Molecular Mechanisms of Basal Cell and Squamous Cell Carcinomas (Medical Intelligence Unit) 0387260463, 9780387260464

Citation preview

MEDICAL INTELUGENCE UNIT

Molecular Mechanisms of Basal Cell and Squamous Cell Carcinomas Jorg Reichrath, Prof. Dr. med. The Saarland University Hospital Clinic for Dermatology, Allergology and Venerology Homburg/Saar, Germany

L A N D E S B I O S C I E N C E / EUREKAH.COM

GEORGETOWN, TEXAS

USA

SPRINGER SCIENCE+BUSINESS M E D I A

NEW YORK, NEW YORK

U.S.A.

MOLECULAR MECHANISMS OF BASAL CELL AND SQUAMOUS CELL CARCINOMAS Medical Intelligence Unit Landes Bioscience / Eurekah.com Springer Science+Business Media, Inc. ISBN: 0-387-26046-3

Printed on acid-free paper.

Copyright ©2006 Landes Bioscience and Springer Science+Business Media, Inc. All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher, except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in the publication of trade names, trademarks, service marks and similar terms even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. While the authors, editors and publisher believe that drug selection and dosage and the specifications and usage of equipment and devices, as set forth in this book, are in accord with current recommendations and practice at the time of publication, they make no warranty, expressed or implied, with respect to material described in this book. In view of the ongoing research, equipment development, changes in governmental regulations and the rapid accumulation of information relating to the biomedical sciences, the reader is u r ^ d to carefiilly review and evaluate the information provided herein. Springer Science+Business Media, Inc., 233 Spring Street, New York, New York 10013, U.S.A. http://www.springer.com Please address all inquiries to the Publishers: Landes Bioscience / Eurekah.com, 810 South Church Street, Georgetown, Texas 78626, U . S A Phone: 512/ 863 7762; FAX: 512/ 863 0081 http://www.eurekah.com http://www.landesbioscience.com Printed in the United States of America. 9 8 7 6 5 4 3 2 1

Library of Congress Cataloging-in-Publication Data Molecular mechanisms of basal cell and squamous cell carcinomas / [edited by] Jorg Reichrath. p. ; cm. ~ (Medical intelligence unit) Includes bibliographical references and index. ISBN 0-387-26046-3 (alk. paper) 1. Skin~Cancer~Genetic aspects. 2. Basal cell carcinoma—Genetic aspects. 3. Squamous cell carcinomaGenetic aspects. I. Reichrath, J. (Jorg), 1962- 11. Title. III. Series: Medical intelligence unit (Unnumbered : 2003) [DNLM: 1. Carcinoma, Basal Cell—genetics, 2. Carcinoma, Basal Cell—physiopathology. 3. Carcinoma, Squamous Cell—genetics. 4. Carcinoma, Squamous Cell—physiopathology. 5. Cytogenetics—methods. 6. Papillomavirus Infections—complications. Q Z 365 M 7 1 8 2006] RC280.S5M65 2006 616.99'477042-dc22

2006005864

About the Editor JORG REICHRATH is Professor of Dermatology and Deputy Director of the Clinic for Dermatology, Allergology and Venerology at the Saarland University Hospital in Homburg/Saar, Germany. Main research interests include carcinogenesis and treatment of non-melanoma skin cancer and melanoma. He is a member of numerous national and international scientific organizations, including the German Dermatological Society ( D D G ) , the Deutsche Krebsgesellschaft (DKG), the German Dermatologic Co-operative Oncology Group (DeCOG), and the European Society of Dermatological Research (ESDR). Jorg Reichrath received his academic degrees (Dr. med., venia legendi) from the Saarland University, Germany.

Dedication I dedicate this book to Univ.-Prof. Dr. med. Wolfgang Tiigen, my teacher in dermato-oncology.

CONTENTS Preface 1. The Epidemiology of Basal Cell and Squamous Cell Carcinoma Hao Wang and Thomas L. Diepgen Incidence of NonmelanomaSkin Cancer (NMSC) Trends in Incidence of BCC and SCC Risk Factors Ultraviolet Radiation (UVR) Ageing Smoking Alcohol Diet Medical Conditions Ionizing Radiation Occupation Chemical Carcinogens History and Precursor Lesions Skin Cancer Prevention Strategies 2.

3.

4.

Histology of Epithelial Skin Tumors Jorg Reichrath andKerstin Querings Squamous Cell Carcinoma Basal Cell Carcinoma Basosquamous Carcinoma UV Damage and D N A Repair in Basal Cell and Squamous Cell Carcinomas Knuth Rass UV Damage and DNA Repair UV-Induced Carcinogenesis in Squamous Cell Carcinoma (SCC) UV-Induced Carcinogenesis in Basal Cell Carcinoma (BCC) Papillomavirus Infections and Cancerogenesis of Squamous Cell and Basal Cell Carcinomas Guido Bens Basal Cell Carcinomas

5. The Immune System and Nonmelanoma Skin Cancers Elma D. Baron Epidemiologic Evidence from Organ Transplant Recipients Evidence from Studies on UV-Induced Immune Suppression and Nonmelanoma Cancer Formation Mechanisms Involved in Immune Suppression and Carcinogenesis of Nonmelanoma Skin Tumors

XV

1 1 3 3 3 4 4 4 4 5 5 5 6 6 6 10 10 15 16

18 19 23 25

31 39 43 43 44 45

6. Cytogenetics of Basal Cell Carcinoma and Squamous Cell Carcinomas MelanieA. CarlesSy Kevin]. Ashton andLyn R Griffiths Cytogenetic Analysis Ploidy Analysis Classical Karyotypic Analysis Fluorescence in Situ Hybridization Analysis and Comparative Genomic Hybridization Analysis Implications of Cytogenetic Findings and Future Developments

49 49 50 51 51 53

7. The Hedgehog Signaling Pathway and Epithelial Skin Cancer Julia Reifenherger The Hedgehog Signaling Pathway Alterations of Hedgehog Signaling in Skin Tumors Mouse Models for Basal Cell Carcinoma Novel Pathogenesis-Based Approaches for Prevention and Therapy of Basal Cell Carcinomas

58

8. p53 Protein and Nonmelanoma Skin Cancer Vladislava O. Melnikova and Honnavara N. Ananthaswamy p53 Tumor Suppressor Gene Induction of p53 Protein by UV and Biologic Consequences A Model of UV-Induced Initiation and Progression of Squamous Cell Carcinomas p53 Mutations in Nonmelanoma Skin Cancers p53 Mutations in Mouse Skin Cancers

66

9. TGF-p Pathway and Cancerogenesis of Epithelial Skin Tumors Miguel Quintanilla, Eduardo Pirez-Gdmez, Diana Romeroy Mar Pons and]aime Renart TGF-p Signalling Role of TGF-P in Epidermal Development Role of TGF-P in Epidermal Proliferation and Homeostasis Role of TGF-p in Epithelial Skin Cancer

80

10. PDGF Pathways and Growth of Basal Cell and Squamous Cell Carcinomas Jingwu Xie Signal Transduction by PDGFs and Their Receptors In Vivo Functions of PDGFs and Their Receptors PDGF Alterations in Cancer The Role of PDGF for Cell Proliferation in BCCs The Role of P D G F A ^ G F Signaling in Angiogenesis of SCCs PDGF Signaling in Dermatofibrosarcoma Protuberans and Giant Cell Fibroblastoma Perspectives

59 60 62 63

66 69 69 69 72

80 82 84 86

94 94 95 97 99 102 104 104

11. Apoptosis and Cancerogenesis of Basal Cell and Squamous Cell Carcinoma Peter Erb, Jingmin Jiy Marion Wemli and Stanislaw A. Buchner Gene Mutations Lead to NMSCs Apoptosis Is Pivotal for the Removal of DNA Damaged and Transformed Cells UV-Light Changes the Expression Pattern of Apoptosis-Inducing and -Preventing Molecules in Skin Epidermis FasL, the Double-Edged Sword Downregulation of FasL on BCC by Antisense Oligonucleotides or Small Interfering RNAs (siRNAs) Outlook 12. The Role of Telomerase for Cancerogenesis of Basal Cell and Squamous Cell Carcinomas Eva-Maria Fahricius Interaction of Telomeres and Telomerase Telomerase Activating in Normal Skin Telomerase Activation in the Nonmelanoma Skin Tumors BCC and s e c and in Their Tumor-Free Margins Index

108 109 110 110 Ill 112 113

115 116 119 121 135

EDITOR Jorg Reichrath The Saarland University Hospital Clinic for Dermatology, Allergology and Venerology Homburg/Saar, Germany Email: [email protected] Chapter 2

CONTRIBUTORS^ Honnavara N. Ananthaswamy Department of Immunology The University of Texas M.D. Anderson Cancer Center Houston, Texas, U.S.A. Email: [email protected] Chapter 8 Kevin J. Ashton Genomics Research Centre School of Health Science Griffith University Gold Coast Bundall, Queensland, Australia Chapter 6 Elma D. Baron Department of Dermatology Skin Study Center University Hospitals of Cleveland Case Western Reserve University Cleveland, Ohio, U.S.A. Email: [email protected] Chapter 5 Guido Bens The Saarland University Hospital Clinic for Dermatology, Allergology and Venerology Homburg/Saar, Germany Email: [email protected] Chapter 4

Stanislaw A. Biichner Department of Dermatology University Hospitals Basel, Switzerland Chapter 11 Melanie A. Carless Genomics Research Centre School of Health Science Griffith University Gold Coast Bundall, Queensland, Australia Chapter 6 Thomas L. Diepgen Department of Clinical Social Medicine, Occupational and Environmental Dermatology University Hospital Heidelberg Heidelberg, Germany Email: [email protected] Chapter 1 Peter Erb Institute for Medical Microbiology University of Basel Basel, Switzerland Email: [email protected] Chapter 11

Eva-Maria Fabricius Clinic for Oral and Maxillofacial Surgery Campus Virchow Hospital Medical Faculty of the Humboldt University of Berlin Charit^ Berlin, Germany Email: [email protected] Chapter 12 Lyn R. Griffiths Genomics Research Centre School of Health Science Griffith University Gold Coast Bundall, Queensland, Australia Email: [email protected] Chapter 6 Jingmin Ji Institute for Medical Microbiology University of Basel Basel, Switzerland Chapter 11 Vladislava O. Melnikova Department of Immunology The University of Texas M.D. Anderson Cancer Center Houston, Texas, U.S.A. Chapter 8 Eduardo P^rez-G6mez Instituto de Investigaciones Biom^dicas Alberto Sols Consejo Superior de Investigaciones Cientfficas-Universidad Aut6noma de Madrid Madrid, Spain Chapter 9 Mar Pons Instituto de Investigaciones Biom^dicas Alberto Sols Consejo Superior de Investigaciones Cientfficas-Universidad Autdnoma de Madrid Madrid, Spain Chapter 9

Kerstin Querings The Saarland University Hospital Clinic for Dermatology, Allergology and Venerology Homburg/Saar, Germany Chapter 2 Miguel Quintanilla Instituto de Investigaciones Biom^dicas Alberto Sols Consejo Superior de Investigaciones Cientfficas-Universidad Aut6noma de Madrid Madrid, Spain Email: [email protected] Chapter 9 Knuth Rass The Saarland University Hospital Clinic for Dermatology, Allergology and Venerology Homburg/Saar, Germany Email: [email protected] Chapter 3 Jtdia Reifenberger Department of Dermatology Heinrich-Heine-Universitat Dusseldorf, Germany Email: reifenbergerj @med. uni-duesseldorf de Chapter 7 Jaime Renart Instituto de Investigaciones Biom^dicas Alberto Sols Consejo Superior de Investigaciones Cientfficas-Universidad Aut6noma de Madrid Madrid, Spain Chapter 9

Diana Romero Instituto de Investigaciones Biom^dicas Alberto Sols Consejo Superior de Investigaciones Cientfficas-Universidad Aut6noma de Madrid Madrid, Spain Chapter 9 Hao Wang Department of Clinical Social Medicine, Occupational and Environmental Dermatology University Hospital Heidelberg Heidelberg, Germany Chapter 1

Marion Wernli Institute for Medical Microbiology University of Basel Basel, Switzerland Chapter 11 Jingwu Xie Sealy Center for Cancer Cell Biology Department of Pharmacology and Toxicology University of Texas Medical Branch Galveston, Texas, U.S.A. Email: [email protected] Chapter 10

PREFACE

R

apid progress in the understanding of carcinogenesis and pathology of epitheUal skin cancer has led to new strategies for the prevention and treatment of these malignancies. The goal of this volume is to comprehensively cover in a highly readable overview our present knowledge of pathogenetic mechanisms and molecular biology of Basal Cell and Squamous Cell Carcinomas. Topics that are discussed in-depth by leading researchers and clinicians range from the newest findings in epidemiology, histology, photobiology, immunology, cytogenetics, and molecular pathology to new concepts for prophylaxis and treatment. Experts in the field as well as health care professionals not intimately involved in these specialized areas are provided with the most significant and timely information related to these topics. It is the aim of this book to summarize essential up-to-date information for clinicians and scientists interested in the biology of Basal Cell and Squamous Cell Carcinomas. The chapters are written by authors who are experts in their respective research areas, and I am gratefixl for their willingness to contribute to this book. I would also like to express my thanks to Ron Landes, Cynthia Conomos, Sara Lord and all the other members of the Landes Bioscience staff for their expertise, diligence and patience in helping me complete this work. Jorg Reichrathy Prof, Dn med.

CHAPTER 1

The Epidemiology of Basal Cell and Squamous Cell Carcinoma Hao Wang and Thomas L. Diepgen* Abstract

B

asal cell and squamous cell carcinoma (nonmelanoma skin cancer = NMSC) are now the most common type of cancer in the Caucasian population, and the incidence of skin cancer has reached epidemic proportions. The highest incidence rates (IR) were reported from population-based studies in Australia with an IR of more than 2% for basal cell carcinoma (BCC) in males (females 1.1%), and 1.3% for squamous cell carcinoma (SCC) (females 0.7%). In this chapter, current epidemiologic data concerning the incidence and its worldwide trends, risk factors, like UV-radiation, ionizing radiation, predisposing host conditions, ageing, smoking, alcohol, diet, medical conditions, occupation, chemical carcinogenes, as well as important aspects of prevention will be discussed.

Incidence of Nonmelanoma Skin Cancer (NMSC) Nonmelanoma skin cancer (NMSC) represents two types of malignant tumors of the skin: basal cell carcinoma (BCC) and squamous cell carcinoma (SCC). They belong to the most common cancers in the world. Both arise from the epidermal tissue of the skin: SCC from epidermal keratinocytes, and BCC from the basal cells of the epidermis. BCC is more common with a ratio of 4:1 to SCC. Although BCC and SCC can be lethal, they are not associated with significant mortality; nevertheless the associated morbidity and therapeutic costs are an increasing burden to the health care system. Of the top-10 health priorities in the U.S.A. for this decade, NMSC ranks as number eight.^ Because of its relatively low mortality, NMSC is not registered in most cancer surveillance systems. The way of reporting to most tumor registries is also not consistent, therefore actual incidence rates are not always easy to estimate from these registries and can more accurately be obtained by population-based studies and surveys. Such studies have mainly been performed in Australia and the U.S.A. Table 1 clearly shows the relatively high incidence rates of NMSC; except for Singapore. These incidence rates are reported from countries or regions which have predominantly a white population. Countries that are closer to the Equator have much higher incidence rates, while incidence rates in males are consistendy higher than in females. In all these, mosdy white, populations BCC is more common. The highest rates are reported from Australia, with about twice as many BCC's compared to SCC's. Annual incidences of NMSC in north Queensland reach more than 2000 BCC's per 100,000 for men and more than 1100 BCC's per 100,000 for women. For SCCs the rates for *CorrespondIng Author: Thomas L. Diepgen—Department of Social Medicine, Occupational and Environmental Dermatology, University Hospital Heidelberg, ThibautStr. 3, 69115 Heidelberg, Germany. Email: [email protected]

Molecular Mechanisms of Basal Cell and Squamous Cell Carcinomas, edited by Jorg Reichrath. ©2006 Landes Bioscience and Springer Science+Business Media.

Molecular Mechanisms of Basal Cell and Squamous Cell Carcinomas

Table 1. Age-standardised incidence rate (per 100,000) at various world locations for men and women Men Period

Standardization (Age)

BCC

SCC

Women BCC SCC

world-standard population world-standard population world-standard population United States population United States population

1173 2058

600

629

1332

1195

407

81

212

228 755 26

309.9 930.3

97.2 356.2

165.5 485.5

32.4 150.4

United States population

935.9

270.6

497.1

112.1

world-standard population world-standard population world-standard population

128

25

105

43.7 53.6

11.2 11.2

31.7 44.0

9 4.4 5.3

Australia TotaP 1995 North Queensland"^ 1997 U.S.A.^ 1994 New Hampshire^^ 1993-1994 Northcentral 1998-1999 New Mexico^ "^ South-eastern 1996 Arizona^ Europe Wales, U.K.2^ 1998 Saarland, Germany^ 1995-1999 Schleswig-Holstein, 1998-2001 Germany^ Eindhoven, 1998-2000 The Netherlands^ ° Vaud, Switzerland^^ 1995-1998 Neuchatel, 1996-1998 Switzerland^'^^ Trentino, Italy^^ 1992-1997 Slovakia^"^ 1993-1995 Asia Chinese Singapore^^ 1993-1997

European standard population 63

58

world-standard population world-standard population

75.1

world-standard population world-standard population

72.7 38.0

23.4

6.7

53.9 29.2

3.8

world-standard population

6.4

3.2

5.8

1.8

28.9

78

66.6

17.1

56 11.2

men in north Queensland are more than 1300 for men and more than 700 for women per 100,000. Also the U.SA. has high incidence rates, with considerably more BCCs.^ Like m Australia, rates are much higher in areas that are closer to the equator.^ The incidence rate of NMSC, i.e., BCC and SCC combined, in the U.S.A. is estimated to be almost similar to the incidence of all other cancers combined.^ From the table it becomes clear that the rates in a number of European countries or regions, which tend to be at a higher latitude than Australia and the U.S.A., are substantially lower. However, it might also be possible that the incidence rate of nonmelanoma skin cancer is underestimated in Europe. Two regions in Germany^'^^ and the nearby Netherlands^^ have per 100,000 an incidence rate of BCC in the order of 43— 63 for men, and 32-58 for women. Rates for SCC are in Germany about 11 for men, and about 5 for women. It is interesting to note that the two areas in Switzerland^^' ^^ and the nearby Trentino region in northern Italy have higher rates than Germany and the Netherlands, probably because these Swiss and Italian regions are further south, i.e., have more sunlight. Incidence rates in Slovakia seem to be relatively low compared to the other European regions.^ Singapore has a predominandy Chinese, i.e., nonwhite, population, which may explain the low incidence rates. Differences in notification or detection of cases may account for some of the variability in the incidence rates. Another source of variability may be the different standard populations that were used for the age-standardisation: the US uses its own US standard population, the Netherlands the European standard, while the other studies use the worlds standard population. It is unlikely that these age adjustments account for major differences.

The Epidemiology of Basal Cell and Squamous Cell Carcinoma

Trends in Incidence of BCC and SCC The incidence of NMSC is rising in a number of countries. In addition to differences in incidence rates, there are also differences in the relative rates of change in the incidence of BCC compared to SCC. The increase of NMSC was obvious in the time span between 1979/80 and 1993/94 in New Hampshire, U.S.A.^^ In particular, the rate of SCC increased considerably: 253% among men and 350% among females. In men and women the rate of BCC increased by more than 80%. These changes were associated with increased exposure to sunlight, with the most prominent increase in incidence occurring among men on the trunk, and among females on the lower limbs. In New Mexico, U.S.A., marked increase in SCC rates were observed over the years between 1977 and 1999.^^ Between 1985 and 1995 incidence rates of NMSC increased in Australia, whereby the increase in incidence of SCC was higher than of BCC. Also very high incidence rates compared to northern parts of the U.S.A. are reported from south-eastern Arizone, U.S.A., whereby it seems that this high incidence is not increasing further. Especially the incidence of SCC declined between 1985 and 1996. Incidence rates for BCC increased steadily among men and women over the years 1976-1998 in the canton Vaud, Switzerland.^^ This region employs a uniform ascertaiment system for NMSC. Interestingly, there was a decline in the rates for SCC since 1990, after a levelling off in the late 1980s. In the study in nearby city Neuchatel this pattern was also reported. ^^ Downward trends of SCC over the past decades are also observed from Singapore, while the incidence of BCC increased on average by 3 % every year over the years 1968 to 1997.^^ A lower increase in the incidence of SCC and rising incidence of BCC is seen in Slovakia, over the year 1978-1995. Age-adjusted incidence rates of BCC have risen in the Netherlands since 1973; this was more pronounced among females. ^^ In males, there was a linear increase in rates, also affecting the younger birth cohorts, with indications that this trend will continue. Increasing risks in white populations are clearly associated with living closer to the equator, which points to exposure to sunlight as a main causal factor. This is supported by the fact that NMSC occurs mainly on sun exposed skin.

Risk Factors A combination of inherited and constitutional factors, with exposure to environmental factors determines the likelihood NMSC will occur in any individual. Skin colour and the response of the skin to sunlight are constitutional factors. This fact is obvious in Caucasians who have a combination of light skin and blue €JQS, and red or blond hair; many of them get a sunburn instead of a tan when they are exposed to direct sunlight. ^^ NMSC is uncommon in black populations, Asians and Hispanic. ^^^^

Ultraviolet Radiation (UVR) The major environmental cause of BCC and SCC is exposure to sunlight,^^ in particular the UV component of sunlight. Within the spectrum of UV, it is mainly the UVB (wavelengths 280-320 nm) that is carcinogenic, while the UVA spectrum (320-400 nm) is carcinogenic to a lesser extent. Clinical studies and studies in migrants have shown the causal link between sunlight (i.e., UV) and NMSCs. NMSC is much less common in white populations who are permanently resident in high latitude regions, where daily exposure to sunlight is low. Those who migrate early in their life from such regions to lower latitudes increase their exposure levels to sunlight and show a higher risk of developing skin cancer. However, different profiles of UV exposure are important for BCC and SCC: For BCC the major risk factors are UV exposure during childhood and intense intermittent UV exposure. For SCC the risk factor is the chronic cumulative UV exposure.^^ This was shown in a recent large-scale population study, showing that a very high cumulative UV dose (> 145,000 kj/m^ within 6-year (1998-2003)) was associated with a doubling of the total numbers of tumors per person and a significantly increased risk of having SCC.^ In this study, participants who received very high doses had a BCC/SCC ratio of 2.1.

Molecular Mechanisms of Basal Cell and Squamous Cell Carcinomas

The face, neck and arms are the most common sites for NMSC, and these sites are the body areas that have the highest exposure to sunlight.^^ Data from Vaud cancer registry show that on the face the incidence rates for SCC are about 120 times higher than on the trunk (which is normally much less exposed to sunlight). For BCC the rates on the face are about 40-50 times higher than on the trunk.

Factors Affecting UV Radiation Changes in the environment and in lifestyle affect the UV radiation that reaches the skin of humans. One major concern is depletion of the ozone layer, because this layer is important in shielding against excess UVB. The ozone layer has decreased by about 2% over the past 20 years.^ For a 1% decrease in total column atmospheric ozone an increase of 2.7% in NMSC was to be expected. It was estimated that the UV radiation over lifetime due to a 2% decrease in ozone concentration will cause a 6-12% increase in NMSC in an exposed population. '^^ White populations living closer to the equator have a higher incidence of NMSC; the incidence of SCC doubles for very 8-10 degrees decline in latitude.

Artificial UV Radiation In white-skinned populations more and more persons, especially young women, are using sun tanning beds, either at home or in tanning-studios. The increased exposure to this kind of artificial UV may increase the risk of NMSC. In a recent study, the risk estimate for any use tanning devices was 2.5 for SCC and 1.5 for BCC.^^

Ageing With increasing age, there is an increasing cumulative exposure to UV radiation and a reduced capacity to repair DNA damage.^® This may be a reason of the exponential increase of the risk of NMSC with age. The incidence of SCC increases more rapidly with age than BCC. Among individuals over 75 years old the incidence of BCC was approximately 5 times higher, and the incidence of SCC was approximately 35 times higher compared to individuals 50-55 years of age.

Smoking Smoking and other types of tobacco use are clearly associated with SCC of the lip. In general, SCC is positively related to cigarette smoking in most studies,^^'^^ but not alL^^The risk of developing SCC by smoking is increased twofold.^ It is not clear whether there is a link between BCC and smoking: the majority of the large-scale studies (Male Health Professionals Follow-up Study, Nurses Health Study, US Radiological Technologists cohort study) have not detected an association.^ ' BCC in young women was associated with past or current history of smoking. In 51 women with BCC of the eyelid there was also an association with smoking. ^^ Smoking was associated with an increased prevalence of BCCs larger than 1.0 cm in diameter. ^^

Alcohol An association of alcohol consumption and BCC could not be demonstrated in case-control studies. However, three large cohort studies reported a significandy increased risk of BCC with increasing daily alcohol intake: the Male Health Professionals Follow-up Study, Nurses Health Study, US Radiological Technologists cohort study.^ ' The relationship between alcohol and SCC has not been specifically investigated.

Diet Dietary factors do not seem to be causally related to the risk of BCC in humans. Experimental studies in mice demonstrated a decreased latent period and increased number of skin tumors when the animals received a diet high in fat or with a high fraction of polyunsaturated fatty acids.^^

The Epidemiology of Basal Cell and Squamous Cell Carcinoma

The relationship between SCC and diet or serum levels of nutrients has been investigated by a few studies. A high intake of n-3 fatty acids was associated with a lower risk of SCC in a case-control study. ^ The incidence of SCC was not influenced in the intervention studies on the effect of beta-carotene supplementation.

Medical Conditions Patients receiving immunosuppressive treatment, especially recipients of a transplant, are known for their highly increased risk of skin cancer, especially SCC. The risk of BCC is about 10 times higher than in the general population. SCC is much more common among transplant recipients, with a SCC/BCC ratio of 4:1. With increasing sun exposure, the time of onset of the skin cancer after the transplantation is proportionally decreased. The cumulative incidence of SCC is about 30% at 10 years and 60% at 20 years after transplantation in Queensland, Australia. ^ In Europe these cumulative incidence rates are achieved about 10 years later. Treatment of skin diseases with PUVA, a combination of psoralen and UVA, increases the incidence of SCC; this is well documented in patients treated for psoriasis. '^ Additionally, several medical conditions such as chronic ulcers, burn scars, human papilloma virus infection, and various syndromes (e.g., xeroderma pigmentosum, albinism, epidermodysplasia verruciformis) can be associated with an increased risk of NMSC, especially SCC.

Ionizing Radiation Ionizing radiation has been shown to cause NMSC. For low-level radiation, an increased risk has been documented in uranium miners and radiologists. Among survivors of the nuclear bomb there is an increased risk of BCC. Risk of BCC is increased among persons exposed to occupational radiation, and among patients receiving therapeutic ionizing radiation before the age of 40. In patients having received ionizing radiation for benign skin disorders the risk of SCC is increased many years after having received this treatment.

