UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

Multidrug resistance: related proteins in primary chorpoidal melanoma and in vitro cell lines

van der Pol, J.P.; Blom, D.J.R.; Flens, M.J.; Luyten, G.P.M.; de Waard-Siebenga, I.;Koornneef, L.; Scheper, R.J.; Jager, M.J.Published in:Investigative Ophthalmology & Visual Science

Link to publication

Citation for published version (APA):van der Pol, J. P., Blom, D. J. R., Flens, M. J., Luyten, G. P. M., de Waard-Siebenga, I., Koornneef, L., Scheper,R. J., & Jager, M. J. (1997). Multidrug resistance: related proteins in primary chorpoidal melanoma and in vitrocell lines. Investigative Ophthalmology & Visual Science, 38, 2523-2530.

General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s),other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, statingyour reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Askthe Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,The Netherlands. You will be contacted as soon as possible.

Download date: 05 Feb 2021

Multidrug Resistance-Related Proteins in PrimaryChoroidal Melanomas and In Vitro Cell Lines

Jeroen P. van der Pol* DerkJ. R. Blom,f Marcel]. Flens,% Ore P. M. Luyten,%Itte de Waard-Siebinga,-f Leo Koornneef* RikJ. Scheper,X and Marline J. Jager-f

Purpose. Metastatic uveal melanoma is strongly resistant to chemotherapy, and multidrugresistance (MDR) may be involved. To investigate the role of MDR, the presence of theMDR-associated proteins P-glycoprotein (Pgp), MRP, and lung resistance protein (LRP) wasdetermined on primary choroidal melanomas and cell lines.

Methods. A panel of primary choroidal melanomas was examined for the presence of MDR-associated proteins by immunohistochemical analysis. In cell lines established from four pri-mary choroidal melanomas and one metastatic choroidal melanoma, die expression of MDR-associated proteins was determined with monoclonal antibodies in cytospin preparations andflow cytometry. In addition, the functional capacities of transporter proteins Pgp and MRPas adenosine triphosphate-driven efflux pumps were determined by measuring the cellularaccumulation and efflux of the fluorescent dyes rhodamine 123 and calcein-AM, with andwithout the presence of specific pump inhibitors PSC833 and probenecid.

Results. Low levels of Pgp and MRP were detected in most primary tumors and in some celllines. Measurable transporter function of Pgp could be determined in cell line OCM-1. Lung-resistance protein was present in all primary tumors and cell lines and showed high expressionlevels.

Conclusions. This study revealed the involvement of LRP and at least a minor role of Pgp andMRP in chemoresistance of choroidal melanoma. Compared with cutaneous melanomas,uveal melanomas appear to express slightly higher levels of Pgp. These findings provideinsights into the drug-resistant phenotype of this disease and can aid in the design of therapeu-tic protocols. Invest Ophthalmol Vis Sci. 1997; 38:2523-2530.

L he prognosis of patients with metastatic choroidalmelanoma is extremely poor. Once the diagnosis ofhepatic metastasis has been made, the median survivaltime is 2 to 7 months.1'2 Immunotherapy and modifiedregimens of drug administration are under investiga-tion, but the resistance of this disease to chemother-apy is one of the major reasons for the current absenceof any effective treatment of metastases.2 Cytotoxic

From the "Department of Ophthalmology, Academic Medical Center, University ofAmsterdam, The Netherlands; ^Department of Ophthalmology, Leiden UniversityHospital, The Netherlands; \Departnient of Pathology, Free University Hospital,Amsterdam, The Nelheiiands; and the ̂ Department of Ophthalmology, UniversityHospital, Rotterdam, The Netherlands.Supported by the Haagsch Oogheelkundig Fonds and Stichting Blinden-Penningand Edmond en Marianne Blauxo Fonds.Submitted for publication February 24, 1997; revised July 14, 1997; accepted July14, 1997.Proprietary intmest category: N.Reprint requests: J. P. van der Pol, Department of Ophthalmology, AcademicMedical Center, University of Amsterdam, PO Box 22700, 1100 DE Amsterdam,The Netherlands.

