Deficiency in the Catatase Activity of Xeroderma Pigmentosum Cell

[CANCER RESEARCH 46, 538-544, February 1986]

Deficiency in the Catatase Activity of Xeroderma Pigmentosum Cell and SimianVirus 40-transformed Human Cell Extracts1

Monique Vuillaume,2 RÃ©gisCalvayrac, Martin Best-Belpomme, Philippe Tarroux, Michelle Hubert, Yves Decroix,

and Alain SarasinU.A. 686 C.N.R.S., Ecole Normale SupÃ©rieure,46, rue d'Ulm, 75230 Paris Â¡M.V., P. T., M. H.]; Laboratoire des Membranes biologiques. UniversitÃ©Paris VII, 2 placeJussieu, 75005 Paris [R. C.Â¡;U.A. 1135 C.N.R.S., UniversitÃ©Paris VI, 7 quai Saint Bernard, 75005 Paris [M. B. B.]; Institut Curie, 26, rue d'Ulm, 75005 Paris [Y. D.]; and

Laboratory of Molecular Mutagenesis, Institut de Recherches Scientifiques sur le Cancer, B. P. No. 8, 94802 Villejuif [A. S.], France

ABSTRACT

It has been previously shown that skin biopsies isolated fromvarious xeroderma pigmentosum (XP) patients present a permanent decline in catalase activity from the onset of the diseaseto the tumor formation. We report here that cultured XP cellstrains are also markedly deficient in the catalase activity withabout only 25% of the activity measured in normal human cells.No direct correlation between catalatic activity and excisionrepair ability has been found, since a XP variant line is as deficientas an XP-C strain. The exact cause of the catalase deficiency is

still unknown but could be due to the synthesis of a modifiedenzyme or to an abnormal regulation leading to a limited enzymesynthesis. Furthermore, simian virus 40 transformation of normaland radiosensitive cells (XP, ataxia telangiectasia) provokes adecrease in catalase activity of about 80% compared to thecontrol derivatives.

Mathematical analysis performed on our data shows a clear-

cut distinction between XP and normal cells while some of theXP hÃ©tÃ©rozygotecells exhibit an intermediate behavior. Althoughmost of the XP syndrome could be explained by the impairmentin the excision repair ability, the decrease in catalase activityleading to a probable increase in intracellular H2O2concentrationand/or to a higher sensitivity to any oxygen-activated species

could represent an additive effect in inducing the carcinogenicprocess.

INTRODUCTION

XP,3 an autosomal recessive human disease, is characterized

by an extreme sunlight sensitivity and a very high incidence ofskin carcinomas (1, 2). Two classes of XP patients have beendetected according to their repair deficiencies after UV irradiation.Fibroblasts isolated from most XP patients are defective in theexcision repair of UV-like induced lesions (1-6). Currently nine

complementation groups have been found using cell fusion experiments (7) which, moreover, display different UV sensitivity(5, 8). Fibroblasts isolated from a few XP patients (XP variant)possess a normal excision-repair pathway, while they are deficient in postreplication repair following UV irradiation (9). Al-

Received 7/11 /85; revised 10/23/85; accepted 10/25/85.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 sotely to indicate this fact.

1This research was supported by grants from trie Association pour le DÃ©ve

loppement de la Recherche sur le Cancer (Villejuif, France) and from the Commission of the European Communities (Brussels, Belgium).

2To whom requests for reprints should be addressed.3The abbreviations used are: XP, xeroderma pigmentosum; AT, ataxia telan

giectasia; BS, Bloom's syndrome; PBS, phosphate-buffered saline; PCA, principal

component analysis.

though the UV sensitivity of the XP cells is correlated with theappearance of skin carcinomas in the XP patients, other mechanisms leading to carcinogenesis may still occur in conjunctionwith DNA repair deficiencies. Particularly, an increase in theconcentration of activated oxygen species or in the H2O2 production can also cause DNA lesions and genetic toxicity (10-

14). These oxygen forms are produced by a wide variety ofphysiological processes and also by various chemical or physicalagents such as ionizing radiations. Free radicals and H202 areknown to be mutagenic (15) and carcinogenic (10, 13, 16-20).

Therefore, it was of interest to determine if abnormal radiationsensitivity of some human cells was not partly due to theaccumulation of such oxygen species owing to a low level ofcatalase activity. Recently, we have shown that skin biopsiesobtained from XP patients present a progressive decline incatalase activity from the onset of the disease to the tumorformation (21).

In this study, we quantify catalase activity in XP cell extractsfrom XP fibroblast lines established in vitro in order to determineif the low level of catalase previously measured in biopsies fromXP patients was retained in an In vitro system. The results clearlyindicate that all tested XP lines are markedly deficient in catalaseactivity measured with two different methods. For some XP celllines, the kinetics of oxygen production due to H2O2 dispropor-tionation is particularly impaired, suggesting the existence of anabnormal catalase function or behavior, towards the H2O2 substrate. After UV irradiation, XP cell extracts are initially moreactive in metabolizing H202 than normal cell extracts, but theircatalase activity becomes rapidly and drastically inactive. Consequently, the ratio of catalase activity to oxygen photocon-sumption decreases strongly from normal to XP cells, while XPhÃ©tÃ©rozygotecell extracts show intermediate levels. Finally, wealso report here that SV40 cell transformation of both normaland radiation-sensitive fibroblasts strongly decreases the cata

lase activity.