Occupation At many working places increased exposure to natural and artificial UV-light theoretically could induce skin cancer. In Europe, the percentage of outdoor workers exposed to increased UV-radiation is estimated to be between 5% and 10% of the work force. In Germany, however, until now, the list of occupational diseases does not recognize skin cancer caused by UV light exposure during work as an occupational disease and therefore UV-light induced skin cancer by work can not be notified nor recognised as an occupational skin disease.^^ Occupational exposure to UV-radiation must be expected for outdoor workers and from technically generated UV radiation. There is a lack of objective evaluation of the occupational UV-radiation (UVR) dose of outdoor workers compared to indoor workers. Thieden et al developed a personal electronic UVR dosimeter in a wristwatch and measured continuously time-related UVR doses in standard erythema dose (SED). They could demonstrated, that the median estimated yearly UVR dose was 132 SEDs for indoor workers but 224 SEDs for gardeners. It has been estimated by the lARC that that outdoor workers had the 2 to 3 times higher UVR dose than indoor workers. In subjects older than 20 years, the UVR exposure is not related to age but to occupation, outdoor sports activities, or being a sun worshipper.^^ Diepgen and Drexler came to the conclusion that there is enough scientific and epidemiological evidence to support the idea to recognize squamous cell carcinoma induced by occupational UV-light exposure as an occupational disease and that for this skin cancer the epidemiological "proof" of an at least doubled risk (RR > 2) due to occupational UV-radiation can be demonstrated.^^ The clear dose response relationship supports these epidemiological findings. For the individual risk assessment an attributive UV-radiation > 40% due to occupational factors has to be demanded. Under those circumstances squamous cell carcinoma should be

Molecular Mechanisms of Basal Cell and Squamous Cell Carcinomas

recognized and compensated as an occupational disease. In summary members of several occupational groups appear to be at a higher risk of skin cancer. There is an increased risk of skin cancer amongst outdoor workers such as farmers, welders, watermen, police officers, physical education teachers, pilots and cabin attendants.^^ Although sun exposure is thought to be a common etiologic factor, some studies have suggested alternative explanations, such as irradiation from the welding arc in welders, nonionizing microwave frequency radiation from radar use in police officers, and cosmetic radiation in pilots and cabin attendants.

Chemical Carcinogens Skin cancers, especially SCC, can be caused by chemical agents, which can act as promoters or as initiators. SCC can typically be caused by exposure to polycyclic aromatic hydrocarbon compounds, which are present in products such as coal tar, pitch, and petroleum oils. (Benz-a-pyrene is one of the most powerful skin carcinogens). A number of chemicals act synergistically with UV radiation: these are photosensitizers. This mechanism has been investigated extensively in the treatment of skin diseases with PUVA, which, as mentioned above, is a combination of oral or topical administration of psoralen with UVA irradiation. There is a synergistic acceleration of the risk of skin cancer through cumulative DNA damage by a combination of exposure to UVA radiation, environmental carcinogens and benz[a]pyrene.^ Exposure to arsenic, not only occupational but also environmental via drinking water, is associated with an increased risk of skin cancer, especially SCC.^^

History and Precursor Lesions The risk of NMSC is increased in individuals with a personal or family history of precursor lesions. BCC arises de novo, which means that there are no known precursor lesions. Precursors for SCC include actinic keratosis, and Bowen's disease (actually SCC in situ). Actinic keratoses (AK) usually occur in fair-skinned individuals (skin type I or II) over 45 years of age. In Australia and in South-western US, relatively more younger people are affected. Approximately 5-20% of actinic lesions are reported to transform into SCC within 10-25 years.^ In individuals who have developed a NMSC the risk of appearance of a new skin cancer is increased. This risk of such a new NMSC is highest within the first year following the appearance of the original skin cancer.^^ Within five years of therapy of the first SCC, 52% of the participants of a follow-up study developed a subsequent NMSC. These findings are the basis of the recommendation to follow-up NMSC patients for at least 6 years.

Skin Cancer Prevention Strategies It is believed that 90% of nonmelanoma skin cancer and two thirds of melanomas may be attributed to excessice exposure to the sun.^^ The aim of primary skin cancer prevention is therefore to limit UV exposure. Campaigns to prevent skin cancer have incorporated numerous messages including the need to avoid sunburn and generally reduce exposure to ultraviolet radiation by staying out of the midday sun (between 11 a.m. and 3 p.m.), wearing protective clothing, seeking shade, and applying sunscreen.^ Another important public health message is that patients should promdy seek medical (dermatological) attention when they notice a suspicous or changing skin lesion. The detection of skin cancer at an early stage when it is most likely to be cured by simple outpatient excision, is classified as secondary prevention. UV-induced skin cancerogenesis is a multistep process that provides an excellent chance for effective prevention strategies to reduce the incidence, morbidity and mortality of skin cancer and its precursor lesions. In Australia, where skin cancer is of epidemic proportions, aggressive public health campaigns have been underway since the 1980s. In addition to to identifying tumors at an early stage, Australia managed an exciting educational program on photodamage prevention and sets standards for a wide variety of sun protective products to include sunscreens, photoprotective apperel, sunglasses, and occupastional standards for sun exposure. There, attitudes have already

The Epidemiology of Basal Cell and Squamous Cell Carcinoma

shifted positively towards avoiding exposure to the sun and away from desire for a tan. In western Europe and the United States, however, a tanned appearance remains fashionable, and also there has been a marked increase in sales of self tanning lotions.^^' The American Academy of Dermatology found in a recent survey that despite having a good understanding of the relation between overexposure to the sun and skin cancer, 8 1 % of Americans still think they look good after beeing in the sun. Risk taking behaviour with respect to exposure to the sun continues. ' Halpern and Kopp found significant differences in skin cancer awareness and sun protection behaviours among Australia, U.S.A. and Europe.^^ In Australia, where the incidence of skin cancer is high, more than 80% of respondents expressed concern over skin cancer. In comparison, Germany (30%) and France (34%) demonstrated the lowest level of concerns about the risk of developing skin cancer. This survey also demonstrated that the main source of information through which awareness was attained was the media and not qualified healthcare representatives, and support the importance of increased patient education by medical professionals in the context of routine medical care. Sunscreen use is the most popular form of sun protection and can significandy reduce the risk of developing skin cancer. However, the results of a recent study from the United Kingdom, France, Italy, Germany and Spain showed that both the general public and the majority of outdoor workers do not regularly apply sunscreens. Concern has also been raised that they may direcdy or indirecdy increase the risk of malignancy, primarily because of poor application and increased exposure to the sun. The thickness of application has been shown to be less than half that officially tested and key exposed sites are often missed completely. For future public health policy it is important to increase skin cancer awareness among European population, together with safe sun practices, such as application of sunscreen, wearing protective clothing, and avoiding the sun during times of peak solar intensity.

References 1. Siiverberg E, Boring CC, Squires TS. Cancer statistics, 1990. CA Cancer J Clin 1990; 40(l):9-26. 2. Anonymous. Health people 2010. Vol 2. Conference Edition ed: U.S. Department of Health and Human Services; 2000. 3. Staples M, Marks R, Giles G. Trends in the incidence of non-melanocytic skin cancer (NMSC) treated in Australia 1985-1995: are primary prevention programs starting to have an effect? Int J Cancer 1998; 78(2): 144-148. 4. Buettner PC, Raasch BA. Incidence rates of skin cancer in Townsviile, Australia. Int J Cancer 1998; 78(5):587-593. 5. Miller DL, Weinstock MA. Nonmelanoma skin cancer in the United States: incidence. J Am Acad Dermatol 1994; 30(5 Pt l):774-778. 6. Harris KB, Griffith K, Moon TE. Trends in the incidence of nonmelanoma skin cancers in southeastern Arizona, 1985-1996. J Am Acad Dermatol 2001; 45(4):528-536. 7. Greenlee RT, Hill-Harmon MB, Murray T et al. Cancer statistics, 2001. CA Cancer J Clin 2001; 51(l):15-36. 8. Stang A, Stegmaier C, Jockel KH. Nonmelanoma skin cancer in the Federal State of Saarland, Germany, 1995-1999. Br J Cancer 2003; 89(7):1205-1208. 9. Katalinic A, Kunze U, Schafer T. Epidemiology of cutaneous melanoma and non-melanoma skin cancer in Schleswig-Holstein, Germany: incidence, clinical subtypes, tumour stages and localization (epidemiology of skin cancer). Br J Dermatol 2003; 149(6):1200-1206. 10. de Vries E, Louwman M, Bastiaens M et al. Rapid and continuous increases in incidence rates of basal cell carcinoma in the southeast Netherlands since 1973. J Invest Dermatol 2004; 123(4):634-638. 11. Levi F, Te VC, Randimbison L et al. Trends in skin cancer incidence in Vaud: an update, 1976-1998. Eur J Cancer Prev 2001; 10(4):371-373. 12. Levi F, Erler G, Te VC et al. Trends in skin cancer incidence in Neuchatel, 1976-98. Tumori 2001; 87(5):288-289. 13. Boi S, Cristofolini M, Micciolo R et al. Epidemiology of skin tumors: data from the cutaneous cancer registry in Trentino, Italy. J Cutan Med Surg 2003; 7(4):300-305. 14. Plesko I, Severi G, Obsitnikova A et al. Trends in the incidence of non-melanoma skin cancer in Slovakia, 1978-1995. Neoplasma 2000; 47(3): 137-142.

Molecular Mechanisms of Basal Cell and Squamous Cell Carcinomas 15. Koh D, Wang H, Lee J et al. Basal cell carcinoma, squamous cell carcinoma and melanoma of the skin: analysis of the Singapore Cancer Registry data 1968-97. Br J Dermatol 2003;148(6):1161-1166. 16. Karagas MR, Greenberg ER, Spencer SK et al. Increase in incidence rates of basal cell and squamous cell skin cancer in New Hampshire, USA. New Hampshire Skin Cancer Study Group. Int J Cancer 1999; 81(4):555-559. 17. Athas WF, Hunt WC, Key CR. Changes in nonmelanoma skin cancer incidence between 1977-1978 and 1998-1999 in Northcentral New Mexico. Cancer Epidemiol Biomarkers Prev 2003; 12(10):1105-1108. 18. lARC, Monographs on the evaluation of carcinogenic risks to humans. Solar and ultraviolet radiation. Vol. 55. Lyon: International Agency for Research on Cancer; 1992. 19. Pennello G, Devesa S, Gail M. Association of surface ultraviolet B radiation levels with melanoma and nonmelanoma skin cancer in United States blacks. Cancer Epidemiol Biomarker Prev 2000; 9:291-7. 20. Whiteman DC, Green AC. Melanoma and sun exposure: where are we now? Int J Dermatol 1999; 38(7):481-489. 21. Armstrong BK, Kricker A. The epidemiology of UV induced skin cancer. J Photochem Photobiol B 2001; 63(l-3):8-18. 22. Ramos J, Villa J, Ruiz A et al. UV dose determines key characteristics of nonmelanoma skin cancer. Cancer Epidemiol Biomarkers Prev 2004; 13(12):2006-2011. 23. Franceschi S, Levi F, Randimbison L et al. Site distribution of different types of skin cancer: new aetiological clues. Int J Cancer 1996; 67(l):24-28. 24. Kripke ML. Impact of ozone depletion on skin cancers. J Dermatol Surg Oncol 1988; l4(8):853-857. 25. Kelfkens G, de Gruijl FR, van der Leun JC. Ozone depletion and increase in annual carcinogenic ultraviolet dose. Photochem Photobiol 1990; 52(4):819-823. 26. Giles GG, Marks R, Foley P. Incidence of non-melanocytic skin cancer treated in Australia. Br Med J (Clin Res Ed) 1988; 296(6614):13-17. 27. Karagas MR, Stannard VA, Mott LA et al. Use of tanning devices and risk of basal cell and squamous cell skin cancers. J Natl Cancer Inst 2002; 94(3):224-226. 28. Stern RS, Nichols KT, Vakeva LH. Malignant melanoma in patients treated for psoriasis with methoxsalen (psoralen) and ultraviolet A radiation (PUVA). The PUVA FoUow-Up Study. N Engl J Med 1997; 336(15):104l-1045. 29. Holme SA, Malinovszky K, Roberts DL. Changing trends in non-melanoma skin cancer in South Wales, 1988-98. Br J Dermatol 2000; 143(6):1224-1229. 30. Doll R. Cancers weakly related to smoking. Br Med Bull 1996; 52(l):35-49. 31. Moore S, Johnson N, Pierce A et al. The epidemiology of lip cancer: a review of global incidence and aetiology. Oral Dis 1999; 5(3):185-195. 32. Green A, Battistutta D, Hart V et al. Skin cancer in a subtropical Australian population: incidence and lack of association with occupation. The Nambour Study Group. Am J Epidemiol 1996; 144(11):1034-1040. 33. De Hertog SA, Wensveen CA, Bastiaens MT et al. Relation between smoking and skin cancer. J Clin Oncol 2001; 19(l):231-238. 34. Hunter DJ, Colditz GA, Stampfer MJ et al. Risk factors for basal cell carcinoma in a prospective cohort of women. Ann Epidemiol 1990; l(l):13-23. 35. van Dam RM, Huang Z, Rimm EB et al. Risk factors for basal cell carcinoma of the skin in men: results from the health professionals follow-up study. Am J Epidemiol 1999; 150(5):459-468. 36. Freedman DM, Sigurdson A, Doody MM et al. Risk of basal cell carcinoma in relation to alcohol intake and smoking. Cancer Epidemiol Biomarkers Prev 2003; 12(12):1540-1543. 37. Boyd AS, Shyr Y, King LE, Jr. Basal cell carcinoma in young women: an evaluation of the association of tanning bed use and smoking. J Am Acad Dermatol 2002; 46(5):706-709. 38. Wojno TH. The association between cigarette smoking and basal cell carcinoma of the eyelids in women. Ophthal Plast Reconstr Surg 1999; 15(6):390-392. 39. Smith JB, Randle HW. Giant basal cell carcinoma and cigarette smoking. Cutis 2001; 67(l):73-76. 40. Black HS. Photocarcinogenesis and diet. Fed Proc 1987; 46(5):1901-1905. 41. Hakim lA, Harris RB, Ritenbaugh C. Fat intake and risk of squamous cell carcinoma of the skin. Nutr Cancer 2000; 36(2): 155-162. 42. Green A, Williams G, Neale R et al. Daily sunscreen application and betacarotene supplementation in prevention of basal-cell and squamous-cell carcinomas of the skin: a randomised controlled trial. Lancet 1999; 354(9180):723-729. 43. Hartevelt MM, Bavinck JN, Kootte AM et al. Incidence of skin cancer after renal transplantation in The Netherlands. Transplantation 1990; 49(3):506-509.

The Epidemiology

of Basal Cell and Squamous

Cell

Carcinoma

44. Berg D , Otley C C . Skin cancer in organ transplant recipients: Epidemiology, pathogenesis, and management. J Am Acad Dermatol 2002; 47(1): 1-17; quiz 18-20. 45. Bouwes Bavinck J N , Hardie DR, Green A et al. T h e risk of skin cancer in renal transplant recipients in Queensland, Australia. A follow-up study. Transplantation 1996; 61(5):715-721. 46. London NJ, Farmery SM, Will EJ et al. Risk of neoplasia in renal transplant patients. Lancet 1995; 346(8972):403-406. A7. Lindelof B, Sigurgeirsson B, Tegner E et al. PUVA and cancer risk: the Swedish follow-up study. Br J Dermatol 1999; 141(1):108-112. 48. Diepgen TL, Mahler V. T h e epidemiology of skin cancer. Br J Dermatol 2002; 146 Suppl 61:1-6. 49. Lichter M D , Karagas MR, Mott LA et al. Therapeutic ionizing radiation and the incidence of basal cell carcinoma and squamous cell carcinoma. T h e N e w Hampshire Skin Cancer Study Group. Arch Dermatol 2000; 136(8):1007-1011. 50. Ron E, Preston DL, Kishikawa M et al. Skin tumor risk among atomic-bomb survivors in Japan. Cancer Causes Control 1998;9(4):393-401. 5 1 . Diepgen T L , Drexler H . [Skin cancer and occupational disease]. Hautarzt 2004;55(l):22-27. 52. Thieden E, Philipsen PA, Heydenreich J et al. U V radiation exposure related to age, sex, occupation, and sun behavior based on time-stamped personal dosimeter readings. Arch Dermatol 2004; l40(2):197-203. 53. Ramirez C C , Federman D G , Kirsner RS. Skin cancer as an occupational disease: the effect of ultraviolet and other forms of radiation. Int J Dermatol 2005; 44(2):95-100. 54. Saladi R, Austin L, Gao D et al. T h e combination of benzo[a]pyrene and ultraviolet A causes an in vivo time-related accumulation of D N A damage in mouse skin. Photochem Photobiol 2 0 0 3 ; 77(4):413-419. 55. Yu R C , Hsu KH, Chen CJ et al. Arsenic methylation capacity and skin cancer. Cancer Epidemiol Biomarkers Prev 2000; 9(11):1259-1262. 56. Marks R, Rennie G, Selwood T S . Malignant transformation of solar keratoses to squamous cell carcinoma. Lancet 1988; l(8589):795-797. 57. Frankel D H , Hanusa B H , Zitelli JA. New primary nonmelanoma skin cancer in patients with a history of squamous cell carcinoma of the skin. Implications and recommendations for follow-up. J Am Acad Dermatol 1992; 26(5 Pt l):720-726. 58. Koh HK, Geller A C , Miller D R et al. Prevention and early detection strategies for melanoma and skin cancer. Current status. Arch Dermatol 1996; 132(4):436-443. 59. Fry A, Verne J. Preventing skin cancer. BMJ 2003; 326(7381):114-115. 60. Livingston PM, White V M , Ugoni AM et al. Knowledge, attitudes and self-care practices related to sun protection among secondary students in Australia. Health Educ Res 2 0 0 1 ; l6(3):269-278. 6 1 . Halpern A C , Kopp LJ. Awareness, knowledge and attitudes to non-melanoma skin cancer and actinic keratosis among the general public. Int J Dermatol 2005; 44(2):107-111. 62. Dermatology' AAo. N e w AAD survey reveals people understand the relationship between overexposure to the sun and skin cancer. www.aad.org/PressRelease/SunSafety/Survey.html. 63. Shoveller JA, Savoy D M , Roberts RE. Sun protection among parents and children at freshwater beaches. Can J Public Health 2002; 93(2):146-148. 64. MacKie R M . Awareness, knowledge and attitudes to basal cell carcinoma and actinic keratoses among the general public within Europe. J Eur Acad Dermatol Venereol 2004; 18(5):552-555. 65. Johnson K, Davy L, Boyett T et al. Sun protection practices for children: knowledge, attitudes, and parent behaviors. Arch Pediatr Adolesc Med 2 0 0 1 ; 155(8):891-896.

CHAPTER 2

Histology of Epithelial Sldn Tumors Jorg Reichrath* and Kerstin Querings Squamous Cell Carcinoma Squamous Cell Carcinoma in Situ Actinic Keratosis (Squamous Cell Carcinoma in Situ, Type Actinic Keratosis; Synonymi Solar Keratosis, Senile Keratosis) lthough this topic is still a matter of debate, actinic keratoses can now be considered to represent early squamous cell carcinomas in situ. However, it appears that only a small proportion of actinic keratoses will develop into an invasive squamous cell carcinoma. Regression of some cases of actinic keratoses has also been reported, most likely as a result of immune mechanisms.^ Metastases after transformation of actinic keratosis into invasive squamous cell carcinoma are very rare except for those tumors that arise on the ear, lip, anus and vulva, which have been reported to be often associated with a more aggressive behaviour. Actinic keratoses are characterized by dysplasia of varying degree (from mild changes through to typical carcinoma in situ), predominandy in keratinocytes of the interadnexal epidermis (Fig. 1). Parakeratosis is usually present. Variants of actinic keratosis include the hyperplastic, the proliferating, the atrophic, the acantholytic, the epidermolytic and the bowenoid subtype. The dermis underlying and adjacent to actinic keratoses typically shows the presence of ectatic vessels and solar elastosis.'^'^

A

Bowen's Disease (Squamous Cell Carcinoma in Situ, Type Bowen's Disease) In Bowens disease, the epidermis typically shows ftdl thickness dysplasia resembling carcinoma in situ, which involves keratinocytes of the entire epidermis including the intraepidermal portions of the cutaneous adnexae (Fig. 2). In addition, very large atypical cells (Bowens cells) and bizarre mitoses^ are usually found (Fig. 2). Characteristically, the stratum corneum is thickened and parakeratosis with diminished or absent stratum granulosum is present. The parakeratotic scales may show prominent and hyperchromatic nuclei. Occasionally, lesions may be markedly glycogenated and verrucous architectural characteristics may also be observed. In many cases, the superficial dermis adjacent to and underlying the tumor shows prominently dilated and proliferating vessels, fibrosis, and a chronic inflammatory infiltrate. The latter is likely to represent a localized immune response. Progression towards invasive carcinoma (Bowens carcinoma) has been reported to be relatively uncommon, occurring in approximately 5% of patients.^ Of these roughly 30% have been reported to have metastatic potential. Bowens disease affecting the penis is referred to as erythroplasia of Queyrat.^^ In a large series of more

*Corresponding Author: Jorg Reichrath—The Saarland University Hospital, Clinic for Dermatology, Allergology and Venerology, Kirrbergerstr., D-66421 Homburg/Saar, Germany. Email: [email protected]

Molecular Mechanisms of Basal Cell and Squamous Cell Carcinomas, edited by Jorg Reichrath. ©2006 Landes Bioscience and Springer Science+Business Media.

Histology of Epithelial Skin Tumors

11

Figure 1. Actinic keratosis (HE-stain, original magnification x200).

Figure 2. Bowens disease (HE-stain, original magnification x400). than 100 cases with erythroplasia of Queyrat, 2 2 % recurred, 8 % progressed to invasive t u m o r and 2 % metastasised.^^

12

Molecular Mechanisms of Basal Cell and Squamous Cell Carcinomas

Figure 3. Superficial cutaneous squamous cell carcinoma (HE-stain, original magnification x200).

Invasive Squamous Cell Carcinoma Squamous cell carcinoma (Fig. 3) is an invasive malignant tumor of epidermal keratinocytes, that retains variable characteristics of uninvolved suprabasal epidermis. In most cases, the tumor arises from an dsyplastic epithelium resembling actinic keratosis (in situ carcinoma) and presents as infiltrating sheets and islands of variably differentiated squamous epithelium. Although the in situ forms may give rise to invasive lesions, an in situ stage is not necessary for the development of invasive squamous cell carcinoma. Degree of mitotic activity and cell pleomorphism of tumor cells may vary. Differentiation is towards keratinisation, that typically results in the formation of so called keratin "pearls", horn cysts and/or scattered keratinised cells. Squamous cell carcinomas can be classified into well differentiated, moderately differentiated and poorly differentiated tumors. Alternatively, Broders system of classification, based on four grades of differentiation, has been used: grade 1 - 75% or more of the lesion well differentiated; grade 2 - 50% or more well differentiated; grade 3 - 25%-50% well differentiated; grade 4 — less than 25% well differentiated. Variants of squamous cell carcinoma include clear cell, spindle cell, acantholytic, and verrucous squamous carcinoma as well as lymphoepithelioma-like carcinoma and Bowen's carcinoma.^'^' In general, poorly differentiated tumors recur and metastasize more frequendy than well differentiated variants.® Neurotropism is associated with high recurrence and metastasis rates.® Perineural spread is particularly common in tumors arising on the head and neck, especially the lip and mid-face. In a large series of such cases, 14% of tumors showed involvement of the perineural space, particularly spindle cell and acantholytic variants.^ Although the genesis of invasive squamous cell carcinoma of the skin is considered to be a complex, multistage process, it has been demonstrated that mutations in the tumor suppressor gene p53 and in the Ha-ras gene are of high importance for the cancerogenesis of squamous cell carcinomas. ' Recent work on human epidermis reveals that sustained Ras signaling coupled with suppression of Ras-induced growth arrest is sufficient to drive the entire process of squamous cell carcinogenesis and that the alpha6beta4 integrin and its laminin 5 ligand are essential components of this process. Additionally, a connection between development of epidermal p53 clones and squamous cell carcinomas has been suggested. It has been

Histology ofEpithelial Skin Tumors

13

demonstrated that reduced levels of apoptotic cells in squamous cell but not in basal cell carcinomas correlate with the detection of cutaneous human papillomavirus. Interestingly, different expression patterns of calpain isozymes 1 and 2 (GAPNl and 2) have been detected in squamous cell carcinomas and basal cell carcinomas. ^^ Calpain, also named CAPN (for calcium-activated neutral protease), is an ubiquitous intracellular cytoplasmic nonlysosomal cysteine endopeptidase that requires calcium ions to exert its activity. Many known substrates of the different calpain isoenzymes, such as the transcription factors c-Fos and c-Jun, the tumour suppressor protein p53, protein kinase C, pp60src, or the adhesion molecule integrin, have been implicated in the pathogenesis of squamous and basal cell carcinomas, suggesting an important role of the calpain isoenzymes in these diseases. CAPNl immunoreactivity is markedly reduced in basal cell carcinomas compared to normal human skin or squamous cell carcinomas, while in contrast CAPNl mRNA levels are markedly elevated in basal cell carcinomas and squamous cell carcinomas compared to normal human skin. No differences are found analysing CAPN2 protein and mRNA expression in normal human skin, basal cell carcinomas and squamous cell carcinomas. Increasing evidence indicates that, besides other factors, the vitamin D system is of importance for the growth characteristics of cutaneous squamous cell carcinomas. Expression of vitamin D receptor (VDR) and of the main enzymes involved in the metabolism and catabolism of the biolocically active vitamin D metabolite, 1,25-dihydroxyvitamin D3 (vitamin D-25-hydroxylase (25-OHase), 25-hydroxyvitamin D-lalpha-hydroxylase (lalpha-OHase), and 1,25-dihydroxyvitamin D-24-hydroxylase (24-OHase)) was detected in squamous cell carcinomas. Modulation of VDR expression and local synthesis or metabolism of vitamin D metabolites may be of importance for growth characteristics of squamous cell carcinomas.

Keratoacanthoma Keratocanthomas can be described as fastly growing, if left untreated often spontaneously regressing tumors. However, one has to keep in mind that perineural invasion^ and in very rare cases intravascular spread'^^ have been reported. In most cases, these tumors affect sun-exposed hair follicle-bearing skin of elderly individuals and they mimic clinically and histopathologically well-differentiated squamous cell carcinomas. Although this topic is still a matter of debate, keratocanthomas can therefore be considered as a histologic variant of squamous cell carcinoma with distinctive clinical and pathologic attributes.^^''^ '^^ Recendy, it has been reported that analysis of telomerase activity, COX-2, and p53 expression provide evidence that keratoacanthoma and squamous cell carcinoma are indeed distinct entities and that this analysis may also help in discriminating these two lesions, which may closely resemble each other on conventional morphology. Additionally, amphiregulin overexpression has been demonstrated in keratoacanthomas (21 of 21 positive, strong immunoreactivity) and conventional squamous cell carcinomas (5 of 6 positive, in general weak staining) while all basal cell carcinomas (6 of 6) analyzed in that study were negative.^'^ Interestingly, it has been reported that expression of VCAM (CD-106) and ICAM (CD-54) adhesion molecules^^ and die level of syndecan-1 may distinguish keratoacanthomas from cutaneous squamous cell carcinomas. In contrast, MIB-1 immunohistometry, although presenting insights into the proliferative potential of keratoacanthomas and conventional squamous cell carcinomas, has been shown to be of only limited value for the differential diagnosis of the two lesions in routine surgical pathology.^^ In general, keratoacanthomas are characterized by a distinctive histologic architecture and, at least in most cases, by spontaneous regression. Typically, these tumors represent an exophytic and endophytic neoplasm with a typical cup-shape that is sharply demarcated from the surrounding epidermis and dermis. The center of the tumor is characterized by a crater that is filled with eosinophilic laminated orthokeratotic scale. This crater is in most cases partially enclosed by a well defined lip that forms the superficial border of the neoplasm. The epithelium of the lip may be hyperplastic, but there is usually no evidence of dysplasia or actinic keratosis in the epithelium adjacent to the tumor. ' '

14

Molecular Mechanisms of Basal Cell and Squamous Cell Carcinomas

Figure 4. Basal cell carcinoma (HE-stain, original magnification xlOO; A: nodular, B: superficial, C: sclerosing variants).