agents are successful in many malignancies (e.g., tes-ticular cancer, leukemia), but these agents have hadlittle success in metastatic uveal melanoma.3 Tumorcells may be resistant to a specific drug or a group ofdrugs with structural or functional similarities, andthere is increasing evidence that several mechanismsallow tumor cells to develop resistance to a variety ofdrugs that differ in structure and function.'1 This iscalled multidrug resistance (MDR). The most fre-quently described mechanism in MDR involves overex-pression of a 170-kDa plasma membrane protein, P-glycoprotein (Pgp), which is encoded by the humanMDR1 gene.5'6 This protein acts as an adenosine tri-phosphate-driven drug efflux pump and has beenshown to cause "classical MDR," conferring resistanceto anthracyclines, epipodophyllotoxins, actinomycinD, vinca alkaloids, and some alkylating agents.6"8 Thepresence of Pgp in normal tissues such as colon, liver,and kidney and at the blood-brain barrier suggests

Investigative Ophthalmology & Visual Science, November 1997, Vol. 38, No. 12Copyright © Association for Research in Vision and Ophthalmology 2523

2524 Investigative Ophthalmology 8c Visual Science, November 1997, Vol. 38, No. 12

that it has a physiological role in the defense againstxenobiotics.9 Increased expression of Pgp has beendetected in various tumors arising from tissues thatnormally express Pgp, as well as in tumors originatingfrom cells that do not (e.g., myelomas, sarcomas).9 Insome cancers (e.g., childhood malignancies, multiplemyeloma, some adult leukemias), clinically relevantcorrelations have been established between Pgp ex-pression and the results of chemotherapy.10

Several lipophilic compounds have been identi-fied that can reverse the Pgp-mediated resistance tochemotherapy.1112 Reversal compounds (e.g., vera-pamil, cyclosporin) have been used with some successto overcome drug resistance.1314 A second MDR-re-lated protein, MRP, was identified in cell lines showingdrug resistance without expression of Pgp.15"17 LikePgp, MRP belongs to the adenosine triphosphate-binding cassette superfamily of membrane transportproteins and also causes resistance by acting as a mo-lecular efflux pump.1819 The drug-resistance patternconferred by the MRP gene is similar but not identicalto that of Pgp.1719

Recently, a third MDR-associated protein, lung re-sistance protein (LRP), was identified.20 This proteinforms the major component of the newly describedcellular organelles named vaults.21 Vaults are distrib-uted mainly through the cytoplasm, and a few of themare localized at the nuclear membrane and nuclearpore complexes. It has been suggested that vaults playa role in drug resistance by regulating both the nucleo-cytoplasmic transport and the cytoplasmic redistribu-tion of drugs.21"22 Lung resistance protein is a strongpredictor of poor response to chemotherapy andprognosis in acute myeloid leukemia and ovarian can-cer.23'24

To understand the mechanisms involved in thehigh resistance of metastatic choroidal melanoma tochemotherapeutic agents, we investigated the expres-sion of Pgp, MRP, and LRP on sections of choroidalmelanomas and on uveal melanoma cell lines. Thefunctional capacities of transporter proteins were de-termined in functional assays.

MATERIALS AND METHODS

Primary choroidal melanomas were examined forMDR-associated proteins using immunoperoxidasestaining of fresh-frozen tissue. Cell lines isolated fromprimary choroidal melanomas and one metastasisfrom choroidal melanoma were studied using immu-nohistochemistry and flow cytometry. Because the de-gree of immunostaining of antigens that are associatedwith functional proteins does not necessarily representthe number of active transporter proteins (antigenson nonfunctional proteins are stained as well), func-tional assays were conducted for the presence of Pgp

and MRP. True functional capacities of transporterproteins Pgp and MDR as adenosine triphosphate-driven efflux pumps were determined by measuringcellular accumulation and efflux of fluorescent dyeswith and without the presence of specific pump inhibi-tors. For LRP, no inhibitor is known, so no functionalassay was available.

Tissue Specimens

Specimens from primary choroidal melanomas wereobtained from storage banks of the departments ofophthalmology of the university hospitals of Amster-dam and Leiden, The Netherlands. All 12 eyes wereenucleated without prior therapy. Part of the tumorwas snap-frozen in OCT compound (Ames Co., Div.of Miles Laboratories, Elkhart, IN) and stored at—70°C. The remainder of the eye was fixed in formalinand embedded in paraffin or celloidin (Leiden) forhistopathologic analysis. All material used in this studywas obtained from routinely stored specimens and didnot interfere in any way with diagnostic or therapeuticprocedures for patients.

Immunohistochemical Analysis

Cryostat sections of tissue samples were cut 4 fim thick,air dried overnight, and fixed in acetone for 10 min-utes at room temperature. The slides were incubatedfor 20 minutes with normal rabbit serum (1:50; DAKO,Copenhagen, Denmark) followed by monoclonal anti-bodies (C21925 [1:10] andJSB-l,26MRPrl,27MRPm6,27

LRP-56,20 and LMR-5 [all 1:20]) and conjugates di-luted in phosphate-buffered saline and bovine serumalbumin. Isotype-matched negative controls wereused.

