quantitative detection of ultraviolet-specific p53 ... · suppressor gene, p53, is the most often...

Vol. 6, 433-438, June 1997 Cancer Epidemiology, Biomarkers & Prevention 433

3 The abbreviations used are: NMSC, nonmelanocytic skin cancer; XP, xero-

derma pigmentosum; AS-PCR, mutant allele-specific PCR.

Quantitative Detection of Ultraviolet-specific p53 Mutations in

Normal Skin from Japanese Patients1

Allal Ouhtit, Masato Ueda, Hisayoshi Nakazawa,Masamitsu Ichihashi, Nicolas Dumaz, Alain Sarasin, andHiroshi Yamasaki2

Unit of Multistage Carcinogenesis, IARC, 69372 Lyon, France [A. 0., H. N.,

H. Y.l; Department of Dermatology, Kobe University School of Medicine,

Kobe 650, Japan [M. U., M. 1.1; and Laboratory of Molecular Genetics,Institut de Recherches Scientifiques sur Ic Cancer, 94800 Villejuif. France

[N. D., A. S.]

Abstract

We have previously developed sensitive methods to detectUV-specific p53 mutations (CC to TT tandem mutations)and have reported that such mutations could be found inthe normal skin cell populations of sun-exposed bodysites, but not in those of covered sites, in Australian

cancer patients. We have now further refined our allele-specific PCR method for detecting CC to TT mutations atcodons 247/248 of the p53 gene to allow quantitativemeasurements. Using DNA containing this mutation from

a tumor as a standard for calibration and 5 �tg ofgenomic DNAIPCR reaction, we could detect 1 mutantallele in about 106 wild-type alleles. It is essential to usepurified primers and 64#{176}Cas the annealing temperaturefor PCR. Our method has been applied in a study of thecorrelation of sun exposure and accumulation of CC toTT mutations in normal skin biopsies from Japanesepatients. There were more p53 mutations in samplestaken from sites that were chronically exposed to the sunthan in those from covered sites. A significant trend ofincreased p53 mutation frequency with increase in age ofsubjects was found, suggesting the cumulative nature ofthe mutation. On the other hand, the p53 mutationfrequency was higher in patients with premalignanttumors or nonmelanocytic skin cancer than in patientswith only benign tumors. These results confirm the utilityof PCR-based p53 gene mutation assays for themeasurement of exposure to UV as well as for predictingthe risk of UV-associated skin cancer.

Introduction

Genetic alterations, which can be detected after exposure tospecific carcinogens and at early stages of the multistep caci-

Received 1 1/15/96; accepted 2/24/97.

The costs of publication of this article were defrayed in part by the payment of

page charges. This article must therefore be hereby marked advertisement in

accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

i Supported in part by EU grants ENV4-CT96-Ol 72 and ENV4-CT96-0194 and

by United States Environmental Protection Agency Grant 824264-01 ; M. U. was

supported by Zaigaikenkyuu Fellowship from the Ministry of Education, Japan.

2 To whom requests for reprints should be addressed, at Unit of Multistage

Carcinogenesis, IARC, 150 Cours Albert-Thomas, Lyon F-69372, France. Phone:

33-72-73-84-85; Fax: 33-72-73-84-42; E-mail: [email protected].

nogenic process, have the potential to serve as molecular mark-

ers of exposure to relevant etiological factors (1 , 2). Among

such alterations are carcinogen-specific point mutations in on-

cogenes and tumor suppressor genes (3-5). Early detection of

such biomakers depends on sensitive techniques for their iden-

tification when present even as a small minority of the cell

population exposed to a given carcinogen. Such techniques

include allele-specific PCR, enriched PCR, and RFLP-PCR,

which are sensitive enough to allow the detection of specific

point mutations at low frequencies (104_106) and can be

used in quantitative assays (6-8).

NMSCs3 are the most common tumors in humans (in

fair-skinned populations), and their incidence has been rising

steadily (9-1 1). Depletion of stratospheric ozone, resulting in

an increased level of solar UV exposure, has been proposed to

be one factor responsible for this increase ( 12). The tumor

suppressor gene, p53, is the most often mutated gene in NMSC

(3, 13-15). The mutations occur mainly at dipyrimidine sites

(16-18), indicating the direct action of UV to form dipyrimi-

dine photoproducts (cyclobutane pyrimidine dimers, pyrimi-

dine-pyrimidone(6-4) photoproducts, and/or Dewar photoiso-

mers; Refs. 16, 19, and 20). Furthermore, in these tumors, the

most notable molecular signatures of UV exposure found in the

p53 gene are tandem CC-i’TT mutations.

