Professional Documents
Culture Documents
& 2006 Nature Publishing Group All rights reserved 0950-9232/06 $30.00
www.nature.com/onc
ORIGINAL ARTICLE
In order to obtain a comprehensive picture of the radiation. UVB, with a wavelength range between 290
molecular events regulating cutaneous photodamage of and 320 nm, represents one of the most important
intact human epidermis, suction blister roofs obtained environmental hazards affecting human skin (Hahn
after a single dose of in vivo ultraviolet (UV)B exposure and Weinberg, 2002). To protect itself against the
were used for microarray profiling. We found a changed DNA-damaging effects of sunlight, the skin disposes
expression of 619 genes. Half of the UVB-regulated genes over highly complicated cellular programs, including
had returned to pre-exposure baseline levels at 72 h, cell-cycle arrest, DNA repair and apoptosis (Brash et al.,
underscoring the transient character of the molecular 1996). Failure in selected elements of these defensive
cutaneous UVB response. Of special interest was our strategies may result in propagation of cutaneous
finding that several of the central p53 target genes cancers. Handling of photodamage by the skin constitu-
remained unaffected following UVB exposure in spite of tes a crucial survival mechanism, and the regulation and
p53 protein accumulation. We next compared the in vivo activity of cutaneous UVB-regulated genes represents an
expression profiles of epidermal sheets to that of cultured attractive model for the functioning of central protective
human epidermal keratinocytes exposed to UVB in vitro. cellular responses under physiological conditions.
We found 1931 genes that differed in their expression The advent of microarray technology allows the
profiles between the two groups. The expression profile in generation of global expression profiles and differential
intact epidemis was geared mainly towards DNA repair, expression of thousands of genes (Park et al., 2002; Joos
whereas cultured keratinocytes responded predominantly et al., 2003; Sellheyer and Belbin, 2004). Several recent
by activating genes associated with cell-cycle arrest and studies have employed microarray profiling to study
apoptosis. These differences in expression profiles might UVB-regulated gene expression in cultured human
reflect differences between mature differentiating kerati- keratinocytes (Becker et al., 2001; Li et al., 2001;
nocytes in the suprabasal epidermal layers versus Murakami et al., 2001; Sesto et al., 2002; Takao et al.,
exponentially proliferating keratinocytes in cell culture. 2002; Dazard et al., 2003; Pisarchik et al., 2004; Lee
Our findings show that extreme care should be taken when et al., 2005), revealing that a wide range of genes
extrapolating from findings based on keratinocyte cultures regulating transcription factors, cytoskeletal proteins,
to changes in intact epidermis. secreted signaling molecules and controlling cellular
Oncogene (2006) 25, 2601–2614. doi:10.1038/sj.onc.1209292; functions such as signal transduction, terminal differ-
published online 23 January 2006 entiation and apoptosis, are induced by UVB. However,
since these studies differ considerably in terms of time
Keywords: UVB; epidermis; p53; in vivo; in vitro; kinetics, UVB dose, light sources and hybridization
microarrays techniques, it is difficult to generate a comprehensive
picture of the complex changes in gene expression taking
place in UVB-irradiated keratinocytes.
An additional drawback common to all these studies
Introduction
is the fact that they were performed on cultured
keratinocytes under in vitro conditions. Skin tissue
The skin constitutes the outer barrier of the human
cultures differ substantially from the intact tissues they
organism towards external stimuli and ultraviolet (UV)
are supposed to imitate: Keratinocytes are usually
cultured in the presence of hormones and growth factors
Correspondence: Dr CD Enk, Department of Dermatology, Hadassah that might affect cell-cycle regulation, stress response
Medical Center, P.O. Box 12000, IL 91010, Israel.
E-mail: enk@md.huji.ac.il
and apoptosis (Gibbs et al., 1998), thereby modifying
Received 11 July 2005; revised 6 October 2005; accepted 28 October the response to UVB. Furthermore, confluence
2005; published online 23 January 2006 in keratinocyte cultures affects the regulation of
UVB-induced gene expression profile of human epidermis
CD Enk et al
2602
programmed cell death, and a significant difference in
sensitivity to UVB-induced apoptosis between subcon-
fluent cultures and exponentially growing cells have
been reported (Gniadecki et al., 1997). Finally, keratino-
cyte cultures constitute a homogeneous population of
cells lacking the additional cellular constituents of intact
epidermis such as Langerhans cells and intraepidermal
inflammatory cells that together with the keratinocytes
play important roles in the response of human epidermis
to UV exposure (Baadsgaard, 1991).