Staining was done using an indirect immunoper-oxidase technique with peroxidase-conjugated rabbitantirat or antimouse immunoglobulin G (IgG) (1:25;DAKO). Monoclonal antibody (MAb) binding was de-tected using biotinylated rabbit antimouse IgG (1:150;DAKO) or biotinylated rabbit antirat IgG (1:100;DAKO) and streptavidin conjugated to horseradishperoxidase (1:500; Zymed, San Francisco, CA). Boundperoxidase was developed with 4 mg (wt/vol) 3-amino-9 ethylcarbazole and 0.02% H2O2 in 0.1 M NaAc (pH5.0). The slides were counterstained with hematoxylinand mounted. Immunoreactivity was scored semi-quantitatively in four categories: no, weak, positive,and strongly positive staining. All slides were scoredby two observers. No major discrepancies (>one cate-gory) occurred, and smaller discrepancies were re-solved after parallel examination of the slides.

Melanoma Cell Lines

Five human choroidal melanoma cell lines (92-1,28

OMM-1,29 OCM-1,30 EOM-3,29 and Mel 20231) wereused. All cell lines were established from primary cho-

Multidrug Resistance in Choroidal Melanomas 2525

TABLE i. Staining of Multidrug Resistant-associated Proteins in aPanel of Primary Choroidal Melanomas

Tumor

123456789

101112

Cell Type

EpithelioidSpindleMixedSpindleEpithelioidMixedMixedSpindleSpindleMixedSpindleMixed

LargestDiameter

17ND17.5121525101511132312

P-Glycoprotein

0O©©O©•

oooo

MRP

•©0©••©

• •••

• •

LRP

• •

•©•

• ••

• •••

• •

Staining intensities were scored according to the following scale: O = no staining; © = weakstaining; • = positive staining; • • = strongly positive staining. ND = not done, could not bedetermined; LRP = lung resistance protein.

roidal melanomas except OMM-1, which originatedfrom a choroidal melanoma metastasis. Cell lines 92-1, EOM-3, and Mel 202 were used at approximately60 passages and cell lines OMM-1 and OCM-1 at ap-proximately 50 passages. In vitro doubling time of cellline EOM-3 was approximately 96 hours, cell lines 92-1 and Mel 202 doubled in 54 hours, and OMM-1 andOCM-1 were the fastest-growing cells, with doublingtimes of 50 hours.

Cell line 92-1 was established in the laboratory ofthe Department of Ophthalmology, Leiden UniversityHospital, The Netherlands; cell lines OMM-1 andEOM-3 came from the Department of Ophthalmol-ogy, Erasmus University, Rotterdam, The Netherlands.Cell line OCM-1 was a generous gift from J. Kan-Mitch-ell (University of California at San Diego). Cell lineMel 202 was kindly provided by B. Ksander (SchepensEye Research Institute, Harvard Medical School, Bos-ton, MA).

Flow Cytometric Analysis

In flow cytometry analysis, cells were permeabilizedwith a 10% lysing solution (Solution G; Becton Dick-inson, San Jose, CA) in distilled water for 15 minutesand incubated for 15 minutes at 4°C in phosphate-buffered saline-bovine serum albumin containing1% normal goat serum for measurements with MAbsC219, MRPrl, MRPm6, LRP-56, and LMR-5 (all stainintracellular antigens). Cell lines were not permeabil-ized for determination of Pgp with MAb MRK-16(stains extracellular epitopes of Pgp).32 Cells (105)were then incubated for 1 hour at 4°C in 500 fi\phosphate-buffered saline-bovine serum albumincontaining one of the Pgp, MRP, or LRP MAbs, or rator mouse isotype-matched control MAbs. Antibodybinding was detected with fluorescein isothiocya-

nate-labeled goat antirat or antimouse immunoglob-ulins (1:100; DAKO). Fluorescence was analyzed ona FACScan II (Becton Dickinson). Results are ex-pressed as the ratio of the two mean fluorescencesof specifically stained cells and controls. Controlswere cell line GLC4, a drug-sensitive small cell lungcancer cell line expressing low levels of MRP andLRP, and the corresponding subline GLC4/ADR,which overexpresses MRP and LRP and is stronglydrug resistant.33 For analysis of Pgp expression, celllines KB3-1 (low Pgp expression; sensitive to doxoru-bicin) and KB8-5 (overexpression of Pgp; twofold re-sistance to doxorubicin) served as controls.34