Histology of Epithelial Skin Tumors

15

Basal Cell Carcinoma Basal cell carcinomas (Fig. 4) are tumors with in general nonmetastasizing behavior that derive from undifferentiated pluripotent epithelial stem cell. They are typically characterized by a fibrous stroma surrounding islands of dependent tumor cells that resemble keratinocytes of the basal layer of the epidermis or hair follicle (Fig. 4). Usually, these tumor cells are fairly regular with rounded haematoxophilic nuclei and litde cytoplasm. Typically, the proliferating cell component of the tumor is found predominandy in so called peripheral "palisades" of cells around the margins of each tumor nest. It has been shown that this phenomenon corresponds to the way -in which basal cell carcinomas grow by slow progressive, local invasion. ^^'^^ Proliferation of basal cell carcinoma cells has been analyzed immunohistochemically. The Ki-67-positive growth fraction displays great variation between tumors belonging to the same subtype (nodular type, 1-G7%\ superficial type, 18-49%; fibrosing type 4-33%). It has been demonstrated that areas with a high Ki-67 labelling index often occur adjacent to rather quiescent strands, suggesting that an individual tumor is not in a uniform state of proliferation. In view of the fact that BCCs are rather slow-growing tumors, the large Ki-67 growth fractions indicate a prolonged duration of the cell cycle or a considerable continuous loss of cells. As the microarchitecture of BCCs is much more complex than would be expected from the location of their IG-67-positive cells, the growth pattern is probably determined to a high degree by the adjacent connective tissue (physical properties and texture of collagen and elastic fibres, enzyme activity of fibroblasts). The close relationship between the tumor cell component and the surrounding fibrous stroma is a very interesting aspect of the growth behavior of basal cell carcinomas. It is well known that the tumor cells cannot develop without the surroimding stroma. This stroma, that is usually loose and rich in mucin (predominantly hyaluronic acid), is particularly abundant in sclerosing basal cell carcinoma and constitutes the major part of the volume of the tumors. One important clue to the diagnosis of basal cell carcinoma is the presence of a constant retraction artefact: the separation of the tumor cells from the underlying stroma. It has been speculated that this effect may be due to defective production of hemidesmosomes, reduced numbers of anchoring fibrils, and dimished expression of bullous pemphigoid antigen by these tumor cells.^ Variants of basal cell carcinoma includefivemain clinical subtypes: nodular/ulcerative (solid) (45-60%), diffuse (infiltrating and morphoeic, sclerosing) (4-17%), superficial (multicentric) (15-35%), pigmented variant (1-7%), and the fibroepithelioma of Pinkus. Rare basal cell carcinomas that exhibit significant nuclear anaplasia and infiltrative growth characteristics have been referred to in the past as metatypical basal cell carcinomas, although the use of this term is now discouraged.^ Neuroendocrine differentiation has been desribed as well, but is uncommon. ' Pigmentation in basal cell carcinomas may be evident in macrophages and dendritic cells. Haemosiderin may also be present in some cases. Basal cell carcinomas are characterized by a nonmetastasizing behavior. Apoptotic cells are present in these tumors. ^^ The integrin profile of basal cell carcinomas does not differ essentially from that of metastasizing tumor varieties and cannot be regarded as a major reason for the nonmetastasizing phenotype of basal cell carcinomas. However, it has been suggested that the very low expression of the receptor for hyaluronic acid (CD44std) may be one of the factors which block the formation of metastases from basal cell carcinomas. Increasing evidence indicates that the vitamin D system is of critical importance for growth characteristics in basal cell carcinomas. RNA expression of vitamin D receptor (VDR) and of main enzymes involved in synthesis and metabolism of calcitriol was detected in basal cell carcinomas and normal skin. Increased VDR-immunoreactivity was demonstrated in basal cell carcinomas using a streptavidin-peroxidase technique.^^'^^ In conclusion, these findings provide supportive evidence for the concept that endogeneous synthesis and metabolism of vitamin D metabolites as well as VDR expression may regulate growth characteristics of basal cell carcinomas. '^^

16

Molecular Mechanisms of Basal Cell and Squamous Cell Carcinomas

Recently, the expression and distribution of the DNA mismatch repair enzyme ^MSH-2 in normal skin and basal cell carcinomas was investigated immunohistochemically.^^ All basal cell carcinomas analysed revealed strong nuclear imunoreactivity that was pronounced in peripheral tumour cells and cells of the palisade. Expression of ^MSH-2 protein was consistently and strongly upregulated in tumour cells of the carcinomas as compared to adjacent unaffected epidermis or epidermis of normal human skin. These findings indicate that upregulation of ^MSH-2 protein expression may be of importance for the genetic stability of basal cell carcinomas in vivo.^^

Basosquamous Carcinoma Basosquamous carcinoma of the skin is a very rare malignancy with specific histopathological features of both basal cell carcinoma and squamous cell carcinoma. ^' ^ Some authors believe that basosquamous carcinoma is a variant of basal cell carcinoma, while others suggest that this tumour may behave more aggressively and represents a distinct entity. In general, the predominant expression is of a basal cell carcinoma with squamous differentiation occurring in larger groups of tumor cells. These tumors are unlikely to metastasize and should therefore be regarded as variants of basal cell carcinoma from the point of view of clinical manage13

ment.^

References 1. Ackerman AB. Solar keratosis is squamous cell carcinoma. Arch Dermatol 2003; 139(9): 1216-7. 2. Fu W, Cockerell CJ. The actinic (solar) keratosis. Arch Dermatol 2003; 139:66-70. 3. Lober BA, Lober CW. Actinic keratosis is squamous cell carcinoma. Southern Medical Journal 2000; 93(7):650-655. 4. Stadler R, Hartig C. Epidermale Tumoren. In: Kerl, Garbe, Cerroni, Wolff, eds. Histopathologic der Haut. 1st ed. Berlin, Heidelberg, New York: Springer, 2003. 5. Marks R, Foley P, Goodman G et al. Spontaneous remission of solar keratoses: The case for conservative management. Br J Dermatol 1986; 115:649-656. 6. Marks R, Rennie G, Selwood TS. Malignant transformation of solar keratoses to squamous cell carcinoma. Lancet 1988; l(8589):795-797. 7. Kirkham N. Intraepidermal keratoses and tumours. In: Kirkham N, ed. Biopsy pathology of the skin. 1st ed. London, New York, Tokyo, Melbourne, Madras: Chapman and Hall Medical, 1991a:32-53. 8. McKee PH. Tumours of the surface epithelium. In: McKee, ed. Pathology of the skin. 2nd ed. London, Philadelphia, St. Louis, Sydney, Tokyo: Mosby International 1999a: 14.1-14.28. 9. Murphy GF, Elder DE. Nonmelanocytic tumors of the skin. In: Rosai, Sobin, eds. Atlas of tumour pathology. Third series, fascicle 1. Washington, DC: American Registry of Pathology, 1991. 10. Kaye V, Zhang G, Dehner LP et al. Carcinoma in situ of penis. Is distinction between erythroplasia of Queyrat and Bowen's disease relevant? Urology 1990; 33:479-482. 11. Graham JH, Helwig EB. Erythroplasia of Queyrat. A clinicopathologic and histochemical study. Cancer 1973; 32:1396-1414. 12. Broders AC. Practical points on the microscopic grading of carcinoma. NY State J Med 1932; 32:667-671. 13. Kirkham N. Epidermal tumours. In: Kirkham N, ed. Biopsy pathology of the skin. 1st ed. London, New York, Tokyo, Melbourne, Madras: Chapman and Hall Medical, 1991b:67-79. 14. Goepfert H, Dichtel W, Medina JE et al. Perineural invasion of squamous cell carcinoma of the head and neck. Am J Surg 1984; 148:542-547. 15. Giglia-Mari G, Sarasin A. TP53 mutations in human skin cancers. Hum Mutat 2003; 21(3):217-28. 16. Mercurio AM. Invasive skin carcinoma—Ras and alpha6beta4 integrin lead the way. Cancer Cell 2003; 3(3):201-2. 17. Backvall H, Wolf O, Hermelin H et al. The density of epidermal p53 clones is higher adjacent to squamous cell carcinoma in comparison with basal cell carcinoma. Br J Dermatol 2004; 150(2):259-66. 18. Jackson S, Ghali L, Harwood C et al. Reduced apoptotic levels in squamous but not basal cell carcinomas correlates with detection of cutaneous human papillomavirus. Br J Cancer 2002; 87(3):319-23.

Histology of Epithelial Skin Tumors

17

19. Reichrath J, Welter C, Mitschele T et al. Different expression patterns of calpain isozymes 1 and 2 (CAPNl and 2) in squamous cell carcinomas (SCC) and basal cell carcinomas (BCC) of human skin. J Pathol 2003; 199(4):509-16. 20. Reichrath J, Rafi L, Rech M et al. Analysis of the vitamin D system in cutaneous squamous cell carcinomas. J Cutan Pathol 2004; 31(3):224-31. 21. Godbolt AM, Sullivan JJ, Weedon D. Keratoacanthoma with perineural invasion: A report of 40 cases. Australas J Dermatol 2001; 42(3): 168-71. 22. Gottfarstein-Maruani A, Michenet P, Kerdraon R et al. Keratoacanthoma: Two cases with intravascular spread. Ann Pathol 2003; 23(5):438-42. 23. Balasubramaniam N, Resnick KS, Ackerman AB, What is your diagnosis? (keratoacanthoma with metastasis). Dermatopathol Pract Concept 1999; 5:51. 24. Hodak E, Jones RE, Ackerman AB. Solitary keratoacanthoma is a squamous-cell carcinoma: Three examples with metastases. Am J Dermatopathol 1993; 15:332, (discussion 343-352). 25. Kane CL, Keehn CA, Smithberger E et al. Histopathology of cutaneous squamous cell carcinoma and 1st variants. Semin Cutan Med Surg 2004; 23(1):54-61. 26. Putti TC, Teh M, Lee YS. Biological behavior of keratoacanthoma and squamous cell carcinoma: Telomerase activity and COX-2 as potential markers. Mod Pathol 2004; 17(4):468-75. 27. Billings SD, Southall MD, Li T et al. AmphireguUn overexpression results in rapidly growing keratinocytic tumors: An in vivo xenograft model of keratoacanthoma. Am J Pathol 2003; 163(6):2451-8. 28. Melendez ND, SmoUer BR, Morgan M. VCAM (CD-106) and ICAM (CD-54) adhesion molecules distinguish keratoacanthomas from cutaneous squamous cell carcinomas. Mod Pathol 2003; 16(1):8-13. 29. Mukunyadzi P, Sanderson RD, Fan CY et al. The level of syndecan-1 expression is a distinguishing feature in behavior between keratoacanthoma and invasive cutaneous squamous cell carcinoma. Mod Padiol 2002; 15(l):45-9. 30. Biesterfeld S, Josef J. Differential diagnosis of keratoacanthoma and squamous cell carcinoma of the epidermis by MIB-1 immunohistometry. Anticancer Res 2002; 22 (5): 3019-23. 31. Grimwood RE, Ferris CF, Mercill DB et al. Proliferating cells of human basal cell carcinoma are located on the periphery of tumor nodules. J Invest Dermatol 1986; 86(2): 191-4. 32. Mitschele T, Diesel B, Friedrich M et al. Analysis of the vitamin D system in basal cell carcinomas (BCCs). Lab Invest 2004; 84(6):693-702. 33. Baum HP, Meurer I, Unteregger G. Ki-67 antigen expression and growth pattern of basal cell carcinomas. Arch Dermatol Res 1993; 285(5):291-5. 34. Jones JC, Steinman HK, Goldsmith BA. Hemidesmosomes, collagen VII, and intermediate filaments in basal cell carcinoma. J Invest Dermatol 1989; 93(5):662-71. 35. McKee PH. Tumours of the epidermal appendages. In: McKee, ed. Pathology of the skin. 2nd ed. London, Philadelphia, St. Louis, Sydney, Tokyo: Mosby International, 1999b:15.32-15.40. 36. Baum HP, Schmid T, Reichrath J. Integrin molecules: A clue to the nonmetastasizing behaviour of basal cell carcinomas? Acta Derm Venereol 1996b; 76(l):24-7. 37. Baum HP, Schmid T, Schock G et al. Expression of CD44 isoforms in basal cell carcinomas. Br J Dermatol 1996a; 134(3):465-8. 38. Mitschele T, Diesel B, Friedrich M et al. Analysis of the vitamin D system in basal cell carcinomas (BCCs). Lab Invest 2004; 84(6):693-702. 39. Rass K, Gutwein P, MuUer SM et al. Immunohistochemical analysis of DNA mismatch repair enzyme hMSH-2 in normal human skin and basal cell carcinomas. Histochem J 2000; 32(2):93-7. 40. Sendur N, Karaman G, Dikicioglu E et al. Cutaneous basosquamous carcinoma infiltrating cerebral tissue. J Eur Acad Dermatol Venereol 2004; 18(3):334-336.

CHAPTER 3

UV Damage and DNA Repair in Basal CeU and Squamous Cell Carcinomas KnuthRass* Abstract

E

xposure of the skin with ultraviolet radiation (UV) is the main cause of skin cancer development. The consistendy increasing incidences of melanocytic and nonmelanocytic skin tumors is associated with recreational sun exposure. Epidemiological data indicate that excessive or cumulative sunlight exposition take place years and decades before the tumor occurs. The most important protection strategies against UV in human skin consists in melanin synthesis and active repair mechanisms. DNA is the major target of direct or indirect UV damage. Low pigmentation capacity in white Caucasians and rare congenital defects in DNA repair are mainly responsible for protection failures. The important function of nucleotide excision DNA repair (NER) to avoid skin cancer becomes obvious by the rare genetic disease xeroderma pigmentosum, in which diverse NER genes are mutated. In animal models it has been demonstrated that UVB is more effective to induce skin cancer than UVA. UV-induced DNA photoproducts are able to cause specific mutations (UV-signature) in susceptible genes for squamous cell carcinoma (SCC) and basal cell carcinoma (BCC). In SCC development UV-signature mutations in the p53 tumor suppressor gene are the most common event, as precancerous lesions reveal - 80% and SCCs > 90% UV-specific p53 mutations. In BCC mutations in Hedgehog pathway related genes, especially PTCH1, represent the most significant pathogenic event in BCC; specific UV-induced mutations can be found only in -- 50% of sporadic BCC. Thus, cumulative UVB radiation is considered not to be the single factor for BCC development. An interesting new perspective in DNA damage and repair research, as recent publications suggest, lies in the participation of mammalian mismatch repair (MMR) in UV damage correction. As MMR enzyme hMSH2 displays a p53 target gene, is induced by UVB radiation and is involved in NER pathways, studies have been initiated to elucidate the physiological and pathophysiological role in skin cancer development.

Introduction Sunlight is an indispensable requirement for life on earth by spending the essential thermal energy and facilitating photosynthesis in plants, which in turn supplies our atmosphere with oxygen. On the other hand ultraviolet radiation of sunlight (UV) can be assumed to be the most important and ubiquitously occurring physical carcinogen inducing melanocytic and nonmelanocytic skin cancer with increasing incidences. The solar UV spectrum consists of *Knuth Rass—The Saarland University Hospital, Clinic for Dermatology, Allergology and Venerology, Kirrbergerstr., 66421 Homburg/Saar, Germany. Email: [email protected]

Molecular Mechanisms of Basal Cell and Squamous Cell Carcinomas, edited by Jorg Reichrath. ©2006 Landes Bioscience and Springer Science+Business Media.

UVDamage and DMA Repair in Basal Cell and Squamous Cell Carcinomas UVC (wavelengths below 280 nm), UVB (280-315 nm) and UVA bands (315-400 nm). The predominant part of the short-wave, high-energy and destructive UV spectrum cannot reach the earths surface: the ozone layer of the outer earth atmosphere absorbs the shorter wavelengths up to --310 nm (UVC and main part of UVB radiation). The remaining transmitted UV spectrum, i.e., a small UVB and the complete UVA band, is responsible for biological effects in human skin. UV light does not penetrate the body any deeper than the skin and is absorbed by the different skin layers in a wavelength dependent manner: UVB is almost completely absorbed by the epidermis; only 10-20% of UVB energy reaches the epidermal stratum basale and dermal stratum papillate. UVA penetrates deeper into the dermis and deposits 30-50% of its energy in dermal stratum papillate. These absorption characteristics in human skin explain, why UVB effects have to be expected predominandy in the epidermis (skin cancer development) and UVA effects in the dermis (solar elastosis, skin ageing). As human skin is continuously exposed to sunlight, the organism developed protection strategies, like melanin synthesis and active repair mechanisms, to avoid structural damage of the most important UV target molecule, the DNA. DNA is a major epidermal chromophore with an absorption maximum of 260 nm and a continuous decrease over the UVB and UVA spectra. Excessive or continuing chronic sun exposition, low pigmentation capacity in white Caucasians and rare congenital defects are responsible for failures of these adaptive mechanisms. UVB—and far less UVA—is able to cause molecular rearrangements of the DNA with formation of specific photoproducts, which are known to be mutagenic. The geno toxic potential of UVA is predominandy due to oxidative DNA damage. If the UV-induced mutations concern particular genes involved in signaling pathways of cell cycle control, proliferation, apoptosis and DNA repair, a malignant tumour can emerge. The fact that skin cancer does not occur immediately after sun exposure but with al latency period of years and decades underlines the theory of multiple genomic hits to establish a malignant phenotype (multistep carcinogenesis). Extending mutations in affected epithelial cells make it subsequendy more probable that these cells gain abilities to progress (activating oncogene mutations) or loose cell growth inhibitory or anti-apoptotic functions (inactivating tumor suppressor gene mutations). Furthermore UV radiation displays immunosuppressive effects and is able to generate tolerance against immunogenetic skin tumors. Thus UV is considered to be a "double-edged sword" causing skin cancer by DNA damage on the one hand and enabling tumor escape from immune surveillance on the other. In the following essay the genotoxic effects of UV radiation, the important repair mechanisms to avoid structural DNA damage and the pathogenic role of UV radiation in nonmelanocytic skin cancer (squamous and basal cell carcinoma) should be elucidated.

UV Damage and DNA Repair In 1928 Gates described for the first time that the bactericidal effect of UV radiation is connected with its absorption by prokaryotic DNA.^ Thirteen years later HoUaender and Emmons observed that the occurrence of mutations in eukaryotic (Fungi) DNA due to UV radiation is wavelength dependent.^ Thus the indispensable importance of DNA concerning cell survival and cell transformation on the one hand and the relevance of UV light to induce potentially lethal genomic disruptions on the other was already known in the first half of the last century (for a review see refs. 3-5). In the 1960's the essential molecular characteristics of DNA alterations caused direcdy by UVC and UVB radiation were uncovered. Beukers and Berends, as well as Sedow and Carrier^ found covalent interactions between two adjacent pyrimidine bases forming cyclobutane pyrimidine dimers (CPD: thymine dimers, cytosine dimers); later on Varghese und Patrick^ described another typical dimer formation at di-pyrimidine sites, the 6-4 photoproducts (6-4 PP: thymine-cytosine dimers). UVC is completely absorbed by the atmospheric ozone layer. Nevertheless, if UVC would reach the earth surface—in consideration of an increasing ozone dismanding—its short wavelength spectrum would hardly allow to penetrate into the human epidermal stratum basale

19

Molecular Mechanisms ofBasal Cell and Squamous Cell Carcinomas

20

Wavelength (nm)

Chromophores

DNA damage

Indirect

Reaction type

Figure 1. UV-induced DNA damage. The reaction type of DNA damage is wavelengdi-dependent: direct induaion of DNA photoproducts (CPD, 6-4 PP) by UVB / UVC and shortwave UVA spearum (315-327 nm); indirect induction ofoxidative DNA disruptions (ssDNA breaks, DNA-protein crosslinks) by longwave UVA spearum (347-400 nm); mixed reaction type with direa and indirect effeas by UVA wavelength 327-347 nm. Oxidative DNA damage is mediated by photosensitizers (Type I reaction) or by photosensitizer-induced reactive oxygen species (ROS, Type II reaaion). with an effective dose. On the other hand it has been shown that squamous cell carcinomas (SCC) and fibrosarcomas coidd be generated by 254 nm UVC irradiation in mice.^ Thus, it is not clear, to what extent UVC would be relevant for the induction of skin cancer in human. The biological effects of UVA on DNA are different from UVB/UVC and are predominantly indirect: UVA energy is mainly absorbed by chromophores other than D N A The energetic activation of those chromophores—endogenous or exogenous "photosensitizers" like NADH or different drugs (psoralen, tetracyclines)—causes reactive oxygen species (ROS) by photochemical interactions in vitro. The energy-enhanced photosensitizers themselves (type I) or those aggressive oxygen molecules (type II) in turn react with the DNA molecule leading to DNA single strand breaks and DNA-to-protein crosslinks. The relevance of these findings for carcinogenesis were supposed by different animal experiments, which demonstrate the induction of SCCs, melanomas and other tumors by UVA alone and an enhancement of the carcinogenic UVB effect by UVA.^^'^^ Beside the indirect effects of UVA on DNA a small part of UVA spectrum (315-327 nm) is considered to be able to generate CPD's and GA PP direcdy. Between 328 and 347 nm UVA irradiation can induce both, direct and indirect reaction types of DNA damage (Fig. 1).^^ UV-induced DNA lesions influence cellular death, aging, mutagenesis and carcinogenesis, if they are not completely rejected by the nuclear DNA repair machinery. The ability to erase DNA photoproducts from bacterial DNA, a process nowadays known as nucleotide excision repair (NER), was discovered in 1964}^ NER is the main repair system responsible for correcting direcdy UV-induced DNA damage. Oxidized (indirect) DNA base lesions are removed by

UVDamage and DNA Repair in Basal Cell and Squamous Cell Carcinomas

21

Figure 2. 9-year-old boy with xeroderma pigmentosum presenting with typical poikilodermia in the light-exposed skin, multiple actinic keratoses, actinic cheilitis, squamous cell carcinoma (bridge of nose) and basal cell carcinomas (right nose-canthus region, upper lip). essentially two types of activity: base excision repair (BER), involving removal of single lesions by a glycosylase action, and NER. In contrast to NER, disruptions in BER as a principle for cancer development are not known so far. NER failure or exceeding of its repair capacity is one important pathogenic step in UV associated skin cancer development as suggested by the rare autosomal recessive disease xeroderma pigmentosum (XP, Fig. 2). Patients sufFering from XP are extremely sun-sensitive with severe sunburns in childhood, are characterized by photo-aged freckled skin and are very prone to sunlight-induced skin tumors (keratoacanthoma, SCC, basal cell carcinoma, melanoma) occurring early in life (childhood - adolescence). As compared to unaffected individuals skin cancer incidence in XP patients is -2000-fold elevated.^^'^^ In 1968 Cleaver^^ initially uncovered the etiology of XP as caused by deficient NER. That's why UV-induced photoproducts accumulate in XP and subsequent mutations can resiJt in malignant phenotypes. Seven different types (complementation groups A — G) and a variant form of XP have been revealed so far. Besides skin cancer development, XP is associated with internal malignancies, neurological and ocular abnormalities. Each complementation group is defined by underlying mutations in genes encoding different NER proteins (XPA-XPG, XPV; (Table 1). In short, NER functions as follows: DNA photoproducts were recognized by different protein complexes, CSA/CSB in

Molecular Mechanisms of Basal Cell and Squamous Cell Carcinomas

22

Table 1, Xeroderma pigmentosum: Complementation groups, genetics, clinical features and distribution Unscheduled Complementation DNA Synthesis (% of Normal) Group A B C D E F G Variant Total

TT tandem transitions. Thymine cyclobutane dimers do not yield a mutation, because noninformative bases on the template DNA strand were substituted by an adenine on the opposite DNA strand ("A-rule"). Concomitant replication errors in different tumor suppressor and growth regulatory genes are supposed to be the genetic mechanism of tumorigenesis in those cells. Mutations in MMR genes are responsible for hereditary nonpolyposis colon cancer (for a review, see ref 29). Functional MMR alterations are furthermore associated with visceral malignancies and the occurrence of sebaceous skin tumors, keratoacanthomas and less frequently squamous cell carcinomas in the rare autosomal dominant Muir-Torre syndrome. Underlying mutations were found in the hMSFi2 and hMLFil gene.^^"^^ MMR and NER seem to be functionally connected: Mellon and Champe could demonstrate, that disruptions of the bacterial DNA mismatch repair genes mutS and mutL reduce NER of the lactose operon in E. coli. Human cells with mutations in particular MMR genes were likewise found to have a deficiency in the repair of UV-induced pyrimidine dimers. Furthermore specific binding of MSH2 / MSH6 (human homologs of mutS) heterodimers to DNA incorporating thymine- or uracil-containing UV light photoproducts^^ and UV-induced activation of transcription of hMSH2 via p53 and c-Jun seem to confirm a significance of MMR pathways for the repair of UV-induced DNA damage. On the other hand Rochette et al^^ previously reported that human cells with homozygous mutations in the DNA mismatch repair genes hMLH 1 (homologs of mutL) or hMSH2 were proficient in NER. Thus the relevance of MMR concerning UV-induced DNA disruptions remains controversial.

UV-Induced Carcinogenesis in Squamous Cell Carcinoma (SCC) The precursor cell of SCC and its precancerous progenitor, the actinic keratosis (AK), is assumed to be the interfoUicular epidermal basal keratinocyte. AK and SCC is strongly related to UV exposure, because predilection sites were the regularly sun exposed skin areas (head, forehead, back of nose, ears, back of hands) in about 90% and the lifetime risk to develop a SCC correlates very closely with the individual cumulative UV dose.^^ Historically, the connection between sunlight and epithelial skin cancer was initially described by Unna^^ and Dubreuilh ^ in the last decade of the 19 century. They observed AK

24

Molecular Mechanisms of Basal Cell and Squamous Cell Carcinomas

and s e c in chronically sun-exposed skin from sailors and vineyard workers. The carcinogenic potential of UV radiation, substantiating the clinical observations by Unna and Dubreuilh, has been proved by animal experiments in the 1920s.^^ 40 years later hairless mice became available and thus an excellent animal model to investigate skin cancer induction by UV with similarity to human skin has been established. By means of hairless mice SCC with AK as precancerous lesion could be generated by UV and later on de Grujil demonstrated a wavelength dependency due to SCC induction effecivity: the murine UV action spectrum peaks at - 300 nm in the UVB band and continuously decreases among the UVA band with a smaller peak at -- 380 nm. Thereupon these findings were applied to human skin estimating the differences in UV transmission characteristics between murine and human skin and it could have been demonstrated, that the relative efficacy to induce SCC by 300 nm UVB irradiaton is - 1000-fold higher than UVA at 380 nm. ^ The estimated human UV action spectrum correlates very closely with the measured concentration of CPD s in human skin in situ following UV radiation with a single peak at - 315 nm.^'"^ In contrast to UVC and UVB, the amount of inducible CPD by UVA irradiation is rather low. On the other hand it is meanwhile well established that UVA is also able, but less effective than UVB, to induce SCC, at least in mice. The small -- 380 nm peak in de Grujil s action spectrum can be explained by the above mentioned UVA mediated indirect DNA damage: UVA predominandy causes photosensitizermediated oxidative DNA damage. In a teratocarcinoma cell line it has been demonstrated that the greatest amount of DNA-protein crosslinks and single strand breaks has been generated by wavelengths between 334 and 405 nm with a peak at 365 nm.^^ As in this part of the UVA spectrum (347-400 nm) direct DNA damage is not observed, it can be assumed that the indirect oxidative UVA effects contribute to the mutagenic potential of UVA in the above mentioned wavelength range. Finally, all these findings make clear that UVB is even more powerfiil than UVA to induce AK and SCC and has to be considered as predominant carcinogen. This is furthermore confirmed by the mutation characteristics in AK and SCC as described below. Numerous studies verify the multistep carcinogenesis model of SCC as precancerous AK can be found in up to 97% adjacent to SCCs. The necessity of further mutational events establishing an invasive SCC from an AK is reflected by the fact that not all AK, which are reversible, result in invasive SCC: the 10-year risk for the development of AK into SCC is not more dian 16%.^^ The most important protein involved in early UV-induced carcinogenesis of SCC appears to be the tumor suppressor p53. p53 is an essential and well defined transcription factor regulating cell cycle control and apotosis. ^ The most well known function is activation of the Gl checkpoint which is believed to provide time for DNA repair prior to entry into S-phase. If the extent of DNA damage exceeds repair capacity and therefore is irreversible, p53 activates pro-apoptotic factors inducing programmed cell death independent from cell cycle control. UV-radiation is known to strongly induce p53 posttranslational stabilization and transcriptional activity by protection of protein degradation via MDM2 oncoprotein inhibition. ^^'^^ Mutations of the p53 gene were found in various tumors and play a general role for carcinogenesis.^ Thus UV specific p53 mutations can be found in 75 - 80% of AK^^ and in more than 90% of cutaneous squamous cell carcinoma. ^^'^^ These UV-fingerprint mutations of p53 mutations appear to correspond predominandy to UVB radiation, as UVA-induced carcinomas in hairless mice reveal p53 mutations only in 15%. ^ The important role of p53 for the protection against UV-induced cancer is furthermore demonstrated by p53-deficient mice: these knockout animals lacking a functional p53 reveal an enormously increased sensitivity to skin cancer by UVB radiation.^ The p53 gene is located on chromosome 17 (17pl3.1).^^ Dysfunction of the p53 protein, similar to other tumor suppressor proteins, depends on truncating mutations in both alleles. As only few dipyrimidine hot spot areas in the p53 gene are susceptible for UV mutations (codons 177, 196, 278, 294, and 342), the probability for corresponding mutations on both alleles is appropriate.^^ Moreover, loss of heterozygosity (LOH) in several chromosomes, including 17p,

UVDamage and DNA Repair in Basal Cell and Squamous Cell Carcinomas have been commonly found in AK.^^ In vitro abrogation of both p53 alleles causes aneuploidy and uncontrolled gene amplification. Such additional events and further unknown UV accumulative mutations in susceptible genes encoding for example other growth control factors or DNA repair enzymes may be responsible for the multistep carcinogenesis from sunburn cells to AK and ultimately to SCC. Recent data suggest that the MMR protein hMSH2 is a novel p53 regulated target gene indicating a direct involvement of p53 in DNA repair mechanisms. Scherer et al. cloned the promoter region of hMSH2 and could detect a site with homology to the p53 consensus binding sequence.^^ Furthermore they demonstrated that purified p53 binds specifically to this hMSH2motif.^^As our group and others were previously able to demonstrate that functionally active MMR protein hMSH2 is detectable in normal human skin, melanocytic and nonmelanocytic tumors, a link between UV radiation, DNA repair and carcinogenesis of skin cancer was supposed. ' ^ Recendy Liang et al. reported that hMSH2 expression is elevated in precancerous skin lesions (AK, Bowen's disease), but not in SCC compared to adjacent normal skin. These observations have been confirmed by our group (Figs. 4A-B) and underline the above mentioned physiological importance of MMR due to UV-induced DNA damage in precancerous skin lesions; diminished hMSH2 expression in SCC could reflect the malignant transformation associated with DNA repair deficiencies following p53 dysfunction (second allelic hit, see above). If down-regulation of hMSH2 is just a side effect or essential for the carcinogenesis of SCC is yet unsolved.