We have previously developed sensitive PCR- and ligase

chain reaction-based methods to detect UV-specific mutations

(CC-+TT) at codons 245 and 247/248 in the p53 gene, using

UV-exposed human skin cell cultures and normal human skin

biopsies (21). Both mutations have been found in human

NMSC at high prevalence (13, 22). These results strongly

suggest that tandem-base mutations are specifically induced in

normal skin by solar UV radiation. We have detected these

mutations in normal skin of Australian skin cancer patients at

sun-exposed sites, such as the shoulder, but not at nonexposed

sites (i.e., buttock; Ref. 21).

We report here further refinement of the allele-specific

PCR assay to give greater sensitivity, so that cumulative p53

mutations can be measured reliably and used in the molecular

epidemiology of UV-induced skin cancer. We have optimized

the conditions so that tandem CC-+’Vl’ mutations at codons

247/248 of the p53 gene can be detected at frequencies as low

as l0_6. We then applied this quantitative assay to normal skin

biopsies collected from patients with benign tumors and also to

biopsies from patients with premalignant tumors or NMSC and

examined its value for assessing cumulative UV exposure and

predicting the risk of skin cancer.

on May 8, 2020. © 1997 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

http://cebp.aacrjournals.org/

1/F

32/M

37/F

42/M

52/F

57/M

58/F

58/F

65/M

60/F

76/F

58/F

75/F

79/F

42/M

671M

Face (+)

Face (+)

Abdomen (-)

44/M

44/M

44fM

434 Quandtation of p53 Mutation in Normal Human Skin

Table 1 Characteristics of normal skin samples taken from s un-exposed (+)

and covered (-) sites of Japanese samples

Anatomical site ofSkin lesions

normal skin biopsyAge (yrs)/Sex

A. Skin from normal tissue surrounding benign tumors

123 Nevus cell nevus Face (+)


138 Dermatofibroma Face (+)

125 Epidermal cyst Face (+)


72 Epidermal cyst Face (+)



86 Benign adenoma Face (+)


96 Eccrine poroma Sole (-)

48 Seborrheic keratosis Face (+)



104 Seborrheic keratosis Scalp (-)

120 Seborrheic keratosis Abdomen (-)

B. Skin from normal tissue of patients with premalignant tumor or NMSC

1 33 Actinic keratosis Face (+) 65/F

2 Actinic keratosis Back of hand (+) 77/F

24 Actinic keratosis Back of hand (+) 87/M

I5n Bowen’s disease Back (-) 83/F

102 Bowen’s disease Mamma (-) 60/F

135 Bowen’s disease Back of finger (+) 651M

122 Bowen’s disease Back of foot (-) 75/F

124 Dermatofibrosarcoma protuberans Face (+) 89/M

137 Sebaceous epithelioma Face (+) 601M

39 BCE� Face (+) 59/F

108 BCC Neck(+) 59/M

17n BCE Face(+) 81/F

76 BCE Face(+) 711M

69 BCE Face (+) 73/M

70 BCE Clavicula (-) 73/M

98 BCC Face (+) 74/F

78 BCE Face (+) 79/F

79 BCE Clavicula (-) 79/F

8 BCE Face (+) 81/F

9n BCE Clavicula (-) 61fF

60 BCE Abdomen (-) 74fM

36 5CC Face(+) 90/F

100 5CC Face (+) 62/F

139 SCC Face (+) 851M

C. XP patient”

8 1 BCE and Lentigo maligna

83 BCE and Lentigo maligna

84 BCE and Lentigo maligna

a BCE, basal cell epithelioma; 5CC, squamous cell carcinoma; BCC, basal cell

carcinoma.

b This patient is not included in our analysis shown in Figs. 3, 4, and 5.

Materials and Methods

Skin Biopsy and DNA Extraction. Full-thickness biopsies ofnormal skin were taken from Japanese patients (patients withbenign tumors and patients with premalignant tumors orNMSC; Table 1); 30 were from chronically sun-exposed sites(face, neck, and hands), and 10 were from covered sites (ab-

domen, mamma, sole, and clavicula). Biopsies were immedi-ately frozen in liquid nitrogen and kept at -80#{176}C.Tissue pelletswere digested by 5 units/ml of proteinase K overnight at 37#{176}C,followed by phenol and chloroform extraction for DNA prep-

aration (21).To prepare a standard for calibration, DNA was extracted

from a basal cell carcinoma tumor obtained from a XP patientthat contained anCCgg-�aaUgg tandem-base mutations at the247/248 codons of the p.5,3 gene corresponding to the Asn-Arg-�Asn-Trp amino acid changes (13). This DNA preparation

was serially diluted (from 10 ‘ to l0�) with normal DNAextracted from lymphocytes isolated from the peripheral nor-mal blood of healthy donors with Ficoll-Paque research grade

solution (Phamacia Biotech).