Immunocytochemical AnalysisCytospin preparations of tumor cell lines were air-dried and fixed for 10 minutes in acetone. The slideswere incubated for 1 hour with MAb diluted in phos-phate-buffered saline containing 1% bovine serum al-bumin. 4E335 (Signet, Dedham, MA) was diluted 1:10and JSB-1, MRPrl, MRPm6, LRP-56, and LMR-5 wereused at dilutions of 1:20. Isotype-matched negativecontrols were used. Monoclonal antibody binding wasdetected using biotinylated rabbit antirat IgG (1:100;DAKO) or antimouse IgG (1:150; DAKO) or strep-tavidin-conjugated horseradish peroxidase (1:500;Zymed). Bound peroxidase was developed with 4 mg(wt/vol) amino-ethyl-carbazole and 0.02% H2O2 in 0.1M NaAc (pH 5.0), counterstained with hematoxylin,and mounted with Aquamount (Gurr, BDH Chemi-cals, Poole, UK). Immunoreactivity was determined asin immunohistochemical analysis.

Functional Assay of Cellular DrugAccumulation and EffluxThe transporter function of Pgp-associated proteinsin cell lines was determined by flow cytometry, using

2526 Investigative Ophthalmology & Visual Science, November 1997, Vol. 38, No. 12

rhodamine 123 (200 ng/ml"1; Sigma, St. Louis, MO)and calcein-AM (0.5 (iM; Molecular Probes, Eugene,OR) as substrates and PSC833 (2 [M; Sandoz, Basel,Switzerland) as an inhibitor of Pgp-mediated trans-port. For MRP, calcein-AM and probenecid (0.5 raM)served as substrate and inhibitor, respectively. P-glyco-protein and MRP functional assays were conductedas described by Feller et al.36 Briefly, tumor cells (fromthe cell lines described already) were loaded with calc-ein-AM in medium A (growth medium without phenolred and bicarbonate buffer but with 20 raM Hepes)plus 10% fetal calf serum. After incubation, cells werewashed in ice-cold medium A plus 10% fetal calf serumand incubated in drug-free medium plus 10% fetalcalf serum, with and without the relevant modulator.The efflux was stopped by centrifuging the cells andadding ice-cold medium. Fluorescence was analyzedon a FACScan II. Results are expressed as the ratio ofthe two mean fluorescences (with and without modu-lator) .

RESULTS

Immunohistochemical Analysis of PrimaryChoroidal Melanomas

The frozen sections of 12 primary choroidal melano-mas were stained with two MAbs against different epi-topes on each protein: Pgp (MAbsJSB-1 and C219),MRP (MAbs MRPm6 and MRPrl), and LRP (MAbsLRP-56 and LMR-5). JSB-1 showed slightly higherstaining intensities, with higher background stainingthan C219 in each specimen; otherwise, no discrepan-cies occurred. Results of semiquantitative scoring (Ta-ble 1) are average scores of two antibodies per protein.P-glycoprotein was expressed weakly on the mem-branes of 4 tumors, and only 1 of 12 lesions showedsignificant staining. In two tumors, blood vesselsstained positively with C219 and JSB-1.

Expression of MRP on tumors was higher: positive

or strongly positive staining was observed on 7 of 12tumors, and only 1 tumor showed no immunoreactiv-ity. Staining was predominantly cytoplasmic, but mem-brane staining was also observed in strongly MRP-posi-tive tumors (Fig. 1). Immunoreactivity of LRP was thestrongest of all MDR-associated proteins and occurredon all lesions, with 9 of 12 tumors scoring positive orstrongly positive. No correlations were found betweenstaining results and cell type or tumor size. Most tu-mors showed considerable heterogeneity of proteinexpression. These tumors were mostly rated "weakstaining" or "positive staining" in Table 1. Lesionsrated "strongly positive staining" showed more uni-form staining. No relation could be determined withnecrotic regions, vascularization, or pigmentation.