UV-Induced Carcinogenesis in Basal Cell Carcinoma (BCC) BCC is the most common malignancy in white people with a worldwide increasing incidence. Exposure to UV radiation is assumed to be the main causative pathogenic factor for BCCs as well, but the precise relation between amount, timing, and pattern of UV exposure and BCC risk is still discussed. Compared to SCC, the correlation of UV with basal cell carcinogenesis is far less obvious and epidemiological data are not completely in line with an impact of cumulative UV dose. Some studies revealed a link between cumulative UV dose and BCC risk, although the relative risks were small with odds ratios of 1.0 to 1.5.^^'^^ Odier investigations failed to demonstrate such an association. Thereupon the predilection sites of BCC do not exacdy correspond with the "sun terraces" as in SCC; BCC predominandy affect the seborrhoical central parts of the face (root of nose — can thus region, nasal flaring, nasolabial fold), the head, the trunk and lower limbs; BCCs on the backs of the hands are absolutely rare. Thus the anatomic distribution is not exclusively explainable with cumulative UV exposition. Parallel to melanoma, sunburns in childhood and adolescence may contribute more to BCC and recreational sun exposure in childhood seems to be an important risk factor. ^' ^ In this context it has been recently demonstrated, that in adulthood sunscreens protect against new formation of SCCs but not BCCs. Another facet underlines, that cumulative UV cannot be the decisive causative agent in BCC: Patients with BCC have a three year risk to develop a subsequent BCC between 33% and 77%, but only a 6% risk to develop a SCC.^ The BCC precursor cell is less well defined than in SCC and is supposed to stem from interfoUicular epidermal basal keratinocytes with retained basal morphology, from follicular outer root sheath or sebaceous gland derived keratinocytes.'''^ The deeper anatomic localization of the BCC originating cells in hair follicles and sebaceous glands may be one explanation for differences in the carcinogenesis of BCC and SCC. BCCs are predominandy sporadic, or appear in persons suff^ering from rare hereditary disorders like nevoid basal cell carcinoma syndrome (NBCCS; Gorlin's syndrome) or XP (see above). NBCCS is an autosomal dominant neurocutaneous disorder characterized by multiple BCCs occurring early in life, development of other tumors (medulloblastoma, ovarian fibroma), jaw cysts, calcification of falx cerebri, skeletal defects, pitting of the palms and the soles of the feet and other abnormalities.^'^ Mutations in the human homologue of Drosophila segment polarity gcnQ patched (PTCHl) are responsible for Gorlin's syndrome^^' and UV exposition does not play a main role for BCC development in this complex disease. Physiologically, PTCH1

25

Molecular Mechanisms ofBasal Cell and Squamous Cell Carcinomas

26

A

•v..

*

'^ - ^

'53 has been found to inhibit angiogenesis through up r^ulation ofThrombospondin 1 (TSPl), Brain-specific angiogenesis inhibitor 1 {BAIl), plasminogen activator inhibitor type 1 (PAI-1) and some other angiogenesis inhibitors.

Loss of Wild-Typep53 Functions Due to Mutations: Mutantp53 Gain ofFunction Normal functions of wild-type p53 are abrogated by mutations in this gene. Several studies have shown that the p53 gene is a common target for carcinogen-specific mutations and that more than 50% of htunan cancers contain point mutations. All mutations identified in tumor-derived p53 gene are point mutations in the DNA-binding domain (amino acids 96-292). Acquisition of point mutation in one allele may be sufficient for transdominant suppression of the wild-type p53 and loss of p53 functions due to defective binding to promoters containing wild-type p53 response elements PuPuPuCA/TA/TGPyPyPy.^^' p53 deficient cells may thus have an impaired ability to execute cell-cycle arrest, DNA repair and apoptosis. ' '^' For example, p53-deficient thymocytes are remarkably resistant to radiation-induced apoptosis. ^'^^ Epidermal keratinocytes with mutated p53 are resistant to UV-induced apoptosis^^'^^ though this effect may depend on their differentiation stage.^^ Most of the p53 knockout mice show gready increased genetic instability and develop malignant diseases by month 6 of age.^ p53 knockout mice lacking one or both copies of the p53 gene are also increasingly susceptible to UV carcinogenesis. In addition to the loss of functions, certain p53 mutants have been shown to exert oncogenic features due to acquisition of novel functions that confer proliferative advantage to cells sustaining genetic damage.^'^^^^ Some tumor-derived p53 mutants, when introduced into p53-null cells, promote tumorigenicity and tissue invasiveness, increases frequency of metastasis and resistance to p53-independent apoptosis induced by chemotherapeutic drugs. We have recendy found that UV-induced mouse p53 mutant proteins are phosphorylated at critical serine residues and accumulate in large amounts in cell nucleus, potentially fulfilling some of the requirements for being a "gain of function" mutants.^

p53 Protein and Nonmelanoma Skin Cancer

69

Induction of p53 Protein by UV and Biologic Consequences Solar ultraviolet radiation causes DNA damage, photoperoxidation of lipids, protein cross-linking, sunburn, immunosuppression, photoaging and cancer. p53 protein acts as molecular sensor for the effects of UV radiation by mediating cell cycle arrest and apoptosis, or sunburn, in damaged epidermal keratinocytes, thereby functioning as a "guardian of tissue".^ ^'^^'^^ Numerous studies demonstrated UV-induced activation of p53 protein both in cell cultures and in human and mouse tissues.^^'^^'^^ Studies by Nelson and Kastan^^ indicated that UV-induced DNA lesions, pyrimidine dimers, when accompanied by excision repair-associated DNA strand breaks, trigger p53 induction. The mechanism by which DNA strand breaks induce p53 is not well understood, but may involve the activation of an ATM or ATR protein kinase activity. Indeed, UV radiation causes phosphorylation of p53 protein at multiple serine residues, including Serl5, Ser20, Ser33, Ser37, Ser46, and Ser392.^^'^^ Evidence exists for the involvement of ATM, ATR, and p38, ERKl/2 and JNK-1 MAP kinases in phosphorylation of various p53 serine residues in response to UV irradiation.^^'^^ Excessive DNA damage induced by UV may trigger apoptosis in a p53-dependent manner. '^ '^ ' ^ Brash and coworkers demonstrated that UV irradiation of normal mouse skin containing wild-type p53 protein induced the formation of sunburn cells (apoptotic keratinocytes), and that p53-null mice were resistant to UV-induced sunburn cell formation.^^ Our studies in Skh-hrl mice showed that acute UV exposure induces expression of p53, followed by induction of p21Wafl/Cipl, and apoptosis.

A Model of UV-Induced Initiation and Progression of Squamous Cell Carcinomas The best characterized model of carcinogenesis is that of the UV-induced development of skin carcinomas. Figure 2 depicts the stages of initiation and progression of SCC, in which mutation-associated inactivation of p53 tumor suppressor gene plays a critical role. ' Analysis of data on gene mutations in human premalignant actinic keratosis (AK) lesions, as well as data from the UV-induced carcinogenesis experiments in mice have suggested that the first step involves acquisition of UV-induced mutations in the p53 by epidermal keratinocytes.^^' '^^ This defect diminishes sunburn cell formation and enhances cell survival allowing retention of initiated, precancerous keratinocytes.^^ Second, chronic exposures to solar UV results in the accumulation of p53 mutations in skin, which confer a selective growth advantage to initiated keratinocytes and allow their clonal expansion, leading to formation of premalignant AK.^^ The expanded cell death-defective clones represent a larger target for additional UV-induced p53 mutations or mutations in other genes, thus enabling progression to carcinomas.

p53 Mutations in Nonmelanoma Skin Cancers UV'Induced DNA Lesions and Mutations in the p53 Gene Analysis of mutations in p53 gene has established an unequivocal connection between UV exposure, DNA damage, and skin carcinogenesis. UVB and UVC radiation induces unique types of DNA damage, producing cyclobutane-type pyrimidine dimers (CPD) and pyrimidine {G-A) pyrimidone or {6-4t) photoproducts.^^'^^^ Both lesions are formed exclusively in runs of tandemly located pyrimidine residues, which are often "hot spots" for UV-induced DNA damage and mutation.^^^'^^ These unique lesions in the DNA give rise to unique mutations. UV radiation induces predominandy C—>T and CC->TT transitions at dipyrimidine sequences, which have become the "signature" of UV-induced mutagenesis.^ The C ^ T and C C ^ T T mutations are hypothesized to arise during semi conservative replication of the DNA due to misincorporation of A residues at non-instructional sites in accordance with the A rule: when DNA polymerase comes across lesions in the DNA template that it cannot interpret, it inserts A residues by default. ^^^ Thus, cytosine-cytosine dimers give rise to C ^ T or CC—>TT mutations because the DNA polymerase inserts A residues opposite C residues. During semi conservative DNA replication, T residues are inserted in the newly synthesized strand.

70

Molecular Mechanisms ofBasal Cell and Squamous Cell Carcinomas

Normal l a i sskin

1

Acute UV

DNA damage, sunburn (apoptosis) mb

i

Chronic UV

p53 mutations, inhibition of apoptosis, clonal expansion

1

Chronic UV

Dysplasia, Actinic keratosis

1 sec Figure 2. Model for UV-induced initiation and progression of SCC. UVB radiation induces DNA damage and apoptosis in epidermal keratinocytes. Generation of DNA photoproducts by UVB and defects in DNA repair and replication leads to accumulation of mutations in p53 tumor suppressor gene in keratinocytes and apoptosis resistance. Upon repeated exposure to UV, p53-mutant cells undergo clonal expansion and accumulate mutations in other key genes. Some clones develop into preneoplastic lesions (actinic keratosis), and some of them progress into SCC.

p53 Mutations in Human Skin Cancers p53 Mutations in Human Pre-Cancerous Skin Lesions The mutations in p53 gene appear to be an early genetic change in the development of UV-induced skin cancers. Thousands of p53-mutant cell clones are found in normal-appearing sun-exposed skin. '^^^ High frequency of p53 mutations has been reported in pre-malignant actinic keratosis lesions, which are considered to be pre-SCCs. For example, in AK study by Ziegler et al,^^ p53 mutations were found at a GG% frequency and a high proportion of them (23/35) were C ^ T transition. Nelson et al^^^ showed that 8 of 15 (53%) AKs had C ^ T transition in p53 gene. Study by Campbell et al^^^ showed that 40% (8 out of 20) of individuals with Bowen's disease carried p53 mutations. These early findings suggested that p53 mutations may be involved in the malignant conversion of precancerous lesions to SCCs and that mutations in p53 and/or p53 over expression associated with it may be used as biomarkers for skin cancer susceptibility. Since that, the presence of UV signature C—>T and CC-^TT mutations in the p53 gene in human and experimental mouse skin cancers has been well documentecl.58-"^'2i p53 Mutations in Human SCC and BCC of the Skin A number of investigators have detected/>53 gene mutations in a large proportion of human squamous cell carcinomas and basal cell carcinomas.^^'^^'^^^'^^^ Initial studies by Brash and coworkers^ revealed p53 mutation in 58% of human SCC. Of 24 SCCs analyzed, 3

p53 Protein and NonmeUnoma Skin Cancer

71

exhibited CC-^TT double base changes, 5 tumors had C—>T transitions. Later studies by Ziegler et al, and Rady et al^^^ have demonstrated ^53 mutations in human BCCs at 56% and 50% frequencies, respectively. In the former study, 11 of 20 mutations detected were C ^ T transitions and 4 were CC - ^ T T double base substitutions, whereas in the latter study 4 of 7 mutations detected were C ^ T transitions. Interestingly, Ziegler et al^^ found diat 45% of human BCCs contained a second point mutation on the other p53 allele. More recendy, Bolshakov et al analyzed 342 tissues from patients with aggressive and nonaggressive BCCs and SCCs for p53 mutations by single-strand conformation polymorphism and nucleotide sequencing. p53 mutations were detected in 33 of 50 aggressive BCCs {GG%), 37 of 98 nonaggressive BCCs (38%), 28 of 80 aggressive SCCs (35%), 28 of 56 nonaggressive SCCs (50%), and 3 of 29 samples of sun-exposed skin (10%). About 7 1 % of the p53 mutations detected in aggressive and nonaggressive BCCs and SCCs were UV signature mutations.^^^ We have also analyzed nonmelanoma skin tumors from the psoriasis patients treated with psoralen+UVA.^^^ Out of 69 tumors tested, 37 (54%) tumors had one or more p53 mutations. Of 37 tumors with mutations, 17 (46%) tumors had only ultraviolet-type mutations, two (5%) tumors had only psoralen + ultraviolet A-type mutations, and 18 (49%) tumors had both types of mutations. Most recendy. Agar et al have examined 8 primary SCCs and 8 pre-malignant solar keratosis lesions for p53 mutations separately, in basal and suprabasal layers of keratinocytes using laser capture microdissection.^ They were able to detect UVA-type mutations (A:T-^C:G transversions) both in SCCs and SC lesions mostly in the basal germinative layer, which contrasted with a predominandy suprabasal localization of UVB-signature mutations in these lesions. This epidermal layer bias was confirmed by immunohistochemical analyses with a superficial localization of UVB-induced CPD contrasting with the localization of UVA-induced 8-hydroxy-2'-deoxyguanine adducts to the basal epithelial layer. The basal location of UVArather than UVB-induced DNA damage and mutation suggests that UVA component of solar radiation is an important carcinogen in the stem cell compartment of the skin. Despite similarly high frequencies of UV-induced p53 mutations in BCCs and SCCs of the skin, some differences exist in the mechanisms of their UV induction. The originating cells may arise from interfoUicular basal cells, hair follicles or sebaceous glands, thus from a deeper zone than the SCC ones, which probably means exposure to different doses or wavelengths of UV. In addition to p53, UV also targets the patched gene {PTCH) for mutations leading to the development of BCC. Mutations in the PTCHgne responsible for hereditary BCCs in Gorlins syndrome, sporadic BCC, and BCCs isolated from xeroderma pigmentosum, although with a lower incidence of "UV signature" (reviewed in ref 123). Besides mutations in the p53 and PTCH genes, a small subset of SCC and BCC of the skin also carries mutations in INK4a/ ART tumor suppressor gene products and ras oncogene. ^'^^ p53 Mutations in NMSC of Patients with Xeroderma Pigmentosum and Renal Allograft Recipients (RAR) p53 mutations have also been found at high frequencies in skin cancers from patients with the genetic disorder Xeroderma Pigmentosum. ' XP patients have an extreme sensitivity to sunlight due to a defect in their ability to repair UV-induced DNA damage and frequendy develop skin cancer. ' Studies by Sato et al^ revealed that 5 of 8 XP skin cancers h a d / 5 3 mutations and of the 6 mutations seen, 2 were C ^ T transitions and 2 were C C - ^ T T double base substitutions. Dumaz et al^^^ showed that/>53 mutations were present in 17 of 43 (40%) skin cancers from XP patients and 6 1 % of these mutations were tandem CC—>TT base substitutions. Immunosuppressed recipients of renal allografts (RAR) are also at much higher risk for skin cancer development. Over-expression of p53 protein and/>53 mutations have been detected in large proportion of SCCs and premalignant lesions in RAR patients. In one study, accumulated p53 was present in 4 1 % of premalignant keratoses, 65% of intraepidermal carcinomas

72

Molecular Mechanisms of Basal Cell and Squamous Cell Carcinomas

and 56% of squamous cell carcinomas from RAR patients.^"^^ McGregor et al ^ have showed similarly high incidence of p53 mutations in nonmelanoma skin tumors from RAR patients and sporadic NMSC from immune competent patients: 48% and 63% respectively. Seventy-five percent of all mutations in transplant patients and 100% mutations in non-transplant patients were UV-signature mutations.^ ^ Some evidence suggest that arginine/arginine genotype at a common polymorphism site 2Xp53 codon 72 may confer a susceptibility to the development of NMSC in RAR patients. Finally, some evidence suggest a role for human papillomavirus (HPV) and its p53 protein-inhibitory activity in skin carcinogenesis within the immunosuppressed population. ^^ p53 Mutational Spectra in Human Precancerous Skin Lesions and NMSC Many studies analyzed the/>53 mutational spectra in human skin cancers. Actinic keratoses are premalignant lesions display/>53 mutations clustering between amino acids 200 and 280.^^ However, SCCs show the majority of mutations in the hot spot region 241 to 280. Since SCCs develop from AK it would appear as if those AKs possessing mutations in the hot-spot region 241-280 confer a clonal advantage toward the progression to malignancy. Hence, AK displays a diverse spectrum of mutations that may result in local, noninvasive, benign proliferation. The process of malignant conversion may select for those tumor cells that have acquired a mutation in the region 241 to 280. Mutations in this region may be either more effective in disruption of p53 function (loss of funaion phenotype) or confer oncogenic phenotype to p53 due to gain of function. BCCs appear to have a major mutational hot spot region at 241 to 280 and a minor region at 161-200. Thus, the advantage in acquisition of mutation at 161-200 may be specific for development of BCC. Recent studies on p53 mutation spectra revealed that noncancerous skin adjacent to tumors harbor/^53 gene mutations that are distinct from those present in the skin cancers. This provides molecular evidence for field cancerization and further suggests that only a subset of UV-induced p53 mutations confer cells with malignant phenotype, while other p53 mutants are not necessarily associated with malignant progression. Ren et al have shown that human epidermal cancer cells and accompanying precursors have identical p53 mutations, which are different fromp53 mutations in adjacent areas of clonally expanded non-neoplastic keratinocytes. Kanjilal et al in their study of NMSC of the head and neck and adjacent nonmalignant skin samples, revealed multiple but distinct p53 mutations ( C ^ T transitions at dipyrimidine sequences in 30% of missense mutations) in the two areas. Finally, one report^^^ compared multiple NMSCs with and without p53 mutations in the same XP patient and found that the former tend to exhibit more rapidly-growing and/or histologically immature clinical features, suggesting that/>53 gene mutations would bring more malignant characteristics to NMSCs.

p53 Mutations in Mouse Skin Cancers Frequency ofp53 Mutations in Murine Skin Cancers In mouse photocarcinogenesis experiments, mutations in the/>53 gene are clearly linked to UVB radiation and the p53 alterations seem to be an essential early event in tumor development. ^^^ The frequency and the location of p53 gene mutations vary somewhat among mouse strains. Kress et al^^^ analyzed UVB-induced skin tumors from four different strains of mice (SKH-l/hr, SENCAR, BALB/c, and C3H/He) and found/>53 mutation in 7 of 35 tumors. Thep53 mutations detected in these tumors were all base substitutions at exons 5, 7, and 8. In addition, SCCs from a different strain of mice displayed p53 mutations at different frequencies. Two of 9 (22%) SCCs from SKH-l/hr mice displayed/>53 mutation, while 8 papillomas (precancerous skin lesions in mice) from this strain did not have any mutation. Similarly, 3 of 6 SCCs (50%) from BALB/c mice and 2 of 7 (29%) SCCs from C3H/He mice harbored/53 mutation. In contrast, studies by Ananthaswamy and co-workers have demonstrated/>53 mutation at an unprecedented 100% frequency in UV-induced C3H mouse skin tumors. They

p53 Protein and Nonmelanoma Skin Cancer

73

also found that several of the UV-induced C3H mouse skin cancers contained multiple mutant/53 alleles, which suggests that these tumors are made up of heterogeneous populations of cells. ^ It is possible that mutant ^53 alleles with single base changes were targets of secondary mutation events, perhaps because of continued exposure to UV radiation during tumor progression. Interestingly, it has been shown that the frequency and spectrum ofp53 mutations of UV-induced skin cancers in XPC knockout mice is similar to that of NER-proficient C3H and SKH-hrl mice. Nevertheless, most of the mutations that were detected in UV-induced mouse skin cancers were C—>T and CC->TT transitions at dipyrimidine tracts as seen in the human skin cancers.

p53 Mutational Spectra in Murine NMSC UV-induced murine skin cancers exhibit hot-spot regions at 121-160 and at 241-280. The former region contains the conserved domain II and appears to be murine-specific in UV-induced skin cancers. In AK, some of the more frequent lesions observed are mutations at 235, 245, 247-248 and 278 sites. Specifically, chronic UV exposure seems to favor proliferation of keratinocytes with/>53^^^ mutant allele./>53^ -mutant keratinocytes have been shown to gready out number those containing other hot spot UV-induced/>53 mutations with respect to the quantity of the clones they form,^^^ as well as incidence of tumors c a r r y i n g / 5 3 ^ a l l e les.5?'i^^

Mechanism of Clonal Expansion ofp53 Mutant Keratinocytes Murine model of UV-induced carcinogenesis allowed a unique opportunity for investigating the fate of p53-mutant keratinocytes during various stages of skin cancer development. In skin of hairless mice, p53 mutations induced by chronic UV exposure could be detected by allele-specific PCR as early as one week after initiation of the experiment, with 100% animals incurring^53 mutations after eight weeks of UV treatment. Two-three weeks after beginning the UV treatment, clones of keratinocytes carrying mutant p53 can be already visualized using immunohistochemical assays. ' As a tumor promoter, UV induces cell proliferation by stimulating the production of various growth factors and cytokines, as well as activation of their receptors. Repeated exposure of skin to UV radiation therefore results in clonal expansion of initiated p53-mutant cells. ^^^'^^'^'^^^ Brash and colleagues have shown that every successive UVB exposure allows p53-mutant keratinocytes to colonize adjacent epidermal stem-cell compartments without incurring additional mutations. ^^^ Two mechanisms are believed to contribute to selective expansion of p53-mutant cells: their resistance to UV-induced apoptosis, and their proliferative advantage over normal keratinocytes in response to stimulation with UV. Indeed, single UV exposure was shown to stimulate the proliferation of/>53-mutant cells while inducing apoptosis in normal keratinocytes in culture and in artificial skin models. ^^'^^'^^ However, chronic UV irradiation of skin quickly induces apoptosis-resistance and stimulates hyperproliferation throughout the epidermis as an adaptive response. The mechanism of selective proliferative advantage of/>53-mutant cells is yet unclear, but it may be a critical factor promoting clonal expansion of initiated cells. One mechanism that may contribute to expansion of initiated keratinocytes is the deregulation of UV-induced Fas/Fas-Ligand mediated apoptosis in skin. Hill et al^^ showed that accumulation of p53 mutations in the epidermis of FasL deficient mice occurred at much higher frequency compared with wild-type mice after chronic UV irradiation. Authors concluded that i^^Z-mediated apoptosis is important for skin homeostasis, and that the dysreguration of Fas-FasL interactions may be central to the development of skin cancer. Ouhtit et al ftxrther found that in skin of chronically-irradiated SKH-hrl mice, the progressive decrease of FasL expression was paralleled by accumulation of p53 mutations and the decrease in a number of apoptotic cells. These findings suggest that chronic UV exposure would induce a loss of FasL expression and a gain in p53 mutations, leading to dysregulation of apoptosis, expansion of mutated keratinocytes, and initiation of skin cancer.

74

Molecular Mechanisms of Basal Cell and Squamous Cell Carcinomas

While patches of p53-mutant keratinoq^es grow in density and size while UV treatment continues, they decline rapidly once the UV exposures are ceased. ^^'^'^^^' Remeynic et al showed that regression of precancerous p53-positive clones occurs due to mechanisms other than antigen-specific immunity, proceeding with similar kinetics in the skin of Ragl antigen-specific immunity incompetent mice and their wild-type counterparts, our preliminary results suggest that elimination of p53-mutated keratinocytes occurs due to normal skin turnover. Both continued and discontinued regiments of chronic UV treatment ultimately result in skin tumor development with 100% incidence, although the kinetics of tumor occurrence is delayed in the later case. De Gruijl and coworkers have used a mathematical model that relates tumor occurrence to the daily dose of UV and the time needed to contract tumors. This model also offers prediction of skin cancer susceptibility depending on the load of p53-mutated keratinocyte clones in skin.^^® Thus these studies suggest that skin cancer development can be delayed but not abrogated upon further avoidance of exposure to UV.