Using a unique model system based on UV-exposed
human epidermis followed by isolation of epidermal
cells from suction blister roofs, we have recently studied
the global expression profile in intact human epidermis
using oligonucleotide microarrays (Enk et al., 2004).
With this in vivo setup that overcomes several of the
limitations outlined above, we identified more than 800
UVB-regulated genes, some of which not previously
known to be UVB sensitive. In the present study, we
expand this investigation by monitoring the differen- Figure 1 Number up- and downregulated genes at different time-
tially expressed genes over a 72 h period following UVB points following UVB irradiation of intact human epidermis.
exposure. We found that the majority of the in vivo
UVB-regulated genes changed their expression at the
24 h time-point, but had returned to background levels
within 72 h. We next compared the expression profile of sion profiles (Kannan et al., 2001). Applying this
the in vivo exposed epidermal cells to that of cultured method to genes that were at least twofold changed in
UVB-exposed normal human epidermal keratinocytes all volunteers at least at one time-point, triggered
(NHEK), thus providing the first direct comparison distinct, well-ordered gene expression and partitioned
between the UV response of intact epidermis and the samples according to interval after UV exposure
keratinocyte cultures. We here report that the expression rather than according to individuals (Figure 2). This
profile in intact epidermis was geared mainly towards profiling pattern testifies to the robustness of the data set
DNA repair, whereas cultured keratinocytes responded and indicates that our results reflect an overall UVB
predominantly by activating genes associated with cell- response. The algorithm showed that the most closely
cycle arrest and apoptosis. related expression profiles were those of the 0, 2 and 72
time-points, whereas the 24 h expression profile dis-
played greatest dissimilarity from the other time-points.
Results and Discussion This suggests that the majority of UVB-induced changes
in epidermal gene expression is transient and has
UVB-induced gene expression reversed to background expression within 72 h. The
To study the global gene expression profile in intact finding that the early UVB response involves a relatively
human epidermis following a single physiological UVB small number of genes, whereas the 24 h response is
exposure, we exposed the inner forearms of three more complex and expresses a larger number of genes
volunteers to 4 MED of UVB in vivo. Suction blisters corresponds well with the recent findings of Lee et al.
were raised at 2, 24 and 72 h following exposure, and (2005) who examined the differentially expressed genes
mRNA extracted from the blister roofs underwent gene of UVB-irradiated immortalized HaCat keratinocytes at
profiling using Affymetrix Human Focus oligonucleotide 0.5, 6 and 24 h.
arrays as described in Methods. Applying an arbitrary The dendogram revealed eight major, stable clusters
filter level of twofold changes in the ratios of gene containing different expression kinetics of up- and
expression in all volunteers at least at one time-point, we downregulated genes at 2, 24 and 72 h after UV
found 619 genes whose expression was modified by UVB exposure (Figure 2). Interestingly, in five of the eight
(Figure 1), of which 246 genes were upregulated and 373 clusters (cluster 1, 2, 5, 7 and 8), representing 316 of the
were downregulated, representing approximately 12% of 619 UVB-regulated genes (51%), the expression pattern
the ‘valid’ genes (1.5–2% of the human genome). In our had returned to pre-exposure baseline levels a 72 h,
list, the majority, 53% of the differentially expressed again underscoring the transient character of the
genes, were found at the 24 h time-point, whereas 21 and genomic changes taking place in human skin following
26% of the UVB-regulated genes were found 2 and 72 h, a single, low-dose, UVB exposure.
respectively, following exposure.
Figure 2 Cluster algorithm showing clustering of the 619 genes with at least twofold expression level changes in all volunteers in at
least one time-point following in vivo UVB exposure of intact human epidermis. The normalized expression level for each gene (rows)
in each sample (columns) is indicated by a color code (green downregulated and red upregulated). When applying an unsupervised
algorithm, the samples were partitioned according to interval after UV exposure and showed greatest similarities among the 0, 2 and
72 h expression patterns. Z-score analysis grouped the genes into eight kinetically distinguishable clusters. The number of genes in each
cluster is indicated. In five of the clusters, representing half of the UVB-modulated genes, the expression pattern had returned to
baseline patterns at 72 h.
selected genes of interest (http://www.hadassah.org.il/ (Maelandsmo et al., 1997). In normal human keratino-
Departments/photobiology; for the full list). The 619 cytes, S100A2 nuclear staining disappears following
differentially expressed genes were found in a wide range treatment with H2O2, and S100A2 protein might be
of functional categories at one or more time-points after involved in keratinocyte responses to ROS (Zhang et al.,
UVB irradiation. 2002). The downregulation of S100A2 expression follow-
ing UVB radiation reported in our study could thus result
from the activity of free radicals on the epidermal cells.
