(CANCER RESEARCH 48, 5597-5603, October I, 1988|

Monitoring of Interaction Products of cis-Diamminedichloroplatinum(II) andci5-Diammine(l,l-cyclobutanedicarboxylato)platinum(H) with DNA in Cellsfrom Platinum-treated Cancer Patients1

Philippe M. A. B. Terheggen, Robert Dijkman, Adrian C. Begg, Ria Dubbelman, Ben G. J. Floot,Augustinus A. M. Hart, and Leo den Engelse2

Divisions of Chemical Carcinogenesis [P. M. A. B. T., R. DL, B. G. J. F., L. d. E.], Experimental Therapy [A. C. B.J, and Clinical Oncology-¡A.A. M. H.J. and ClinicalResearch Unit [R. Du.], The Netherlands Cancer Institute (Antoni van Leeuwenhoek Huis), Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands

ABSTRACT

The formation and stability of interaction products between the anti-cancer drug c/Vdiamminedichloroplatinum(II) (c/s-DDP) and DNA werestudied in buccal epithelial and urinary cells from ten cancer patientswho received c/î-DDP-basedtherapy. Buccal cells were collected l hbefore and 1-2 h after i.v. infusions with cw-DDP. The interactionproducts were visualized in an immunocytochemical peroxidase assay,using an antiserum against c/s-DDP-modified calf thymus DNA. Thenuclear staining density was measured by microdensitometry. Nuclearstaining densities in buccal cells after infusions of =20 mg/m2 c/s-DDP

were always higher than pretreatment values. Repeated sampling fromindividual patients treated for 2-5 consecutive days with daily doses of20-70 mg/m2 cis-DDP indicated that </v DDI'-DN A binding in buccalcells increased in proportion to the cumulative total dose of c/s-DDP.The variation in dose-density response between patients was 17%. Apparent adduct loss in buccal cells from four patients, as measured 8-17days after the last infusion, amounted to 67-86%. Platinum-inducedDNA modifications could also be detected in buccal cells from two m-diammine(l,l-cyclobutanedicarboxylato)platinum(II)-treated patients.In vitro experiments with human buccal cells and lymphocytes indicatedlinear relationships between DNA modification and either cis-DDP concentration or incubation time. Nuclear staining densities in pretreatmentbuccal cells from ten cancer patients treated in vitro with 33 *iMcis-DDPfor l h revealed that interpatient variation in in vitro DNA modificationby cis-DDP was low. No quantitative correlation was found between insitu and in vitro DNA modification.

INTRODUCTION

The drug m-DDP3 is well known for its efficacy against

several clinical malignancies, but the mechanisms underlyingthe cytotoxicity of this and other platinum coordination compounds have not yet been elucidated. Evidence suggesting thatthe cytostatic effects result from platinum-DNA interactionshas been reviewed (1). Interaction products with DNA includemonoadducts and cross-links with either protein, glutathione,or DNA (1-7). Recently, immunochemical assays became available to study m-DDP-induced DNA modifications in isolatedDNA, using antibodies against normal nucleosides (8), platin-ated DNA (9, 10), or m-DDP-modified nucleotides (11).

Poirier et al. (12) and Reed et al. (13, 14) reported a correlation between platinum-DNA binding, measured by enzyme-linked immunosorbent assay, in leukocytes from platinum-treated patients and clinical response, although many discrepancies were seen. Fichtinger-Schepman et al. (15, 16) showed

Received 3/7/88; revised 6/29/88; accepted 7/5/88.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 inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

' This work was supported by Grant NKI86-11 from the Koningin Wilhelmina

Fonds, The Netherlands Cancer Foundation.2To whom requests for reprints should be addressed.3The abbreviations used are: c/s-DDP, m-diamminedichloroplatinum(II);

CBDCA, r»-diammine(l,l-cyclobutanedicarboxylato)platinum(II); FCS, fetalcalf serum; CT-DNA. calf thymus DNA; GAR, goat anti-rabbit immunoglobulin;NRS, normal rabbit serum: PAP, peroxidase-(rabbit)antiperoxidase complex;PBS, phosphate-buffered saline; c.v., coefficient of variation.

the formation of ministrami cross-links at predominantlypGpG and pApG sequences in DNA of human leukocytes afterm-DDP-based chemotherapy. It seems unlikely, however, thatthe kinetics of adduct formation in leukocytes and solid tumorswill be the same. Therefore, it will be worthwhile to study cis-DDP-induced DNA modifications in the tumor cells themselves, e.g., in needle biopsies obtainable early during therapy.As a first step we investigated whether it would be feasible toquantify m-DDP-DNA binding in some easily obtainable human cell types. In previous papers we have shown that immunocytochemical methods can be used to localize m-DDP-in-duced DNA modifications in tissues, including tumors, fromm-DDP-treated rats and mice (17), and in cultured cells afterin vitro exposure to m-DDP ( 18). Here we report the visualization and quantification of m-DDP-DNA binding at the single-cell level in buccal epithelial and exfoliated urinary bladdercells from m-DDP-treated cancer patients and in normal buccalcells and lymphocytes after in vitro exposure to cis-DDP. Buccalcells were chosen because they could be collected from patientsundergoing chemotherapy and from healthy individuals withoutany inconvenience or risk. The results show that, when therapyconsisted of repeated daily m-DDP infusions, m-DDP-DNAbinding in buccal cells accumulated linearly with dose.

MATERIALS AND METHODS

Patients and Drugs. The cancer patients studied were all nursed inthe Netherlands Cancer Institute (Antoni van Leeuwenhoek Huis) anddiffered in sex, tumor type, stage, and chemotherapeutic treatment.They were treated for cancer of the testis (A-F), ovary (G-I), endome-trium (J), bladder (K, male), or stomach (L, male). With the exceptionof patient H, who received m-DDP (Platino!; Bristol-Myers, Weesp,The Netherlands) 1 year before this study, none of the patients had anyprior chemotherapy.