References 1. Lane DP, Crawford LV. T antigen is bound to a host protein in SV40-transformed ceils. Nature 1979; 278:261-263. 2. Linzer DI, Levine AJ. Characterization of a 54K dalton cellular SV40 tumor antigen present in SV40-transformed cells and uninfected embryonal carcinoma cells. Cell 1979; 17:43-52. 3. Eliyahu D, Michalovitz D, Eliyahu S et al. Wild-type p53 can inhibit oncogene-mediated focus formation. Proc Natl Acad Sci USA 1989; 86:8763-8767. 4. Levine AJ, Momand J, Finlay CA. The p53 tumour suppressor gene. Nature 1991; 351:453-456. 5. Vogelstein B, Kinzler KW. p53 function and dysfunction. Cell 1992; 70:523-526. 6. Kuerbitz SJ, Plunkett BS, Walsh WV et al. Proc Nad Acad Sci USA 1992; 89:7491-7495. 7. Zahn Q, Carrier F, Fornace Jr AJ. Induction of cellular p53 activity by DNA-damaging agents and growth arrest. Mol Cell Biol 1993; 13:4242-4250. 8. Yonish-Roauch E, Reznitzky D, Lotem J et al. Wild type p53 induces apoptosis of myeloid leukemic cells that is inhibited by IL-6. Nature 1991; 352:345-347. 9. el-Deiry WS. p21/p53, cellular growth control and genomic integrity. Curr Top Microbiol Immunol 1998; 227:121-137. 10. Sionov RV, Haupt Y. The cellular response to p53: The decision between life and death. Oncogene 1999; 18:6145-6157. 11. Harris CC. p53: At the crossroads of molecular carcinogenesis and risk assessment. Science 1993; 262:1980-1981. 12. Tokino T, Thiagalingam S, el-Deiry WS et al. p53 tagged sites from human genomic DNA. Hum Mol Genet 1994; 3:1537-1542. 13. Ko LJ, Prives C. p53: Puzzle and paradigm. Genes Dev 1996; 10:1054-72. 14. Levine AJ. p53, the cellular gatekeaper for growth and division. Cell 1997; 88:323-331. 15. Jin S, Levine AJ. The p53 functional curcuit. J Cell Science 2001; 114:4139-4140. 16. DuUc V, Kaufmann WK, Wilson SJ et al. p53-dependent inhibition of cycUn-dependent kinase activities in human fibroblasts during radiation-induced Gl arrest. Cell 1994; 76:1013-1023. 17. Harper JW, Adami GR, Wei N et al. The p21 cdk-interacting protein Cipl is a potent inhibitor of Gl cyclin-dependent kinases. Cell 1993; 75:805-816. 18. El-Deiry WS, Tokino T, Velculesce VE et al. WAFl, a potential mediator of p53 tumor suppression. Cell 1993; 75:817-825. 19. Cross SM, Sanchez CA, Morgan CA et al. A p53-dependent mouse spindle checkpoint. Science 1995; 267:1353-6. 20. Chan TA, Hermeking H, Lengauer C et al. l4-3-3Sigma is required to prevent mitotic catastrophe after DNA damage. Nature 1999; 401:616-20. 21. Ohki R, Nemoto J, Murasawa H et al. Reprimo, a new candidate mediator of the p53-mediated cell cycle arrest at the G2 phase. J Biol Chem 2000; 275:22627-22630. 22. Taylor WR, Stark GR. Regulation of the G2/M transition by p53. Oncogene 2001; 20:1803-1815. 23. Utrera R, Collavin L, Lazarevic D et al. EMBO J 1998; 17:5015-5025. 24. Sheikh MS, Chen YQ, Smith ML et al. Role of p21Wafl/Cipl/Sdil in cell death and DNA repair as studied using a tetracycline-inducible system in p53-deficient cells. Oncogene 1997; 14:1875-82. 25. Hwang BJ, Ford JM, Hanawalt PC et al. Expression of the p48 xeroderma pigmentosum gene is p53-dependent and is involved in global genomic repair. Proc Nad Acad Sci USA 1999; 96:424-428.

p53 Protein and Nonmelanoma

Skin Cancer

75

26. Amundson SA, Patterson A, D o K T et al. A nucleotide excision repair master-switch: p 5 3 regulated coordinate induction of global genomic repair genes. Cancer Biol Ther 2002; 1:145-149. 27. Carrier F, Georgel PT, Pourquier P et al. G a d d 4 5 , a p53-responsive stress protein, modifies D N A accessibility on damaged chromatin. Mol Cell Biol 1999; 19:1673-1685. 28. Smith ML, Seo YR. p53 regulation of D N A excision repair pathways. Mutagenesis 2002; 17:149-156. 29. Z h o u J, Ahn J, Wilson SH et al. A role for p53 in base excision repair. E M B O J 2 0 0 1 ; 20:914-23. 30. Miyashita T , Reed J C . T u m o r suppressor p 5 3 is a direct transcriptional activator of the h u m a n bax gene. Cell 1995; 80:293-299. 3 1 . Nakano K, Vousden K H . PUMA, a novel proapoptotic gene, is induced by p 5 3 . Mol Cell 2 0 0 1 ; 7(3):683-694. 32. O d a K, Arakawa H , Tanaka T et al. p 5 3 A I P l , a potential mediator of p53-dependent apoptosis, and its regulation by Ser-46-phosphorylated p 5 3 . Cell 2000; 102:849-862. 3 3 . O d a E, Ohki R, Murasawa H et al. Noxa, a BH3-only member of the Bcl-2 family and candidate mediator of p53-induced apoptosis. Science 2000; 288(5468):1053-1058. 34. Polyak K, Xia Y, Zweier J et al. Nature 1997; 389:300-305. 35. MuUer M, Wilder D , Bannasch D et al. p 5 3 activates the C D 9 5 (APOl/Fas) gene in response to D N A damage by anticancer drugs. J Exp Med 1998; 188:2033-2045. 36. Bennet M, McDonald K, Chan S W et al. Cell surface trafficking of Fas: A rapid mechanism of p53-mediated apoptosis. Science 1998; 282:290-293. 37. W u GS, Burns T F , McDonald III ER et al. KILLER/DR5 is a D N A damage-inducible p53-regulated death receptor gene. Nat Genet 1997; 17(2): 141-143. 38. Lin Y, M a W, Benchimol S. Pidd, a new death-domain-containing protein, is induced by p53 and promotes apoptosis. N a t Genet 2000; 26:122-127. 39. Hill LL, Ouhtit A, Loughlin SM et al. Fas ligand: A sensor for D N A damage critical in skin cancer etiology. Science 1999; 285:898-900. 40. Fiucci G, Beaucourt S, Duflaut D et al. Siah-lb is a direct transcriptional target of p 5 3 : Identification of the fimctional p53 responsive element in the siah-lb promoter. Proc Natl Acad Sci USA 2004; 101:3510-3515. 4 1 . Buckbinder L, Talbott R, VelascoMiguel S et al. Induction of the growth inhibitor IGF-binding protein 3 by p 5 3 . Nature 1995; 377(6550):646-649. 42. Dameron KM, Volpert O V , Tainsky M A et al. Science 1994; 265:1582-1584. 4 3 . Nishimori H , Shiratsuchi T, Urano T et al. A novel brain-specific p53-target gene, BAIl, containing t h r o m b o s p o n d i n type 1 repeats inhibits experimental angiogenesis. O n c o g e n e 1 9 9 7 ; 15:2145-2150. 44. Kunz C, Pebler S, Otte J et al. Differential regulation of plasminogen activator and inhibitor gene transcription by the tumor suppressor p 5 3 . Nucleic Acids Res 1995; 23:3710-3717. 45. HoUstein M , Sidransky D , Vogelstein B. p53 mutations in human cancers. Science 1991; 253:49-53. 46. Unger T, Mietz JA, Scheffiier M et al. Mol Cell Biol 1993; 13:5186-5194. 47. Kastan MB, Zhan Q, El-Deiry W S et al. A mammalian cell cycle checkpoint pathway utilizing p53 and G A D D 4 5 is defective in ataxia-telangiectasia. Cell 1992; 71:587-597. 48. Rowan S, Ludwig RL, H a u p t Y. Specific loss of apoptotic but not cell-cycle arrest function in a human tumor derived p53 mutant. E M B O J 1996; 15:827-838. 49. Lowe SW, Schmitt EM, Smith S W et al. Nature 1993; 362(6423):847-849. 50. Clarke AR., Purdie CA, Harrison DJ et al. Thymocyte apoptosis induced by p53-dependent and independent pathways. Nature 1993; 362(6423):849-852. 5 1 . Ziegler A, Jonason AS, Leffell DJ et al. Sunburn and p53 in the onset of skin cancer. Nature 1994; 372:773-776. 52. Mudgil AV, Segal N , Andriani F et al. Ultraviolet B irradiation induces expansion of intraepithelial tumor cells in a tissue model of early cancer progression. J Invest Dermatol 2003; 121:191-197. 53. T r o n VA, Trotter MJ, T a n g L et al. p53-regulated apoptosis is differentiation dependent in ultraviolet B-irradiated mouse keratinocytes. A m J Pathol 1998; 153:579-585. 54. Donehower LA, Harvey M , Slagle BL et al. Mice deficient for p53 are developmentally normal but susceptible to spontaneous tumours. Nature 1992; 356(6366):215-221. 55. Jiang W, Ananthaswamy H N , MuUer H K et al. Oncogene 1999; 18(29):4247-4253. 56. Levine AJ, M o m a n d J, Finlay CA. T h e p53 tumour suppressor gene. Nature 1991; 351:453-456. 57. Sigal A, Rotter V. Oncogenic mutations of the p53 tumor suppressor: T h e demons of the guardian of the genome. Cancer Res 2000; 60:6788-6793. 58. Greenblatt M S , Bennett W P , HoUstein M et al. Mutations in the p53 tumor suppressor gene: Clues to cancer etiology and molecular pathogenesis. Cancer Res 1994; 54(18):4855-4878. 59. Dittmer D , Pati S, Zambetti G et al. Gain of fimction mutations in p 5 3 . N a t Genet 1993; 4:42-46. 60. Michalovitz D , Halevy O , Oren M. p53 mutations: Gains or losses? J Cell Biochem 1991; 45:22-29.

76

Molecular Mechanisms of Basal Cell and Squamous Cell Carcinomas

61. Hsiao M, Low J, Dorn E et al. Gain-of-function mutations of the p53 gene induce lymphohematopoietic metastatic potential and tissue invasiveness. Am J Pathol 1994; 145:702-714. 62. Gloushankova N, Ossovskaya V, Vasiliev J et al. Changes in p53 expression can modify cell shape of ras-transformed fibroblasts and epitheliocytes. Oncogene 1997; 15:2985-2989. 63. Kremenetskaya OS, Logacheva NP, Baryshnikov AY et al. Distinct effects of various p53 mutants on differentiation and viability of human K562 leukemia cells. Oncol Res 1997; 9:155-166. 64. Li R, Sutphin PD, Schwartz D et al. Mutant p53 protein expression interferes with p53-independent apoptotic pathways. Oncogene 1998; 16:3269-3277. 65. Liu G, McDonnell TJ, Montes de Oca Luna R et al. High metastatic potential in mice inheriting a targeted p53 missense mutation. Proc Natl Acad Sci USA 2000; 97:4174-4179. Ci(i. Lotem J, Sachs L. A mutant p53 antagonizes the deregulated c-myc-mediated enhancement of apoptosis and decrease in leukemogenicity. Proc Natl Acad Sci USA 1995; 92:9672-9676. 67. Peled A, Zipori D, Rotter V. Cooperation between p53-dependent and p53-independent apoptotic pathways in myeloid cells. Cancer Res 1996; 56:2148-2156. 68. Sun Y, Nakamura K, Wendel E et al. Progression toward tumor cell phenotype is enhanced by overexpression of a mutant p53 tumor-suppressor gene isolated from nasopharyngeal carcinoma. Proc Natl Acad Sci USA 1993; 90:2827-2831. 69. Wolf D, Harris N, Rotter V. Reconstitution of p53 expression in a nonproducer Ab-MuLVtransformed cell line by transfection of a functional p53 gene. Cell 1984; 38:119-126. 70. Matas D, Sigal A, Stambolsky P et al. Integrity of the N-terminal transcription domain of p53 is required for mutant p53 interference with drug-induced apoptosis. EMBO J 2001; 20:4163-4172. 71. Frazier MW, He X, Wang J et al. Activation of c-myc gene expression by tumor-derived p53 mutants requires a discrete C-terminal domain. Mol Cell Biol 1998; 18:3735-3743. 72. Melnikova VO, Santamaria AB, Bolshakov SV et al. Mutant p53 is constitutively phosphorylated at serine 15 in UV-induced mouse skin tumors: Involvement of ERKl/2 MAP kinase. Oncogene 2003; 22:5958-5966. 73. Wang XW, Forrester K, Yeh H et al. Hepatitis B virus X protein inhibits p53 sequence-specific DNA binding, transcriptional activity and association with transcriptional factor ERCC. Proc Natl Acad Sci USA 1994; 91:2230-2234. 74. Yamaizumi M, Sugano T. UV-induced nuclear accumulation of p53 is evoked through DNA damage of actively transcribed genes, independent of the cell cycle'. Oncogene 1994; 9:2275-2284. 75. Brash DE. Cellular proofreading. Nat Med 1996; 2:525-526. Id. Gujuluva CN, Back JH, Shin KH et al. Effect of UV-irradiation on cell cycle, viability and the expression of p53, gaddl53 and gadd45 genes in normal and HPV-immortalized human oral keratinocytes. Oncogene 1994; 7:1819-1827. 17. Maltzman W, Czyzyk L. UV irradiation stimulates levels of p53 cellular tumor antigen in nontransformed mouse cells. Mol Cell Biol 1984; 4:1689-1694. 78. Healy E, Reynolds NJ, Smith MD et al. Dissociation of erythema and p53 protein expression in human skin following UVB irradiation, and induction of p53 protein and mRNA following application of skin irritants. J Invest Dermatol 1994; 103:493-499. 79. Ouhtit A, Muller HK, Davis WD. Ananthaswamy: Temporal events in skin injury and the early adaptive responses in ultraviolet-irradiated mouse skin. Am J Pathol 2000; 156:201-207. 80. Ouhtit A, Corny A, Muller HK et al. Loss of Fas-Ligand expression in mouse keratinocytes during UV carcinogenesis. Am J Pathol 2000; 157:1975-1981. 81. Hall P, McKee PH, Menage HP et al. High levels of p53 protein in UV-induced normal human skin. Oncogene 1993; 8:203-207. 82. Campbell C, Quinn AG, Angus B et al. Wavelength specific patterns of p53 induction in human skin following exposure to UV radiation. Cancer Res 1993; 53:2697-2699. 83. Nelson WG and Kastan MB. DNA strand breaks: The DNA template alterations that trigger p53-dependent DNA damage response. Mol Cell Biol 1994; 14:1815-1823. 84. Siliciano JD, Canman CE, Taya Y et al. DNA damage induces phosphorylation of the amino terminus of p53. Genes Dev 1997; 11:3471-3481. 85. Chehab NH, Malikzay A, Stavridi ES et al. Phosphorylation of Ser-20 mediates stabilization of human p53 in response to DNA damage. Proc Natl Acad Sci USA 1999; 96:13777-13782. 86. Kapoor M, Lozano G. Functional activation of p53 via phosphorylation following DNA damage by UV but not gamma radiation. Proc Natl Acad Sci USA 1998; 95:2834-2837. 87. Kapoor M, Hamm R, Yan W et al. Cooperative phosphorylation at multiple sites is required to activate p53 in response to UV radiation. Oncogene 2000; 19:358-364. 88. Keller DM, Lu H. p53 serine 392 phosphorylation increases after UV through induction of the assembly of the CK2.hSPT16 SSRPl complex. J Biol Chem 2000; 277:50206-50213.

p53 Protein and Nonmelanoma

Skin Cancer

77

89. Saito S, Yamaguchi H , Higashimoto Y et al. Anderson Phosphorylation site interdependence of h u m a n p 5 3 p o s t - t r a n s l a t i o n a l m o d i f i c a t i o n s in response to stress. J Biol C h e m 2 0 0 3 ; 278:37536-37544. 90. Bannin S, Moyal L, Shieh SY et al. Science 1998; 281:1674-1679. 91. Tibbetts RS, Brumbaugh KM, Williams JM et al. A role for ATR in the D N A damagp-induced phosphorylation of p53. Genes Dev 1999; 13:152-157. 92. Milne D M , Campbell LE, Campbell D C et al. p 5 3 is phosphorylated in vitro and in vivo by an ultraviolet radiation-induced protein kinase characteristic of the c-Jun kinase, J N K l . J Biol Chem 1995; 270:5511-5518. 93. She Q B , Chen N , D o n g Z. ERKs and p38 kinase phosphorylate p53 protein at serine 15 in response to U V radiation. J Biol Chem 2000; 275:20444-20449. 94. Brash DE, Ziegler A, Jonason AS et al. Sunlight and sunburn in human skin cancer: p 5 3 , apoptosis, and tumor promotion. J Investig Dermatol Symp Proc 1996; 1:136-142. 95. Smith ML, Fornace Jr AJ, p53-mediated protective responses to U V irradiation. Proc Natl Acad Sci USA 1997; 94:12255-12257. 96. Brash DE, Rudolph JA, Simon JA et al. A role for sunlight in skin cancer: UV-induced p53 mutations in squamous cell carcinoma. Proc Natl Acad Sci USA 1991; 88:10124-10128. 97. Leffell DJ, Brash DE. SunUght and skin cancer. Sci A m 1996; 275:52-53, 56-59. 98. Ananthaswamy H N , Loughlin SM, Cox P et al. Sunlight and skin cancer: Inhibition of p53 mutations in UV-irradiated mouse skin by sunscreens. N a t Med 1997; 3:510-514. 99. Setlow RB, Carrier WL. Pyrimidine dimers in ultraviolet-irradiated DNA's. J Mol Biol 1966; 17:237-254. 100. Mitchell DL. T h e relative cytotoxicity of (6-4) photoproducts and cyclobutane dimers in mammaUan cells. Photochem Photobiol 1988; 48:51-57. 101. Mitchell D L , Nairn RS. T h e biology of the 6-4 photoproduct. Photochem Photobiol 1989; 49:805-819. 102. Sage E. Disribution and repair of photolesions in D N A : Genetic consequences and the role of sequence context. Photochem Photobiol 1993; 57:163-174. 103. Brash DE. U V mutagenic photoproducts in Escherichia coli and human cells: A molecular genetics perspective on human skin cancer. Photochem Photobiol 1988; 49:59-66. 104. Mitchell DL, Jen J, Cleaver JE. Sequence specificity of cyclobutane pyrimidine dimers in D N A treated with solar (ultraviolet B) radiation. Nucleic Acid Res 1992; 20:225-229. 105. Brash D E , Rudolph JA, Simon JA et al. A role for sunlight in skin cancer: UV-induced p53 mutations in squamous cell carcinoma. Proc Natl Acad Sci USA 1991; 88:10124-10128. 106. Nakazawa H , EngUsh D , Randell PL et al. U V and skin cancer: Specific p53 gene mutation in normal skin as a biologically relevant exposure measurement. Proc Natl Acad Sci USA 1994; 91:360-364. 107. Ren ZP, H e n d r u m A, Potten F et al. Oncogene 1996; 12:765-773. 108. Jonason AS, Kunala S, Price GJ et al. Frequent clones of p53-mutated keratinocytes in normal human skin. Proc Natl Acad Sci USA 1996; 93:14025-14029. 109 Nelson MA, Einspahr J G , Alberts DS et al. Analysis of p 5 3 gene in human precancerous actinic keratosis lesions and squamous cell cancers. Cancer Lett 1994; 85:23-29. 110. Campbell C, Q u i n n AG, Ro YS et al. p53 mutations are common and early events that precede tumor invasion in squamous cell neoplasia of the skin. J Invest Dermatol 1993; 100:746-748. 111. Rady P, Scinicariello F, Wagner RF et al. p53 mutations in basal cell carcinomas. Cancer Res 1992; 52:3804-3806. 112. Ziegler A, Lefi^ell DJ, Kunala S et al. Mutation hotspots due to sunlight in the p53 gene of nonmelanoma skin cancers. Proc Natl Acad Sci USA 1993; 90:4216-4220. 113. Pierceall W E , Mukhopadhyay T , Goldberg L H et al. Mutations in p53 tumor suppressor gene in human cutaneous squamous cell carcinomas. Mol Carcinog 1991; 4:445-449. 114. Moles JP, Moyret C, Guillot B et al. p53 gene mutations in human epithelial skin cancers. Oncogene 1993; 8:583-588. 115. Dumaz N , Drougard C, Sarasin A et al. Specific UV-induced mutation spectrum in the p53 gene of skin tumors in D N A repair deficient xeroderma pigmentosum patients. Proc Natl Acad Sci USA 1993; 90:10529-10533. 116. Sato M , Nishigori C, Zghal M et al. Ultraviolet-specific mutations in the p53 gene in skin tumors in xeroderma pigmentosum patients. Cancer Res 1993; 53:2944-2946. 117. Van der Riet P, Karp D , Farmer E et al. Progression of basal cell carcinoma through loss of chromosome 9q and inactivation of a single p53 allele. Cancer Res 1994; 54:25-27. 118. Bolshakov S, Walker C M , Strom SS et al. p53 mutations in human aggressive and nonaggressive basal and squamous cell carcinomas. Clin Cancer Res 2003; 9:228-234.

78

Molecular Mechanisms

of Basal Cell and Squamous

Cell Carcinomas

119. Stern RS, Bolshakov S, Nataraj AJ et al. p53 mutation in nonmelanoma skin cancers occurring in psoralen ultraviolet a-treated patients: Evidence for heterogeneity and field cancerization. J Invest Dermatol 2002; 119:522-526. 120. Kress S, Sutter C, Strickland P T et al. Carcinogen-specific mutational pattern in the p53 gene in ultraviolet B radiation-induced squamous cell carcinomas of mouse skin. Cancer Res 1992; 52:6400-6403. 121. Kanjilal S, Pierceall W E , C u m m i n g s KK. H i g h frequency of p 5 3 mutations in ultraviolet radiation-induced murine skin tumors: Evidence for strand bias and tumor heterogeneity. Cancer Res 1993; 53:2961-2964. 122. Agar N S , HalHday C M , Barnetson RStC et al. T h e basal layer in human squamous tumors harbors more UVA than UVB fingerprint mutations: A role for UVA in human skin carcinogenesis. Proc N a d Acad Sci USA 2004; 101:4954-4959. 123. Matsumura Y, Ananthaswamy H N . Molecular mechanisms of photocarcinogenesis. Front Biosci 2002; 7:d765-783. 124. Kraemer K H , Lee M M , Scotto J. Xeroderma pigmentosum: Cutaneous ocular and neurologic abnormalities in 830 published cases. Arch Dermatol 1987; 123:241-250. 125. Cleaver JE, Kraemer K H . Xeroderma pigmentosum. In: C R Scriver, AL Beaudet, SS Sly, D Valle, eds. T h e Metabolic Basis of Inherited Disease. New York: McGraw-Hill, 1989; 2949-2971. 126. Cleaver JE. Defective repair replication of D N A in xeroderma pigmentosum. Nature 1968; 218:652-656. 127. Stark LA, Arends MJ, McLaren KM et al. Accumulation of p53 is associated with tumour progression in cutaneous lesions of renal allograft recipients. Br J Cancer 1994; 70(4):662-7. 128. McGregor J M , Berkhout RJ, Rozycka M. p53 mutations implicate sunlight in post-transplant skin cancer irrespective of human papillomavirus status. Oncogene. 1997; 15(14):1737-1740. 129. McGregor J M , Harwood CA, Brooks L et al. Relationship between p 5 3 codon 72 polymorphism and susceptibility to sunburn and skin cancer. J Invest Dermatol 2002; 119(l):84-90. 130. Purdie KJ, Pennington J, Proby C M et al. T h e promoter of a novel human papillomavirus (HPV77) associated with skin cancer displays U V responsiveness, which is mediated through a consensus p53 binding sequence. E M B O J 1999; 18(19):5359-5369. 131. Kanjilal S, Strom S, Clayman G et al. p53 mutations in nonmelanoma skin cancer of head and neck: Molecular evidence for field cancerization. Cancer Res 1995; 55:3604-3609. 132. Matsumura Y, Sato M, Nishigori C et al. High prevalence of mutations in the p53 gene in poorly differentiated squamous cell carcinomas in xeroderma pigmentosum patients. J Invest Dermatol 1995; 105:399-401. 133. Van Kranen HJ, Gruijl FR. Mutations in cancer genes of UV-induced skin tumors of hairless mice. J Epidemiol 1999; 9(6 Supple):S58-65. 134. Ananthaswamy H N , Ouhtit A, Evans RL et al. Persistence of p53 mutations and resistance of keratinocytes to apoptosis arc associated with the increased susceptibility of mice lacking the XPC gene to U V carcinogenesis. Oncogene 1999; 18:7395-7398. 135. Zhang W , Remenyik E, Zelterman D et al. Escaping the stem cell compartment: Sustained UVB exposure allows p53-mutant keratinocytes to colonize adjacent epidermal proliferating units without incurring additional mutations. Proc N a d Acad Sci USA 2001; 98:13948-13953. 136. You YH, Szabo PE, Pfeifer G P . Cyclobutane pyrimidine dimers form preferentially at the major p53 mutational hotspot in UVB-induced mouse skin tumors. Carcinogenesis 2000; 21:2113-2117. 137. Berg RJW, van Kranen HJ, Rebel H G et al. Early p53 alterations in mouse skin carcinogenesis by UVB radiation: Immunohistochemical detection of mutant p53 protein in clusters of preneoplastic epidermal cells. Proc N a d Acad Sci USA 1996; 93:274-278. 138. Rebel H , Mosnier O , Berg RJW et al. Early p53-postitive foci as indicator of tumor risk in ultraviolet-exposed hairless mice: Kinetics of induction, effects of D N A repair deficiency, and p53 heterozygosity. Cancer Res 2 0 0 1 ; 61:977-983. 139. De-Metys P, Urso B, Christoffcrsen C T et al. Mechanism of insulin and IGF-1 receptor activation and signal transduction specificity. Receptor dimer cross-linking, bell-shaped curves and sustained versus transient signaling. Ann NY Acad Sci 1995; 766:388-401. 140. Rosette C, Karin M. U V and osmotic stress: activation of the JNK cascade through multiple growth factor and cytokine receptors. Science 1996; 274:1194-1197. 141. Bender K, Blattner C, Knebel A et al. UV-induced signal transduction. J Photochem Photobiol B Biology 1997; 37:1-17. 142. Kuhn C, Hurwitz SA, Kumar M G et al. Activation of the insulin-like growth factor-a receptor promotes the survival of human keratinocytes following ultraviolet B irradiation. Int J Cancer 1999; 80:431-438.

p53 Protein and Nonmelanoma Skin Cancer

79

143. Jost M, Kari C, Rodeck U. The EGF-receptor - an essential regulator of multiple epidermal functions. Eur J Dermatol 2000; 10:505-510. 144. Peus D, Vasa RA, Meves A et al. UVB-induced epidermal growth factor receptor phosphorylation is critical for downstream signaling and keratinocyte survival. Photochem Photobiol 2000; 72:135-140. 145. Ullrich SE. The effect of ultraviolet radiation on the immune response. In: Kydonieus AF, Wille JJ, eds. Biochemical Modulation of Skin Reactions. CRC Boca Raton Fl, 2000:281-300. 146. Walterscheid JP, Ullrich SE, Nghiem DX. Platelet-activating factor, a molecular sensor for cellular damage, activates systemic immune suppression. J Exp Med 2002; 195:171-179. 147. Coffer PJ, Burgering BM, Peppelenbosch MP et al. UV activation of receptor tyrosine kinase activity. Oncogene 1995; 11:561-569. 148. Remenyik E, Wikonkal NM, Zhang W. Antigen-specific immunity does not mediate acute regression of UVB-induced p53-mutant clones. Oncogene 2003; 22:6369-6376. 149. de Gruijl FR, van der Leun JC. Development of skin tumors in hairless mice after discontinuation of ultraviolet irradiation. Cancer Res 1991; 51:979-984.

CHAPTER 9

TGF'P Pathway and Cancerogenesis of Epithelial Skin Tumors Miguel Quintanilla,* Eduardo P^rez-G6mez, Diana Romero, Mar Pons and Jaime Renart Introduction

T

he TGF-P family of cytokines, including TGF-Ps themselves, activins/inhibins and bone morphogenetic proteins (BMPs), regulate many fundamental processes during embryonic development and in adult tissues, such as cell growth, differentiation, remodelling of the extracellular matrix, cell migration and adhesion, angiogenesis and the immune response.^ In mammals, there are three major members of the family: TGF-Pi, -p2 and -P3, which are encoded by different genes but share similar biological activities and signal through the same receptor system. TGF-Ps are secreted as inactive, "latent", complexes, formed by a TGF-P dimer non-covalently bound to two TGF-p pro-domains, named as latency associated peptides (LAPs). Latent TGF-P is sequestered in the extracellular matrix, and release of the highly stable, active, TGF-p dimer is mediated by plasmin and collagenases MMP-9 and MMP-2. '^ Because of urokinase- and tissue-type plasminogen activators (uPA, tPA), that convert plasminogen into plasmin, and collagenases are often up-regulated in tmnors at sites of migration/invasion, '^ it is likely that malignant cells are exposed to significant levels of active TGF-P. Since its discovery in 1981, '^ TGF-P was early recognized as a potent inhibitor of epithelial cell growth.® Subsequent studies showed that some carcinoma cell lines escape from TGF-P growth inhibition, and reported a direct role for TGF-p as an autocrine stimulator of tumor cell invasion and metastasis. ^'^^ In the past decade, a large body of experimental evidence has accumulated suggesting a dual role for TGF-P in cancer (see refs. 11-13 for reviews). Thus, it is now widely accepted that TGF-P can act as a tumor suppressor at early stages of tumorigenesis, and, also, as a potent driver of malignant progression, invasion and metastasis at later stages. Of the three classical members of the TGF-p family, TGF-pi is most frequendy up-regulated in tumors^^ and is the focus of most studies on the role of TGF-p in cancerogenesis.