S100. Several skin disorders have been associated with
the S100 family of calcium-binding proteins, including
disorders of keratinization, proliferation and differentia- Serine protease inhibitors. Our data show that a
tion. The S100A7, S100A8 and S100A9 protein genes are number of serine proteinase inhibitors were induced
absent or minimally expressed in normal epidermis, but following UVB irradiation. The squamous cell carci-
are markedly overexpressed in psoriatic lesions, in noma antigen (SCCA1 or SERPINB3) plays a role in
keratinocytes with abnormal differentiation and in cuta- differentiation of squamous epithelium (Crombach
neous tumors (Boni et al., 1997; Alowami et al., 2003; et al., 1989; Kato, 1996) and significantly inhibits certain
Broome et al., 2003). S100A8 and S100A9 gene expression forms of apoptosis (Suminami et al., 2000). In human
is upregulated after injury of murine and human epidermis skin, SCCA is expressed in epidermis overlying inflam-
(Thorey et al., 2001), and Grimbaldeston et al. (2003) matory lesions such as psoriasis and dermatitis (Hor-
demonstrated that S100A8, but not S100A9, was over- iuchi et al., 1994; Takeda et al., 2002). There is no
expressed in the epidermis of UVA-irradiated BALB/c consensus regarding the expression in normal epidermis
mice in response to reactive oxygen species (ROS). We (Duk et al., 1989; Cataltepe et al., 2000; Takeda et al.,
here report upregulation of S100A6, S100A7, S100A8, 2002). Our data show that SCCA1 is dramatically
S100A9 and S100P protein genes in human epidermis increased 24 and 72 h after UVB exposure. Of specific
following UVB radiation. All S100 protein genes showed interest is our finding of moderate but significant
the same pattern of expression with upregulation at 24 h upregulation of serpinB13 (hurpin/headpin/PI13). This
and sustained activity at 72 h. S100A2 is expressed in new member of the human ov-serpin family has
benign nevi but not in metastatic melanomas, and loss of previously been demonstrated in the human immor-
S100A2 activity might play a role in tumorigenesis and talized HaCaT keratinocyte cell line, in cultured normal
constitute an early event in melanoma development human keratinocytes, in lesional skin from psoriatic
Oncogene
UVB-induced gene expression profile of human epidermis
CD Enk et al
2604
Table 1 Gene expression changes in intact human epidermis following in vivo UVB irradiation
GenBank Symbol Gene/protein Fold change from
nonirradiated
2h 24 h 72 h
AF119873.1 SERPINA1 Serine (or cysteine) proteinase inhibitor, clade A (alpha-1 antiproteinase, 3.21 7.22 2.19
antitrypsin), member 1
AI554300 SERPINB1 Serine (or cysteine) proteinase inhibitor, clade B (ovalbumin), member 1 0.45 3.95 0.99
AF169949.1 SERPINB13 Serine (or cysteine) proteinase inhibitor, clade B (ovalbumin), member 13 1.03 1.97 2.69
BC005224.1 SERPINB3 Serine (or cysteine) proteinase inhibitor, clade B (ovalbumin), member 3 0.89 13.11 9.91
Cell cycle
NM_001211.2 BUB1B BUB1 budding uninhibited by benzimidazoles 1 homolog beta (yeast) 1.01 0.63 2.24
NM_001237.1 CCNA2 Cyclin A2 2.31 1.67 8.50
NM_004701.2 CCNB2 Cyclin B2 1.46 0.92 5.90
NM_001759.1 CCND2 Cyclin D2 0.93 1.49 4.00
AW134535 CCNG2 Cyclin G2 0.92 2.70 1.05