For patients A-D and F the treatment consisted of daily infusionsof 20 mg/m2 c/s-DDP at days 1-5 of a 21-day cycle; this cycle wasrepeated 2 (F) or 3 (A-D) times. Patient E received infusions of 50mg/m2 c/s-DDP at days 1 and 2 of a 10-day cycle; this cycle wasrepeated 3 times. Patient G received an infusion of 75 mg/m2 cw-DDPat day 1 of a 21-day cycle; this cycle was repeated 3 times. Two patients(H, I) were treated with an infusion of 350 mg/m2 (H) or 800 mg/m2(I) CBDCA (Carboplatin; Bristol-Myers). Patients J, K, and L receivedinfusions of 50 mg/m2 (J), 70 mg/m2 (K), or 40 mg/m2 (L) c/s-DDP at

days 1 and 8 (L), days 1 and 21 (J), or days 1 and 28 (K). Theadministration of cis-DDP or CBDCA was combined with bleomycin(A-E), etoposide (A-C, F, and L), vinblastine sulfate (D), vincristinesulfate/Iphosphamide (F), methotrexate (K), doxorubicin hydrochlo-

ride (J, L), or cyclophosphamide (G, H, and J).Collection of Cells. Buccal cells were collected (after informed con

sent) from 11 cancer patients and from 6 healthy volunteers by repeatedly wiping the inner side of the cheek with cotton swabs. The cellswere transferred to cold (4°C)RPMI 1640 (Gibco, Paisley, United

Kingdom). Generally, buccal cells were collected at 3 different timepoints: 1-2 h before the first treatment; 1-2 h after c/s-DDP therapy;and 24 h after CBDCA therapy. Pretreatment samples from 9 of these

5597

Research. on August 21, 2018. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from

PLATINUM-DNA ADDUCTS IN CANCER PATIENTS

patients were split into two: one part was used as negative control,whereas the other was incubated for l h at 37°Cin RPMI 1640

containing 33 MMcw-DDP (positive control). From patients A, D, E,and G, buccal cells were also collected at additional time points duringtherapy. Urine was collected over a 24-h period prior to (patients Eand K) or immediately after (patients, E, F, J, and K) therapy. Exfoliated urinary bladder cells were collected by spinning 200-400 ml urinefor 10 min at 200 x g. Whole blood from healthy volunteers was usedto prepare lymphocytes. This was done by gradient centrifugation onFicoll-Paque (Pharmacia, Uppsala, Sweden) of whole blood (The Netherlands Red Cross Blood Transfusion Service, Amsterdam, The Netherlands). Cytospin slides were made on glass slides coated with eggalbumin [100 n\ 0.5% egg albumin (Sigma Chemical Co., St. Louis,MO) in demineralized water per 2.6 x 6-cm slide, quickly dried on a60°Cmetal plate, and baked overnight at 120°C].Cytospin slides wereimmediately fixed by successive incubations in precooled (-20°C)meth-

anol (10 min) and acetone (2 min) and air-dried. Cytospin slides werewrapped in aluminum foil and stored at -20°C.

In Vitro Experiments with Buccal Cells and Lymphocytes. Buccalcells from healthy volunteers ( l O2-1OVperson) were washed twice inRPMI 1640, incubated for l h with 0-33 MMcis-DDP or for 0-6 hwith 6.7 tiM cis-DDP, and washed twice in PBS. Cytospin slides weremade directly or after postincubation in medium. In the latter experiments, cells were resuspended in RPMI 1640 plus 10% PCS (Sera-Lab,Ltd., Sussex, United Kingdom) and kept for 0-25 h at either 37°Cor4°C.Lymphocytes from healthy volunteers were washed twice in PBSand resuspended in RPMI 1640 plus 10% PCS (37°C;air plus 5% CO2)

containing 0, 67, or 167 MMcis-DDP. After 1 h, the cells were washedtwice in PBS and cytospin slides were prepared.

Antisera. The preparation of the rabbit (NKI-A59) antiserum againstas-DDP-modified CT-DNA has been described previously (17). Preimmune NRS was used as a control. The anti-cw-DDP-DNA antiserumand the NRS were applied without further purification and were usedin an optimal dilution of 1:1800. In a control experiment, the anti-ci's-

DDP-DNA antiserum was preincubated in 3 successive wells coatedwith cw-DDP-modified CT-DNA (drug/nucleotide ratio, 9 x 10~2)forl h at 37°C.Antisera used for the peroxidase staining procedure were:

GAR (against whole IgG; Campro Benelux, Elst, The Netherlands);and PAP (American Qualex, La Miranda, CA). These antisera wereused in the following dilutions: GAR, 1:640; PAP, 1:3200.

Immunocytochemical Assay. The immunostaining procedure was carried out as described earlier (17, 19-21) with some adaptations to singlecell preparations. The general outline of the procedure was as follows.Cytospin slides were treated with methanol-HiOi (to inactivate endogenous peroxidases), RNases A and TI (to remove m-DDP-modifiedRNA), 1 M KC1, proteinase K, ethanol-NaOH (the last 3 steps servedto increase the accessibility of the cw-DDP-DNA adducts to the anti-serum; compare Ref. 22), 10% PCS (to reduce nonspecific antibodybinding), and the anti-m-DDP-DNA antiserum. NK1-A59 antibodiesbound in the last step were visualized by double PAP staining, i.e., bysequential incubation of the sections in GAR, PAP, and GAR andPAP. The color reaction was carried out using the peroxidase substrate3,3'-diaminobenzidine-HCI (Sigma), resulting in a brown precipitate.