AS-PCR. AS-PCR was carried out as described previously(21). This is a nested PCR that consists oftwo successive PCRs.In the first PCR, the entire exon 7 of the p.53 gene was amplified

using primers 5’-ACTGGCCTCATCTTGGGCCT-3’ and 5’-

TGTGCAGGGTGGCAAGTGGC-3’. In the second nestedPCR, a short sequence containing the CC to TI’ mutation at

codons 247/248 was then selectively amplified using the[y-32P]ATP end-labeled primers 5’-CTGCATGGGCGGCAT-

GAAI�[-3’ (specific for CC to ‘IT tandem-base mutations) and5’-CAAGTGGCTCCTGACCTGGA-3’.

For the first PCR, each reaction mixture (100 pi) con-

tamed s ,�g of genomic DNA, 1 X PCR buffer (BoehningerMannheim), 400 ,.LM each deoxynucleotide triphosphate, 2 units

of Taq polymerase (Boehninger Mannheim), and primers. Am-plification was initiated by denaturation at 94#{176}Cfor 2 mm and

by the hot start method (70#{176}C),followed by 35 cycles of 94#{176}Cfor 30 5, 60#{176}C for I mm, and 72#{176}Cfor 30 s. The amplified exon7 of the p53 gene was purified by column chromatography(Microcon 30; Amicon) and lyophilized to be used as a tem-plate for the second PCR. The reaction mixture for this secondAS-PCR consisted of the purified p53 exon 7 (PCR Product),1 X PCR buffer, 400 �.LM each deoxynucleotide triphosphate,

units of Taq polymerase, and 200 p.M each of the 5’ [-y-32P}ATP

end-labeled primers. PCR was performed by DNA denaturation

at 94#{176}Cfor 2 mm and by the hot start method at 70#{176}C,followedby 30 cycles of 94#{176}Cfor 30 s, 64#{176}Cfor 1 mm, and 72#{176}Cfor

30 s. Amplified DNAs were lyophilized and subjected to 6%PAGE. Amplified CC to TI’ mutations (the band corresponding

to 75 bp) were detected by autoradiography using a Phosphor-Imager system (445 SI; Molecular Dynamics).

Quantitative Analysis. Gels were scanned by Phosphorlm-

ager (Scanner control program). Using the Image-Quant pro-gram (version 1 . 1 ; Molecular Dynamics), the background cor-rection was carried out; automatic substraction of background

from each mutant band (-75 bp) was performed using thewild-type DNA band as control. The bands from the calibrationstandard dilutions were then analyzed by measuring their signalintensity, as well as those of the investigated samples. From the

calibration values (lO_6 to l0�), the volume value for each

sample was transformed to the concentration value correspond-ing to the cumulative p53 mutation frequency. We could detect1 mutant allele in the presence of 106 wild-type alleles. All

statistical significances were verified by Student t test, Wil-coxon and Fisher’s r to z tests, using StatView program (ver-sion 4.5; Abacus).

Results

Improvement of the AS-PCR Technique. Using reference

DNA from a basal cell carcinoma with a 247/248 CC to iTmutation in the p53 gene as a calibration standard (mutant p53

diluted with 10- to 107-fold wild-type p53), several assays were

carried out to optimize the experimental conditions of the PCRreaction. AS-PCR was carried out at different temperatures

(60#{176}C to 67#{176}C)to determine the optimal annealing temperature.As seen in Fig. 1A, the optimal annealing temperature was64#{176}C,at which the mutant band was detected at a lO_6 dilution.



A i0��iO�’ 1O4WT

a

B

A

�-‘ 10�

. -

� .2.2 �

5.

E� 0.1.