MATERIALS AND METHODS

CellsDiploid human fibroblasts were grown in Ham's F-10 medium supple

mented with 15% fetal calf serum in a 5% CO2 incubator at 37Â°C.The

origins of the various cell lines are described in Table 1. Some XP celllines have been established in our laboratory from child skin biopsies. Inthe absence of the knowledge of the complementation group, the DNArepair deficiency has been determined by unscheduled DNA synthesis(1,9).

Preparation of Cell Extracts

Confluent monolayers of diploid human fibroblasts were harvested bytrypsination followed by two washes in PBS. The cell pellet was resus-pended in PBS at a final concentration of about 106 cells/ml. This

CANCER RESEARCH VOL. 46 FEBRUARY 1986

538

Research. on December 22, 2018. © 1986 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

LOW CATALASE ACTIVITY IN XP CELLS

Table 1

Origin of cellstrainsNos.used

inthispaper12345

678g10111213141516171819202122232425PatientCell

strainHF19MRC5AS3HG1376AS8

AS9AS20AS23XP1BRXP4PAAS10XPSC8XP1PAAS15

AS16XP30ROAT3BIAT4BIAT5BIHG1332MRC5-V2HG1376-SVXP12RO-SVXP4PA-SVAT5BI-VAClinical

descriptionNormalNormalNormalNormalXP

hÃ©tÃ©rozygoteXPhÃ©tÃ©rozygoteXPhÃ©tÃ©rozygoteXPhÃ©tÃ©rozygoteXPXPXPXPXPXP

i sameXP tpatientXPATATATBSNormalNormalXPXPATComments|

Parents ofAS10Father

ofXPSC8Fatherof XP1PAGroup

DGroupCGroupeGroup

H and cockaynesyndromeGroupCNon-sunlight-exposed

skinSunlight-exposedskinVariantMRC5-SV40

transformedHG1376-SV40transformedXP12RO

(Group A) SV40transformedXP4PA(Group C) SV40transformedAT5BI-SV40

transformedAgeFetalEmbryo303141Embryo48555304618FetalAdultEmbryo18SexMFMFMMFMMFM

MMMMMMMIn

vitropop-ulstionuou-

WingrangeduringstudyP12-P16P10-P14P16-P21P16P4-P7

P5P8P4-P5P16P6-P21P4-P6P13-P15PSP7-P12

P6-P21P15P10-P13P14P16P12-P18SourceR.

Cox(Manchester)R.CoxJ.

Boue(Paris)J.German (NewYork)J.Boue

J.BoueJ.BoueJ.BoueC.

Artett(Brighton)J.BoueJ.BoueJ.BoueJ.Bouej

A. Sarasin(Villejuif)D.

Bootsma(Rotterdam)C.ArietiC.ArtettC.ArtettJ.GermanHolliday

(London)A.SarasinD.BootsmaA.SarasinC.

Artett

suspension was sonicated at 8,000 cycles with a M.S.E, sonteator in theice for two 30-s cycles. After a 300 x g centrifugation, the supernatant

was used to measure both catalatic activity and protein content.

Catalatic Activity Measurements

The catalatic activity of the catatase, as defined by Chance (22),corresponds to the disproportionation of H2O2 into water and molecularoxygen and was measured by two different methods after the additionof various amounts of H2O2(10"4 to 10"5 M).

Measure of the Oxygen Burst. The oxygen produced by the disproportionation of H2O2by the catalase in cell extracts was quantified witha Ysi Clark oxygen electrode in a 1.8-ml electrode chamber maintainedat 25Â°C.The oxygen pressure was continuously recorded (21), and the

catalatic activity was expressed in /Â¿molof O2 produced per h and per108 cells, or per mg of proteins.

Spectrophotometric Assay. The initial H2O2 concentration corresponds to an absorbance of A2W = 1. After the addition of cell extracts

at room temperature, the decrease of H2O2was followed at 240 nm asa function of time (23). The values of the initial slopes were used forquantifying the catalase activity which was expressed in unit A240min*1

per 10e cells.

In both methods, the catalase activity was totally inhibited by 10 mwCN", sodium azide, or aminotriazole.

Irradiation of Cell Extracts and Measure of the UV-Oxygen Photocon-

sumption

In some cases, cell extracts were irradiated before use with UV light(Zeiss Hg vapor lamp, high pressure HBO) at an incident energy of 5-10~2 W/cm2 measured with an ISCO spectroradiometer. Oxygen con

sumption determined by using the Clark electrode was measured afterirradiation of cell extracts. The value for UV-oxygen photoconsumption

was the difference between the O2 consumption after UV irradiation andthe O2 consumption in the dark before irradiation and was expressed inÂ¿imolO2 per h per 106 cells.

Data Analysis

The different measures made on the cell extracts were compared by

using the PCA technique (24). Four variables have been defined: Vi,catalatic activity; V2, UV-Oz photoconsumption; V3, V,:V2; VÂ«,the shape

of the O2 kinetics curves determined after UV irradiation. [Vt is equal to1 or 0, depending upon the shape of O2 production after H2O2addition.Zero corresponds to "normal" shape, and 1, to an "XP-like" shape (Fig.