Modifications of the previously described protocol (17) were: the introduction of a preincubation for l h at 37°Cin prewarmed 50 mM Tris-

HC1, pH 7.2—1 M KC1 plus 0.3% Triton X-100; introduction ofpreincubation for 4 min in prewarmed (37°C)20 mM Tris-HCl, pH

7.4-5 HIMEDTA containing 20 Mg/ml proteinase K (Merck-Schuchard,Munich, Federal Republic of Germany); 0.042 MNaOH in 80% ethanolfor 2 min at room temperature instead of 0.07 M NaOH in 30%ethanol; preincubation with 10% FCS in PBS containing 0.04% TritonX-100 instead of preincubation in 10% normal goat serum.

Quantification of the Staining Intensity. The staining intensity ofindividual nuclei was measured microdensitometrically with a Knott(Munich, Federal Republic of Germany) light-measuring device (beamdiameter, 0.5 Mm), coupled to a Leitz Orthoplan microscope. Thisscanning cytophotometry equipment was linked to an Atari ST [Atari,Sunnyvale, CA] microcomputer, programmed with an adapted versionof the Histochemical Data Acquisition System [Hidacsys (23), purchased by Microscan, Leiden, The Netherlands). The selected density

(the integrated absorbance of a selected area, i.e., one whole nucleus,expressed in arbitrary units) was determined in 25-30 nuclei/slide.

Statistics. Levels of significance were calculated using Student's t

test. P < 0.05 was considered indicative of a significant differencebetween groups. The existence of linearity between dose and nuclearstaining density in buccal cells from patients was tested with multiplelinear regression analysis, using software from BMDP Statistical Software, Inc. (Los Angeles, CA). The mean slope and interpatient variationin the dose-density curves were calculated with a mixed-model analysisof variance; analysis of interassay and intrasample variation was calculated with a fixed-model analysis of variance.

RESULTS

DNA Modification after in Vitro Exposure to «\-DDP

Controls of the Immunocytochemical Assay. Nuclear stainingwas almost absent from untreated lymphocytes or buccal cells,or from cw-DDP-treated buccal cells (33 ^M cw-DDP for 1 h)when the anti-cw-DDP-DNA (NKI-A59) antiserum was substituted by NRS or had been preabsorbed with c/s-DDP-modifiedCT-DNA. Almost absent means that the selected density wasS10 units, 7 times lower than the nuclear staining density afterincubation of buccal cells with 6.7 jtM cw-DDP for 1 h.

Intrasample Variation. Substantial differences in nuclearstaining density were visible between cells from the same cytospin slide, i.e. after the same treatment and staining procedure.The staining intensity was independent of nuclear size or thelocation of cells on the cytospin slide. The variation was reflected in the microdensitometry data; c.v. values from 20 slidesfrom three in vitro experiments (Figs. 1-3) range between 20and 90% (mean, 41). The observed variation cannot be explained by some kind of a Poisson-type process, because thiswould have led to increasing c.v. values with decreasing meandensities.

Reproducibility. To determine the reproducibility of the im-munocytochemical assay, duplicate cytospin slides from a singlesample of buccal cells, treated for 1 h with cis-DDP (13 UM),were stained in 6 independent assays at different times. Meannuclear densities ranged between 195 and 226 units (mean, 200;c.v., 5.8%), indicating minor interassay variation. This variationfits within the range of the intrasample variation.

Dose and Time Response. Significant nuclear staining couldbe seen in buccal cells after incubation with 13-33 ¿¿Mc/s-DDPfor 1 h, while only minor cytoplasmic staining was observed.Microdensitometry revealed a linear relation between the CM-

CISDOP IpM)

Fig. 1. Nuclear staining density ±SE (hurx) of human buccal cells (•)andhuman lymphocytes (O) treated in vitro with 0-167 »iMci's-DDP (1 h, 37*0, air

plus 5% CO:) immunostained for c/s-DDP-DNA binding. Cells were obtainedfrom healthy volunteers. Each point represents 2 simultaneously stained slideswith IS nuclei each.

5598

Research. on August 21, 2018. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from

PLATINUM-DNA ADDUCTS IN CANCER PATIENTS

3 6

INCUBATIONTIME(hi

Fig. 2. Nuclear staining density ±SE (bars) of human buccal cells fromhealthy volunteers, treated in vitro for 1 to 6 h with 6.7 MMcis-DDP (37'C, airplus 5% COi) and immunostained for c/'j-DDP-DNA binding. Each point repre

sents 2 simultaneously stained slides with IS nuclei each.

37"C

0 10 20

TIME AFTER cisDDP (h)

Fig. 3. Nuclear staining density ±SE (bars) of buccal cells from healthyvolunteers, at different times after in vitro exposure to c/s-DDP. Cells wereincubated with 13 pM cis-DDP for 1 h, washed twice, incubated for 0-25 h at37°C(O) or 4°C(•),and immunostained for c/i-DDP-DNA binding. Each point

represents 2 simultaneously stained slides with 15 nuclei each.

DDP dose and the mean nuclear density (Fig. 1, closed symbols).The nuclear staining density of human lymphocytes incubatedin vitro with 0, 67, or 167 /J.Mc/s-DDP for 1 h is also shown inFig. 1 (open symbols). The staining density of buccal cell nucleiafter continuous exposure to m-DDP (6.7 ßM)for 1-6 h isdepicted in Fig. 2.

Stability of c/s-DDP-induced DNA Modification in BuccalCells. Fig. 3 indicates that c/s-DDP-DNA adducts were stableat 4°Cduring a 25-h period immediately after c/s-DDP exposure (l h, 13 MMc/s-DDP at 37°C).When, however, buccal cellswere kept at 37°C(air plus 5% COz), the nuclear stainingdensity decreased markedly with time after c/s-DDP exposure.The number of cells did not significantly change under theseconditions. The apparent half-life of c/s-DDP-DNA binding inbuccal cells kept at 37°Cwas calculated to be 16 ±2 h.