E

a.e�0.x

1.0

0.01 10

Dilution 10-s

Fig. 2. Intercalibration variations using the AS-PCR assay for quantitative de-

tection of the CC to iT p53 mutation in excess of wild-type alleles. AS-PCR

assay was carried out for samples in which DNA containing the mutation were

serially diluted. The data were taken from six independent experiments. Phos-

phorlmager analysis (see ‘Materials and Methods”) showed linear reproducibility

of the quantitative assay. Taking the frequency of mutation detected in the 10

dilution as the reference value of I .0. we found that the average value was

0.092 ± 0.006 at 10� dilution and 0.01 1 ± 0.002 at l0� dilution. The estimated

numbers of mutant alleles present in each tube of assay are 80. 8. and 0.8 for

dilutions of � � and l0�. respectively.

0.1

Fig. I . Detection of CC to Tt mutations at the 247/248 codons of the p53 gene

by AS-PCR under various conditions. A, AS-PCR was carried out at different

annealing temperatures to determine the optimal temperature. WT. wild-type. B.

AS-PCR performed with purified (P) or unpurified (NP) mutant-specific primers.

Cancer Epidemiology, Biomarkers & Prevention 435

63#{176}C�

64#{176}C

65#{176}C�

66#{176}C�

67#{176}CI

iO-�� 1O� ________P P NP P NI”

Under the same PCR conditions (64#{176}Cas the optimal annealingtemperature), AS-PCR was carried out using purified and un-purified primers. The best results were obtained with purifiedprimers; the purity of primers used for the second PCR reaction

seemed to be critical for obtaining quantitatively reliable results

(Fig. 1B).

Quantitative Validation of the AS-PCR Method. The AS-PCR reaction was carried out under the optimal conditions

determined above on six dilutions of mutated DNA. By Phos-phorlmager scanning, mutant p53 was detected at l0_6. Theresults illustrated in Fig. 2 show a linear reproducibility of thequantitative measurements. Taking the frequency of mutation

detected in the 1O�� dilution as the reference value of 1.0, we

found that the average value was 0.092 ± 0.006 at the i05dilution and 0.01 1 ± 0.002 at the l06 dilution. Therefore, up

to the 1O_6 dilution, our assays seem to be linear. At a lO_6

dilution overall, the average value is roughly linear, but theresults from individual experiments vary significantly. This was

expected, because, at the lO6 dilution, most reaction tubescontain only one mutant allele and others contain none (esti-mated numbers of mutant alleles present in each tube of assayare 80, 8, and 0.8 for dilutions of lO�, lO�, and lOo,

respectively).

Measurement of CC-�TT Mutation at Codons 247/248 ofthe p53 Gene in Japanese Normal Skin Samples; Relation-ship to Anatomical Site of Biopsy, Sex, Age, and the Type of

Skin Lesions of Patients. A quantitative assay was performedon normal skin samples from benign skin tumors and patientswith premalignant tumors or NMSC in Japan, and the values

obtained from Phosphorlmager analysis were converted to real

io� io�’

concentrations (mutation frequencies) by using the calibration

standard.Based on the results described above (calibration stand-

ard), we have taken the value of lO� as the threshold ofdetection for analysis of the subsequent data. On the other hand,when we examined the images (which were obtained after

scanning the gels by Phosphorlmager) by naked eyes, thesample judged to be positive (A. 0. and H. Y., consensusjudgment) contained a mutation frequency of more than 5.0 X

lO�. This is likely to occur, because at the mutation frequencyof 5 .0 x 10 � �, we expected that at least one of three assay tubes

would contain a mutant. Therefore, we also used this as a limit

of detection in our calculation.The p53 mutation frequency in samples from sun-exposed

and nonexposed sites for males and females is shown in Fig. 3.Statistical analysis indicated a significantly higher p53 muta-tion frequency in the samples taken from sites that are chron-

ically exposed to the sun than in those from covered sites(Student t test, P = 0.005) when we set the threshold at 5.0 X

io�. When we set the threshold at l0_6, there was a tendencyof an increased p53 mutation frequency with sun exposure to

the anatomic site (Wilcoxon test, P = 0.01).Taking 5.0 x � as a threshold, there was also a sig-

nificant increase in p53 mutation frequency with the age ofsubjects (Fisher’s r to z test, P 0.02 and correlation coeffi-cient = 0.47), suggesting the accumulation of the CC to iTmutation as a result of chronic sun exposure over the lifetime of

the patient (Fig. 4). However, when we set a threshold at I0�6,

only samples from subjects more than 70 years of age containedp53 mutation.

To see whether normal skin from patients with skin cancerand from those with only benign tumors had accumulated p53gene mutations to different degrees, we compared the p53mutation frequency found in samples taken from patients with

premalignant tumors or NMSC and in patients with only benign



3.0

‘-4

x

�0)

0)ha

4.