1)]. These initial objects (Vi. V2, V3, and V4) can be plotted in an n-dimensional space, where n is the number of initial variables (here n =

4). We are considering that starting variables are not independent fromone another. Then the computer program calculates a new axis system(factors) from the initial data assay, in such a way that these factors nowbecome independent. Figures represented here show the initial pointprojection on each plane of this new space (Fig. 4). So, we get a pictureon which the information borne by the variations of the initial variables ismaximal. This representation maximizes the separation of the objectsinto distinct groups and allows one to plot the position of the initialvariables (V,, V2, V3, V4) in the factors. The main interest ot this mathematical description of the data is to give the broadest distribution ofsubsets of the cell population through a space of n (here 4) dimensions(24). Calculations were carried out at the Computer Center of the EcoleNormale SupÃ©rieure(Paris) on an IBM-4341-2 computer.

RESULTS

Catalase Activity in XP Cells versus Nonna! Cells

After H2OÂ¡additionto the cell extract, the catalaseactivitypresentinhumandiploidfibroblastswasquantifiedbymeasuringeither the oxygenburst with the Clark electrodeor the H202consumptionby spectrophotometry.The comparisonof thesetwo techniquesshowedus that the Clarkelectrodemethodisthe more sensitiveand quickerassay.After sonicationof thetrypsinizedcells, supematantsof cell extracts centrifugeaatvariousspeedshavebeentestedfor catalaticactivity.Thebestandmorereproducibleresultshavebeenobtainedon a 300 x gsupernatantwhichhas beenthereforeusedfor routineassays.Finally,sinceXPcellsareknownto behypersensitiveto UVlight,


539




we have then measured catalatic activity with and without a UVpreirradiation.

Qualitative Analysis. Typical kinetic values of 02 productionmeasured with the Clark electrode after two successive additionsof exogenous H2O2 are shown in Fig. 1 for diploid fibroblastsisolated from a normal individual (Fig. 1, a and b) and from a XPpatient (Fig. 1, c and d) with and without UV irradiation prior tomeasurement. The first 02 burst is almost linearly proportionalto short durations after H202 addition. The slope of the tangentat the origin of the curve is used to quantify the catalatic activityof cell extracts. After UV irradiation, the overall 02 kinetics is ofthe same shape, although the absolute values for the initialslopes are smaller for normal fibroblasts (Fig. 1, a and b). Incontrast, for all classical XP cells, the initial slopes are higherafter UV irradiation than before, but a plateau for the 02 burst isquickly reached (Fig. 1, c and d). After UV irradiation, the shapeof the first O2 burst is characteristic for XP cells and has thereforebeen called "XP-type shape" in opposition to the "normal-typeshape" seen with control cells, the XP variant line (XP30RO),

and the XP12RO-SV line (compare b and d of Fig. 1). The secondH2O2 addition in UV-irradiated XP cell extracts reveals the ab

sence of detectable catalatic activity (Fig. 1d), while in the other

Time Cmin-)

Fig. 1. Schematic representationof kinetics of O2burst (catalatic activity) measured by the Clark electrode after addition of H2O2in human diploid cell extracts,a and b, catalatic activities of normal fibroblasts. a, measured in darkness (D),without UV preirradiation;b, measuredafter UVtreatment (1.5 min with 5 â€¢10r2W/cm2)followed by 1.5 min of darkness, c and d, catalatic activities of XP fibroblasts(dotted lines), c. measured in darkness, without UV preirradiation; d. measuredafter UV treatment (the same as in b).

In order to be in the same experimental ranges to measure the 02 burst, thecrude extracts from normal fibroblasts have beendiluted out 3- to 4-fold comparedto XP cell extracts.H202arrow corresponds to the addition of 5 â€¢10~5M H202final.

Note the rapid breakdown of the 02 kinetics burst in UV-irradiatedXP cell (d).The shapeof this curve has been named"XP-type shape" as comparedto "normal-type shape" (c) and taken as the VÂ«variable in the mathematical analysis (see"Materials and Methods").

cases (Fig. 1, a to c), we can see a slope with a value twicehigher than for the first H2O2 addition, indicating that the enzymatic reaction is still linear to the substrate concentration. Consequently, the differences in catalatic activities measured between normal and XP fibroblasts are the greatest without UVpreirradiation of cell extracts (Table 2). However, since almostall the results described in this paper have been obtained afterUV irradiation of cell extracts, we probably overestimated thecatalatic activities in XP cells.

Quantitative Analysis. The catalatic activities quantified bythe Clark electrode were measured on various normal humanfibroblast strains, on different XP cell lines developed from skinbiopsies or from amniotic fluids, and on strains established fromskin biopsies of hÃ©tÃ©rozygoteXP parents. Results, detailed foreach cell line, show that catalatic activity is about 3 to 4 timeslower in XP cell lines than in controls (Fig. 2). Calculationsperformed on all our experiments indicate an average catalaticactivity of 4.7 Â±0.7 /Â¿molO2/h/106cells (or 3.6 Â±0.6 nmo\ 02/h/mg protein) for normal cell lines and 1.2 Â±0.2 /tmol O2/h/106

cells (or 0.84 Â±0.13 Â¿Â¿molO2/h/mg proteins) for XP cell lines,the differences being statistically significant (P < 0.01) by the f

Table 2Effect of UVpreexposureon catalatic activity of cell extracts

(normal,XP,AT cells)The catalaseactivity has beenmeasuredby the Clark electrodemethod. Results

are expressed in Â¿imol02/h/106 cells. The numbers in parentheses indicating thecell strains used correspond to those given in Table 1.