TGF-p Signalling TGF-P-related factors signal through a cell-surface protein complex formed by two different transmembrane serine/threonine kinase receptors (TGFpRI andTGFpRII). Ligand binding induces the formation of tetraheteromeric TGFpRI-TGFPRII complexes whereby TGPPRII phosphorylates a regulatory region within TGFpRI known as the GS domain, resulting in kinase activation (see Fig. lA). The TGF-p receptor complex also contains two ^Corresponding Author: Miguel Quintanilla—Institute de Investigaciones Bionnedicas Alberto Sols, Consejo Superior de Investigaciones Cientfficas-Universidad Autonoma de Madrid, Arturo Duperier4, 28029-Madrid, Spain. Email: [email protected]

Molecular Mechanisms ofBasal Cell and Sqtuimous Cell Carcinomasy edited by Jorg Reichrath. ©2006 Landes Bioscience and Springer Science+Business Media.

TGF'P Pathway and Cancerogenesis ofEpithelial Skin Tumors

81

Figure lA. Schematic representation of TGF-p receptor activation and downstream signalling pathways. Smad transduction pathway. auxiliary receptors named endoglin and betaglycan^ They are disulfide-linked homodimers containing large extracellular domains and cytoplasmic regions without consensus signaling motifs. Endoglin (also known as CD 105) is involved in angiogenesis.^^ It binds TGF-P, TGF-P3, and several members of the activin and BMP family in the presence of TGFpRI andTCppRII, and appears to modulate cellular responses to TGF-p by an as yet undefined mechanism. ' '^ Smad proteins have been identified as the key intracellular mediators that transmit TGF-p signals from the membrane to the nucleus^' (Fig. lA). Upon receptor activation, TGPPRI specifically recognizes and phosphorylates receptor-regulated Smads (R-Smads). The R-Smads Smad2 and Smad3 are activated by the TGF-p and activin type I receptors, while Smadl, Smad5 and SmadS bind to the BMP type I receptors and mediate BMP signalling. Activated R-Smads form heteromeric complexes with Smad4, a shared partner (or Co-Smad) for all R-Smads. Smad4 is not phosphorylated by the receptors but appears to be essential for transactivation of target genes. Phosphorylated Smad2/Smad3 forms dimers or trimers with Smad4, which upon translocation to the nucleus interact with other transcription factors as well as co-activators and co-repressors to modulate transcription of TGF-P target genes. A third group of Smads, Smad6 and Smad7, are inhibitory (I-Smads). In contrast to R-Smads, expression of Smad6 or Smad7 is highly regulated by extracellular signals, including EGF, TNF-a, IFN-y and TGF-P/BMP themselves. Induction of I-Smad expression by TGF-P appears to represent an auto-inhibitory feed-back mechanism to attenuate Smad activation. Besides Smad-mediated signalling, TGF-p triggers other signalling pathways independendy of Smad activation^^ (Fig. IB). Thus, TGF-P can activate mitogen-activated protein kinase

82

Molecular Mechanisms of Basal Cell and Squamous Cell Carcinomas

Figure IB. Schematic representation ofTGF-p receptor aaivation and downstream signalling pathways. Smad-independent pathways triggered by TGF-p (adapted from ref. 18). (MAPK) pathways, such as ERK, JNK and p38 MAPKs, by mechanisms that are poorly characterized. TGF-p induces in keratinocytes and other epithelial cells a rapid and transient activation of Ras, which can account for TGF-p-induced ERK MAPK signalling. ^^'^^ The activation of Rho-like small GTPases, including RhoA, Racl and Cdc42, as well as of phosphatid7linositol-3-kinase (PI3K), possibly through regulation of RhoA, has been implicated in TGF-p-induced changes in cytoskeletal organization and epithelial to mesenchymal transitions (reviewed in ref 18).

Role of TGF-P in Epidermal Development The epidermis and its appendages derive from the ectodermal layer. After neurulation, the cells that cover the embryo constitute the presumptive epidermis, a single layer of cells^^ that very soon (at embryonic day E9.5 during mouse embryogenesis) starts to stratify becoming a two-layered structure. The outer layer, the periderm, is a temporary cover that is discarded once the inner layer differentiates into a true epidermis. The epidermis is fully differentiated at El8.5, and hair follicle morphogenesis is initiated between E14 and El6 in response to signals derived from dermal cells. The signalling cascades responsible for these processes are not fully understood, but likely involve the cooperation of a broad range of signal transduction pathways and transcription factors mediating epithelial-mesenchymal interactions. Among these pathways, TGF-p signalling appears to play a key role in the development of epidermis and epidermal appendages.^^ Recendy, it has been shown that p63 may be the molecular switch required for epithelial stratification.^^ The p63 gene encodes isoforms that contain (TA) or lack (AN) a transactivation domain. p63 is already detectable at E8.5, before the expression of keratin 14 (K14) is observed (starting at E9.5). At this stage, the epithelium is still single-layered and expresses K18, which continues to be expressed in simple epithelia all life. TAp63 is the

TGF-P Pathway and Cancerogenesis of Epithelial Skin Tumors

83

Table 1. Knock-out mouse models for TGF-p growth factors and TGF-p signallmg components showing skin developmental defects Gene

Phenotype

Refs.

TGF-Pi

Slightly advanced hair follicle morphogenesis Hyperproliferation in epidermal basal layer Die at 3-4 weeks of age Delayed hair follicle morphogenesis Delayed hair follicle morphogenesis No effect in normal skin (accelerated wound healing) Die at 1 -10 months of age

31,32

TGF-p2 Noggin Smad3

31 33 34

first form expressed, and is necessary for initiation and commitment to stratification. ' These TA isoforms, however, block terminal differentiation, and have to be competed by ANp63 isoforms to allow cells to respond to differentiation signals. In fact, TAp63 is expressed in the basal layer of epidermis, which retains the proliferative capacity, whereas ANp63 is expressed in suprabasal layers ongoing terminal differentiation, and ectopic expression ofTAp63 in a simple epithelium results in squamous metaplasia"^ Whereas no connection has been established between TGF-P signaling and p63 expression, other transcription factors implicated in epidermal development appear to be regulated by TGF-P. For example, Msxl and Msx2 are homeobox genes that are expressed in the mouse hair follicle. Mice null for these genes have defects in hair follicle development. ' Conversely, mice overexpressing Msx2 show reduced proliferation and premature differentiation of hair follicles. Msx transcription factors are involved in a complex regulatory loop involving TGF-p signalling. BMP4 regulates epidermal induction in Xenopus in a pathway dependent on Msxl ,^^ and Smad4 is necessary for the activation of the Msx2 promoter. ^ Other reports have emphasized cross-talks between TGF-P signalling and other transduction pathways involved in epidermal/appendage development. Thus, TGF-p signalling may interact with the Wnt/P-catenin/Lef-1 and Sonic hedgehog (Shh) signal transduction pathways during hair follicle development (reviewed in ref. 22). Despite eliciting common downstream pathways, the three classical TGF-P isoforms can exert distinct biological functions in the skin due, in part, to different expression patterns. ^^ TGF-Pi andTGF-P3 proteins appear to be expressed in the epidermis at E14.5-16.5 and later in the epithelium of invaginating hair follicles, while TGF-P2 protein was found in the epidermis and dermis at day El 5,5. In vivo studies in knock-out mice suggest distinct roles of TGF-P isoforms during epidermal and hair follicle development (Table 1). TGF-P2 appears to be an inducer of hair follicle morphogenesis, since TGF-P2 null mice exhibit a profound delay of hair follicle morphogenesis and a drastic reduction in the number of hair follicles.^^ In contrast to hair follicles, growth and differentiation of interfoUicular epidermis was not affected in these mice. Unlike TGF-P2 null mice, mice with a disruption of the TGF-pi gene showed slightly advanced hair follicle formation, as well as hyperproliferation in the basal layer of the epidermis, without evidence of suprabasal proliferation or hyperplasia. These data suggest that TGF-p 1 may counteract TGF-p2 in hair follicle development by inhibiting keratinocyte proliferation. On the other hand, TGF-P3 null mice were found to have no morphological alteration in their skin and hair follicles.^^ A differential expression of BMP members has also been observed during mouse embryogenesis, and several lines of evidence suggest that BMPs may be involved in epithelial-mesenchymal interactions during hair follicle development.^^ Mice niJl for BMP2 or BMP4 die during embryonic development and, therefore, the roles of these BMPs on epidermal/appendage development have been difficult to address by conventional knock-out strategies. However, embryos deficient for Noggin, a BMP2/BMP4 antagonist which binds to BMPs preventing their association with the BMP type I receptor, have delayed hair

84

Molecular Mechanisms of Basal Cell and Squamous Cell Carcinomas

morphogenesis and develop significandy reduced number of hair follicles. Homozygous disruption of Smadl, Smad2, Smad4 and Smad5 genes are also ledial during embryogenesis. However, Smad3 null mice are viable, but they do not show any alteration in normal skin, although exhibit accelerated cutaneous wound healing, compared with wild-type and heterozygous mice.^ These results indicate an involvement of TGF-p signalling in the control of tissue repair.

Role of TGF-|3 in Epidermal Proliferation and Homeostasis In vitro studies strongly suggest that TGF-ps are potent growth inhibitors of keratinocyte proliferation (see below). Consistent with this idea, disruption ofTGF-p signalling by transgenic expression of a dominant-negative TGFpRII in basal and suprabasal layers of the epidermis under the control of a truncated loricrine promoter (L-ATGFPRII mice) induced epidermal hyperproliferation, hyperplasia and hyperkeratosis. However, controversial results were obtained when the transgene was directed to a different epidermal compartment. Thus, targeted expression of the ATGFpRII receptor to the basal cell compartment and follicular keratinocytes, under the control of a bovine K5 promoter (K5-ATGF_RII mice), did not disturb proliferation or differentiation in normal epidermis (see Table 2). These observations suggested that TGF-P signalling in suprabasal keratinocytes, and not in basal cells, might be more important for maintaining homeostasis in normal epidermis. Contradictory results were also obtained in transgenic mouse models in which TGF-pi was overexpressed in different epidermal compartments.^^ When expression of TGF-pi was directed to all layers of the epidermis by a truncated human Kl promoter, a severe phenotype was produced with suppression of all epidermal proliferation and neonatal lethality. In contrast, when TGF-pi was expressed in differentiating keratinocytes of suprabasal layers using a KIO promoter, or a truncated K6 promoter inducible by the phorbol ester 12-0-tetradecanoylphorbol-13-acetate (TPA), increased proliferation was observed in the basal layer of the epidermis, without signs of hyperplasia or hyperkeratosis.^^' Nevertheless, TPA-induced hyperplasia was reduced in both transgenic mouse models with respect to control mice. The unexpected finding of an increased mitotic rate in the epidermis by overexpression of TGF-P i led to the authors of these reports to postulate a bimodal function of this growth factor acting as a positive regulator of keratinocyte proliferation in quiescent epidermis and as a negative regulator during hyperplasia following TPA treatment. These controversial observations highlight the complex nature of the biological responses toTGF-P, and suggest that differences on the transgene expression levels and/or the differentiation stage of keratinocytes where these genes are expressed can produce significantly distinct phenotypes in the epidermis. To overcome the problem of neonatal lethality by high constitutive TGF-pi levels in the epidermis, and address the question of timing and levels of transgene expression, two laboratories have produced bitransgenic mice capable of conditional expression ofTGF-pi.^i'^^ij^^l^g first of these models, TGF-p i expression targeted to basal and suprabasal layers of the epidermis by a loricrine promoter was controlled by topical application of an antagonist of progesterone (L-PRDT.TGF-pi mice). Induction ofTGF-p i at high levels inhibited keratinocyte proliferation in both the neonatal and adult epidermis, and produced resistance to TPA-induced hyperplasia, while lower TGF-Pi levels did not affect epidermal proliferation. In the second mouse model, TGF-P 1 was targeted to the basal layer of the epidermis and hair follicles with a K5 promoter by utilizing a tetO conditional expression system allowing for induction or suppression of exogenous TGF-Pi by doxycycline (K5-rTADT.TGF-Pi and K5-tTADT.TGF-pi mice). Acute induction of TGF-Pi blocked epidermal and hair follicle proliferation in newborn and adult bitransgenic mice. In addition, hair follicle morphogenesis and hair cycling were also affected. However, chronic expression of exogenous TGF-pi in adult mice caused epidermal and follicular hyperproliferation, as well as severe alopecia, dermatitis and inflammation. The hyperproliferative phenotype could be associated, at least in part, with up-regulation of the inhibitory Smad7 in the epidermis and hair follicles. In fact, transgenic mice

TGF-P Pathway and Cancerogenesis of Epithelial Skin Tumors

'+-

o u

^

m vD rn

85

00

cr> o ^ rs| m Tf T;f ^

ro ro ""^

Iu^ U

1

.^ ^'

1

s^

\ 1

_C Q.

^— ^ •fS

to Q.

o

CD C

0) CJO

S ^^ ^^

o

^Q . o E 0)

22

Q.

1

.s^

1

^

1

* ^

^ -oc

c

o

O CD

U

(\3

"ra

^

C

,9

c o 2 2 i2 v5 CD U

CD

"o

C

Q.

.9 ^O 15

'(/) '«« ^ Ic I>> >« 1^ c I iz c: E o .^^ .9 c: § .9 o .O o o ft

1

CD

11

% ^ o o ^ I>. Z I>- h ^ a; Q .

i

Q.

CD

"TO

to o

i5 "re C

O

CD

I>. I>- Z

Q.

O

c .9

Q .

Si

a;

1

C

15 :3 S I n u -c > > >c >

1

O

1 1 *^ ? 1

.3

fO

b

1 1 ^^ 1 1 .c^ 1

Q.

ig

rt

1 s^^ 1 1 ^ 1 *c 1 11 1 1 1

.^

CD

^y

_6

_c "c

-s*^

.^

CD

_ra

3

_y ~o ^ B "c

_r3

3

^y "o "t B c

J5

iS

3

=j

^y

_y

~o

"5

t

a; c

o

M— "D C oi

^CD C

_£S

J^

Z3

13

.y .y "o "o ^ "t ^

c

Q^

c

o

M— -D

*4— •u

c r3

c

as

iS i 5 _co _« 3

3

=3

D

,y ^y .y .y ~o "5 "o "o

^B

c

"t=

CD C

^CD CD ^ C

C

^

1 -^

1 .C^ 1 -^ 1 ^ 1 cvj 1 1

*•«

1 -2^ 1 ^

1 £

;

V C 9i DO (A

c

°$^ P b^ K ^

-ij

LT)

^

^s

"A ^

fe ^^ s s p::

S ci

4 ^

s

SS

S S d Kp 'Q ir, vi J^j ^IT) :i^ 2^

86

Molecular Mechanisms of Basal Cell and Squamous Cell Carcinomas

overexpressing Smad7 in the interfoUicular and follicular epidermis (K5-Smad7) exhibited hyperplasia and hyperproliferation of the neonatal epidermis and sebaceous glands, as well as delayed and aberrant hair follicle morphogenesis. Most of K5-Smad7 mice die before day 10 post partum due to severe pathological alterations in multiple tissues. Thus, Smad7, which is the only inhibitory Smad expressed in the epidermis, appears to be a potent in vivo inhibitor for TGF-P signal transduction in the epidermis and other epithelial tissues. Interestingly, the phenotypes of K5-rTADT.TGF-pi and K5-tTADT.TGF-Pi mice present similarities with that of K10-BMP6 transgenic mice, which display a progressive alopecia and dermatitis, and with that of Kl4-activin PA transgenic mice, which exhibit epidermal hyperplasia, altered differentiation, and dermal fibrosis. These results highlight the importance of TGF-p signalling during development and maintenance of epidermal homeostasis. Particularly, the hair follicle appears to be a crucial target for the action of TGF-P in the epidermis, and follicular alterations observed in the transgenic mice expressing TGF-Pi support the idea that one of the major functions of TGF-Pi is the regulation through apoptosis of the anagen-catagen transition during the hair cycle. On the other hand, these results indicate that the diverse phenotypes observed in the transgenic mouse models result from specific levels and timing in the expression of the transgene. They also reflect activation of an overlapping set of genes and biological responses in epidermal keratinocytes and dermal cells.

Role of TGF-P in Epithelial Skin Cancer The Two-Stage Mouse Skin Carcinogenesis Model System Multistage mouse skin chemical carcinogenesis has provided a paradigm to study the genetic and epigenetic events that contribute to the development of squamous cell carcinomas (SCCs).^^The most common chemical carcinogenesis protocol is two-stage induaion, which involves the topical application of a single dose of a carcinogen initiator; i.e., the polycyclic aromatic hydrocarbon 7,12-dimethylbenz(^i)anthracene (DMBA), followed by once or twice weekly treatment with the tumor promoter TPA. This protocol results in the development of multiple benign papillomas, most of which are promoter-dependent and regress rapidly when TPA treatment is finished. However, a small proportion of papillomas (about 5-10%) do not regress in the absence of promotion and spontaneously progress to malignant SCCs. SCCs can vary in their histological grade, from well differentiated, with clear basal and suprabasal layers, to poorly differentiated, with little structural organization. The latest stage of tumor progression is the development of spindle cell carcinomas (SpCCs), a highly malignant type of tumor formed by cells that have lost the epithelial phenotype and acquired mesenchymal characteristics. '^^ Spindle carcinoma cells have down-regulated the expression of keratinocyte specific proteins, such as cytoskeletal keratins^ ^ and the cell-cell adhesion molecules E- and P-cadherin,^^ and are characterised by the expression of fibroblastic cell markers, such as vimentin. Initiation involves a specific oncogenic mutation in the H-Ras gene, ' and additional changes in the H-Ras mutated allele leading to overexpression of H-Ras oncogenic protein are associated with malignant progression.^^'^^'^'^^ Other alterations associated with tumor progression lead to deregulation of the cell cycle machinery and promote cell growth and, possibly, genetic instability. Among these genetic alterations, the most relevant are Ras-dependent overexpression of cyclin Di,^^ inactivation of the p53 gene,^^ and deletion or altered regulation of the INK4 locus encoding the cyclin-dependent kinase (cdk) inhibitors pi 5, p i 6 and pi9, the latter associated with the late spindle stage.

Changes in the Expression of TGF-P and Smad Proteins during Mouse Skin Carcinogenesis Several laboratories have reported significant alterations in the expression of TGF-p growth factors (and other components of the TGF-p system) during mouse skin carcinogenesis. Nevertheless, definitive conclusions about the expression patterns of individual TGF-P isoforms

TGF'P Pathway and Cancerogenesis of Epithelial Skin Tumors

87

are complicated by the fact that diflFerent methods of detection, such as in situ hybridization and immunohistochemistry, have been indistinctly utilized in these studies, and by the use of antibodies with different immunoreactivities. TGF-pi protein appears to be expressed at low levels by basal keratinocytes of normal mouse epidermis,^^'^^ but it is increased in suprabasal layers after TPA-induced hyperplasia, ' and in human epidermis after wounding. TGF-P2 is expressed by differentiating keratinocytes both in vitro and in vivo. ' Dysplastic skin papillomas and SCCs express high levels of TGF-Pi mRNA, but do not express TGF-pi protein, ' suggesting that tumor cells evade the growth inhibitory effects of TGF-p by loss of this growth factor during progression to SCCs. Nevertheless, the growth factor is abundandy synthesized in the stroma of mouse skin carcinomas, ' and, thus, carcinoma cells are surrounded by a high concentration of stromal TGF-pi. Some data, however, indicate that stromal TGF-pi could have a significant different effect on tumor cell growth than autocrine TGF-Pi. For example, subcutaneous injection of TGF-Pi enhances skin tumor promotion. On the other hand, the current evidence from studies on human and experimental cancers suggest that TGF-Pi stimulates stromal cells to produce pro-tumorigenic cytokines and promote angiogenesis. ' While mutations in TGF-P receptor and Smad genes have not been found during mouse skin carcinogenesis, changes in Smad protein expression have been also associated with tumor development. Smadl to Smad5 mRNAs are constandy expressed in the epidermis regardless of the state of differentiation and stages of carcinogenesis, but protein levels were highly reduced or completely lost in SCCs. In contrast, the mRNA levels of the inhibitory Smad7 were substantially elevated in both papillomas and SCCs. These results suggest that changes in Smad protein expression could also contribute to the loss of growth inhibition mediated by TGF-P, resulting in tumor progression.

A Double Role for TGF-P1 in Skin Carcinogenesis Studies on Cultured Keratinocytes and Carcinoma Cell Lines In vitro studies have documented that TGF-pi is a potent growth inhibitor for primary and immortalized keratinocytes, whereas some carcinoma cell lines are less sensitive to the inhibitory effect of the growth factor. ^ ^ ' ' A complete loss of the anti-proliferative response appears to occur late during malignant progression, associated with the spindle stage. .The mechanism for TGF-Pi-induced growth arrest in keratinocytes involves down-regulation of c-myj" and induction of the cdk inhibitors p21 and p l 5 . '^^ Whereas these events appear to be Smad-dependent,^^ the ERK signalling pathway appears also to be involved in TGF-P 1-induced growth inhibition and induction of p21 in transformed keratinocytes. TGF-pi also promotes apoptosis in cultured keratinocytes, while TGF-P3 protects keratinocytes against TPA-induced cell death.^ Carcinoma cells do not only exhibit an attenuated response to growth inhibition, but they are also stimulated to malignancy by the growth factor. Thus, we showed that chronic exposure of transformed keratinocytes to TGF-Pi induced a reversible epithelial-mesenchymal transition^^ (EMT). When these TGF-pi-treated fibroblat-like cells were transplanted onto athymic mouse skin they developed SpCCs, while epithelial, untreated cells formed well differentiated SCCs, indicating that TGF-Pi could be involved in the SCC-SpCC transition that occurs in vivo during carcinogenesis. Interestingly, pre-malignant keratinocytes cultured under the same conditions were growth arrested and committed to cell death. Further reports demonstrated that the phenotypic changes induced by TGF-pi in transformed keratinocytes were associated with increased invasive and metastatic abilities,^ and with up-regulation of extracellular matrix degrading proteases, such as uPA and MMP-9.^^'^^ TGF-p signalling appears to be essential for EMT induced on SCC cell lines, since expression of a dominant-negative ATGFPRII prevent them from transition to SpCC.^^ TGF-Pi-mediated EMTs have also been reported in several human epithelial cell lines. '^^'^^ The signalling pathways by which TGF-Pi triggers this phenotypic response are poorly characterized and might depend on the specific epithelial cell type. However, activation of the Ras/MAPK transduction

Molecular Mechanisms of Basal Cell and Squamous Cell Carcinomas

88

Table 3. Behavior of genetically modified mice after two-stage skin carcinogenesis

Mouse Model

Tumor Yield

Frequency of Papilloma-SCC Conversion

SCC-SpCC Transition

Ref.

L-ATGFPRII K5-ATGFPRII KW-TGF-Pi Ke-TCF-Pi

Increased Normal Reduced Reduced Reduced

Increased Increased Increased Increased Increased

Normal Normal Increased Increased Increased

88 36 89 89 92

pathway appears to be a key event for TGF-pi stimulation of invasiveness and metastasis in keratinocytes/^'^^'^^ Furthermore, we have reported a cross-talk between Ras and Smad signalling pathways by which Smad4 attenuates Ras activity in keratinocytes expressing a Ras oncogene. Blockade of Smad4 function in these cells led to hyperactivation of the Ras/ERK transduction pathway and progression to undifferentiated carcinomas.^^ The requirement of Ras for TGF-pl-mediated EMT has also been reported in other epithelial cell systems. Thus, TGF-Pi must cooperate with a Ras oncogene for induction of EMT in mammary epithelial cells,^'^ and TGF-pl synergizes with ERK signalling activity to promote EMT in MDCK cells.^^ ERK and PI3K signalling pathways are also involved in TGF-pi-mediated induction of Snail, a repressor of E-cadherin gene transcription and inducer of EMT, in epithelial cells. Interestingly, a recent report from Oft and colleagues suggests that increased threshold levels of oncogenic H-Ras and activated Smad2 drive EMT and malignant progression in transformed keratinocytes.^^ In Vivo Studies The combination of two-stage chemical mouse skin carcinogenesis and transgenic/knock-out approaches has provided the experimental framework to demonstrate a double and paradoxical role of TGF-P signalling in carcinogenesis in vivo. Initial studies were devoted to demonstrate a suppressor role of TGF-p in carcinogenesis. Since TGF-p is a potent growth inhibitor for keratinocytes in vitro, it was believed that papillomas with a high-risk for malignant conversion should overcome TGF-P growth inhibitory effects before progression can occur. Thus, Glick and co-workers reported that grafts ofTGF-pi-null keratinocytes transduced with a viral H-Ras oncogene onto the dorsum of athymic mice progressed rapidly to multifocal SCCs within displastic papillomas, while grafts of viral H-Ras-initiated TGF-pi-heterozygous or wild type keratinocytes produced mostly well differentiated papillomas. Similar results have recendy been obtained with viral H-Ras-transduced Smad3-null keratinocytes.^"^ Consistent with these observations, disruption of TGF-P signalling by expression of a dominant-negative type-II TGFP receptor in the epidermis of transgenic mice (L- or K5-ATGFPRII mice) resulted in increased sensitivity to DMBA/TPA carcinogenesis with respect to control mice (Table 3). These mice showed a higher frequency of malignant conversion from papillomas to SCCs.^ '^^ Furthermore, TPA promotion alone induced papilloma formation in L-ATGFPRII mice, and most of TPA-induced papillomas did not exhibit H-Ras mutations, suggesting that loss of TGF-p signalling can serve as an initiating event in skin carcinogenesis. Nevertheless, the most striking result on the role of TGF-P in skin carcinogenesis was obtained by Cui and co-workers utilizing two-stage chemical carcinogenesis in transgenic mice with TGF-Pi expression targeted to the epidermis. Overexpression of TGF-Pi in epidermal keratinocytes inhibited benign tumor formation, but enhanced conversion of papillomas to SCCs, as well as the squamous to spindle cell transition.^^ Taken together, these studies support a direct role for TGF-Pi as a tumor suppressor at early stages of carcinogenesis, yet it also behaves as a promoter of malignancy at later stages. A double phenotype, identical to that of Cui and colleagues was

TGF-p Pathway and Cancerogenesis of Epithelial Skin Tumors

89

obtained in our laboratory utilizing mice heterozygous for the TGF-P co-receptor endoglin (Eng^ mice). Endoglin-null embryos die at midgestation due to vascular and cardiac defects, '^^ yet Eng^' mice develop normally, although they exhibit clinical signs of hereditary hemorrhagic telangiectasia, depending on the genetic background. Although most of endoglin studies have focused on its role on vascular alterations and angiogenesis, endoglin is also expressed in basal keratinocytes of interfoUicular epidermis and in hair follicles.^^ We used Eng^ mice on a mixed background that did not show any visible vascular abnormality. Decreased endoglin expression inhibited benign tumor outgrowth, but accelerated malignant conversion and development of highly undifferentiated carcinomas. ^^ These results suggest that endoglin attenuate TGF-P signalling in keratinocytes, and point to this co-receptor as an important component within the TGF-p receptor complex to modulate epidermal homeostasis and carcinogenesis. In addition to its direct role on tumor cells by inhibiting cell growth or promoting an EMT associated with development of highly aggressive spindle tumors, TGF-p profoundly affects the tumor microenvironment. TGF-P acts on several cell types that are in the proximity to the tumor, including fibroblast, endothelial and immune cells. Thus, TGF-P has potent extracellular matrix remodelling, immunosuppressor and angiogenic effects, creating an environment that facilitates tumor growth, invasion and metastasis.^ It is likely that the complex and paradoxical effects of TGF-p on tumor development implicates an intricate network of autocrine and paracrine responses on keratinocytes and stromal cells involving TGF-p as well as other cytokines. Thus, overexpression of theTGF-Pi transgene in the epidermis enhances angiogenesis, apparently not by a paracrine mechanism involving diffusion of the growth factor into the dermis, but rather by stimulating vascular endothelial growth factor (VEGF) expression in keratinocytes.^^ Similarly, chemically-induced tumors on ATGFPRII mice showed enhanced angiogenesis, which was associated with increased endogenous TGF-Pi expression.^^

TGF'P and Basal Cell Carcinoma Although SCCs and basal cell carcinomas (BCCs) are thought to come from a epidermal stem-cell progeny that gives rise to keratinocytes of the interfoUicular epidermis and hair follicles, they show different characteristics and behaviour, likely as a result of the specific genes altered as well as of the particular cells in which mutations occur.^^'^ SCCs develop from benign precursor lesions as a result of a multi-step process involving several genetic and epigenetic alterations that, likely, affect a number of distinct pathways. SCCs are thought to arise from the interfoUicular epidermis, since they show characteristics of interfoUicular epidermal differentiation,^ although studies with transgenic mice in which initiating Ras oncogenes were targeted to different epidermal compartments suggest that most malignant SCCs arise from the hair follicle region.^ BCCs, on the other hand, appear to develop from constitutive activation of the Shh/Gli signalling pathway, which is involved in the development of normal hair follicles^ '^^ (see also the Chapter of Reifenberger and co-workers in this book). BCCs are unusual carcinomas in which they rarely metastasize, although can be locally invasive.^ There is some evidence suggesting that the expression of TGF-P is regulated by the hedgehog signalling pathway in some developmental systems. Thus, TGF-P2 produced by epithelial placode during chick feather bud development is downstream to Shh,^^ and the product of the decapentaplegic (dpp) gene (a TGF-P ortholog in Drosophila) is controlled by hedgehog signalling during development of the fly. ' Since constitutive activation of Shh/Gli signalling appears to be a hallmark for all BCCs, it would be expected that up-regulation of TGF-P could be a characteristic of this type of cancer. In contrast, several studies have shown markedly reduced or negative expression for TGF-Ps and Smad proteins in BCCs compared with normal epidermis,^^^ while expression of TGF-p and its receptors TGFpRI and TGFPRII were enhanced in the peritumoral stroma. ^^ ' These data indicate a potential growth inhibitory escape mechanism for BCCs by down-regulating TGF-p in tumor ceUs. They also suggest a possible role for TGF-P signalling in stromal cells that could contribute to tumor local invasion. However, to date, the evidence for an involvement of TGF-P signaUing in the development of BCCs is merely circumstantial.