0

..�

40 1.0’

A

A

A

3.0

2.5

2.0

1.5

1.0

0.5

x

U

0)

0)

0

A

A

A

I

0

0

0

0

0

0

0

SC

Covered site Exposed site

Fig. 3. Measurement of the CC-s’I’T mutation at codons 247/248 of the p53

gene in Japanese skin samples; relationship to the anatomical site of biopsy and

the sex of the patients. The quantitative AS-PCR assays were applied to DNA

extracted from the normal skin of exposed and covered sites taken from patients

of various ages.

Patients with

benign tumors

0

3.0

0‘--4

x 2.5

U� 2.00)

0�� 1.5

‘4-4

.� 1.0

�0.5

0

00

436 Quantitation of p53 Mutation in Normal Human Skin

A MaleA Female

O�OO.5 , I ‘ I� ‘ ��1’ ‘ #{182}�M�.? �

20 40 60 80 100

Age of patients (years)

Fig. 4. Measurement of the CC-sTT mutation at codons 247/248 of the p53

gene in Japanese skin samples; relationship to the age of patients.

tumors (Fig. 5). Setting the threshold at 5.0 X i0�, the patientswith premalignant tumors or NMSC [(8.6 ± 1.0) X 10�]

showed a significantly higher average of p53 mutation fre-quency than those with benign [(5.3 ± 0.2) X i0�] tumors(Student t test, P 0.009). On the other hand, when we set a

threshold at 10_6, there was a tendency of increased p53mutation frequency in patients with premalignant tumors orNMSC (Wilcoxon test, P = 0.01).

Discussion

Several approaches have been developed and successfully used

for analysis of mutant alleles that are present at low frequencies.The frequency of mutations in normal tissue is expected to bevery low, because mutations are present in only a very small

fraction of the overall cell population, in contrast to tumortissues, in which mutant alleles are expected to appear at veryhigh frequency (1, 23, 24). Thus, when PCR is performed onDNA from normal tissues, the majority of the amplified mate-

Patients withpremalignant tumors

or NMSC

Fig. 5. Measurement of the CC-siT mutation at codons 247/248 of the p53

gene in Japanese skin samples; relationship to the type of lesions of subjects. See

Table 1 for information about samples with respect to the type of skin tumors.

rial is expected to be normal rather than mutant. This underlinesthe need for sensitive methods that permit selection of themutant alleles over the normal ones, if we have to identify

carcinogen-induced mutations at a very early stage of carcino-genesis.

0 In the present study, we have improved and refined the

0 AS-PCR method to detect the UV-specific aaCCgg-�aaiTgg0 tandem-base mutation at codons 247/248 of the p53 gene,

corresponding to an Asn-Arg-�Asn-Trp amino acid change(21). We chose to detect this particular mutation because it is

found frequently in human NMSC (21, 24). Furthermore, wechose a tandem-base mutation rather than a single-base muta-tion because these are known to be specifically induced by UVradiation (16, 17, 22 25, 26) and because they can be moreaccurately and specifically detected by PCR than single-base

substitutions.We considered several PCR parameters in refining the

method. Besides the usual parameters that affect conventionalPCR (27) such as the polymerase enzyme sources, designing ofspecific primers by computer program, and optimization of theannealing temperature (in our study, 64#{176}C),we found that theuse of high-performance liquid chromatography-purified prim-ers clearly gave increased sensitivity of detection comparedwith that of nonpurified primers.

These scrupulous refinements enabled us to establish op-timal conditions for the AS-PCR quantitative assay, permitting

the detection of p53 mutations at frequencies as low as l0_6.The amount of template DNA used in the PCR was limited to

5 ,.tg, corresponding to the amount of DNA in about 8.0 X i0�cells. Therefore, our assay was in fact optimized up to thetheoretical limit of detection (at 10_6 dilutions, which contains0.8 mutant); there should be no detectable mutant after furtherdilution. At the early phase of this study, we examined theimages (obtained after scanning of gels by Phosphorlmager) bynaked eyes and judged whether the sample was positive or

negative. Such analysis indicated that those samples judged tobe positive (A. 0. and H. Y., consensus judgment) contained amutation frequency of more than 5.0 X i07. Because thereaction tube contained a DNA amount equivalent to no more