DNA Modification after in Situ Exposure to c/s-DDP

Buccal Cells from Cancer Patients. Nuclear staining densitiesin buccal cells collected 1 to 2 h after c/s-DDP-based therapy

were always higher than the corresponding pretreatment staining densities; a typical example is illustrated in Fig. 4. Thenuclear staining density proved to be strongly dose and timedependent (Fig. 5). Fig. 5 (top) shows mean nuclear stainingdensities from patient D during the first 26 days after the startof the treatment. During days 1-5, when he received dailyinfusions of 20 mg/m2 c/s-DDP, nuclear densities increased,

reaching values 16 times control level at day 5. At day 16 (11days after the first series of c/s-DDP infusions), the meannuclear density was reduced to 5 times control level. At day 22,when nuclear staining density was only twice the control level,a new cycle of five daily infusions was initiated. At the end ofeach of these c/s-DDP infusions, the nuclear staining densityhad increased markedly. Similar patterns of c/s-DDP-inducedDNA modification were observed in patients E and G (Fig. 5).In all three patients (D, E, and G) the nuclear staining densityin buccal cells increased during each treatment cycle and decreased again when the administration of c/s-DDP was stopped.There was no indication for a nonlinear dose-density relationship (calculated by multiple linear regression analysis). Measurements during the treatment cycle, as performed in patientsD and E, showed a linear increase of the nuclear staining densitywith dose. The nuclear density found 8-17 days after treatmentranged between 14 and 33% of the density immediately afterthe last treatment and was always significantly higher (as shownby multiple linear regression analysis) than pretreatment valuesof the same patient. Fig. 5 also indicates that the rate of adduciaccumulation was very similar during successive treatments.Multiple linear regression analysis confirmed this observationbut showed significant differences in the rate of adduci accumulation between patienl D and Ihe two olher palienls. Whenall available dala [from 9 palienls (Fig. 6)] were combined, ilwas also found lhal Ihe nuclear slaining densily increasedlinearly wilh c/s-DDP dose wilhin each of Ihe ireatmenl cycles.Fig. 6 includes only cumulative c/s-DDP doses which wereoblained in a single Irealment cycle; the lolal c/s-DDP doseadministered lo a palienl could be appreciably higher (forinstance, 300 mg/m2 in palienl E). The mean slope in Fig. 6 is0.87 unil/mg/m2 wilh an inlerpalienl varialion of 0.17.

Intra- and Interpatient Heterogeneity. Varialion of nuclearslaining densily was also seen belween cells from the samecytospin slide; c.v. values from 24 slides ranged between 10 and104% (mean, 46%). There was also a significant variation inbuccal cell mean nuclear density belween c/s-DDP-lrealed palienls, who received Ihe same Irealmenl. Four palienls wilh aleslicular carcinoma and subjected lo an idenlical therapeuticregimen (5 daily infusions of 20 mg/m2 c/s-DDP) were com

pared. When buccal cells from ihese palients, collected at day5, were stained and measured in the same experiment, thenuclear slaining densilies amounled lo 108 ±13 (SE), 51 ±6,103 ±14, and 81 ±4, respeclively. Inlerpalienl helerogeneilyis also expressed by the 20% interpalienl varialion of Ihe meanslope in the dose density curves of Fig. 6.

In Vitro versus in Situ DNA Modification in Buccal Cells.When buccal cells from len cancer patients who had not beenexposed to platinum compounds before (excepl palienl H) wereIrealed in vitro wilh 33 UMc/s-DDP for 1 h and slained for c/s-DDP-induced DNA modifications (see Fig. 7), Ihe mean nuclear slaining density ranged between 234 and 322 units (mean,272; c.v., 38%). Differences belween palienls were not significant (P > 0.05), suggesting that interpalienl varialion wilhregard lo in vitro DNA modification is low. No correlalionappeared to exisl belween the exlenls of in situ and in vitroDNA modificalion.

5599

Research. on August 21, 2018. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from

PLATINUM-DNA ADDUCTS IN CANCER PATIENTS

Fig. 4. Immunocytochemical visualizationof nuclei with c/s-DDP-DNA adducts in buccalcells from a testis cancer patient, treated withtwo daily infusions of 50 mg/m2 c/'j-DDP (patient E; see "Materials and Methods"). /. l hbefore c/s-DDP-based chemotherapy; 2, l hafter infusion of 50 mg/m2 c/'i-DDP at day 1:3, I h after infusion of 50 mg/m2 c/s-DDP atday 2. Bright-field optics; bar, 10 urn.

".-

-

LJJ

a

<t—«/)

ce s

LU

patient D

Hm Hm

H +t H1 patient E

patient G

10 20 30

DAYS DURING THERAPYFig. 5. Nuclear staining density ±SE (bars) of buccal cells from 3 c/s-DDP-

treated patients at different times after initiation of therapy. Cells were isolatedI h before and immediately after <iVDI >!' infusions, fixed within l h after cellisolation, and immunostained for c/i-DDP-DNA binding. Top: patient D received20 mg/m2 cis-DDP on days 1-5 and on days 22-26. Middle: patient E receivedinfusions of 50 mg/m2 a's-DDP on days 1 and 2, days 10 and 11, and days 21and 22. Bottom: patient G received 75 mg/m2 rá-DDP at day 1 and at day 18.

Each point represents 2 independently stained slides with 15 nuclei each. Eacharnm indicates a single infusion of c/j-DDP.