Normal cellsCatalaseactivity HF190) MRC5(2)

XP cellsXP4PA

(10)

AT cells

AT3Bi(17) AT4Bi(18)Without UV pre- 5.0 Â±0.5a 7.1 Â±0.1 0.24 Â±0.05 4.3 2.0 Â±0.2

irradiationAfter UV 3.8 Â±0.3 5.85 Â±0.10 0.7 + 0.1 3.3 1.5 Â±0.1

irradiation" Mean Â±SE.

Helero-Normal zygotes XP AT Bloom

<fi{â€¢^Ã³11-^.-I1*I

i.a1*5Mu{tTÂ°Â»l1â€¢

D

123 8 9 1011 1213 M15 lÂ«

Cell strains17 18192O

Fig.2. Catalatic activities of the different cell strains. All the extracts weresubmitted to the 1.5-min UV light before the analysis in the Clark electrode cuvet.Results are expressed in ^mol O2/h/106cells.0,1 to 4, normal fibroblasts; â€¢5 to8, hÃ©tÃ©rozygotefibroblasts; â€¢,9 to 16, XP fibroblasts; A, 17 to Ã•9,AT fibroblasts;D, 20, BS fibroblasts.

The numbers indicatingthe cell strains used correspond to those given in Table1. Points, mean; oars, SE, calculated from 3 experiments for lines 4 and 17, 5experiments for lines 8 and 20, and 10 to 20 experiments Â¡orthe other lines.


540




test versus control group. The catalatic activity in hÃ©tÃ©rozygotecell lines is almost identical to that in control cells. The case ofLine 3 which presents a catatase activity level intermediatebetween normal lines and XP cells will be discussed later (seeUV-02 photoconsumption and "Discussion").

Inside the group of XP cells, we did not find any obviouscorrelation between catalase activity, excision repair ability, and/or genetic complementation group. For instance the XP variantline XP30RO which shows an almost normal excision repairability has as low catalase activity as the fetal Group C line(XP4PA), which is 90% deficient in the excision repair mode (25).

Absence of Effect of the Cell Growth State on the CatalaseLevel

Although the XP and normal cell lines we have used havemore or less the same growth characteristics and all the experiments were carried out on confluent cell monolayers, we testedif variations in the confluent state might explain the differencesin catalatic activity observed between XP and control cells.Therefore, 1 wk before the experiments, the cell cultures wereseeded at various dilutions from 1:1 to 1:6. Results reported inTable 3 show that the same catalatic activity is observed whatever the confluent state at which cell extracts have been prepared. Thus, the tow level in catalatic activity of XP4PA cellscompared to the normal HF19 line was observed for both veryconfluent monolayers and sparse ones (Table 3). Moreover, noeffect of the population doubling numbers on the catalatic activityhas been found. For example, the same enzyme activity is foundfor the XP4PA and AS16 lines at population doublings of 6 and21.

Absence of Catalase Inhibitor in XP Cell Extracts

In order to determine if the tow catalatic activity measured inXP ceils was due to the presence of a specific cellular inhibitor,we mixed XP and normal cell extracts in various proportions.Results show that XP cell extracts do not inhibit the catalaticactivity of normal ceHextracts (Table 4). In fact, the total activityof the two extracts is always greater than the expected values.The experiments were carried out in conditions in which thecatalatic activity of the XP or normal extracts was directlyproportional to the cell extract concentration.

Catalatic Activity in Other Cell Lines

We have also determined catalase activity in cells isolatedfrom other human disorders for which DNA repair pathways are

TablesCatalalic activities and Ot photoconsumptions in XP and normal cell extracts

prepared from different confluence states of the cell monolayersConfluent state"

CeÂ«strains 1:1 1:2 1:2.5 1:3 1:4

Catalatic activitiesHF19(1) 52-5 5.15 5 5XP4PA(10) 0.6-0.62 0.62-0.5 0.62

UV-Ozphotoconsumptions''

HF19(1) 0.08 0.06 0.16 0.08XP4PAQO) 0.22 0.16-0.24 0.2

* The confluent state is arbitrarily expressed as the dilution factor used to seed

the cells 1 wk before the experiment.0 Measured by the Clark electrode and expressed in ^mol Oz/h/10Â»cells.

Table 4

Catalatic activities measured in mixtures of XP and normal cell extracts

Extracts from XP and normal cells are mixed at various ratios before the measureof catalatic activity with the Clark electrode. Results are expressed in Â¿imolO?/h/10*06118.

XP cells Normal Expected(XP4PA) cells VsXP+ %XP+ y2XP+ by Effective

(10) (HF19)(1) % normal '/anormal '/inormai calculation rise

0.83.88.35.942.81.82.3x3X3.4X1.7

impaired, such as in AT and BS patients (26, 27). We concludefrom Fig. 2 that AT and BS cell extracts have a lower catalaticactivity than normal cells, although one AT cell line (AT3BI) showsa normal enzyme activity, according to the data of Sheridan efal. (28). However, in all cases, UV irradiation of AT cells alsodecreases the catalase activity in a similar manner as for normalcells (Table 2).

Catalatic Activity in Various SV40-transformed Cell Lines

In order to study the effect of viral transformation on catalaseactivity, we have measured the enzyme level in various normal,XP, and AT cell lines and in their SV40-transformed derivatives.Results have been obtained by using both the Clark electrodeand spectrophotometric measurements.

Clark Electrode Analysis. SV40 transformation strongly decreases the catalatic activity measured in the cellular extractsfrom both normal, XP, and AT cell lines (Table 5). This decreaseleads to almost identical levels of enzyme activity for all SV40-transformed normal and XP cells. The UV preirradiation exerts asynergistic effect with SV40 transformation in decreasing thecatalase activity of control cells (Table 5).