Cancer Epidemiology, Biomarkers & Prevention 437

than 106 cells, it seems that we pushed the assay to the theo-retical limit of detection. This is therefore an extremely sensi-

tive mutation assay that has the potential to detect an early

genetic alteration occurring in normal cells.We applied the assay to study the correlation between

sun exposure and accumulation of CC to TT mutations in

normal skin from Japanese patients with benign tumors andpatients with premalignant tumors or NMSC. The resultsconfirm the positive association between UV exposure of

body sites and p53 mutation frequency, which probably

constitutes an early event in the multistage process of car-

cinogenesis (28, 29). However, the incidence of CC to TTtandem-base mutations at codons 247/248 of the p53 gene in

Japanese patients was much lower than that found in Aus-tralian patients (12% versus 63.6% in exposed sites and 0%

versus 0% in covered sites), even though the samples were

taken from chronically exposed sites (face, neck, and hands)in Japanese patients and possibly less-exposed sites (shoul-

ders) in Australian subjects. In the calculation, we defined

Japanese samples with the mutation frequency of > 10_6 aspositives. Because this p53 mutation can be considered a

molecular signature of UV radiation, this difference in in-

cidence may be due to the high level of solar UV exposurein the Australian subjects (9, 12) and/or to a genetic differ-

ence in susceptibility to UV-induced mutation including skin

pigmentation and repair capacity of DNA photodamage (14,20, 26, 30, 3 1 ). It is well established that the prevalence of

skin cancer is low among Japanese in comparison to Cau-casians (32). A high frequency ofp53 mutation (>l0_6) was

found only in normal skin from chronically sun-exposed

sites of patients more than 70 years of age, suggesting that

these p53 mutations accumulate as a result of chronic sunexposure during daily life (28, 33). In addition, photodamage

may be more slowly repaired in aged skin, leading to fasteraccumulation of lesions and thus mutations (34, 35).

In this study, one of six XP patients (who are extremely

susceptible to UV and deficient in DNA repair capacity) Un-

expectedly showed a high prevalence of CC to iT p53 muta-

tions not only at exposed body sites but also at a covered site

(data not shown). Tandem CC to iT mutations can be induced

by oxidants (28, 36, 37). Such oxidative damage and the pa-

tient’s inability to repair it may explain the accumulation of p53mutations at the covered site. On the other hand, this particular

subject was diagnosed as a XP-variant patient when he was 44years old, and he had been practicing outdoor sports, especially

golf, before his diagnosis. Therefore, it is possible that his

abdomen (biopsy site) had also been exposed to UV.We have already proposed that the measurement of p53

mutations can be a useful and biologically relevant measure of

UV exposure in humans and a possible predictor of risk for skincancer (2 1). Two observations from our laboratories suggest

that CC to iT mutations at codons 247/248 of the p53 gene may

serve as a good skin cancer predictor but not as a good measureof UV exposure: (a) the frequency of this mutation increases in

UV-exposed cultured cells after serial passages, suggesting thatcells containing this mutation have a growth advantage over

normal cells (21); and (b) in the present study, this mutation

was found at a higher frequency in patients with premalignant

tumors or NMSC than in subjects with only benign tumors. Itis, however, necessary to conduct a better controlled case-control study before concluding that p53 mutation frequency is

a good indicator for predicting the risk for NMSC. In addition,it is necessary to develop a similar method to detect p53

mutations that do not result in a growth advantage or mutations

in housekeeping genes that can then serve as a method to assessintrinsic UV exposure per Se.

Acknowledgments

We thank Nicole Martel for technical assistance, Dr. John Cheney for editorial

reading, Dr. Jacques Esteve for helpful discussion on statistical analysis of data,

and Chantal D#{233}chauxfor secretarial help.

References

I . Yamasaki, H., Galiana, C., and Nakazawa, H. Do genetic alterations found in

tumors reflect exposure to carcinogen? In: 0. H. Iversen (ed), Cancer Causation:

Exploring New Frontiers, pp. 153-166. Washington. DC: Taylor & Francis, 1993.

2. Mao L., and Sidransky, D. Cancer screening based on genetic alterations in

human tumors. Cancer Res. 54: 1939-1940, 1994.

3. Rees J. L. p53 and the origins of skin cancer. J. Invest. Dermatol., 104:

883-884, 1995.

4. Weinberg. R. A. Oncogenes, tumour suppressor genes, and cell transforma-

tion: trying to put it all together. In: J. Brugge. T. Curran, E. Harlow, and F.MacCormick (eds.), Origins of Human Cancer, pp. 1-16. Cold Spring Harbor,

NY: Cold Spring Harbor Laboratory, I 991.