DNA Modification in Exfoliated Urinary Bladder Cells, cis-DDP-induced DNA modifications could also be visualized incells collected from posttreatment urine from patients E and K(Fig. 8). The nuclear staining densities in pretreatment urinarycells from these patients were very' low. Fig. 8 suggests that

there might be a relationship between nuclear staining densityand c/s-DDP dose. The nuclear staining observed in posttreatment samples from patients J and F (no pretreatment sampleswere collected) strengthens this suggestion.

CBDCA-induced DNA Modification

In situ CBDCA-DNA binding was demonstrated in buccalcells from two ovarian carcinoma patients (Fig. 9). The nuclearstaining density after a single infusion of 350 mg/m2 CBDCA

C O E G J KL

cisDDP Img/m?)

Fig. 6. Nuclear staining density of buccal cells from 9 patients (A-E, G, and.11.). after (repeated) administration of m-DDP. Standard immunostaining forci'i-DDP-DNA binding. Each bar on abscissa represents a dose scale from 0-100mg/m2 c/'i-DDP. •,cells isolated during the first treatment cycle; O, patients D,

G, the second cycle; A, patient E. third cycle. No pretreatment cells were availablefrom patient K (ml. not determined). Each point represents 2 independentlystained slides with 15 nuclei each.

1J.11JL1JL,111

E G H I JPATIENTS

L

Fig. 7. Nuclear staining density ±SE (ears) of pretreatment buccal cells from10 patients. The cells were incubated in vitro with 33 JIMcis-DDP for 1 h (37"C,air plus 5% COj) and immunostained for c/s-DDP-DNA binding. Each pointrepresents 2 simultaneously stained slides with 15 nuclei each.

(patient H) was higher than pretreatment nuclear staining density values (patients H and I) and lower than the nuclear stainingdensity after a single infusion of 800 mg/m2 CBDCA (patientI). These results suggest a dose-dependent extent of DNAmodification. Comparison of Figs. 6 and 9 reveals that 14-18times more CBDCA than cw-DDP (on a molar basis) is required

to obtain the same level of nuclear staining.5600

Research. on August 21, 2018. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from

PLATINUM-DNA ADDUCTS IN CANCER PATIENTS

I/)zUJo

«I-

100

eis DOP2n°°21

Fig. 8. Nuclear staining density ±SE (bars) of exfoliated urinary bladder cellsbefore (2 patients) and after (4 patients) treatments with cis-DDP. Cells wereisolated from pooled urine, collected during 24 h prior to or during 24 h followingcis-DDP infusions. Standard immunostaining for c/s-DDP-DNA binding. Eachpoint represents 2 simultaneously stained slides with 15 nuclei each.

Si

ÕH

H.I

0 500 1000

CBDCA (mg/m2)

Fig. 9. Nuclear staining density ±SE (bars) of buccal cells after infusions with350 mg/m2 (patient H) or 800 mg/m2 CBDCA (patient I). Standard immunostaining for Pt-DNA binding, using the NKI-A59 anti-c/i-DDP-DNA antiserum.Each point represents 2 simultaneously stained slides with l S nuclei each.

DISCUSSION

Our recent observation that interaction products of the anti-tumor drugs cis-DDP and CBDCA with DNA in rodent tissuescan be visualized by an immunocytochemical peroxidase technique (17) suggested the feasibility of c/s-DDP-DNA bindinganalysis at the level of the individual cell in cancer patientssubjected to platinum-based chemotherapy. We now report thatquantitative single-cell analysis of DNA modification in cancerpatients exposed to relatively low doses of cis-DDP (S20 mg/m2) or CBDCA (S350 mg/m2) has been achieved. A linear

relationship was shown to exist between the nuclear stainingdensity in buccal cells from cancer patients and the dose of cis-DDP administered. This linear relationship between dose andadduci formation is in agreement with the suggested (24, 25)linear interdependency between the serum platinum concentration in humans and the dose of cis-DDP or CBDCA administered. It is also consistent with the linear relationship betweenthe c/s-DDP concentration in growth medium and the extentof DNA-DNA interstrand cross-link formation observed incultured Chinese hamster ovary cells (26). The sensitivity of thepresent immunocytochemical method for monitoring c/s-DDP-DNA binding is indicated by the increased staining density inbuccal cell nuclei after a single infusion of 20 mg/m2 c/s-DDP,the lowest dose of c/s-DDP applied in this study. When chemotherapy consisted of repeated daily c/s-DDP infusions, each ofthese resulted in an increase in the nuclear staining density inbuccal cells. This indicates that during treatment c/s-DDP-DNA adduci accumulation was faster lhan adduci loss. Obser

vations in three patients showed that the rale of c/s-DDPbinding lo DNA apparently remained unchanged during successive treatment cycles (Fig. 5). Hence, we have no indicationthai buccal cells change Iheir response lo either m-DDP exposure or c/s-DDP-DNA binding within three cycles of chemotherapy (compare Ref. 18). Due to cell proliferation and turnover of buccal cells, the in vivo half-life of c/s-DDP-DNAbinding in these cells cannot be calculated.

The relationship between c/s-DDP dose and nuclear stainingdensity in urinary cells was less obvious than it was in buccalcells; this was due to the lower number of samples. The stainingdensity of urinary cell nuclei was higher than that of buccalcells (patients E and J; see Figs. 6 and 8), probably as a resultof the relatively high concentrations of free c/s-DDP in thehuman urine (27). The morphology of the urinary cells wasrelatively poor; hardly any fully intacl cell could be seen incylospin slides after immunocytochemislry. Urinary cells,therefore, seemed less appropriate than buccal cells for thedetection of c/s-DDP-DNA binding at Ihe single-cell level.