Spectrophotometric Measurements. Experiments were carried out with the same cell extracts as the ones used for theClark electrode measurements but without UV preirradiation.The results presented in Tabte 5 confirm that the catalaticactivities are decreased 2.6- to 6.7-fold in XP cells as compared

to control cells. The transformation by SV40 also decreases thecatalatic activities in control cells, whereas it has almost no effectin XP cells which are already very tow in catalase activity (Table5).

It is noticeable that the two different methods for measuringcatalase activity give similar results for both XP versus controlceHs and for SV40 transformation.

UV-O2 Photoconsumption in Cell Extracts

The UV-O2 photoconsumption in various cell extracts had beenmeasured with the Clark electrode in darkness following a 1.5-min period of UV irradiation (Fig. 3). The results show that UVirradiation induces a photoconsumption of human cell extractsdetected by a decrease in the O2 level (see Fig. 1, b and d). Withthe exception of Line 3, control cells (Lines 1, 2, and 4) presenta tow OÃphotoconsumption level (0.11 Â±0.01 /imol 0Â¡>/h/106

cells) as compared to the XP cell lines which show a higheraverage activity (0.29 Â±0.03 /tmol Oz/h/106 cells), this difference

being statistically significant (P < 0.01 ) by the t test. The obviouspeculiarity of the control une 3 will be considered in the mathematical analysis and 'Discussion" sections. The XP hÃ©tÃ©rozygote

cells have a behavior quite similar to XP cells, white AT and BScells display a very low level of UV-O2 consumption (Fig. 3).


541




TaWe5

Spectrophotometricmethod8Cell

strains(2f

MRC5(21)MRC5-V2(4)

HG 1376(22)HG1376-SV(10)

XP4PA(24)XP4PA-SV(19)

AT5BÃŒ(25) AT5BÃŒ-VACatalase

activity0.537

0.1220.210

0.1680.080

0.083%

of decreaseinSV40-transformed

cells77200Catalase

activitymeasured

withoutpreirradiation6.4

34.3

3.6Clark

electrodemethod6%

of decreasein SV40-

transformedcells5316Catalase

activitymeasured afterUV irradiation5.8

0.80.4

0.31.8

0.2%

of decreasein SV40-

transformedcells862588

* Data are the means of 3 independent experiments measured by the Spectrophotometric method and expressed in arbitrary units as explained in 'Materials andMethods."

6 Results expressed in ^mol Cyh/IO* cells.c The numbers in parentheses indicating the cell strains used correspond to those given in Table 1.

Normal HÃ©tÃ©rozygotes XP AT Bloom

I**

T *

* *

i i * Ã¨Â« * Â«n n M Â»Cell Â«nkn

17 M W JO

Fig. 3. UV-O2 photoconsumption measured with the Clark electrode. Samesymbols as in Fig. 2. Results are expressed in Â»/mol02/h/106 cells.

Data Analysis

Mathematical analysis was carried out on the various reportedresults by using the PCA method which allowed us to perform amathematical description of all our data by giving the broadestdistribution of subsets of the studied cell populations. Fourvariables have been used: Vit catalatic activities; V2, UV-O2photoconsumption; V3,the ratio of catalatic activities versus UV-

02 photoconsumptions (equal to vVVz); V4, the shape of O2kinetics curves ("normal" or "XP-type" determined from Fig. 1).

The projection of the observation from the first three PCA planes(95% of the initial information) shows a partition of the analysisspace into two subsets: the first one corresponding to thecharacteristics presented by the normal fibroblasts and the second one to the cell lines issued from patients (Fig. 4, a to c).

On the figure defined by the 1-2 plane (Fig. 4a), the normal

cell points aggregate with V, and V3. On the left, we find all thecells isolated from patients (the different XP strains, AT, and BSlines), named here "abnormal cells" in opposition to the normal

human fibroblasts. Although abnormal cells are clearly separated

from the normal cells, the abnormal area shows some heterogeneity. This area could be split into two subsets, centered eitheron V4 or on V2. It could be noted that all the SV40-transformed

cell lines, even the one derived from normal cells, are closelyrelated to the abnormal area.

These observations are confirmed on the 1-3 plane (Fig. 4Â£>)

which allows the distinction of two parallel parts: on the rightside concerned by Vi and V3, all the normal cells, the AS9hÃ©tÃ©rozygotestrain, and the AT3 Bl line, while on the left sidebetween V2 and V4, all the abnormal cells (XP, AT, BS cells). Inthe middle and related to the normal area are cells from hÃ©tÃ©rozygote (AS8 and AS23), from one "normal fetus" (AS3), and

from a variant XP (XP30RO). Finally, at the bottom of the graph,we find the XP-SV40-transformed line (XP12RO-SV). This 1-3

plane seems to show a periodic organization, also observed inFig. 4c. Indeed in the 1-4 plane (Fig. 4c), a regular periodicity

can be observed which is known in PCA analysis as the Guttmaneffect (29). Then a gradual repartition can be followed whichreflects a kind of evolution from normal to abnormal cells via Line3 (AS 3 line) and some XP hÃ©tÃ©rozygotesand towards the SV40-

transformed strains.In conclusion, PCA analysis made on our experimental data

confirms that the XP lines have characteristics completely different from normal fibroblasts. Some of the hÃ©tÃ©rozygotecells andthe Line 3 strain possess intermediate properties between XPlines and normal lines. One line of AT (AT3 Bl) is normal, whileAT4 Bl, AT5 Bl, and BS strains appear very different from normalin our conditions.