5. Ziegler. A., Jonason, A. S., Leffell, D. J., Simon, J. A., Sharma, H. W.,

Kimmelman, J., Remington, L., Jaks, T., and Brash, D. E. Sunbum and p53 in the

onset of skin cancer. Nature (Lond.), 372: 773-776, 1994.

6. Hussain, S. P., Aguilar, F., and Cerutti, P. Mutagenesis of codon 248 of the

human p53 tumor suppressor gene by N-ethyl-N-nitrosourea. Oncogene. 9: 13-

18, 1994.

7. Pourzand, C.. and Cerutti, P. Genotypic mutation analysis by RFLP/PCR.

Mutat. Rca., 288: 1 13-121, 1993.

8. Ronai, Z., and Yakubovskaya, M. PCR in clinical diagnosis. J. Clin. Lab.

Anal., 9: 269-283, 1995.

9. Armstrong. B. K., and Kricker, A. Skin cancer. Dermatol. Clin., 13: 583-594,

I 995.

10. Green, A. Changing pattems in incidence of non-melanoma skin cancer.

Epithelial Cell Biol., I: 47-51, 1992.

I I . IARC Monographs on the Evaluation of Carcinogenic Risks to Humans:

Solar and Ultraviolet Radiation, pp. 43-279. Lyon. France: IARC, 1992.

12. Armstrong. B. K. Stratospheric ozone and health. Int. J. Epidemiol.. 23:

873-885, 1994.

13. Dumaz, N., Drougard, C., Sarasin, A., and Daya-Grosjean, L. Specific

UV-induced mutation spectrum in the p53 gene of skin tumors from DNA

repair-deficient xeroderma pigmentosum patients. Proc. NatI. Acad. Sd. USA,

90: 10529-10533, 1993.

14. Dumaz, N., Stary, A., Soussi, T., Daya-Grosjean, L., and Sarasin, A. Can we

predict solar ultraviolet radiation as the causal event in human tumours by

analysing the mutation spectra of the p53 gene. Mutat. Res., 307: 375-386, 1994.

15. Kanjilal, S., Strom, S. S., Clayman, G. L., Weber, R. S., El-Naggar, A. K.,

Kapur. V.. Cummings, K. K., Hill, L. A., Spitz. M. R., Kripke, M. L., and

Ananthaswamy. H. N. p53 mutations in nonmelanoma skin cancer of the head andneck: molecular evidence for field cancerization. Cancer Res., 55: 3604-3609,

1995.

16. Brash, D. E., Rudolph, J. A., Simon, J. A., Lin, A., MacKenna, 0. J., Baden,

H. P., Halperin, A. J., and Pont#{233}n,J. A role for sunlight in skin cancer: UV-

induced p53 mutations in squamous cell carcinoma. Proc. NatI. Acad. Sci. USA,

88: 10124-10128, 1991.

17. Rady, P., Scinicariello, F., Wagner, R. F., and Tyring, S. K. p53 mutations in

basal cell carcinomas. Cancer Res., 52: 3804-3806, 1992.

18. Tomaletti, S., Rozek, D., and Pfeifer, G. P. The distribution of UV photo-products along the human p53 gene and its relation to mutations in skin cancer.

Oncogene, 8: 2051-2057, 1993.

19. Amstad, P., Hussain, S. P., and Cerutti, P. Ultraviolet B light-induced

mutagenesis of p53 hot-spot codons 248 and 249 in human skin fibroblasts. Mol.

Carcinog., 10: 181-188, 1994.

20. Sage. E. Distribution and repair of photolesions in DNA: genetic conse-

quences and the role of sequence context. Photochem. Photobiol., 57: 163-174,

1993.

21 . Nakazawa, H., English, D., Randell, P. L., Nakasawa, K., Martel, N.,Armstrong, B. K., and Yamasaki, H. UV and skin cancer: specific p53 gene

mutation in normal skin as a biologically relevant exposure measurement. Proc.

NatI. Acad. Sci. USA, 91: 360-364, 1994.

22. Ziegler, A. M., Leffell, D. J., Kunala, S., Sharma, H. W., Gailani, M., Simon.

J. A., Halperin, A. J., Baden, H. P., Shapiro, P. E., Bale, A. E., and Brash, D. E.



438 Quantitation of p53 Mutation in Normal Human Skin

Mutation hot spots due to sunlight in the p53 gene of nonmelanoma skin cancers.

Proc. Nail. Acad. Sci. USA, 90: 4216-4220, 1993.

23. de Fromentel, C., and Soussi, T. The p53 tumor suppressor gene: a model forinvestigating human mutagenesis. Genes Chromosomes & Cancer, 4: 1-4, 1992.