The in vitro data concerning the linearity between drug exposure and microdensitometric values valídaleIhose from cancer pat it-ins. The mean nuclear staining density of in vitro cis-DDP-treated buccal cells or lymphocytes was a linear functionwith dose (Fig. 1), and the rate of DNA modification in buccalcells in vitro remained constant over a 6-h incubation period(Fig. 2). To oblain a certain level of nuclear staining densityafter in vitro treatment, lymphocytes required five limes higherconcenlralions of c/s-DDP lhan buccal cells (see Fig. 1). Ap-parenlly, relatively low levels of c/'s-DDP-DNA adduci are

formed in lymphocytes. This is consistenl wilh an observalionin c/s-DDP-treated rats, where relalively low levels of DNA-DNA inlraslrand cross-links were observed in leukocyles4 in

spile of high plasma levels of c/s-DDP immediately after ad-

minislralion (24, 28).The variation in staining density of buccal cell nuclei within

a single cytospin slide was relatively high both after in vitro andin situ exposure lo c/s-DDP. Inlrasample variation of DNA

modification in cultured cells seems to be a common phenomenon, since il has also been reported for immunofluorescencedeleclion of benzo(a)pyrene adducts to nucleic acids in culluredmurine keralinocyles (29) and after indirecl immunoauloradi-ographic analysis of jY-aceloxy-2-acelylaminofluorene-DNAbinding in cullured human fibroblasls (22). These dala suggeslthat a substantial varialion occurs in response lo a toxin between cells from Ihe "same" populalion, allhough il cannol be

lolally excluded lhal Ihe described helerogeneily in nuclearslaining densily resulled from unknown lechnical artifacts. InIhe presenl experimenls, Ihe inlerpalienl varialion in Ihe exlenlof in situ c/s-DDP-DNA binding in buccal cells after Ihe samedose ranged up lo a faclor 3 (see Fig. 6). Inlerindividual varialion in the level of c/s-DDP-DNA adducts has also been reported for rat tissues (30) and human leukocyles (12-15). Ourresulls are in agreemenl wilh Ihose of Fichlinger-Schepman etal. (15), which indicale lhal the extenl of c/s-DDP-pGpG cross

link formalion in leukocyles from differenl cancer patients maydiffer up lo 4.5-fold (assuming linearity belween dose andadduci level; cf. Refs. 24 and 25). Fichlinger-Schepman et al.( 15) also reported a posilive correlalion belween Ihe exlenls ofin vitro and in vivo formalion of c/s-DDP-pGpG cross-links inDNA of human leukocyles. In our presenl experimenls noquantitative relationship appeared lo exisl belween Ihe in vitroand in situ binding of c/s-DDP lo buccal cell DNA.

* A. M. J. Fichtinger-Schepman, personal communication.

5601

Research. on August 21, 2018. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from

PLATINUM-DNA ADDUCTS IN CANCER PATIENTS

The anti-m-DDP-DNA antiserum used in the present studiesrecognizes c/s-DDP-modified DNA in rodent tissues (17) andin cell suspensions (Ref. 18; this report). This rabbit antiserumwas elicited against in vitro modified CT-DNA complexed withmethylated bovine serum albumin. In a competitive enzyme-linked immunosorbent assay, the antiserum showed affinity notonly for the m-DDP-reacted polymers poly(dG)-poly(dC),poly[d(G,C)].poly[d(G,Q], and poly(dG), but also for CT-DNA modified by CBDCA or TN01 (crs-dichloro[l,l-bis(aminomethyl)cyclohexane]platinum(II), NSC 269608). Norecognition could be demonstrated for m-DDP, frans-DDP,fraws-DDP-DNA, m-DDP-reacted poly(dA)-poly(dT), or cis-DDP-modified nucleotides.5 It is concluded that the anti-m-DDP-DNA antiserum is specific for ds-platinum coordinationcomplexes bound to DNA and predominantly recognizes cis-DDP-modified guanine residues. This observation is in linewith guanine being the primary site of interaction between cis-DDP and DNA (3-7, 31). A detailed characterization of theantiserum will be published elsewhere. Although our antiserumhad been raised in the same way as that of Poirier et al. (9), thetwo antisera have partly different specificities. Whereas ourantiserum recognizes the c/s-DDP-modified alternating heter-opolymer poly[d(G,C)]-poly[d(G,C)], this is not the case withthe antiserum of Poirier et al. (32). The recognition of CBDCA-induced DNA modifications by our anti-m-DDP-DNA anti-serum indicates at least partial epitopic similarity of the DNAinteraction products of m-DDP and CBDCA (33).

The minimal dose of m-DDP after which significant nuclearstaining could be observed was S0.5 mg/kg [rodents (17)] ora20 mg/m2 m-DDP [humans (this report)]. To determine the

corresponding degree of DNA modification, rats were giveninjections of increasing doses of m-DDP (up to 15 mg/kg).Quantitative immunocytochemistry of liver sections was performed and the results were compared with data from atomicabsorption spectroscopic analysis of DNA, isolated from thesame livers. This comparison indicated that the detection limitof our immunocytochemical assay amounted to 2 x IO4 cis-DDP-DNA adducts/1010 nucleotides. Similar experiments withcultured cells (33)5 indicated a detection limit of about 3-5 xIO4 platinum atoms/genome (10'°nucleotides). This indicates

that the detection limit of our immunocytochemical assay forsingle cells is about the same as that for tissue sections.

Future research will concentrate on adduct formation intumors in experimental animals (18) and in human cancersafter in situ exposure to m-DDP. Tumor cell heterogeneity,resulting in the presence of drug-resistant subpopulations, is alimiting factor in chemotherapy (34). Hence, single-cell analysisof the formation and persistence of DNA adducts in cancerpatients might be helpful in explaining and/or predicting drug-induced toxicity and resistance of tumor cells against m-DDPand other platinum compounds.