AL524035 CDC2 Cell division cycle 2, G1 to S and G2 to M 0.92 0.46 4.19
NM_001255.1 CDC20 CDC20 cell division cycle 20 homolog (Saccharomyces cerevisiae) 2.29 1.17 6.12
NM_021873.1 CDC25B Cell division cycle 25B 1.71 1.36 3.61
AF213033.1 CDKN3 Cyclin-dependent kinase inhibitor 3 (CDK2-associated dual specificity 1.70 0.94 5.40
phosphatase)
NM_001826.1 CKS1B CDC28 protein kinase regulatory subunit 1B 1.42 0.89 2.02
NM_001953.2 ECGF1 Endothelial cell growth factor 1 (platelet-derived) 1.27 4.40 4.85
NM_022346.1 HCAP-G Chromosome condensation protein G 1.05 0.96 2.52
NM_006101.1 HEC Highly expressed in cancer, rich in leucine heptad repeats 4.15 3.77 21.36
NM_002358.2 MAD2L1 MAD2 mitotic arrest deficient-like 1 (yeast) 1.66 0.97 3.86
NM_004526.1 MCM2 MCM2 minichromosome maintenance deficient 2, mitotin (S. cerevisiae) 1.38 1.07 3.13
NM_002388.2 MCM3 MCM3 minichromosome maintenance deficient 3 (S. cerevisiae) 1.34 0.85 2.26
AA807529 MCM5 MCM5 minichromosome maintenance deficient 5, cell division cycle 46 1.72 0.72 3.22
(S. cerevisiae)
NM_005915.2 MCM6 MCM6 minichromosome maintenance deficient 6 (MIS5 homolog, S. pombe) 1.52 1.56 3.13
(S. cerevisiae)
NM_002592.1 PCNA Proliferating cell nuclear antigen 1.77 1.78 3.38
BC001422.1 PGF Placental growth factor, vascular endothelial growth factor-related protein 0.23 13.02 9.11
NM_005030.1 PLK Polo-like kinase (Drosophila) 1.44 1.52 3.89
NM_000946.1 PRIM1 Primase, polypeptide 1, 49 kDa 2.08 0.91 2.43
NM_021000.1 PTTG1 Pituitary tumor-transforming 1 1.29 0.79 3.76
BC003186.1 Pfs2 DNA replication complex GINS protein PSF2 1.93 0.82 4.69
BC001866.1 RFC5 Replication factor C (activator 1) 5, 36.5 kDa 1.78 1.20 2.29
NM_006397.1 RNASEH2A Ribonuclease H2, large subunit 2.11 0.87 2.83
BC005264.1 RPA3 Replication protein A3, 14 kDa 1.71 1.15 3.05
BC001886.1 RRM2 Ribonucleotide reductase M2 polypeptide 1.81 0.77 8.55
NM_014264.1 STK18 Serine/threonine kinase 18 0.80 0.67 2.49
NM_003236.1 TGFA Transforming growth factor, alpha 0.54 2.80 1.54
AF098158.1 TPX2 TPX2, microtubule-associated protein homolog (Xenopus laevis) 1.26 0.41 2.57
NM_001952.2 E2F6 E2F transcription factor 6 1.01 0.63 2.24
NM_003620.1 PPM1D Protein phosphatase 1D magnesium-dependent, delta isoform 2.31 1.67 8.50
AF022375.1 VEGF Vascular endothelial growth factor 1.46 0.92 5.90
NM_002094.1 GSPT1 G1 to S phase transition 1 0.93 1.49 4.00
NM_014574.1 STRN3 Striatin, calmodulin binding protein 3 0.92 2.70 1.05
NM_002748.1 MAPK6 Mitogen-activated protein kinase 6 0.92 0.46 4.19
NM_002467.1 MYC v-myc myelocytomatosis viral oncogene homolog (avian) 2.29 1.17 6.12
NM_002266.1 KPNA2 Karyopherin alpha 2 (RAG cohort 1, importin alpha 1) 1.71 1.36 3.61
NM_006732.1 FOSB FBJ murine osteosarcoma viral oncogene homolog B 1.70 0.94 5.40
Oncogene
UVB-induced gene expression profile of human epidermis
CD Enk et al
2605
Table 1 (continued )
GenBank Symbol Gene/protein Fold change from
nonirradiated
2h 24 h 72 h
Repair
D21089.1 XPC Xeroderma pigmentosum, complementation group C 2.52 2.39 1.81
U64315.1 ERCC4 Excision repair cross-complementing rodent repair deficiency, complementation 0.71 0.80 1.42
(XPF) group F
Histones
BC002649.1 HIST1H1C Histone 1, H1c 2.31 1.22 2.57