24. Hollstein, M., Sidransky, D., Vogelstein, B., and Harris, C. p53 mutations in

human cancers. Science (Washington, DC), 253: 49-53, 1991.

25. Pierceall, W. E., Mukhopadhyay, T., Goldberg, L. H., and Ananthaswamy,H. N. Mutations in the p53 tumor suppressor gene in human cutaneous squamous

cell carcinomas. Mol. Carcinog., 4: 445-449, 1991.

26. Sato M., Nishigori, C., Zghal, M., Yagi, T., and Takebe, H. UV-specific

mutations in p53 gene in skin tumors in XP patients. Cancer Res., 53: 2944-2946,

1993.

27. Gelfand. D. H., and While, T. J. Thermostable DNA Polymerases in PCR

Protocols: A Guide to Methods and Applications. p. 129. New York: Academic

Press, 1990.

28. Berg, R. J. W., Van Kranen, H. J., Rebel, H. G., Dc Vries, A., Van Vloten,

W. A., Van Kreijel, C. F., Van der Leun, J. C., and Gruijl, F. R. Early p53

alterations in mouse skin carcinogenesis by UVB radiation: immunohistochem-

ical detection of mutant p53 protein in clusters of preneoplastic epidermal cells.

Proc. Nail. Acad. Sci. USA, 93: 274-278, 1996.

29. Ren, Z-P., Hedrum, A., Pont#{233}n,F., Nist#{233}r,M., Ahmadian, A., Lundeberg, J.,

Uhl#{233}n,M., and Pont#{233}n,J. Human epidermal cancer and accompanying precursorshave identical p53 mutations different from mutations in adjacent areas of

clonally expanded non-neoplastic keratinocytes. Oncogene, 12: 765-773, 1996.

30. Sage, E., Lamolet, B., Brulay, E., Moustacchi, E., Ch#{226}teauneuf, A., and

Drobetsky, A. Mutagenic specificity of solar UV light in nucleotide excision

repair-deficient rodent cells. Proc. Natl. Acad. Sci. USA, 93: 176-180, 1996.

31. Tomaletti, S., and Pfeifer, G. P. Slow repair ofpyrimidine dimers at p53 mutation

hot spots in skin cancer. Science (Washington, DC), 263: 1436-1438, 1994.

32. Chuang, T. Y., Reizneir, G. T., Elpern, D. J., Stone, J. L., and Farmer, E. R.

Nonmelanoma skin cancer in Japanese ethnic Hawaiians in Kauai, Hawaii: an

incidence report. J. Am. Acad. Dermatol.. 33: 422-426, 1995.

33. Jonason, A. S., Kunala, S., Price, G. J., Restifo, R. J., Spinelli, H. M., Persing,

J. A., Leffell, D. J., Tarone, R. E., and Brash, D. E. Frequent clones of p53-mutated keratinocytes in normal human skin. Proc. Natl. Acad. Sci. USA, 93:

14025-14029, 1996.

34. Green, A. C. Premature ageing of the skin in a Queensland population. Med.

J. Aust., 155: 473-474, 1991.

35. Urano, Y., Asano, T., Yoshimoto, K., Iwahana, H., Kubo, Y., Kato, S.,

Sasaki, S., Takeuchi, N., Uchida, N., Nakanishi, H., Anise, S., and Itakura, M.

Frequent p53 accumulation in the chronically sun-exposed epidermis and clonal

expansion of p53 mutant cells in the epidermis adjacent to basal cell carcinoma.

J. Invest. Dermatol., 104: 928-932, 1995.

36. Reid, T., and Loeb, L. Mutagenic specificity of oxygen radicals produced by

human leukemic cells. Cancer Res., 52: 1082-1086, 1992.

37. Reid, T., and Loeb, L. Tandem double CC to TT mutations are produced

by reactive oxygen species. Proc. Natl. Acad. Sci. USA, 90: 3904-3907,

1993.



1997;6:433-438. Cancer Epidemiol Biomarkers Prev A Ouhtit, M Ueda, H Nakazawa, et al. normal skin from Japanese patients.Quantitative detection of ultraviolet-specific p53 mutations in

Updated version

http://cebp.aacrjournals.org/content/6/6/433

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://cebp.aacrjournals.org/content/6/6/433To request permission to re-use all or part of this article, use this link



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

mailto:[email protected]



quantitative detection of ultraviolet-specific p53 ... · suppressor gene, p53, is the most often...

Documents