Detection of MDR-Associated Proteins on CellLines in CytospinsThe presence of the MDR-associated proteins was de-termined in five uveal melanoma cell lines. Stainingwas more homogeneous in cell lines than in tumorsections. Results of immunocytochemical analysis per-formed on cell lines are shown in Table 2 as averagescores of staining results of two antibodies per cellline. On cytospin preparations, antibodies (4E3 andJSB-1) to Pgp showed little staining; Pgp was weaklydetectable in three of five cell lines (92-1, OCM-1,OMM-1). MRP staining was observed on all cell lines,with a higher reactivity on EOM-3 than on the othercell lines. As in immunohistochemical analysis of thesolid tumors, LRP expression was high: four cell linesshowed high immunoreactivity, and only one cell line(Mel 202) showed weak reactivity. Staining of cytos-pins with LRP-56 showed a cytoplasmic granular pat-tern (Fig. 2), in accordance with reports of the cyto-plasmic distribution of vaults.21'22

Detection of MDR-Associated Proteins on CellLines in Flow Cytometric AnalysisResults of flow cytometric analyses are shown in Table2. Monoclonal antibody MRK-16 was used to detect

FIGURE l. Immunohistochemical detection of multidrug resistance-associated proteins infrozen sections of primary choroidal melanomas. Positive antigens are stained red. Cyto-plasmic staining of lung resistance protein (LRP, top left) and multidrug resistance-associ-ated protein (MRP, top right). There was no immunoreactivity to P-glycoprotein (Pgp) inthe same specimen (bottom left). Negative (isotype-matched antibody) control is shown atbottom right. (Magnification, X200.)

FIGURE 2. Detection of multidrug resistance-associated proteins on cell line EOM-3 in cytos-pin preparations. Positive antigens are stained red. Strong immunoreactivity of lung resis-tance protein (LRP) with a granular cytoplasmic pattern was observed (top left). Staining ofmultidrug resistance-associated protein (MRP) was predominantly cytoplasmic, with somestaining of cell membranes (top right). No immunoreactivity of antibodies to P-glycoproteinwas observed in EOM-3 (bottom left). Negative (isotype-matched antibody) control is shownat bottom right. (Magnification, X400.)

Multidrug Resistance in Choroidal Melanomas 2527

2528 Investigative Ophthalmology & Visual Science, November 1997, Vol. 38, No. 12

TABLE 2. Cell Lines: Results of Immunocytochemical and Flow Cytometric Measurements

Cell Line

92-1MEL202EOM-3OCM-1OMM-1SensitivityResistance

P-Glycoprotein

Cyto-spin

90O

©

o••

Flow Cytometry

1.171.221.142.522.541.00*

13.30f

Cyto-spin

QQ•

• •

MRP

.F/orw Cytometry

1.421.791.601.001.322.90J

15.3§

Cyto-spin

•9

• •• •9O•

LRP

Flow Cytometry

2.051.75

15.904.802.203.75|7.95§

Staining intensities of tumor cells in cytospin samples were scored according to the following scale: O = no staining; © = weak staining;• = positive staining; • • = strongly positive staining. Flow cytometry = ratios of mean fluorescences of specifically stained cells andnegative controls. For interpretation of the results, ratios of known sensitive (limited expression of multidrug resistant [MDR]-associatedproteins) and drug-resistant cell lines (high MDR-associated protein expression) are shown.* Cell line KB3-1, sensitive to doxyrubicin and limited expression of P-glycoprotein.t Cell line KB8-5 overexpresses P-glycoprotein and is twofold resistant to doxyrubicin.% Cell line GLC4 has limited expression of MRP and lung resistance protein (LRP).§ Cell line GLC4/ADR overexpresses MRP and LRP and is strongly drug-resistant.

the presence of Pgp on cell lines. Low levels of Pgpon cell lines OCM-1 and OMM-1 were observed, butthe other cell lines showed no significant staining.MRP could not be detected on cell line OCM-1, andonly weak immunoreactivity was demonstrated by flowcytometry in the other cell lines. In contrast, highlevels of LRP were demonstrated on flow cytometricanalyses. In cell line EOM-3, fluorescence ratios ofLRP markers were twice as high as in the drug-resistantcell line GLC4/ADR, which was used as a positive con-trol. Flow cytometry scatterplots showed variation inthe size of tumor cells.

Detection of MDR-Associated Proteins on CellLines in Functional Assay

Some specific inhibition of efflux of rhodamine 123and calcium-AM in the presence of PSC833 was de-tected in the functional assay of OCM-1, demonstra-ting a measurable transporter function of Pgp. Nosignificant function of MRP could be detected in dyeefflux measurements. Results are shown in Table 3.