NM_021052.1 HIST1H2AE Histone 1, H2ae 6.86 2.41 1.76
NM_003523.1 HIST1H2BE Histone 1, H2be 4.09 3.67 2.48
NM_003522.1 HIST1H2BF Histone 1, H2bf 4.67 2.98 1.42
NM_003525.1 HIST1H2BI Histone 1, H2bi 3.43 2.76 2.14
BC000893.1 HIST1H2BK Histone 1, H2bk 1.32 2.62 1.34
AA451996 HIST2H2AA Histone 2, H2aa 3.83 2.00 0.94
NM_003528.1 HIST2H2BE Histone 2, H2be 6.16 1.90 0.71
Apoptosis
NM_001168.1 BIRC5 Baculoviral IAP repeat-containing 5 (survivin) 2.22 0.47 7.54
NM_000633 BCL2 B-cell CLL/lymphoma 2 1.24 0.49 0.66
U66879 BAD BCL2-antagonist of cell death 1.39 0.71 1.37
NM_004281 BAG3 BCL2-associated athanogene 3 0.49 0.66 0.75
NM_001188 BAK1 BCL2-antagonist/killer 1 0.88 1.87 1.37
AA457021 BAG5 BCL2-associated athanogene 5 0.76 0.63 0.45
AF060922 BNIP3L BCL2/adenovirus E1B 19 kDa interacting protein 3-like 0.85 0.81 2.06
NM_004346 CASP3 Caspase 3, apoptosis-related cysteine protease 0.59 0.88 0.64
U25804 CASP4 Caspase 4, apoptosis-related cysteine protease 1.31 1.64 1.48
BF439983 CASP8 caspase 8, apoptosis-related cysteine protease 1.33 1.97 1.28
NM_003655.1 CBX4 Chromobox homolog 4 (Pc class homolog, Drosophila) 1.85 0.40 0.66
NM_015322.1 FEM1B fem-1 homolog b (Caenorhabditis elegans) 0.24 0.99 0.81
NM_002015.2 FOXO1A Forkhead box O1A (rhabdomyosarcoma) 0.40 0.53 0.70
NM_001562.1 IL18 Interleukin 18 (interferon-gamma-inducing factor) 1.16 0.35 1.55
NM_007350.1 PHLDA1 Pleckstrin homology-like domain, family A, member 1 0.20 2.60 2.55
NM_014456.1 PDCD4 Programmed cell death 4 (neoplastic transformation inhibitor) 1.15 0.21 0.78
NM_006378.1 SEMA4D Sema domain, immunoglobulin domain (Ig), transmembrane domain (TM) and 0.93 0.36 0.83
short cytoplasmic domain, (semaphorin) 4D
NM_004760.1 STK17A Serine/threonine kinase 17a (apoptosis-inducing) 0.18 1.26 0.53
NM_004226.1 STK17B Serine/threonine kinase 17b (apoptosis-inducing) 0.27 0.42 0.50
BF224259 SPF30 Splicing factor 30, survival of motor neuron-related 0.35 0.90 0.54
NM_006290.1 TNFAIP3 Tumor necrosis factor, alpha-induced protein 3 0.41 0.62 0.49
NM_006048.1 UBE4B ubiquitination factor E4B (UFD2 homolog, yeast) 1.32 0.45 0.66
NM_002467.1 MYC v-myc myelocytomatosis viral oncogene homolog (avian) 0.43 0.16 0.31
NM_004052.2 BNIP3 BCL2/adenovirus E1B 19 kDa interacting protein 3 1.31 1.86 3.27
NM_001279.1 CIDEA Cell death-inducing DFFA-like effector a 0.89 1.36 2.99
AF053641.1 CSE1L CSE1 chromosome segregation 1-like (yeast) 1.43 2.23 2.54
NM_004938.1 DAPK1 Death-associated protein kinase 1 1.14 5.41 2.45
NM_004944.1 DNASE1L3 Deoxyribonuclease I-like 3 0.92 7.81 3.13
NM_002462.1 MX1 Myxovirus (influenza virus) resistance 1, interferon-inducible protein p78 (mouse) 1.61 4.07 5.14
BC001808.1 NME6 Nonmetastatic cells 6, protein expressed in (nucleoside-diphosphate kinase) 1.17 2.99 2.22
NM_007350.1 PHLDA1 Pleckstrin homology-like domain, family A, member 1 0.20 2.60 2.55
AF016266.1 TNFRSF10B TRAIL, tumor necrosis factor receptor superfamily, member 10b 0.34 2.33 0.78
Oncogene
UVB-induced gene expression profile of human epidermis
CD Enk et al
2606
Table 1 (continued )
GenBank Symbol Gene/protein Fold change from
nonirradiated
2h 24 h 72 h
Genes with at least twofold changes in expression in all volunteers at least at one time-point are listed.