DISCUSSION

Our data show that catalatic activity quantified by two differentmethods is decreased by more than 50% in XP cells comparedto the control cells. No direct correlation between catalatic activity and excision DNA repair ability is observed; as a matter offact, the XP variant cell line (XP30RO) is as much deficient incatalase as an XP Group C for which excision repair representsonly 5 to 10% of controls (25). Since this deficiency is alsoobserved in the fetal XP cell line (XP4PA), it seems that agenetica! abnormality could be responsible for the catalase impairment. Although the catalase activity in XP hÃ©tÃ©rozygotecells


542




A â€¢

oÂ«I

Xnooo

" IÂ»25

Atl

It

â€¢A'Â«â€¢Â«A

**â€¢*. *â€¢â€¢

30

-Â»x-

Fig. 4. PCA analysis, a, 1-2 plane. Observations of each type of cells are

reported in the factor space using their coordinate on each factorial axis calculatedby the PCA program. The ftrsi axis maximizes the relative distance between theobservations (65% of the initial information). The second axis acts in the same wayusing the initial information (20%). b, 1-3 plane. Same as 1-2 plane with the thirdaxis using 10% of the initial information, c, 1-4 plane. Same as 1-2 plane with the

fourth axis using 5% of the initial information.The four variables (V, to V4)are defined in "Materials and Methods." The numbers

and the symbols of the cell strains used are identical to those in Fig. 2. x, SV40-transformed cells; 8, SV40-transformed cells analyzed after UV preirradiation (1.5min).

is normal, the mathematical analysis seems to show that according to our specific criteria, hÃ©tÃ©rozygotecells appear betweennormal and XP cells (Fig. 4). From the PCA analysis, the controlLine 3 lies between the normal and XP cell lines (Fig. 4, a to c).This observation is especially consistent with the abnormal UV-

O2 photoconsumption observed with this cell line (Fig. 3). Although its catalatic activity is relatively low, the shape of its O2kinetics is of the normal type. Indeed, this line comes from asystematical analysis of chromosomal abnormality in pregnantwomen older than 40. This embryonic cell shows a normal UVrepair as measured by unscheduled DNA synthesis analysis. Thepeculiar behavior of this normal line might be explained by anabnormal O2 metabolism independent of the XP phenotype.Whatever the cellular defect is, the PCA analysis leads us toremove this cell Line 3 from the control group.

The exact cause of the catalatic deficiency is still unknown. Itcould be due to either the synthesis of an abnormal enzymeinhibited by the H2O2substrate, as suggested by the kinetics ofO2 production (Fig. 1), or an abnormal catalase regulation leadingto a limited enzyme production. A structural mutation in the Cs"

mouse decreasing the catalase activity level due to an abnormalassociation between monomers has recently been described(30). The low catalase activity observed in all the SV40-trans-formed cell lines could illustrate a gene repression process (31).It is indeed known that SV40 transformation modifies the cellulargenetic program (32) and induces the expression of a set of newgenes and the repression of other ones (31, 33, 34); in fact,more than 50 important proteins have been shown to be severelyrepressed after SV40 transformation in REF 52 cells (31). Thecatalase gene may well be inhibited after SV40 transformationas 06-methylguanine transferase (35) and Superoxide dismutase

genes (10, 34) are.However, the enzyme impairment is not due to the presence

of a specific cellular inhibitor, since mixed extracts do not displayany decrease in catalatic activity (Table 4). On the contrary, thecatalase activities measured in mixed extracts are always higherthan the expected level, which could indicate that cofactors oractive subunits from control cells may be used by the XP cellextracts to exhibit normal catalase activity. These results remindus of those showing a recovery of DNA repair ability in XP cellsafter cell fusion (5, 8) or microinjection of normal human cellextract (36).

We have also determined catalase activity in three AT cellstrains and one BS line taken as controls. Two of the AT lines(AT4 Bl and AT5 Bl) show a very low level of catalatic activity,and the kinetics of O2 burst after UV irradiation has a shapesimilar to the XP one. However, the AT3 Bl strain has a normalcatalatic activity as previously reported (26, 28).

From the mathematical analysis, a certain heterogeneity canbe observed for both normal and XP cells, although the XPdispersion among the 1-2 plane seems to be larger (Fig. 4a).

Since we have intentionally used human strains from variousorigins, this observed heterogeneity is not surprising. The resultsobserved in the 1-3 and 1-4 planes (Fig. 4, b and c) can beinterpreted in the PCA analysis as a typical "Guttman effect"

(29). This pattern occurs when the evolution of the variableswhich characterize each observation covers a gradual relationship. In our case, this gradual effect could be attributed to theprogressive acquisition of the abnormal state and represents agood description of the appearance of the pathological catalase


543




metabolism. For example, the position of some of the hÃ©tÃ©rozygote points (Fig. 4c), during the evolution from normal to abnormal cells, stresses their intermediate phenotypic expression.