90

Molecular Mechanisms

of Basal Cell and Squamous

Cell Carcinomas

Acknowledgements Work in our laboratory on TGF-p and skin cancer has been funded by the Spanish Ministry of Science and Technology (grant SAF2001-2362) and Medical Research Fund (ISCIII, RTICCC C03/10). We thank Cristina Gonzdlez for skilful technical assistance.

References 1. Massagu^ J. T G F - p signal transduction. Annu Rev Biochem 1998; 67:753-791. 2. Lyons R M , Gentry LE, Purchio AF et al. Mechanism of activation of latent recombinant transforming growth factor P l by plasmin. J Cell Biol 1990; 110:1361-1367. 3. Yu Q , Stamenkovic I. Cell surface-localized matrix metalloproteinase-9 proteollitically activates T G F - P and promotes tumor invasion and angiogenesis. Genes Dev 2000; 14:163-176. 4. Andreasen PA, KjoUer L, Christensen L et al. T h e urokinase-type plasminogen activator system in cancer metastasis: a review. Int J Cancer 1997; 72:1-22. 5. Dano K, Romer J, Nielsen BS et al. Cancer invasion and tissue remodelling—Cooperation of protease systems and cell types. APMIS 1999; 107:120-127. 6. Moses HL, Branum EB, Proper JA et al. Transforming growth factor production by chemically transformed cells. Cancer Res 1981; 41:2842-2848. 7. Roberts AB, Anzano MA, Lamb LC et al. New class of transforming growth factors potentiated by epidermal growth factor: isolation from non-neoplastic tissues. Proc Natl Acad Sci USA 1981; 78:5339-5343. 8. Moses H L , Tucker RF, Leof EB et al. In: Feramisco J, Ozanne B, eds. Cancer Cells 3, Cold Spring Harbor, NY: Cold Spring Harbor Press, 1985:65-71. 9. Wright JA, Turley EA, Greenberg A H . Transforming growth factor beta and fibroblast growth factor as promoters of tumor progression to malignancy. Grit P.ev Ocog 1993; 4:473-492. 10. Newman MJ. Transforming growth factor beta and the cell surface in tumor progression. Cancer Metast Rev 1993; 12:239-254. 11. Siegel PM, Massagu^ J. Cytostatic and apoptotic actions of T G F - P in homeostasis and cancer. Nat Rev Cancer 2003; 3:807-820. 12. Akhurst RJ, Derynck R. T G F - P signalUng in cancer—^A double-edged sword. Trends Cell Biol 2001; 11:S44-S51 13. Derynck R, Akhurst RJ, Balmain A. T G F - p signaling in tumor suppression and cancer progression. Nat Genet 2 0 0 1 ; 29:117-129. 14. Massagu. FHKirUa l« Ijiistimi M"Vs

KI4 i':pM«?rjBP

MXl

BCCI

.,..™.^. ^CO

TJiPliiptiy 'IIWIP' ^88W8pPy ^»!^sw^?^!" MXM JiCO MtT6 B€T7

EC'C'S

Figure 3. Regulation of PDGF/PDGFRa in BCCs high expression of PDGFRa and its ligand PDGF-A in BCCs of mice and humans. Expression of PDGFRa and PDGF-A is detected by immunohistochemistry (A, B and D-F) or immunoblotting (C and G) using specific antibodies. Both PDGFRa and PDGF-A express in mouse BCCs. In human BCCs, PDGFRa is frequendy detected in the tumor nest but sometimes in the stromal (3 out of 14). PDGF-A expression is in the tumor nest in all cases examined. cells, thus resembling the PTCHl status in human BCCs. This cell line contains a high level of PDGFRa (Fig. 3). After treatment with anti-PDGFRa neutralizing antibodies, DNA synthesis index is reduced by 70% (Fig. 4,B). In contrast, FGF neutralizing antibody or purified goat IgG do not affect DNA synthesis. Furthermore, a downstream MEK inhibitor U0126 inhibits DNA synthesis in this cell line (Fig. 4). Therefore, increased expression of PDGFRa and activation of the ERK pathway appears to be important for cell proliferation in the ASZOOl cells. To test for a direct link between the hedgehog pathway and PDGFRa, PTCH1 is reexpressed in Ptchl null ASZOOl cells, which leads to down-regulation of PDGFRa protein (Fig. 4C). This reduced expression of PDGFRa is correlated with reduced cell proliferation (Fig. 4). Thus, down-regulation of the hedgehog pathway can reduce the expression of PDGFRa and can inhibit cell proliferation. All these data indicate that PDGFRa can be regulated by Glil and that PDGFRa mediates GUI-induced Ras/Erk activation. The expression of PDGFRa is up-regulated in BCCs of mice and humans. In the mouse BCC cell line ASZOOl, perturbation of PDGFRa function, whether direcdy by neutralizing antibodies or indirecdy by PTCFil, leads to decreased cell proliferation (Fig. 5). Therefore, up-regulation of PDGFRa appears to be an important mechanism by which hedgehog signaling induces basal cell carcinomas. Consistent with those findings, we have found that inhibition of the hedgehog signaling by SMO antagonist cyclopamine decreases the levels of PDGFRa and phospho-Erk, resulting apoptosis in ASZOOl cell (Fig. 6A). (Athar et al, 2004, submitted). Because cyclopamine inhibits PDGFRa expression, PDGF-A is unable to prevent cyclopamine-mediated cell death. By contrast, EGF, which can activate the Ras-Erk pathway independent of PDGFRa, inhibits cyclopamine-mediated cell death (Fig. 6B). Thus, it appears that SMO antagonist cyclopamine inhibits PDGFRa expression and down-regulates Ras-Erk signaling in BCC cells.

Molecular Mechanisms ofBasal Cell and Squamous Cell Carcinomas

102

Vector

P'lCHl PlXiFR t Actin

Figure 4. Requirement of PDGFRa for cell proliferation in BCCs PDGFRa expression is associated with cell proliferation of BCCs. A hedgehog signaling activated BCC cell line, ASZOO1, which derives from Gli 1 expressing BCC tumor in Ptchl* mice, contains a high level of PDGFRa (see Fig. 3C). Inhibition of PDGFRA fimaion using neutralizing antibodies prevents DNAsynthesis (A) and the reduaion is over 70% (B). Reexpression of PTCH1 into this cell line inhibits hedgehog signaling and decreases PDGFRa protein level (C), which is accompanied by inhibition of cell proliferation (D, colony formation assay).

The Role of PDGFATEGF Signaling in Angiogenesis of SCCs The role of PDGF for tumor development of squamous cell carcinomas is not well studied. However, there is a good correlation of angiogenic cytokine secretion with the microvessel density in the primary tumors of SCC. In both BCCs and SCCs, PDGF-A is elevated, suggesting the PDGF/PDGFR signaling axis may be a common mediator of epidermal hyperproliferation. In head and neck SCCs, VEGF and PDGF-AB are secreted in high amounts.^^ Keratinocytes are a major source of cutaneous PDGF whereas human dermal fibroblasts do not produce any detectable PDGF. The role of PDGF for angiogenesis is demonstrated by Dr. Fusenig s group using HaCaT cells.^^ After transfection with PDGF-B cDNA, HaCaT cells overexpress PDGF-B but are negative for the PDGF receptors alpha and beta (mRNA). Thus, they do not exhibit autocrine growth stimulation in vitro, but proliferation of cocultured fibroblasts is enhanced and this eflfect can be inhibited by neutralizing antibodies to PDGF-BB. After subcutaneous injection into nude mice, the transfected cells maintain high PDGF expression and form rapidly proliferating cysts, classified as benign tumors. During early tumor development (up to 2 months), PDGF-B transfectants induce marked mesenchymal cell proliferation and angiogenesis, yet this effect vanished at later stages (2-6 months) concomitantly with increased epithelial cell

PDGF Pathways and Growth of Basal Cell and Squamous Cell Carcinomas

103

Figure 5. A model of PDGF signaling in hedgehogmediated cell proliferation in BCCs aaivation of the sonic hedgehog pathway, through loss-of-function mutation of PTCH1 or gain-of-function mutation of SMO, functionally activates PDGFRa. As a result of PDGFRa activation, the Ras-Erk pathway, a pathway frequendy associates with cell proliferation, is activated, resulting in elevated cell proliferation and BCC formation.

P«l1

41 m\

\ \ PDGFRa

•I +1 Mek

Erk

i

A|^ Cell proliferation

Apoptosis \

/ PDGFRa ^^^ ^^ ^

Fhospho-Erik

^^mm^^^^

Total Erk

Cydopamin« Sub-G1 01 S QZm

6.6% SO.6% 19.7% 23.1%

DNA Content

Figure 6. Cyclopamine-mediated apoptosis in BCCs inhibition of the hedgehog pathway reduces the PDGFRa protein level and apoptosis in a BCC cell line. A) SMO antagonist cyclopamine reduces PDGFRa expression and decreases the level ofphosphorylated Erk. B) Cyclopamine-mediated apoptosis in BCC cells. EGF, whose receptor is not affected by cyclopamine (B), inhibits cyclopamine-mediated accumulation of sub-Gl cell population. In contrast, the ectopic expression of PDGFRa, but not addition of the ligand PDGF-A, affect cyclopamine-mediated cell death (data not shown), indicating that PDGFRa is an important effector of the hedgehog pathway.

104

Molecular Mechanisms of Basal Cell and Squamous Cell Carcinomas

proliferation and enhanced tumor growth. These results demonstrate that an activated stromal environment can promote tumorigenic conversion of nontumorigenic keratinocytes by inducing sustained epithelial hyperproliferation. Thus, PDGF-BB appears to promote tumor growth by inducing angiogenesis and stromal formation, and PDGF-activated stromal cells maintain elevated keratinocyte proliferation via a paracrine mechanism.

PDGF Signaling in Dermatofibrosarcoma Protuberans and Giant Cell Fibroblastoma Dermatofibrosarcoma protuberans (DFSP) and giant cell fibroblastoma (GCF) are recurrent, infiltrative skin tumors of intermediate malignancy, presents specific features such as reciprocal translocations t(17;22)(q22;ql3) and supernumerary ring chromosomes derived from the t(17;22). This chromosomal rearrangement generates afixsiongene of the collagen type I to the PDGF-B.^^ This gene fusion deletes exon 1 of PDGF and released the growth factor from its normal regulation and is capable of transforming cultured cells.^ PDGF receptors are activated in primary cultures derived from DFSP and GCF tumors. Autocrine PDGF receptor stimulation is therefore predicted to contribute to DFSP and GCF development. Evidence suggests that the primary cultured cells of DFSP and GCF are sensitive to treatment of PDGF receptor inhibitor Sleevec,^^ resulting in reduced cell proliferation. Treatment of DFSP tumor-bearing mice with this inhibitor also decreases tumor growth. These findings suggest targeting of PDGF receptors as a novel treatment strategy for DFSP patients. In fact, clinical trials of STI571 have some good responses in DFSP and GCF patients, ftxrther supporting the role of PDGF signaling in the development of DFSP and GCF.

Perspectives The platelet-derived growth factors (PDGF) are a pleotrophic family of peptide growth factors that signal through cell surface, tyrosine kinase receptors (PDGFR) and stimulate various cellular functions including growth, proliferation, and differentiation. Ever since the discovery of the transforming retroviral v-sis oncogene, PDGF signaling has been an interesting target for cancer treatment. To date, PDGF expression has been demonstrated in a number of different solid tumors, from glioblastomas to prostate carcinomas. Only in recent years, the role of PDGF for proliferation of nonmelanoma skin cancers has been established. The autocrine growth stimulation of tumor cells by PDGF is well demonstrated in many types of tumors including DFSP and BCCs of the skin. In addition, PDGF can regulate stromal cells through a paracrine mechanism, which is observed in skin SCCs and melanomas. Improved methods for detection of activated PDGF receptors would be most usefiil for screens of PDGF signaling activated tumors. Just as the discovery of ErbB2/Nu in breast cancer leads to a revolution of breast cancer treatment, clinically useftil PDGF receptor antagonists, like Gleevec (STI571/ Glivec), now allows for an evaluation of the importance of PDGF receptor signaling in a variety of cancers, including BCCs and SCCs of the skin. Thus, discovering the importance of PDGF signaling for the development of BCCs not only provides a molecular basis of hedgehog signaling-mediated tumorigenesis, but also has significant therapeutic implications. Thus far, there are several emerging possibilities for BCC treatments in addition to cryosurgery: suppression of the sonic hedgehog pathway (Athar et al, 2004, submitted), interferon alpha treatment^^ and inhibition of PDGF signaling. Additional research of PDGF signaling in other cutaneous malignancies will certainly help us design better ways to treat these cancers.

Acknowledgements This work was supported in part by National Institute of Health (NIH) grant ROl- CA94160 and Department of Defense grant PC030429. Due to space limitation, only selected references are listed in this review.

PDGF Pathways and Growth of Basal Cell and Squamous

Cell Carcinomas

105

References 1. Heldin C H , Ostman A, Ronnstrand L. Signal transduction via platelet-derived growth factor receptors. Biochim Biophys Acta 1998; 1378(1):F79-113. 2. Ostman A, Heldin C H . Involvement of platelet-derived growth factor in disease: Development of specific antagonists. Adv Cancer Res 2 0 0 1 ; 80:1-38. 3. Yu J, Ustach C, Kim H R . Platelet-derived growth factor signaling and human cancer. J Biochem Mol Biol 2 0 0 3 ; 36(l):49-59. 4. Hoch RV, Soriano P. Roles of P D G F in animal development. Development 2003; 130:4769-4784. 5. Betsholtz C. Biology of platelet-derived growth factors in development. Birth Defects Res Part C Embryo Today 2003; 69(4):272-285. 6. Li X, Ponten A, Aase K et al. P D G F - C is a new protease-activated ligand for the P D G F alpha-receptor. Nat Cell Biol 2000; 2(5):302-309. 7. LaRochelle WJ, Jeffers M, McDonald W F et al. P D G F - D , a new protease-activated growth factor. Nat Cell Biol 2 0 0 1 ; 3(5):517-521. 8. Bergsten E, Uutela M , Li X et al. P D G F - D is a specific, protease-activated ligand for the P D G F beta-receptor. N a t Cell Biol 2 0 0 1 ; 3(5):512-516. 9. Soriano P. T h e P D G F alpha receptor is required for neural crest cell development and for normal patterning of the somites. Development 1997; 124(l4):2691-2700. 10. Bostrom H , Willetts K, Pekny M et al. P D G F - A signaling is a critical event in lung alveolar myofibroblast development and alveogenesis. Cell 1996; 85 (6):863-873. 11. Clarke I D , Dirks PB. A human brain tumor-derived PDGFR-alpha deletion mutant is transforming. Oncogene 2003; 22(5):722-733. 12. Uhrbom L, Hesselager G, Nister M et al. Induction of brain tumors in mice using a recombinant platelet-derived growth factor B-chain retrovirus. Cancer Res 1998; 58(23):5275-5279. 13. Dai C, Celestino J C , Okada Y et al. P D G F autocrine stimulation dedifferentiates cultured astrocytes and induces oligodendrogHomas and oligoastrocytomas from neural progenitors and astrocytes in vivo. Genes Dev 2 0 0 1 ; 15(15):1913-1925. 14. Kilic T, Alberta JA, Zdunek PR et al. Intracranial inhibition of platelet-derived growth factor-mediated glioblastoma cell growth by an orally active kinase inhibitor of the 2-phenylaminopyrimidine class. Cancer Res 2000; 60(18):5143-5150. 15. Lokker NA, Sullivan C M , HoUenbach SJ et al. Platelet-derived growth factor (PDGF) autocrine signaling regulates survival and mitogenic pathways in glioblastoma cells: Evidence that the novel P D G F - C and P D G F - D ligands may play a role in the development of brain tumors. Cancer Res 2002; 62(13):3729-3735. 16. Heinrich M C , Corless CL, Duensing A et al. P D G F R A activating mutations in gastrointestinal stromal tumors. Science 2003; 299(5607):708-710. 17. Sommer G, Agosti V, Ehlers I et al. Gastrointestinal stromal tumors in a mouse model by targeted mutation of the kit receptor tyrosine kinase. Proc N a d Acad Sci USA 2003; 100(11):6706-6711. 18. Heinrich M C , Corless CL, Demetri C D et al. Kinase mutations and imatinib response in patients with metastatic gastrointestinal stromal tumor. J Clin Oncol 2003; 21(23):4342-4349. 19. Golub TR, Barker GF, Lovett M et al. Fusion of P D G F receptor beta to a novel ets-like gene, tel, in chronic myelomonocytic leukemia with t(5;12) chromosomal translocation. Cell 1994; 77(2):307-316. 20. Carroll M, Tomasson M H , Barker GF et al. T h e TEL/platelet-derived growth factor beta receptor (PDGF beta R) fiision in chronic myelomonocytic leukemia is a transforming protein that self-associates and activates P D G F beta R kinase-dependent signaling pathways. Proc N a d Acad Sci USA 1996; 93(25):14845-14850. 2 1 . Baxter EJ, Hochhaus A, Bolufer P et al. T h e t ( 4 ; 2 2 ) ( q l 2 ; q l l ) in atypical chronic myeloid leukaemia fiises BCR to PDGFRA. H u m Mol Genet 2002; 11(12):1391-1397. 22. Andrae J, Molander C, Smits A et al. Platelet-derived growth factor-B and -C and active alpha-receptors in meduUoblastoma cells. Biochem Biophys Res C o m m u n 2002; 296(3) :604-611. 23. Gilbertson RJ, Clifford SC. PDGFRB is overexpressed in metastatic meduUoblastoma. Nat Genet 2003; 35(3):197-198. 24. Johnson RL, Rothman AL, Xie J et al. H u m a n homolog of patched, a candidate gene for the basal cell nevus syndrome. Science 1996; 272(5268): 1668-1671. 25. Hahn H , Wicking C, Zaphiropoulous PG et al. Mutations of the human homolog of Drosophila patched in the nevoid basal cell carcinoma syndrome. Cell 1996; 85 (6):841-851. 26. Xie J, Q u i n n A, Zhang X et al. Physical mapping of the 5 M b D9S196-D9S180 interval harboring the basal cell nevus syndrome gene and localization of six genes in this region. Genes Chromosomes Cancer 1997; 18(4):305-309.

106

Molecular Mechanisms

of Basal Cell and Squamous

Cell

Carcinomas

27. Quinn AG, Epstein Jr E. Patched, hedgehog, and skin cancer. Methods Mol Biol 2003; 222:85-95. 28. Bale AE, Yu KP. T h e hedgehog pathway and basal cell carcinomas. H u m Mol Genet 2 0 0 1 ; 10(7):757-762. 29. Toftgard R. Hedgehog signaling in cancer. Cell Mol Life Sci 2000; 57(12):1720-1731. 30. Wicking C, Bale AE. Molecular basis of the nevoid basal cell carcinoma syndrome. Curr Opin Pediatr 1997; 9(6):630-635. 3 1 . Goodrich LV, Milenkovic L, Higgins KM et al. Altered neural cell fates and meduUoblastoma in mouse patched mutants. Science 1997; 277(5329):1109-1113. 32. Hahn H , Wojnowski L, Zimmer AM et al. Rhabdomyosarcomas and radiation hypersensitivity in a mouse model of Gorlin syndrome. Nat Med 1998; 4(5):619-622. 33. Aszterbaum M, Epstein J, Oro A et al. Ultraviolet and ionizing radiation enhance the growth of BCCs and trichoblastomas in patched heterozygous knockout mice. Nat Med 1999; 5(11): 1285-1291. 34. Xie J, Murone M, Luoh SM et al. Activating smoothened mutations in sporadic basal-cell carcinoma. Nature 1998; 391 (6662) :90-92. 35. Lam C W , Xie J, T o KF et al. A frequent activated smoothened mutation in sporadic basal cell carcinomas. Oncogene 1999; 18(3):833-836. 36. Reifenberger J, Wolter M , Weber RG et al. Missense mutations in S M O H in sporadic basal cell carcinomas of the skin and primitive neuroectodermal tumors of the central nervous system. Cancer Res 1998; 58(9):1798-1803. 37. Xie J, Johnson RL, Zhang X et al. Mutations of the P A T C H E D gene in several types of sporadic extracutaneous tumors. Cancer Res 1997; 57(12):2369-2372. 38. Berman D M , Karhadkar SS, Maitra A et al. Widespread requirement for Hedgehog ligand stimulation in growth of digestive tract tumors. Nature 2003; 425:846-51. 39. Thayer S et al. Hedgehog is an early and late mediator of pancreatic cancer tumorigenesis. Nature 2003; 425:851-6. 40. Watkins D N , Berman D M , Burkholder SG et al. Hedgehog signaling within airway epitheUal progenitors and in small-cell lung cancer. Nature 2003; 422(6929)'315-517. 4 1 . Wechsler-Reya RJ, Scott M P . Control of neuronal precursor proliferation in the cerebellum by Sonic Hedgehog. Neuron 1999; 22(1):103-114. 42. Wallace VA. Purkinje-cell-derived Sonic hedgehog regulates granule neuron precursor cell proliferation in the developing mouse cerebellum. Curr Biol 1999; 9(8):445-448. 4 3 . Oliver T G , Grasfeder LL, Carroll AL et al. Transcriptional profiling of the Sonic hedgehog response: A critical role for N-myc in proliferation of neuronal precursors. Proc Natl Acad Sci USA 2003; 100(12):7331-7336. 44. Allen M , Grachtchouk M , Sheng H et al. Hedgehog signaling regulates sebaceous gland development. Am J Pathol 2003; 163(6):2173-2178. 45. Niemann C, Unden AB, Lyle S et al. Indian hedgehog and beta-catenin signaling: Role in the sebaceous lineage of normal and neoplastic mammalian epidermis. Proc Natl Acad Sci USA 2003; 100(Suppl 1):11873-11880. 46. Mill P, M o R, Fu H et al. Sonic hedgehog-dependent activation of Gli2 is essential for embryonic hair follicle development. Genes Dev 2003; 17(2):282-294. 47. Milenkovic L, Goodrich LV, Higgins KM et al. Mouse patched 1 controls body size determination and limb patterning. Development 1999; 126(20):4431-4440. 48. Xie J, Aszterbaum M, Zhang X et al. A role of PDGFRalpha in basal cell carcinoma proliferation. Proc Natl Acad Sci USA 2 0 0 1 ; 98(16):9255-9259. 49. Ninck S, Reisser C, Dyckhoff G et al. Expression profiles of angiogenic growth factors in squamous cell carcinomas of the head and neck. Int J Cancer 2003; 106(l):34-44. 50. Gleich LL, Srivastava L, Gluckman JL. Plasma platelet-derived growth factor: Preliminary study of a potential marker in head and neck cancer. Ann Otol Rhinol Laryngol 1996; 105(9):710-712. 5 1 . Zhang J Z , Maruyama K, O n o I et al. Production and secretion of platelet-derived growth factor AB by cultured human keratinocytes: Regulatory effects of phorbol 12-myristate 13-acetate, etretinate, 1,25-dihydroxyvitamin D 3 , and several cytokines. J Dermatol 1995; 22(5):305-309. 52. Mueller M M , Fusenig N E . Tumor-stroma interactions directing phenotype and progression of epithelial skin tumor cells. Differentiation 2002; 70(9-10):486-497. 53. Simon M P , Pedeutour F, Sirvent N et al. Deregulation of the platelet-derived growth factor B-chain gene via fusion with collagen gene C O L l A l in dermatofibrosarcoma protuberans and giant-cell fibroblastoma. Nat Genet 1997; 15(l):95-98. 54. Greco A, Fusetti L, Villa R et al. Transforming activity of the chimeric sequence formed by the fusion of collagen gene C O L l A l and the platelet derived growth factor b-chain gene in dermatofibrosarcoma protuberans. Oncogene 1998; 17(10):1313-1319.

PDGF Pathways and Growth of Basal Cell and Squamous Cell Carcinomas

107

55. Sjoblom T, Shimizu A, O'Brien KP et al. Growth inhibition of dermatofibrosarcoma protuberans tumors by the platelet-derived growth factor receptor antagonist STI571 through induction of apoptosis. Cancer Res 2001; 61(15):5778-5783. 56. Rubin BP, Schuetze SM, Eary JF et al. Molecular targeting of platelet-derived growth factor B by imatinib mesylate in a patient with metastatic dermatofibrosarcoma protuberans. J Clin Oncol 2002; 20(17):3586-3591. 57. Li CX, Chi S, He N et al. INF alpha induces Fas expression and apoptosis in hedgehog pathway activated BCC cells through inhibiting Ras-Erk signaling. Oncogene 2004; 23(8):1608-17.

CHAPTER 11

Apoptosis and Guicerogenesis of Basal Cell and Squamous Cell Carcinoma Peter Erb, Jingmin Ji, Marion Wernli and Stanislaw A. Buchner Abstract

B

asal cell carcinoma (BCC) and squamous cell carcinoma (SCC) are the most frequent tumors in the Caucasian population. The formation of these tumors is a consequence of long term UV-exposure of the skin. UV-light induces DNA damage in cells. If the damaged DNA cannot be repaired or the DNA damaged cell is not eliminated by apoptosis (so-called sunburn cells), cell transformation and tumor development can be the outcome. Fas-ligand (FasL), a member of the tumor necrosis superfamily, is a key molecule involved in the elimination of sunburn cells. FasL is expressed in normal skin epidermis, preferentially in the basal layer. Regulation of FasL expression has a dual effect on cancerogenesis. On the one hand, FasL expression is downregulated in skin epidermis by UV irradiation leading to the loss of its sensor function and thereby increasing the risk of cell transformation and skin tumor development. On the other hand, once BCC or SCC have developed, FasL is strongly up-regulated. High expression of FasL may now serve to protect the tumor from the attack of immune effector cells. To prove the immune escape hypothesis in vivo, the prevention or downregulation of FasL expression in tumor tissue is required. Two approaches were successfully applied to silence the FasL gene in BCC tissues ex vivo, the antisense technology and RNA interference with small interfering RNA duplexes. With both techniques FasL expression can be efFiciendy downregulated in BCC tissues pathing the way to test the immune escape hypothesis in vivo.