patients and only weakly expressed in normal skin (Abts expression at 24 h and sustained upregulation at 48 h
et al., 1999). Hurpin overexpression in HaCaT cells after 2 MED of solar simulator radiation. In our study,
confers resistance to UV-induced apoptosis (Welss et al., TIMP-1 was upregulated already at 2 h and remained
2003). Our finding of overexpression of hurpin in UVB- high for 72 h with a distinct peak at 24 h. It is intriguing
irradiated normal human epidermis suggests a role for that a single low dose of UVB induces long-lasting
hurpin in the hyperproliferative state conferred to UV- activity of MMP-1 with deleterious effects on dermal
irradiated epidermis (Lee et al., 2002). extracellular matrix. However, the ratio of MMP-1 and
TIMP-1 is considered to be relevant for determining the
amount of active MMP-1 (Sudel et al., 2003). Calculat-
Extracellular matrix. The matrix metalloproteinases ing these ratios with our data (0.15, 3.38 and 0.75 for
(MMP) form a family of structurally and functionally the 2, 24 and 72 h time-points) indicates that MMP-1
related zinc endopeptidases capable of degrading activity following a single UVB dose is short lived and is
structural proteins in connective tissue with implications abolished at 72 h. SKALP/PI3, a recently described
on tissue plasticity and migration of tumor cells proteinase inhibitor absent from normal epidermis but
(Birkedal-Hansen et al., 1993; Kahari and Saarialho- found in hyperproliferative skin such as psoriasis and
Kere, 1997). MMPs are tightly regulated by a special healing wounds, and regulated by UVB and proinflam-
class of tissue inhibitors of matrix protease, TIMP-1. matory cytokines (Pfundt et al., 2000; Tanaka et al.,
MMP induction by UV has been implicated in the 2000b), was strongly upregulated at 24 h and remained
qualitative and quantitative alterations in the composi- high at 72 h in our list.
tion of the dermal extracellular matrix, a hallmark of
photoaged skin (Scharffetter et al., 1991; Fisher et al.,
1996). In our list, MMP-1 and MMP-3 expression was Cell cycle. Cell-cycle progression and proliferation is a
dramatically induced after 24 h and remained upregu- tightly controlled process induced by a wide range of
lated though to a lower degree at 72 h. These findings exogenous events including UVB radiation that is
are in accord with those of Fisher et al. (1996) regulated by a complex series of endogenous processes
who demonstrated induction of MMP-1 and MMP-3 (Celis et al., 1987). We found that most of the cell-cycle
at 16–24 h in human skin biopsies following 2 MED regulator genes were strongly upregulated at 72 h
of UVB radiation, and those of Lahmann et al. (2001) but only partially at earlier time-points, indicating a
and Sudel et al. (2003) who found a peak of MMP-1 late burst of cell-cycle progression and proliferation.
Oncogene
UVB-induced gene expression profile of human epidermis
CD Enk et al
2607
Interestingly, cyclins A2, B2, D2 and G2 as well as exposure. The interaction of photoproducts with his-
PCNA were upregulated, consistent with the lack of tones results in modulation of the nucleosome structure
GADD45 and WAF1/CIP1/p21 upregulation (Table 2). causing altered cromatin structure with increased
accessibility to repair enzymes (Brown et al., 1993; Liu
Repair. UVB causes DNA lesions in epidermal cells, et al., 2000). Interestingly, an early and strong upregu-
predominantly cyclobutane pyrimidine dimers (CPD) lation of eight different histone genes was included in
and pyrimidine–pyrimidone [6–4] photoproducts (Sage, our list.