Since 1968, XP cells have been shown to be deficient in theexcision repair of UV-like lesions (1), this defect being very wellcorrelated with the tumor development on sun-exposed skin of

these patients (2, 5). Although XP cells show a normal behaviortowards ionizing radiations, they are also sensitive to sometreatments involving the Superoxide ion and H2O2 production.XP cells are thus sensitive to bleomycin (37) and to asbestos(38), which are known to produce O2 reduced species. Accumulation of H2O2due to a low level of catalase activity could beresponsible for cell sensitivity and tumorigenesis, since H2O2 isa very toxic intermediate and has been shown to induce DNAlesions (10, 13, 14, 17, 20, 39) and to be mutagenic andcarcinogenic in mammals (16,18,40). For example, the inductionof duodenal tumors by H2O2in mice has been reported (11).

Although it is clear that most of the features of XP cells andXP patients are due to the lack of UV-induced DNA lesion repair,

the decrease in catalase activity and therefore the increase inH2O2 cellular concentration and/or the sensitivity to any freeradical inducer (including sun light) could be an additive effect forinducing the carcinogenic process. This could be particularlyimportant for XP variants, for which excision repair is quitenormal, but sun-exposed tumorigenesis is still very high. Finally,

observations suggesting a transitory stabilization of the tumordevelopment of XP patients by local catalase application (21)indicate that, as shown by our in vitro analysis, there exists arelationship between catalase decrease and XP tumorigenesis.

ACKNOWLEDGMENTS

The authors are very grateful to the various laboratories listed in Fig. 1 for thegift of the cell lines. They are particularly thankful to Dr. L. Daya-Grosjeanwhotransformed some of our cell lines with SV40. We also thank A. Benoit and A.Margot for their technicalassistance,and Dr. G. Renaultfor his help in the statisticalanalysis.

REFERENCES

1. Cleaver,J. E. Defective repair replication of DNA in xeroderma pigmentosum.Nature (Lond.),218: 652-656,1968.

2. Cleaver, J. E., and Bootsma, D. Xeroderma pigmentosum: biochemical andgenetic characteristics. Annu. Rev. Genet., 9:19-38, 1975.

3. De Weerd-Kastelein,E. A., Keijzer,W., and Bootsma,D. Geneticheterogeneityof xeroderma pigmentosum demonstrated by somatic cell hybridization. Nat.New Biol., 238: 80-83, 1972.

4. Hanawalt, P. C., Cooper, P. K., Ganesan,A. K., and Smith, C. A. DNA repairin bacteria and mammaliancells. Annu. Rev. Biochem.,48: 786-836,1976.

5. Kraemer, K. H., Coon, H. G., Petinga, R. A., Barrett, S. F., Rahe, A. A., andBobbins,J. H. Geneticheterogeneityin xeroderma pigmentosum:complementation groups and their relationship to DNA repair rates. Proc.Nati. Acad. Sci.USA, 72: 59-63,1975.

6. Sarasin,A. R., Smith, C. A., and Hanawalt,P.C. Repairof DNA in humancellsafter treatment with activated aflatoxin B1. Cancer Res., 37: 1786-1793,1977.

7. Fischer,E., Keijzer, W., Thielmann,H. W., Popanda,O.,Bohnert, E., Edler, L.,Jung, E. G., and Bootsma, D. A ninth complementationgroup in xerodermapigmentosum XPI. MutÃ¢t.Res., 745. 217-225,1985.

8. Matsukuma, S., Zeli, B., Keijzer, W., Berends, F.,and Bootsma, D. Differentrates of restoration of the repair capacity in complementing XP cells afterfusion. Exp. Cell. Res., 134:103-112,1981.

9. Lehmann,A. R., Kirk-Bell,S., Artett, C. F., Paterson,M. C., Lohman, P. H. M.,De Weerd-Kastelein,E. A., and Bootsma, D. Xeroderma pigmentosum cellswith normal levelsof excision repair have a defect in DNA synthesis after UV-irradiation. Proc. Nati. Acad. Sci. USA, 72: 219-223, 1975.

10. Cerutti, P. A. Prooxidant states and tumor promotion. Science (Wash. DC),227:375-381,1985.

11. Ito, A., Watanabe, H., Naito, M., Naito, Y., and Kawashima, K. Correlation

between induction of duodenal tumor by hydrogen peroxyde and catalaseactivity in mice. Gann, 75:17-21,1984.

12. Me Cord, J. M., and Fridovich, l. Superoxidedismutase:an enzymatic functionfor erythrocuproprotein (hemocuproprotein).J. Bid. Chem., 244: 6049-6057,1969.

13. Oberiey, L. W. Oncology overview. In: Selected Abstracts on Modulation ofCancerby SuperoxideDismutase.L. W. Obertey(ed.),Washington,DC: UnitedStates GovernmentPrinting Office, 1983.

14. Phillips,H. J., James,T. E. B., and Anderson,D. Geneticdamage in CHO cellsexposed to enzymically generated active oxygen species. MutÃ¢t.Res., 726:265-271,1984.

15. Mac Cann,J., Choi, E.,Yamasaki,E., andAmes,B. N. Detectionof carcinogensas mutagens in the Sa/mone//a/microsometest: assayof 300 chemicals.Proc.Nat).Acad. Sci. USA, 72: 5135-5139,1975.

16. Greentein, J. P., and Andervort, H. B. Kidney and blood catalase activity oftumor-bearinganimals.J. Nati. Cancer Inst, 2: 589-593,1981.

17. Henriksen, T., Mete, T. B., and Saxebd, G. Free radical formation andprotection from radiation damage. In: W. A. Pryor (ed.), Free Radicals inBiology, Vol. 2, pp. 213-254. New York: Academic Press, 1976.