DISCUSSION

Metastatic uveal melanoma shows a notably poor re-sponse to chemotherapy,2 and the mechanisms con-ferring drug resistance to tumor cells in this diseaseare unknown. The discovery of MDR-associated trans-port proteins in other tumor cells has brought pre-dictors of patient survival and has led to new possibilit-ies for patient stratification, choice of drugs, and theaddition of drug transport modulators to therapeuticprotocols.10"14 These exciting studies demonstrate theclinical relevance of determining the specific drug-resistance phenotype of each different tumor. Expres-sion of Pgp on choroidal melanomas was previously

described in a minority of tumors, in accordance withour results.37 We report here for the first time on theexpression of MRP and LRP on primary choroidalmelanomas and cell lines. Our results show that thephenotype of uveal melanoma varies with regard tothe expression of MDR-related proteins in type andlevel of expression. In general, LRP was very stronglyexpressed, Pgp and MRP weakly. Lung resistance pro-tein was easily detected by all methods used; immuno-reactivity of Pgp and MRP was strongest in immunocy-tochemical and immunohistochemical analyses.

Results obtained from in vitro cultured cell linesmust be viewed critically because the phenotype of thetumor cells can change with the increasing number ofpassages. Therefore, results from sections of primarymelanomas are probably more meaningful. In thisstudy, the protein expression found in cell lines wascomparable to that found in tumor sections.

The detection of these proteins in functionalassays was not so evident; this could be expected be-cause this technique is less sensitive than immunocyto-chemical analyses.3638 In a functional assay, some spe-cific inhibition of efflux of rhodamine 123 and cal-cium-AM by adding PSC833 was demonstrated inOCM-1, showing that on some uveal melanomas, Pgphas a measurable function as an adenosine triphos-phate-driven pump. Compared with cutaneous mela-nomas, uveal melanomas appear to express slightlyhigher levels of Pgp.39 Because in clinical studies ofother tumors low levels of Pgp expression have dem-onstrated a significant influence on drug resistanceand survival, this finding may be of clinical interest.14

For confirmation of such low levels of Pgp expression,however, determination of messenger RNA in tumorcells using the reverse transcription-polymerasechain reaction may be warranted.38

Multidrug Resistance in Choroidal Melanomas 2529

TABLE 3. Cell Lines: Results of FunctionalAssays for P-Glycoprotein and MRP

P-Glycoprotein

Cell Line Rho Ca-AMMRP(Ca-AM)

92-1MEL202EOM-3OCM-1OMM-1

1.121.011.011.511.13

1.111.011.091.271.15

1.141.121.041.061.18

Functional assay of P-glycoprotein: results (ratios of meanfluorescence) of two functional assays are given with rhodamine(Rho) and Calcein (Ca-AM) as substrates with and withoutPSC833 as inhibitor of pump function. Functional assay of MRP:results of Ca-AM measurements inhibited by probenicid.

The present finding of high levels of expressionof LRP in uveal melanoma (cell lines) suggests a rolefor LRP in the poor response of uveal melanoma tochemotherapy. LRP has been identified as the majorhuman vault protein. Vaults are vesicle-associated, 13-MDa ribonucleoprotein particles composed largely ofthe (104-kDa) major vault protein, in addition to three(55-, 195-, and 210-kDa) minor vault proteins and vaultRNA. These cellular organelles, discovered only 10years ago, were found to be broadly distributed andhighly conserved among diverse eukaryotic cells. Vaultparticles are thought to mediate transmembranetransport of various substrates, possibly in a fashionsimilar to that of proteosomes.21'22 Thus, vault organ-elles may mediate drug resistance by redistributingdrugs away from the nucleus and extruding the drugsfrom the cell through exocytotic vesicles.

In this study, the intrinsic expression of MDR-related proteins varied greatly among primary uvealmelanomas and also between different cell lines. Thisfinding suggests a variation in drug resistance in thesetumors and may in part explain incidental remissionsby drug therapy in some patients. Patients may bestratified according to the presence or absence ofMDR-related proteins on tumor cells. For clinical pur-poses, however, several issues must be resolved. Thelevels of expression of MDR-related proteins arestrongly stimulated by exposure to cytotoxic drugs.This acquired resistance could be a significant factorin the clinical resistance of metastatic choroidal mela-noma. Further, the phenotype of metastases may dif-fer from that of the primary choroidal tumors de-scribed here.

This study indicates that different MDR-relatedproteins are intrinsically present on most primary cho-roidal melanomas, in accordance with known clinicalresistance to cytostatic agents. However, our knowl-edge of drug resistance in metastatic choroidal mela-noma is still limited, and additional studies are war-

ranted for better understanding of the high resistanceof this disease to chemotherapy.