ACKNOWLEDGMENTS

We thank Drs. E. Scherer and J. van Benthem for stimulatingdiscussions and Dr. J. G. McVie for critical reading of the manuscript.

REFERENCES

1. Rosenberg. B. Fundamental studies with cisplatin. Cancer (Phila.), 55:2303-2316, 1985.

2. Eastman, A., and Barry, M. A. Interaction of rra/is-diamminedichloropla-t inmuíII )with DNA: formation of monofunctional adducts and their reactionwith glutathione. Biochemistry, 25: 3303-3307, 1987.

* P. M. A. B. Terheggen et al., unpublished results.

3. Eastman, A. Réévaluationof interaction of cis-dichloro(ethylenediam-mine)platinum(II) with DNA. Biochemistry, 25: 3912-3915, 1986.

4. Fichtinger-Schepman, A. M. J., Van der Veer, J. L., Den Hartog, J. H. J.,I oilman. P. H. M.. and Reedijk, J. Adducts of the antitumor drug CMdiamminedichloroplatinum(Il) with DNA: formation, identification, andquantitation. Biochemistry, 24: 707-713, 1985.

5. Lippard, S. J. New chemistry of an old molecule: cii-[Pt(NHj)2Clj). Science(Wash. DC), 218: 1075-1082, 1982.

6. Zwelling, L. A., and Kohn, K. W. Mechanism of action of ru-diamminedi-chloroplatinum(II). Cancer Treat. Rep.. 63: 1439-1444, 1979.

7. Reedijk, J., Fichtinger-Schepman, A. M. J., Van Oosterom, A. T., and Vande Putte, P. Platinum amine coordination compounds as anti-tumor drugs.Molecular aspects of the mechanism of action. Struct. Bonding, 67: 53-89,1987.

8. Sundquist, W. I., and Lippard, S. H. Binding of cis- and rrafli-diamminedi-chloroplatinum(II) to deoxyribonucleic acid exposes nucleosides as measuredimmunochemically with anti-nucleoside antibodies. Biochemistry, 25:1520-1524, 1986.

9. Poirier, M. C., Lippard. S. J., Zwelling, L. A., Ushay, H. M., Kerrigan, D.,Thill, C. C., Santella, R. M., Grunberger, D., and Yuspa, S. H. Antibodieselicited against cii-diamminedichloroplatinum(II)-modified DNA are specificfor m-diamminedichloroplatinum(II)-DNA adducts formed in vivo and invitro. Proc. Nati. Acad. Sci. USA, 79: 6443-6447, 1982.

10. Sundquist, W. I.. Lippard, S. J., and Sfollar, B. D. Monoclonal antibodiesto DNA modified with cis- or rranj-diamminedichloroplatinum(II). Proc.Nati. Acad. Sci. USA, 84: 8225-8229, 1987.

11. Fichtinger-Schepman, A. M. J., Baan, R. A., Luiten-Schuite, A., Van Dijk,M., and Lohman, P. H. M. Immunochemical quantitation of adducts inducedin DNA by cu-diamminedichloroplatinum(II) and analysis of adduct-relatedDNA-unwinding. Chem.-Biol. Interact., 55: 275-288, 1985.

12. Poirier, M. C., Reed, E., Zwelling, L. A., Ozols, R. F., Litterst, C. L., andYuspa, S. H. Polyclonal antibodies to quantitate a'i-diamminedichloropla-

tinum(II)-DNA adducts in cancer patients and animal models. Environ.Health Perspect., 62:89-95, 1985.

13. Reed, E., Yuspa, S. H., Zwelling, L. A., Ozols, R. F., and Poirier, M. C.Quantitation of rú-diamminedichloroplatinum II (cisplatin)-DNA-intra-strand adducts in testicular and ovarian cancer patients receiving cisplatinchemotherapy. J. Clin. Invest., 77: 545-550, 1986.

14. Reed, E., Ozols, R. F., Tarone, R., Yuspa, S. H., and Poirier, M. C. Platinum-DNA adducts in leukocyte DNA correlate with disease response in ovariancancer patients receiving platinum-based chemotherapy. Proc. Nati. Acad.Sci. USA., 84: 5024-5028, 1987.

15. Fichtinger-Schepman, A. M. J., Van Oosterom, A. T., Lohman, P. H. M.,and Berends, F. Interindividual human variation in cisplatin sensitivitypredictable in an in vitro assay? Mutât.Res., 790: 59-62, 1987.

16. Fichtinger-Schepman, A. M. J., Van Oosterom, A. T., Lohman, P. H. M.,and Berends, F. c/5-Diamminedichloroplatinum(II)-induced DNA adducts inperipheral leukocytes from seven cancer patients: quantitative immunochem-ical detection of the adduct induction and removal after a single dose of cis-diamminedichloroplatinum(II). Cancer Res., 47: 3000-3004, 1987.

17. Terheggen, P., Floot, B. G. J., Scherer, E., Begg, A. C., Fichtinger-Schepman,A. M. J., and Den Engelse, L. Immunocytochemical detection of interactionproducts of cis-diamminedichloroplatinum(II) (m-DDP) and «\diariimine(l,l-cyclobutanedicarboxylato)platinum(II) (CBDCA) with DNA in rodent tissue sections. Cancer Res., 47:6719-6726, 1987.

18. Terheggen, P.. Begg, A. C., Floot, B., Emondt, J., and Den Engelse. L.Visualization of cisplatin-DNA and carboplatin-DNA adducts in tissues andcultured cells. In. M. Nicolini and G. Sandolini (eds.), Proceedings of theFifth International Symposium on Platinum and Other Metal CoordinationCompounds in Cancer Chemotherapy, pp. 139-143. The Hague: MartinusNijhoff Publishing. 1988.

