kuzumaki, N. Marked reduction of (K14) in C13*/pCKI7 13.1±3.2 0.025 40 cisplatin-resistant human lung squamous-carcinoma cell lines. Biochem. Pharmacol., a p values were determined by the t test. 45: 1703—1710,1993. @ b decrease in IC50 = 100 —(IC50 transfected clone/IC50 2008/C13* X 100). 9. Banjar, Z. M., Hnilica, L. S., Briggs, R. C., Stein, 1., and Stein, 0. cis- and CNS,notsignificant. trans-diamminedichloroplatinum(II)-mediated cross-linking of chromosomal DNA non-histone proteins to DNA in HeLa cells. Biochemistry, 23: 1921—1926,1984. 10. Ward, W. S., Schmidt, W. N., Schmidt, C. A., and Hnilica, L. S. Association of despite an increase in cytokeratin 18 expression, the IC50 was only cytokeratin p39 with DNA in intact Novikoff hepatoma cells. Proc. Natl. Acad. Sci. 20—40%of that observed with the cisplatin-resistant 2008/C13* cells USA, 81: 419—423, 1984. (data not shown). Different turnover rates of the newly synthesized 11. Wedrychowski, A., Schmidt, W. N., Ward, W. S., and Hnilica, L. S. Cross-linking of cytokeratins to DNA in vivo by chromium salt and cis-diamminedichloroplatinum(II). cytokeratin 18 in the different clones and the possibility that the newly Biochemistry, 25: 1—9,1986. synthesized cytokeratin 18 forms intermediate filaments with differing 12. Moll, R., Franke, W. W., Schiller, D. L., Geiger, B., and Krepier, R. The catalog of degrees of efficiency in the individualclones may explain why not all of human cytokeratins: pattems of expression in normal epithelial, tumors and cultured cells. Cell, 31: 11—24,1982. the clones exhibit a significantly increased sensitivity to cisplatin. 13. Mills, J. W., and Mandel, L. J. Cytoskeletal regulation of membrane transport events. The major cellular lesion responsible for cisplatin toxicity is thought to FASEBJ., 8: 1161—1165,1994. be the generation of DNA intrastrand cross-links, although interstrand 14. Christen, R. D., Jekunen, A. P., Jones, J. A., Thiebaut, F., Shalinsky, D. R., and Howell, S. B. In vitro modulation of cisplatin accumulation in human ovarian carcinoma cells by and protein-DNA cross-links have also been observed. Nonhistone pro pharmacologicalalterationof micmtubules.J. Clin. Invest.,92: 431—440,1993. teins were found to be cross-linked to DNA in cells or isolated nuclei 15. Kulesh, D. A., and Oshima, R. G. Cloning of human keratin 18 gene and its treated with cisplatin, and these proteins were later identffied as cytok expression in nonepithelial mouse cells. Mol. Cell. Biol., 8: 1540—1550,1988. 16. Parekh, H., and Simpkins, H. Cross-resistance and collateral sensitivity to natural eratins (9—11). Several other studies also support the notion ofan intimate product drugs in cisplatin-sensitive and -resistant rat lymphoma and human ovarian association of intermediate filaments with the cell nucleus (13). Thus, it carcinoma cells. Cancer Chemother. Pharmacol., in press, 1995. is plausible that the formation of cytokeratin-DNA cross-links is associ 17. Bradford, M. M. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem., abedwith cisplatin cytotoxicity and that the reduced expression of cytok 72: 248—254,1976. eratin 18 in the 2008/C13* cells can lead to a reduced formation of 18. Sambrook, J., Fritsch, E. F., and Maniatis, T. In: C. Nolan (ed), Molecular Cloning: protein-DNA cross-links, and thus decreased sensitivity to cisplatin. A Laboratory Manual, Vol. 1—3,pp.7.3—9.58.Plainview,NY: Cold Spring Harbor Laboratory Press, 1989. The association of cytokeratins and anticancer drug toxicity is not 19. Maro, B., and Borens, M. Reorganization of HeLa cell cytoskeleton induced by an limited to cisplatin. Dalton and colleagues (27, 28) have demonstrated a uncoupler of oxidative phosphorylation. Nature (Land.), 295: 334—336, 