Acknowledgements

The authors thank H.J. Broxterman and P. van der Valkfor their assistance and advice.

Key Words

cell lines, immunohistochemistry, melanoma, multidrug re-sistance, uvea

References

1. Seddon JM, Albert DM, Lavin PT, Robinson N. A prog-nostic factor study of disease-free interval and survivalfollowing enucleation of disseminated uveal mela-noma. Arch Ophthalmol. 1983; 101:1894-1899.

2. Kath R, HayungsJ, Bornfeld N, Sauerwein W, HoffkenK, Seeber S. Prognosis and treatment of disseminateduveal melanoma. Cancer. 1993;72(7):2219-2223.

3. De Vita VT Jr. Principles of chemotherapy. In: DeVita VT, Hellman S, Rosenberg SA, eds. Principles andPractice of Oncology. Philadelphia: JB Lippincott;1989:276-300.

4. Gottesman MM. How cancer cells evade chemother-apy. Cancer Res. 1993; 53:747-754.

5. Kartner N, Evernden-Porelle D, Bradley G, Ling V.Detection of P-glycoprotein in multidrug-resistantcelllines by monoclonal antibodies. Nature. 1985; 316:820

6. Ueda K, Cornwell MM, Gottesman MM, et al. Themdrl gene, responsible for multidrug resistance, codesfor P-glycoprotein. Biochem Biophys Res Comm.1986; 141:956.

7. Willingham MC, Richert ND, Cornwell MM, et al. Im-munocytochemical localization of PI 70 at the plasmamembrane of multidrug-resistant human cells./Histo-chem Cytochem. 1987;35:1451.

8. Gottesman MM, Pastan I. Biochemistry of multidrugresistance mediated by the multidrug transporter. AnnRev Biochem. 1993;62:385-427.

9. Cordon-Cardo C, O'Brien JP, BocciaJ, Casals D, Ber-tino JR, Melamed MR. Expression of the multidrugresistance gene product (P-glycoprotein) in humannormal and tumor tissues, f Histol Cytol. 1990; 38:1277.

10. Rischin D, Ling V. Multidrug resistance in leukemia.In: Freireich EJ, Kantarnan H, eds. Leukemia: Advances,Research and Treatment. Amsterdam: Kluwer AcademicPublishers; 1993:269-293.

11. Georges E, Sharom FJ, Ling V. Multidrug resistanceand chemosensitization: Therapeutic implications forcancer chemotherapy. Adv Pharmacol. 1990;21:185-220.

12. Ford JM, Hait WN. Pharmacology of drugs that altermultidrug resistance in cancer. Phamiacol Rev.1990; 42:155-199.

13. Dalton WS, Grogan TM, Meltzer PS, et al. Drug resis-tance in multiple myeloma and non-Hodgkin'slymphoma: Detection of P-glycoprotein and potentialcircumvention by addition of verapamil to chemother-apy. JClin Oncol. 1989;7:415-424.

14. Sonneveld P, Dune BGM, Lokhorst HM, et al. Modu-

2530 Investigative Ophthalmology & Visual Science, November 1997, Vol. 38, No. 12

lation of multidrug-resistant multiple myeloma bycyclosporin. Lancet. 1992; 340:255-259.

15. Slovack ML, Pelkey HJ, Bhardwaj G, Kurz EU, DeeleyRG, Cole SPS. Localization of a novel multidrug resis-tance-associated gene in the HT1080/DR4 andH69AR human tumor cell lines. Cancer Res. 1993;53:3221-3225.

16. Zaman GLR, Versantvoort CHM, SmitJJM, et al. Anal-ysis of the expression of MRP, the gene for a newputative transmembrane drag transporter, in humanmultidrug-resistant lung cancer cell lines. Cancer Res.1993;53:1747-175O.

17. Grant CE, Valdimarsson G, Hipfner DR, Almquist KC,Cole SPC, Deeley RG. Overexpression of multidrugresistance-associated protein (MRP) increases resis-tance to natural product drugs. Cancer Res. 1994;54:357-361.

18. Childs S, Ling V. The MDR superfamily of genes andits biological implications. In DeVita VT, Hellman S,Rosenberg SA, eds. Important Advances in Oncology.Philadelphia: JB Lippincott; 1994:21-36.

19. Zaman GLR, Flens MJ, van Leusden MR, et al. Thehuman multidrug resistance-associated protein(MRP) is a plasma membrane drug efflux pump. ProcNatlAcad Sci USA. 1994;91:8822-8826.