19. Heyting, C., Van der Laken, C. J., Van Raamsdonk, W., and Pool, C. W.Immunohistochemica! detection of 0*-ethyldeoxyguanosine in the rat brainafter in vivo applications of W-ethyl-A'-nitrosourea. Cancer Res., 43: 2935-2941. 1983.

20. Menkveld, G. J., Van der Laken, C. J., Herrasen, T., Kriek, E., Scherer, E.,and Den Engelse, L. Immunohistochemical localization of O'-ethyldeoxy-guanosine and deoxyguanosin-8-yl-(acetyl)aminofluorene in liver sections ofrats treated with diethylnitrosamine, ethylnitrosourea or /V-acetylaminoflu-orene. Carcinogenesis (Lond.). 6: 263-270, 1985.

21. Scherer, E., Van Benthem, J., Terheggen, P., Vermeulen, E., Winterwerp,H. H. K., and Den Engelse. L. Immunocytochemical analysis of DNA adductsat the single cell level: a new tool for experimental carcinogenesis, chemotherapy and molecular epidemiology. In: Proceedings of the Conference onDetection Methods for DNA-damaging Agents in Man, Helsinki, Finland,September 1987.

22. Muysken-Schoen, M. A., Baan, R. A., and Lohman, P. H. M. Detection ofDNA adducts in Ar-acetoxy-2-acetylaminofluorene-treated human fibroblastsby means of immunofluorescence microscopy and quantitative immunoau-toradiography. Carcinogenesis (Lond.), 6: 999-1004, 1985.

23. Van der Ploeg, M., Van der Broek, K., Smulders, A. W. M., Vossepoel, A.M., and Van Duin, P. Hidacsys: computer programs for interactive scanningcytophotometry. Histochemistry, 54: 273-288, 1977.

24. Vermerken, J. B., Van der Vijgh, W., Klein, I., Gall, H., Van Groeningen,C., Hart, G., and Pinedo, H. M. Pharmacokinetics of free and total platinumspecies after rapid and prolonged infusions of cis-DDP. Clin. Pharmacol.Ther., 59:136-144. 1986.

25. Newell, D. R., Siddik, Z. H., Gumbrell, L. A., Boxall, F. E., Gore, M. E.,

5602

Research. on August 21, 2018. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from

PLATINUM-DNA ADDUCTS IN CANCER PATIENTS

Smith, I. I ., and Calvert, A. H. Plasma free platinum pharmacokinetics inpatients treated with high dose carboplatin. Eur. J. Cancer Clin. Oneol., 23:1399-1405, 1987.

26. Plooy, A. C. M., Van Dijk, M.. and Lohman, P. H. M. Induction and repairof DNA cross-links in Chinese hamster ovary cells treated with variousplatinum coordination compounds in relation to platinum binding to DNA.cytotoxicity. mutagenicity and antitumor activity. Cancer Res., 44: 2043-2051. 1984.

27. Hrushesky, W. J. M. Selected aspects of cisplatin nephrotoxicity in the ratand man. In: M. P. Hacker, E. B. Douple, and I. H. Krakoff, (eds.), PlatinumCoordination Complexes in Cancer Chemotherapy, pp. 165-186. The Hague:Martinus Nijhoff Publishing, 1984.

28. Litterst, C. L., LeRoy, A. F.. and Guarino, A. M. Disposition and distributionof platinum following parenteral administration of a's-dichlorodiammine-

platinum(II) to animals. Cancer Treat. Rep.. 63: 1485-1492. 1979.29. Poirier. M. C., Stanley, J. R.. Beckwith, J. B., Weinstein, I. B.. and Yuspa.

S. H. Indirect immunofluorescent localization of benzo(a)pyrene adducted

to nucleic acids in cultured mouse kcratinocyte nuclei. Carcinogenesis(Lond.), 3: 345-348. 1982.

30. Reed, E.. Litterst, C. L.. Thill. C. C., Yuspa. S. H.. and Poirier, M. C. cis-Diamminedichloroplatinum(II)-DNA adduci formation in renal, gonadal andtumor tissues of male and female rats. Cancer Res., 47: 718-722, 1987.

31. Recdijk, J. The mechanism of action of platinum anti-tumor drugs. PureAppi. Chem.. 59: 181-192, 1987.

32. Lippard. S. J.. Ushay, H. M., Merkel, C. M., and Poirier. M. C. Use ofantibodies to probe the stereochemistry of antitumor platinum drug bindingto deoxyribonucleic acid. Biochemistry, 22: 5165-5168, 1983.

33. Knox, R. J.. Friedlos, F., Lydall. D. A., and Roberts, J. J. Mechanism ofcytotoxicity of anticancer platinum drugs: evidence that ris-diamminedichlo-roplatinum(ll) and iv.v-diammine (l.l-cyclobutanedicarboxylato)platinum(ll)differ only in the kinetics of their interaction with DNA. Cancer Res., 46:1972-1979, 1986.

34. Kaye, S., and Merry, S. Tumour cell resistance to anthracyclines—a review.Cancer Chemother. Pharmacol., /¿'96-I03, 1985.

5603

Research. on August 21, 2018. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from

1988;48:5597-5603. Cancer Res   Philippe M. A. B. Terheggen, Robert Dijkman, Adrian C. Begg, et al.   in Cells from Platinum-treated Cancer Patients-Diammine(1,1-cyclobutanedicarboxylato)platinum(II) with DNA

cis-Diamminedichloroplatinum(II) and cisMonitoring of Interaction Prodcuts of

  Updated version

  http://cancerres.aacrjournals.org/content/48/19/5597

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/48/19/5597To request permission to re-use all or part of this article, use this link

Research. on August 21, 2018. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from