1982. binding of mitoxantrone to cytokeratins. Furthermore, a mouse fibroblast 20. Oshima, R. G., Trevor, K., Shevinsky, L. H., Ryder, 0. A., and Cecena, G. Identi fication of the gene coding for the Endo B murine cytokeratin and its methylated, cell line transfected with and expressing cytokeratins 8 and 18 was found stable inactive state in mouse nonepithelial cells. & Dcv., 2: 505—516,1988. to be multidrug resistant, although the transfected cells did not exhibit 21. Darmon, M., Nicolas, i-F., and Lamblin, D. 5-Azacytidine is able to induce the decreased intracellular drug accumulation (29). The authors suggested conversion ofteratocarcinoma derived mesenchymal cells into epithelial cells. EMBO J., 3: 961—967,1984. that the cytokeratin network may influence the intracellular drug distri 22. Darmon, M. Coexpression of specific acid and basic cytokeratins in teratocarcinoma bution in such a manner that the nuclear targets are spared, resulting in a derived fibroblasts treated with 5-azacytidine. Dcv. Biol., I 10: 47—52,1985. resistant phenotype. It has been also demonstrated that, compared to the 23. Stellmach, V., Leask, A., and Fuchs, E. Retinoid-mediated transcriptional regulation of keratin genes in human epidermal and squamous cell carcinoma cells. Proc. NatI. cisplatin-sensitive 2008 cells, the cisplatin-resistant 2008/C13* cells Acad. Sci. USA, 88: 4582—4586,1991. (which express a 6-fold lower level of cytokeratin 18) are hypersensitive 24. Crowe, D. L. Retinoic acid mediates post-transcriptional regulation of to the cytotoxic effects of taxol, vincristine, and vinblastine (16). This mRNA levels. I. Cell Sci., 106: 183—188,1993. 25. Guidice, G. J., and Fuchs, E. The transfection of epidermal keratin genes into correlates, in an inverse manner, with the observation of Bauman et a!. fibroblasts and simple epithelial cells: evidence for inducing a type I keratin by a type (29,) who showed that mouse fibroblast cells expressing higher levels of II gene. Cell, 48: 453—463,1987. cytokeratins 8 and 18 become relatively resistant to the mitotic spindle 26. Domenjoud, L., Jorcano, J. L., Breuer, B., and Alonso, A. Synthesis and fate of keratins 8 and 18 in nonepithelial cells transfected with cDNA. Exp. Cell Res., I 79: poisons, taxol, and the anthracyclines. The identification of cytokeratins 352—361,1988. as proteins that have the potential to modulate the effects of anticancer 27. Cress, A. E., Roberts,R. A., Bowden, G. T., and Dalton, W. S. Modificationof keratin drugs opens a new area of investigation. by the chemotherapeutic drug mitoxantrone. Biochem. PharmacoL, 37: 3043-3046, 1988. 28. Roberts, R. A., Cress, A. E., and Dalton, W. S. Persistent intracellular binding of mitoxantrone in a human colon carcinoma cell line. Biochem. Pharmacol., 38: Acknowledgments 4283—4290,1989. 29. Bauman, P. A., Dalton, W. S., Anderson, J. M., and Cress, A. E. Expression of We thank Dr. R. Oshima for providing the cytokeratin 18 cDNA. We also cytokeratin confers multiple drug resistance. Proc. Natl. Acad. Sci. USA, 91: 5311— thank Dr. Paul Andrews (Georgetown University, Washington, DC) and Dr. T. 5314, 1994. 5206

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1995 American Association for Cancer Research. The Differential Expression of Cytokeratin 18 in Cisplatin-sensitive and -resistant Human Ovarian Adenocarcinoma Cells and Its Association with Drug Sensitivity

Hemant K. Parekh and Henry Simpkins

Cancer Res 1995;55:5203-5206.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/55/22/5203

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

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/55/22/5203. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1995 American Association for Cancer Research.