18. Kaplan, J. H., and Groves, J. N. Liver and Woodcatalase activity of tumor-bearing mice. Cancer Res., 32:1190-1194,1972.

19. Pryor, W. A. The role of free radical reactions in biological systems. In: FreeRadicalsin Biology, Vol. 1, pp. 1-50. New York: AcademicPress, 1976.

20. Ts'o, P. 0. P., Caspary, W. J., and Lorentzen, R. The involvement of free

radicals in chemical carcinogenesis. In: W. A. Pryor (ed.), Free Radicals inBiology, Vol. 3, pp. 251-305. New York: Academic Press, 1977.

21. Vuillaume,M., Deoroix, Y., Calvayrac,R., Vallot, R., and Best-Belpomme, M.Xeroderma pigmentosum: H202 formation and catalatic activity during theevolution of the skin cells: applicationto an early diagnosis and to a tentativetreatment. C. R. Acad. Sci. Paris,296: 845-850,1983.

22. Chance,B., Stes,H., and Boveris,A. HydroperoxkJemetabolismin mammalianorgans. Physio!.Rev., 59: 527-605,1979.

23. Beers, R. F., and Sizers, I. W. A spectrophotometric method for measuringthe breakdown of hydrogen peroxide by catalase. J. Biol. Chem., 795: 133-139,1952.

24. Kendal, H. S., and Stuart, H. Canonical variables. In: C. Griffin (ed.), TheAdvancedTheory of Statistic, pp. 243-306. London: Griffin, 1973.

25. Haltey,D. J. J., Keijzer,W., Jaspers, N. G. J., Niermeijer,M. F., Kteijer,W. Y.,Boue, J., Boue, A., and Bootsma, D. Prenataldiagnosisof XP(Group C) usingassays of unscheduledDNA synthesis and postreplicationrepair. Clin. Genet.Ã•6:137-146,1979.

26. Abielovich,D., and Cohen, H. M. Normalactivity of nucteosidephosphorylase,Superoxidedismutase, and catalase in skin fibroblasts cultured from Ataxia-telangiectasiapatients. Isr. J. Med. Sci., 74: 284-287, 1978.

27. Emerit, I., and Cerutti, P. A. Clastogenic activity from Bloom syndrome fibro-blast cultures. Proc. Nat. Acad. Sci. USA, 78:1868-1872,1981.

28. Sheridan, R. B., and Huang, P. C. Superoxidase dismutase and catalaseactivities in ataxia telangiectasia and normal fibroblast cell extracts. MutÃ¢t.Res., 67:381-386,1979.

29. Guttman, L. Metricizing rank-ordered or unordered data for a linear factoranalysis. Sankhya,27: 257-268,1959.

30. Lewis, W. H. Establishmentof mouse cell lines homozygotes for temperature-sensitive mutation in catalase gene. Somatic Cell Mol. Genet., 77: 319-324,1985.

31. Pranza, B. R., and Garrete, J. I. Transformation-sensitiveprotein of REF52cells detected by computer-analyzedtwo dimensionalgel electrophoresis. In:A. J. Levine, G. F. Van de Wollde, W. C., Topp, and J. D. Watson (eds.),Cancer Cells, pp. 137-146. Cold Spring Harbor, NY: Cold Spring HarborLaboratory, 1984.

32. Huschtscha, L. I., and Holliday, R. Umited and unlimited growth of SV40-transformed cells from human diptoid MRC-5 fibroblasts. J. Cell. Sci., 63: 77-99, 1983.

33. Sarasin, A. SOS response in mammaliancells. Cancer Invest., 3: 163-174,1985.

34. Mariheus, F., Nicole, A., and Sinet, P. M. Lowered level of translatablemessenger RNAs for manganeseSuperoxidedismutase in human fibroblaststransformedby SV40.Btochem.Biophys.Res.Commun., 729:300-305,1985.

35. Yarosh, D. B. The role of O'-methylguanine-DNA methyltransferase in cellsurvival, mutagenesisand carcinogenesis.MutÃ¢t.Res., 745:1-16,1985.

36. De Jonge, A. Y. R., Bermeulen, W., Klein, B., and Hoeijmakers, J. H. J.Microinjection of human cell extracts corrects XP defect. EMBO J., 2: 637-641, 1983.

37. Hurt, M. M., and Robb, E. M. Abnormal response of XP cells to bleomycin.Cancer Res., 44: 4396-4402,1984.

38. Yang, L. I., Kouri, R. E., and Curren, R. D. XP fibroblasts are more sensitiveto asbestos fibers than are normal human fibroblasts.Carcinogenesis(Lond.),5:291-294,1984.

39. Hoffmann, M. E. and Meneghini, R. Action of hydrogen peroxide on humanfibroblast in culture. Photochem. Photobiol., 30:151-155,1979.

40. Saito, T., Kuraski, M., Kaji, H., and Saito, K. Deficiencyof erythrocyte super-oxide dismutase and catalase activities in patients with malignant lymphomaand acute myeloid leukemia.Cancer Lett., 24:141-146,1984.


544



1986;46:538-544. Cancer Res Monique Vuillaume, Régis Calvayrac, Martin Best-Belpomme, et al. Cell and Simian Virus 40-transformed Human Cell ExtractsDeficiency in the Catalase Activity of Xeroderma Pigmentosum

Updated version

http://cancerres.aacrjournals.org/content/46/2/538

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

[email protected] 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/46/2/538To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



Documents

Deficiency in the Catatase Activity of Xeroderma Pigmentosum Cell