20. Scheper RJ, Broxterman HJ, Scheffer GL, et al. Over-expression of a 110-kD vesicular protein in non-P-gly-coprotein-mediated multidrug resistance. Cancer Res.1993;53:1475-1479.

21. Scheffer GL, Wijngaard PLJ, Flens MJ, et al. The drugresistance-related protein LRP is a major vault protein.Nature Med. 1995; 1:578-582.

22. Scheper RJ, Scheffer GL, Flens MJ, et al. Role of LRP-major vault protein in multidrug resistance. In: GuptaS, Tsuruo T, eds. Multidrug Resistance in Cancer Cells:Cellular, Biochemical, Molecular and Biological Aspects.New York: John Wiley & Sons; 1996:109-119.

23. List A, Spier CS, Grogan TM, et al. Overexpressionof the major vault transporter protein LRP predictstreatment outcome in AML. Blood. 1995; 87:2464-2469.

24. Izquierdo MA, van der Zee AGJ, Vermorken JB, et al.Drug resistance-associated marker LRP for predictionof response to chemotherapy and prognoses in ad-vanced ovarian carcinoma. / Natl Cancer Inst.1995;87:1230-1237.

25. Georges E, Bradley G, Gariepy J, Ling V. Detection ofP-glycoprotein isoforms by gene-specific monoclonalantibodies. Proc Natl Acad Sci USA. 1990;87:152-156.

26. Scheper RJ, Bulte JWM, Brakkee JGM, et al. Mono-

clonal antibody JSB-1 detects a highly conserved epi-tope on the P-glycoprotein associated with multidrugresistance. Int] Cancer. 1988;42:389-394.

27. Flens MJ, Izquierdo MA, Scheffer GL, et al. Immuno-chemical detection of MRP in human multidrug-resis-tant tumor cells by monoclonal antibodies. Cancer Res.1994; 4:4557-4563.

28. Waard-Siebinga I de, Blom DJR, Griffioen M, et al.Establishment and characterization of an uveal mela-noma cell line. Int J Cancer. 1995; 62:155-162.

29. Luyten GPM, Naus NC, Mooy CM, et al. Establishmentand characterization of primary and metastatic uvealmelanoma cell lines. Int J Cancer. 1996;66:l-8.

30. Kan-Mitchell J, Mitchell MS, Rao N, Liggett PE. Char-acterization of uveal melanoma cell lines that growas xenografts in rabbit eyes. Invest Ophthabnol Vis Sci.1989; 30:829-834.

31. Ksander BR, Rubsamen PE, Olsen KR, Cousins SW,StreileinJW. Studies of tumor-infiltrating lymphocytesfrom a human choroidal melanoma. Invest OphthabnolVis Sci. 1991;32:3198-3208.

32. Hamada H, Tsuruo T. Functional role for the 170- to180-kDa glycoprotein specific to drug-resistant tumorcells revealed by monoclonal antibodies. Proc NatlAcad Sci USA. 1986;83:7758-7789.

33. Zijlstra JG, de Vries EGE, Mulder NH. Multifactorialdrug resistance in an adriamycin-resistant humansmall cell lung carcinoma cell line. Cancer Res.1987;47:178O-1784.

34. Shen DW, Fojo A, Chin JE, et al. Human multidrug-resistant cell lines: Increased mdrl expression can pre-cede gene amplification. Science. 1986; 232:643-645.

35. Arceci RJ, Stieglitz K, BrasJ, Schinkel A, Baas F, CroopJ. Monoclonal antibody to an external epitope of thehuman mdrl P-glycoprotein. Cancer Res. 1993;53:310—317.

36. Feller N, Kuiper CM, Lankelma J, et al. Functionaldetection of MDR1/P170 and MRP/P190-mediatedmultidrug resistance in tumor cells by flow cytometry.Br] Cancer. 1995; 72:543-549.

37. McNamara M, Clyne M, Dunne B, et al. Multidrugresistance in ocular melanoma. Br J Ophthabnol. 1996;80:1009-1012.

38. Broxterman HJ, Lankelma J, Pinedo HM. How toprobe clinical tumor samples for P-glycoprotein andmultidrug resistance protein. Eur J Cancer. 1996;32A1024-1033.

39. Schadendorf D, Makki A, Stahr C, et al. Membranetransport proteins associated with drug resistance ex-pressed in human melanoma. Am J Pathol. 1995;147:1545-1552.