Introduction Skin cancers such as melanoma and nonmelanoma skin tumors (NMSCs) account for one third of newly diagnosed cancers in the United States, making it the most common human malignancy at all. NMSCs predominandy consist of basal cell carcinoma (BCC) and squamous cell carcinoma (SCC), the former being about four times more common. In fact, BCC is the most frequent tumor in the Caucasian population. Conservative estimates for the year 2001 are well over 1 million cases in the United States, including approximately 200'000 cases of SCC. The incidence in Europe is 300/100*000 per year for BCC, and 150/100'000 per year for SCC. BCC is rare below the age of 30 and is more frequent in elderly individuals. BCCs are most commonly manifest on the face and 85% of all BCCs arise on the sun-exposed areas of the head and neck. Nodular BCC is the most frequent clinical subtype of BCC. The greatest danger of BCC results from local invasion. Although less common than BCC, SCC carries a risk of metastasis. Most SCCs develop from precursor lesions such as actinic keratosis. The *Corresponding Author: Peter Erb—Institute for Medical Microbiology, University of Basel, Petersplatz 10, CH-4003 Basel, Switzerland. Email: [email protected]

Molecular Mechanisms of Basal Cell and Squamous Cell CarcinomaSy edited by Jorg Reichrath. ©2006 Landes Bioscience and Springer Science+Business Media.

Apoptosis and Cancerogenesis of Basal Cell and Squamous Cell Carcinoma

ptoh mytatlon

I

DNA . ^ repair M damage apoptosis

-

109

p53 mutation \i

constitutive actfvatlofi ^ of sonic hedgehog signalling pathway

Loss of cell cycle regulation ^ of apoptosis

sec

^ AetlPie teratosis

Figure 1. Simplified scheme of how UV light causes BCC or SCC formation. majoriiy of SCCs are diagnosed on sun-exposed areas such as head, neck and dorsal hands. Metastases, when they occur, are generally to regional lymph nodes, and are detected 1 to 3 years after initial diagnosis.

Gene Mutations Lead to NMSCs Risk factors for BCC and SCC have been well characterized and include ultraviolet light (UV) exposure, white skin, blue tyos, red hair, Celtic ancestry, and inability to tan. Sunlight acts like a carcinogen and its underlying UVB radiation is most effective in inducing NMSC. UVA (320-400 nm) has been shown to be less carcinogenic than UVB. However, UVA, when added to UVB radiation, may accelerate carcinogenesis. UV exposure can induce immunosuppression and DNA alteration mainly at neighboring pyrimidines causing characteristic C-T or CC-TT base substitutions (Fig. 1). DNA repair represents an important defense mechanism in the pathogenesis of NMSC. The p53 tumor suppressor gene is regarded as the "guardian of the genome" as it protects DNA integrity in response to injury such as UV radiation.^ If DNA repair is not possible, the DNA damaged keratinocytes are eliminated by p53-dependent apoptosis (sunburn cells). Persons with a lower DNA repair capacity have a higher risk not only of developing NMSC but also of having a greater number of skin tumors. Xeroderma pigmentosum, an autosomal recessive disease, is characterized by extreme sun sensitivity and development of multiple skin cancers on sun-exposed part of the body due to a defective DNA repair of UV-induced lesions. Patients with xeroderma pigmentosum have a greater than 1000-fold increased risk of BCC and SCC. However, the p53 gene is itself a major target of UV radiation (Fig. 1). Mutations in this gene can lead to uncontrolled cell proliferation and loss of apoptosis of DNA damaged cells. As a consequence, BCC or SCC can develop. Indeed, mutations in the p53 gene are detected in about 56% of BCC^ and in >90% of SCC.^ Mutation in another gene, the ptch gene plays also a major role in BCC formation (Fig. 1). The identification of mutations in the ptch gene in patients with basal cell nevus syndrome (Gorlin syndrome) expanded our understanding of the genetic basis ofBCC.^'^ Basal cell nevus syndrome is an autosomal dominant disorder characterized by the development of

110

Molecular Mechanisms of Basal Cell and Squamous Cell Carcinomas

multiple BCCs at an early age, of central nervous system tumors (meduUoblastomas and meningiomas) and skeletal abnormalities. The ptch gene is mutated in patients with Gorlin syndrome and thus strongly linked to development of BCCs. Most tumors where ptch is inactivated display a truncating mutation in one allele and deletion of the other allele, i.e., loss of heterozygosity. About 30-40% ptch mutations are found in sporadic BCC. The ptch gene codes for the protein patched which is the cell membrane receptor for the hedgehog protein family. Binding of hedgehog to patched induces the release and activation of smoothened, another cell membrane protein, which leads to the induction of a number of proteins via the Gli transcription factors. These proteins exert a variety of effects on cell activities and function. Overexpression of Gli2 in keratinocytes has been direcdy linked to BCC formation in mice. Thus, UV-induced mutation of the ptch gene leads to an aberrant activation of the hedgehog signalling pathway eventually evoking BCC formation although the specific downstream effectors in this pathway are not known.

Apoptosis Is Pivotal for the Removal of DNA Damaged and Transformed Cells Two major apoptotic pathways exist. The extrinsic pathway is mediated by certain members of the TNF- and TNF-receptor (TNF-R) family (reviewed by Bhardwaj and Aggarwal).^^ For example, Fas-ligand (FasL) or TRAIL induce caspase-dependent apoptosis of cells upon interaction with their corresponding receptors. Fas orTRAIL-receptors (TRAIL-Rs). Cell death can be inhibited by a family of proteins that include decoy receptors, cellular Flice Inhibitory Protein, c-FLIP, survivin and others. The intrinsic, mainly mitochondrial-dependent pathway is executed by members of the bcl-2 family. Bax, Bak and others induce while bcl-2 and bcl-xL inhibit apoptosis. ^^ Both, the extrinsic and the bcl-2 controlled pathways are linked by bid, which is activated and translocated to mitochondria by caspase-8-mediated cleavage. ^^ As mentioned above, UV-induced mutations in the p53 gene results in the loss of apoptosis of DNA damaged cells, i.e., the formation of sunburn cells.^^'^ It has been clearly demonstrated, that FasL is central for the formation of sunburn cells.

UV-Light Changes the Expression Pattern of Apoptosis-Inducing and -Preventing Molecules in Skin Epidermis In normal UV-protected skin epidermis FasL and TRAIL are mainly expressed in the basal layer and somewhat less in the upper layers, while Fas is absent in the whole epidermis (Table 1, Fig. 2). TRAIL-Rl and -R2 are also not expressed in the basal but in the upper layers. TRAIL-R3 (also called decoy receceptor DcRl) and the anti-apoptotic FLIP are strongly expressed in the basal layer and only weakly in the upper layers. Following a single UV-light exposure (sunburn dosis), the epidermis demonstrates dramatic changes of the expression pattern of some of these molecules. FasL is downregulated over the next days at the protein and mRNA level, while Fas is upregulated. TRAIL, TRAIL-Rs and FLIP are transiently downregulated. Apoptotic cells, which are not detected in UV-light protected skin epidermis, are found in the UV-exposed skin. In chronic UV-light exposed skin epidermis, FasL expression is completely absent, whereas Fas expression is present (Fig. 2) as are TRAIL and FLIP, while TRAIL-R expression is low (Table 1). A very similar expression pattern is found in actinic keratosis (AK), a precursor lesion from which SCC can develop (Fig. 1). However, once BCC or SCC have developed, a reverse expression pattern becomes manifest. FasL, but not Fas is now strongly expressed in both tumor types (Fig. 2). TRAIL and FLIP are also strongly expressed, whereas TRAIL-R expression is absent (Table 1). Apoptosis is not found in both tumors, which is in line with the fact, that the relevant death-receptors (Fas, TRAIL-Rs) are not expressed.

Apoptosis and Cancerogenesis ofBasal Cell and Squamous Cell Carcinoma

Fas

FasL

111

TRAIL-R1

UV protected skin UV exposed skin (chronically)

Actinic keratosis

sec

/^^^;

-;••'•:, " > - .

Hs/^

.*sf

Z-^:

Figure 2. Distribution of FasL, Fas and TRAIL-Rl in UV-protected and UV-exposed skin, in actinic keratosis and SCC. Biopsies of a nonsunlight exposed skin area, of long-term sun-exposed skin, actinic keratosis and SCC were made, cryosections were prepared and immunostained with antibodies to FasL, Fas and TRAIL-RL

FasL, the Double-Edged Sword In normal UV-protected skin epidermis FasL serves as an important sensor and preserves skin integrity. ' Thus, FasL not only prevents the influx of inflammatory cells from the dermis, it also eliminates DNA damaged cells which have the potential to transform. ^^ UV-light diminishes this sensor function increasing the risk that aberrant cells are not efficiently eliminated and may develop to tumors. However, as soon as the tumor has been formed FasL takes an opposite role, it now seems to protect the tumor from the attack of effector cells. In other words, BCC or SCC may use FasL for their immune escape. The role of FasL for tumor immune escape in general is unresolved with lots of evidence for and against (reviewed by Igney and Krammer^'^). Most likely, there is no generally valid rule and immune escape has to be evaluated and identified for each tumor type. Which evidence for immune escape is available for BCC or SCC? So far, no direct, but indirect proof is available. The observation that BCCs are frequendy surrounded by activated T cells suggests that the host initiates an immune response which could potentially control tumor development. However, only isolated T cells and macrophages are found infiltrating into the tumor mass. In addition, cells immediately adjacent to BCC are undergoing apoptosis, whereas cells, e.g., T cells, more distant to tumor cells remain viable.^^'^^ The strong expression of TRAIL and FasL in BCCs or SCCs may prevent the attack of effector cells that express their corresponding receptors, and the lack of TRAIL-Rl and -R2 and Fas as well as high levels of FLIP prevent suicide and make the tumor cells resistant to apoptosis. Secondly, FasL on BCC and SCC is functional, i.e., it induces apoptosis in Fas positive celk'^-'* Thirdly and most convincingly, intralesional treatment of BCC with interferon-alpha evokes the intratumoral expression of Fas and the induction of FasL-mediated apoptosis leading to tumor regression. ' Thus, it is likely, but not proven that FasL serves for immune escape of BCC and SCC.

112

Molecular Mechanisms of Basal Cell and Squamous Cell Carcinomas

Table 1, Expression of death-ligands and -receptors in normal skin epidermis and skin tumors

Expression of FasL Fas TRAIL TRAIL-R1 TRAIL-R2 TRAIL-R3 (DcRI) FLIP

UV-Light Protected Skin

Chronic UV Exposed Skin

Actinic Keratosis

+

-

-

-

-

+2

-/+

n.t.

+

+

+/-

+ +' +/-

+ + +/-

BCC or SCC

+ +

^ except basal layer; ^ mainly basal layer; +: medium to high expression; +/- low expression; -/+ very low expression; -: no expression; n.t.: not tested

Downregulation of FasL on BCC by Andsense Oligonucleotides or Small Interfering RNAs (siRNAs) One way to directly proof the immune escape theory is to *knockout' the FasL gene in vivo in BCC and evaluate whether tumor regression takes place. Two approaches were chosen for FasL gene silencing, the transfection of cells with phosphorothiated antisense oligonucleotides (ASO) and with small interfering RNAs (siRNA). ASOs are short, traditionally 15-25 bases long, single stranded DNA fragments, which are in complementary orientation (antisense) to their target mRNA (sense). To enhance their nuclease resistance ASO are usually phosphorothioated or otherwise modified (2^^ and 3 generation ASOs).^^ For successful antisense inhibition, specific Watson-Crick base pairing between an antisense oligonucleotide and its target mRNA is essential. ASOs inhibit the expression of genes mainly by RNase H activation which cleaves the RNA strand of a RNA-DNA duplex, although other mechanisms are also possible.^^ SiRNAs are 21-23 nucleotide (nt) duplexes with 2 nt overhangs at the 3' end.^^'^^ Transfected into cells they are incorporated into a multiprotein RNA-induced silencing complex, RISC, where the duplex siRNA is unwound, leaving the antisense strand to guide RISC to its homologous target mRNA. This mRNA is then cleaved at a single site in the center of the duplex region between the guide siRNA and the target mRNA, 10 nt from the 5' end of the siRNA (see some of the most recent reviews^'^% While die in vitro transfection of mammalian cells and cell lines of various origin with siRNAs has been well established, successful siRNA transfection of human tissues has not been reported so far. Several ASOs directed against different regions of the human FasL mRNA were designed and evaluated.^'^ The ASOs were first tested on FasL-expressing HEK293 cells and demonstrated different inhibitory activities on FasL expression in these cells. The optimally active ASOs had a length of 20 bases and specifically inhibited FasL expression in a broad range of concentration applied for at least 4 days. The most potent antisense oligonucleotide, ASOS, specifically downregulated FasL expression at the protein and mRNA level >80%. Moreover, FasL downregulation strongly reduced the cytotoxic effector function of HEK293 cells towards Fas positive target cells. The ASO results also allowed to select the best target-regions on the FasL gene for gene silencing with siRNAs. The chemically synthesized siRNAs chosen dose-dependendy downregulated FasL in HEK293 cells between 30% and 70% at the protein as well as mRNA level. Having acquired the proof of concept in cultured cells downregulation

Apoptosis and Cancerogenesis of Basal Cell and Squamous Cell Carcinoma

113

BCC tissue transfected with

PBS

FasL-specific AS08

^i

sense ooritrol

Figure 3. Specific ASO inhibit FasL expression in human BCC. Surgically removed BCC was cut into small pieces. Each of them was transfected with either PBS, 500nM FasL-specific AS08 or its sense control and cultured for 4 days. After 2 days the transfeaion was repeated. Cryosections were prepared and immunostained with mouse monoclonal anti-human FasL G247-4. Strong FasL expression is seen on the cell membrane in the PBS and sense control sections, but not in the AS08 section. of FasL was tested in split skin tissue as well as in tissues obtained from surgically excised B C C . Indeed, it was not only possible to transfect the tissues with the reagents under study b u t also to efficiently downregulate FasL expression (Fig. 3). In B C C tissues the FasL downregulation ranged between 7 0 % and 7 5 % with either ASOs or siRNAs.

Outlook Based on the results demonstrating the successful transfection of B C C tissue with siRNAs or ASOs leading to efficient donwregulation of FasL expression, the i m m u n e evasion hypothesis becomes now feasible to test in vivo. BCCs of the skin are particular good targets for this study as the tumors are very easily accessible and of limited size. If the hypothesis proves correct, a therapeutic application of these reagents becomes practicable. In addition, the study could be extended to other FasL expressing tumors and other putative tumor-inducing or -supporting molecules could be evaluated as targets for such a gene silencing approach.

114

Molecular Mechanisms of Basal Cell and Squamous Cell Carcinomas

References 1. Tran H, Chen K, Shumack S. Epidemiology and aetiology of basal cell carcinoma. Br J Dermatol 2003; l49(Suppl 66):50-2. 2. Grossman D, Leffell DJ. The molecular basis of nonmelanoma skin cancer: New understanding. Arch Dermatol 1997; 133(10):1263-70. 3. Levine AJ. p53, the cellular gatekeeper for growth and division. Cell 1997; 88:323-31. 4. Soehnge H, Ouhtit A, Ananthaswamy ON. Mechanisms of induction of skin cancer by UV radiation. Front Biosci 1997; 2:D538-D51. 5. Ziegler A, Jonason AS, Leffell DJ et al. Sunburn and p53 in the onset of skin cancer [see comments]. Nature 1994; 372(6508):773-6. 6. Hahn H, Wicking C, Zaphiropoulous PC et al. Mutations of the human homolog of Drosophila patched in the nevoid basal cell carcinoma syndrome. Cell 1996; 85(6):84l-51. 7. Johnson RL, Rothman ALy Xie J et al. Human homolog of patched, a candidate gene for the basal cell nevus syndrome. Science 1996; 272(5268): 1668-71. 8. Stone DM, Hynes M, Armanini M et al. The tumour-suppressor gene patched encodes a candidate receptor for Sonic hedgehog. Nature 1996; 384(6605): 129-34. 9. Grachtchouk M, Mo R, Yu S et al. Basal cell carcinomas in mice overexpressing Gli2 in skin. Nat Genet 2000; 24(3):216-7. 10. Bhardwaj A, Aggarwal BB. Receptor-mediated choreography of life and death. J Clin Immunol 2003; 23(5):317-32. 11. Kirkin V, Joos S, Zornig M. The role of Bcl-2 family members in tumorigenesis. Biochim Biophys Acta 2004; 1644(2-3):229-49. 12. Li HL, Zhu H, Xu CJ et al. Cleavage of BID by caspase 8 mediates the mitochondrial damage in the Fas pathway of apoptosis. Cell 1998; 94(4):491-501. 13. Hall PA, McKee PH, Menage HD et al. High levels of p53 protein in UV-irradiated normal human skin. Oncogene 1993; 8(l):203-7. 14. Henseleit U, Zhang J, Wanner R et al. Role of p53 in UVB-induced apoptosis in human HaCaT keratinocytes. J Invest Dermatol 1997; 109(6):722-7. 15. Hill LL, Ouhtit A, Loughlin SM et al. Fas ligand: A sensor for DNA damage critical in skin cancer etiology. Science 1999; 285(5429):898-900. 16. Bachmann F, Buechner SA, Wernli M et al. Ultraviolet light downregulates CD95 ligand and trail receptor expression facilitating actinic keratosis and squamous cell carcinoma formation. J Invest Dermatol 2001; 117(l):59-66. 17. Igney FH, Krammer PH. Immune escape of tumors: Apoptosis resistance and tumor counterattack. J Leukocyte Biol 2002; 71(6):907-20. 18. Buechner SA, Wernli M, Harr T et al. Regression of basal cell carcinoma by intralesional interferon-alpha treatment is mediated by CD95 (Apo-l/Fas)-CD95 ligand-induccd suicide. J Clin Invest 1997; 100(11):2691-96. 19. Gutierrez-Steil C, Wrone-Smith T, Sun X et al. SunHght-induced basal cell carcinoma tumor cells and ultraviolet-B- irradiated psoriatic plaques express Fas ligand (CD95L). J Clin Invest 1998; 101(l):33-9. 20. Buechner SA, Wernli M, Bachmann F et al. Intralesional interferon in basal cell carcinoma. Rcecent Results in Cancer Research 2002; 160:246-50. 21. Kurreck J. Antisense technologies - Improvement through novel chemical modifications. Eur J Biochem 2003; 270(8): 1628-44. 22. Varga LV, Toth S, Novak I et al. Antisense strategies: Functions and applications in immunology. Immunol Lett 1999; 69(2):217-24. 23. Elbashir SM, Lendeckel W, Tuschl T. RNA interference is mediated by 21- and 22-nucleotide RNAs. Genes Dev 2001; 15(2): 188-200. 24. Elbashir SM, Harborth J, Lendeckel W et al. Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature 2001; 4ll(6836):494-8. 25. Dykxhoorn DM, Novina CD, Sharp PA. Killing the messenger: Short rnas that silence gene expression. Nat Rev Mol Cell Biol 2003; A{GyA')7'G7. 26. Shi Y. Mammalian RNAi for the masses. Trends Genet 2003; 19(1):9-12. 27. Ji JM, Wernli M, Buechner S et al. Fas ligand downregulation with antisense oligonucleotides in cells and in cultured tissues of normal skin epidermis and basal cell carcinoma. J Invest Dermatol 2003; 120(6): 1094-99. 28. Ji JM, Wernli M, Klimkait T et al. Enhanced gene silencing by the application of multiple specific small interfering RNAs. Febs Lett 2003; 552(2-3):247-52.

CHAPTER 12

The Role of Telomerase for Cancerogenesis of Basal Cell and Squamous Cell Carcinomas Eva-Maria Fabricius* Abstract

B

eside UV exposure and odier exogenous and endogenous factors, the activation of telomerase plays an important role in the cancerogenesis of both skin tumors. Model studies indicate that telomerase activation occurs early in the cancerogenesis of the skin. Telomerase activity (TA) has also been found in the proliferating basal cells of normal, predominantly of sun-exposed skin. Studies have demonstrated its presence in 167 of 186 (90%) basal cell carcinomas (BCC) and in 58 of 75 (77%) squamous cell skin carcinomas (SCC). While the potential for malignancy is higher in SCC, telomerase activation is more frequent in BCC. The period without recurrence is no shorter in patients with telomerase activity in a tumor than it is in patients without telomerase activity in the tumor. Proof of telomerase cannot be used as a prognostic marker in these tumors. However, it is important to prove telomerase activity in tumor-free marginal tissue, and this has a different impact on each of the tumor types. Telomerase activation originating in a BCC tumor-free margin would indicate field cancerization in the surrounding tissue. It could be used as a prognostic marker because the interval without recurrence in patients with TA positive BCC marginal tissue is significandy shorter than that in patients with a TA negative BCC margin. In contrast, telomerase activation in skin and in oral SCC tumor-free margins is associated with a significantly prolonged interval with no recurrence. The telomerase activation in this case can more probably be interpreted as an immune signal and lymphocytic defense function in the tumor margin, which can be confirmed by an in-situ demonstration of telomerase in the tumor margin tissue. The findings in histopathologically tumor-free margin tissue must be supported in further studies with a larger patient population.

Introduction In recent years the incidence of the two most frequent nonmelanoma skin tumors, basal cell carcinoma (BCC) and squamous cell carcinoma (SCC), has been rising world-wide.^'^ Epidemiological data show clearly that ultraviolet (UV) radiation, particularly UVB, is an essential factor along with other exogenous and endogenous noxae in the cancerogenesis of skin tumors. UV exposure is a chronic oxidative stress giving rise to DNA damage with specific mutations of suppressor genes such as p53,^'^ ' releasing various oncogenes^^'^ and / or activating viruses. ^' Molecular changes in the skin are associated with local cell proliferation. This process is depicted graphically by Alam and Ratner^^ in Figure 1.

*Eva-Maria Fabricius—Clinic for Oral and Maxillofacial Surgery, Campus Virchow Hospital, Augustenburger Platz 1, Medical Faculty of the Humboldt University of Berlin Charite, D-13353 Berlin, Germany. Email: [email protected]

Molecular Mechanisms of Basal Cell and Squamous Cell CarcinomaSy edited by Jorg Reichrath. ©2006 Landes Bioscience and Springer Science+Business Media.

Molecular Mechanisms ofBasal Cell and Squamous Cell Carcinomas

116

1

uv

UV

V OiKi |iS3 mutation

V

1

1

K

•

•

IWo i : ^ mutstikms

Apoplodb

^

^

^

^5

&JL^ MMI i l M ^ m M?lll

a

wo

rt#

'

1 1 1

Unoontrali0d oMutor prolttwrntiow

.iii%^ Figure 1. Significance of UV cancerogenesis in skin tumors (reproduced from Alam and Ratner,^^ widi permission from the publishers). According to Lacour,^^ the epithelial zones affected by the UV-induced cancerogenesis of basal cell carcinomas are deeper than those affected with squamous cell carcinoma. DNA damage through UV radiation is part of a complex multistage process which can lead to the immortalization of radiation-damaged cells and to cancerogenesis, and one important step in this process is the activation (or reactivation) of telomerase.^^'^^

Interaction of Telomeres and Telomerase The most important function of telomerase consists in elongating the tandemly repeated TTAGGG-DNA sequences of the chromosome ends (telomere; tele = end). In the absence of telomerase, telomere DNA is lost after each successive cell division. This leads to the progressive shortening of the telomeres, because the DNA polymerase of eukaryotic cells is not able to completely copy the DNA at the ends of the linear chromosomes. This is known as the end replication problem. ' ' The telomeres are the protective caps on the chromosomes, responsible for chromosome stability and protection against fusions, enzymatic degradation or

The Role ofTelomerasefor Cancerogenesis of Basal Cell and Squamous Cell Carcinomas

117

%

$«n. so-

1 ^^i

X

X

J^x

X

X

i 60-

X

«

X

1 ^-"

IS 5

3

X

X

x

40XX

1

30-

^

20-.

X X

», 86-89, 94, 97-99,103,109,110,115,119

N NBCCS 25,60-62 Nonmelanoma skin cancer (NMSC) 1-6, 18, 19, 43, 44, 46, 50, GG, 69, 71-73, 94, 96, 97, 104, 108, 109 Normal skin 15, 16, 25, 45, 7A, 83, 84, 108, 112,119-121,128 Nucleotide excision repair (NER) 18, 20-23, 27, 37, 38, 45, G7. 73

o Oncogene 19, 31, 33, 35-37, 38, 53, 54, GG, 71,88,89,94,97,104 Oral s e c 99,115,122-128 Oxidative stress GGy 115

p53 12, 13, 18, 22-27, 31-33, 36-39, 52, 66-74,86, 109,110,115,119 p53 mutation 18, 24, 27, 62, 69-73. 119 PapUlomavirus 5, 13, 31, 32, 35, 36, 72 Paracrine 89, 95, 96, 104 Patched 25, 52-54, 59, 71, 99, 110 PDGF pathways 94, 96 PDGF receptor 94, 97, 102, 104 PDGF receptor alpha 102 PDGF receptor beta 97, 102 Ploidy 50, 53 Polymerase chain reaction (PCR) 31, 35, 36, 39,40,73,118,120,127 Population-based studies 1 Prevention 1, 6, 27, 58, 63, GA, 108 Prognostic marker 115, 121, 124, 126, 128, 129 Progression 10, 27, 33, 36, 38, 39, 43, AG, 49, 53, 55, 59, G7. 69, 70, 72, 73, 80, 86-88, 118, 119 PTCH 18, 25-27, 52, 58-64, 71, 94, 99, 101-103 Public health 6,7,43

137

Index

R Retinoid 63, 64 Risk factor 1, 3, 25, 43, 61, 109 RNA 15,108,112,117,118

Serine/threonine kinase receptors 80 Signaling pathway 19, 26, 58-60, 81, 82, 87-89,110 Skin cancer 1-3, 5-7, 18-25, 27, 35, 38, 40, 43-46, 49, 50, 55, 58, 66, 69-73, 74, 86, 90, 94, 96, 97, 104, 108, 109 Skin cancer prevention strategies 6 Skin carcinogenesis 6, 22, 27, 58, 69, 72, 86-88,119 Smad 81, 82, 86-89 Small interfering RNA (siRNA) 108, 112, 113 SMOH 58,61-64 Smoothened 26, 59, 99, 110 Solar UV radiation 44, 45 Squamous cell carcinoma (SCC) 1-6, 10, 12, 13, 16, 18, 20, 21, 23-27, 31, 34-40, 43-46, 49-55, 58, 62, GA, 69-72, 86-89, 94, 98, 99, 102, 104, 108-112, 115, 116, 121-129 Structural rearrangement 51-54 Sunburn cell 25, 69, 108-110

Telomerase activation (TA) 68, 82, 83, 115, 117-124,126-129 Telomerase activity 13, 115, 117-128 Telomere GG, 116-119 Telomere shortening GGy 117 TGF-p 80-90 Tissue-type plasminogen activator 80 Transcription 13, 22-24, 31-33, 37-39, 59-62, 66,67,81-83,88,97,110 Transmembrane serine/threonine kinase receptor 80 Transplant recipients 5, 35, 43, 44, 46 Tumor 1, 3, 4, 6, 10-13, 15, 16, 18-27, 31, 33-40,43-46, 52, 55, 58-64, 66-74, 80, 86-89, 94, 97, 99-102, 104, 108-113, 115, 116, 118, 121, 122, 124-129

Tumor initiation 36, 39, 55 Tumor promoter 73, 86 Tumor-free margin 115, 121, 122, 124, 126-129 Tumorigenesis 22, 23, 43, GA, 80, 94, 104, 119

U Ultraviolet (UV) 1, 3-6, 18-25, 27, 31, 38-40, 43-46, 55, 61-63, GG, GS-74, 99, 108-112, 115, 116, 119-122, 128 Urokinase-type plasminogen activators 80 UV exposure 3, 6, 23, 25, 45, 46, 61, 69, 73, 74,109,115,122,128 UV fingerprint mutations 27 UV light 5, 19, 23, 108-112, 120, 121 UV signature 27, 70, 71 UV spectrum 18, 19

Vascular endothelial growth factor (VEGF) 46, 54, 89, 94, 97, 102 Vitamin D receptor (VDR) 13, 15

Xeroderma pigmentosum (XP) 5, 18, 21, 22, 25, 27, 44, 45, 51, 61, G7y 71, 72,109