1993). The biological significance of these DNA lesions
depends upon the capacity of the cell to repair the Apoptosis. UV-induced apoptosis is a highly complex
damage before it can be incorporated permanently into process in which different molecular pathways such as
the genome. Typically, DNA damage is rapidly repaired DNA damage, activation of the tumor suppressor gene
at a relative high rate by inducing upregulation of p53, triggering of cell death receptors, and ROS are
nucleotide excision repair (Young et al., 1996). In our involved (Aragane et al., 1998; Wehrli et al., 2000). In
list, XPC, which is involved in DNA excision repair with our list, genes belonging to the death receptor pathway
special affinity to DNA UV photoproducts (Kosmoski such as TRAIL-R2 and DAPK-1 were upregulated at
et al., 2001), was clearly upregulated already 2 h after the 24 h time-point, and were shut down at 72 h,
Table 2 p53 target gene expression changes in intact human epidermis following in vivo UVB irradiation
Symbol Gene product 2h 24 h 72 h
APAF1 Apoptotic protease-activating factor 0.9 0.7 0.7 0.3 0.9 1.6 0.5 1.1 0.3
BAI1 Brain-specific angiogenesis inhibitor 1 1.2 0.5 0.6 0.5 0.6 0.6 0.3 0.2 0.4
BAX BCL2-associated X protein 1.3 1.0 1.5 1.9 2.0 2.0 1.9 0.2 0.4
BID BH3 interacting domain death agonist 2.5 1.1 0.6 6.5 1.9 0.7 4.0 1.1 0.8
BTG2 BTG family, member 2 1.4 1.0 1.4 1.0 0.5 0.9 0.8 0.6 0.8
CASP6 Caspase 6, apoptosis-related cysteine protease 1.7 3.2 1.6 0.5 1.1 0.6 1.0 1.9 0.9
CCNG1 Cyclin G1 0.6 1.0 0.8 1.6 2.6 1.5 1.4 2.0 1.6
CDKN1A Cyclin-dependent kinase inhibitor 1A (p21, Cip1) 0.5 0.6 0.8 1.0 0.9 1.1 0.7 0.7 0.7
GADD45A Growth arrest and DNA-damage-inducible, alpha 1.1 0.9 0.9 1.1 0.6 0.8 1.0 0.5 0.5
GADD45B growth arrest and DNA-damage-inducible, beta 3.2 1.4 1.1 0.6 0.6 0.7 0.8 0.7 0.9
GADD45G Growth arrest and DNA-damage-inducible, gamma 1.1 1.9 3.0 0.6 2.6 2.3 1.4 0.4 2.1
GGA1 Golgi associated, gamma adaptin ear-containing, 1.3 1.1 1.1 1.1 1.1 1.1 1.1 1.3 1.1
ARF-binding protein 1
GML GPI anchored molecule-like protein 10.6 6.5 0.4 4.9 12.1 2.0 8.0 3.0 0.9
HERC1 Hect (homologous to the E6-AP (UBE3A) carboxyl 1.2 0.9 1.1 0.6 0.4 0.4 0.6 0.7 0.9
terminus) domain and RCC1 (CHC1)-like domain (RLD) 1
IGFBP3 Insulin-like growth factor-binding protein 3 0.5 0.3 0.6 1.0 0.6 0.5 1.5 0.6 0.5
MDM2 Mdm2, transformed 3T3 cell double minute 2, p53-binding 0.5 1.4 0.8 1.1 1.4 1.9 1.4 2.3 0.7
protein (mouse)
MDM4 Mdm4, transformed 3T3 cell double minute 4, p53-binding 1.7 3.7 0.3 1.2 0.5 0.3 16.0 0.7 0.8
protein (mouse)
NOX1 NADPH oxidase 1 8.0 2.3 0.9 1.0 1.4 2.3 2.3 0.8 0.4
NOX3 NADPH oxidase 3 1.3 0.7 1.7 0.8 0.5 1.2 1.3 1.1 1.2
NOX4 NADPH oxidase 4 1.0 1.5 2.0 0.4 1.5 4.6 1.3 1.1 2.3
NOX5 NADPH oxidase, EF hand calcium-binding domain 5 1.9 5.7 0.5 0.9 9.2 1.1 1.2 6.5 0.8
P53AIP1 p53-regulated apoptosis-inducing protein 1 1.5 2.3 1.9 0.4 0.4 0.3 1.1 2.0 3.0
PERP p53-induced protein PIGPC1 1.1 1.2 1.1 0.8 0.9 1.1 0.9 1.1 1.0
PLAGL1 Pleiomorphic adenoma gene-like 1 1.1 1.2 1.0 0.7 0.4 0.8 0.6 1.1 1.1
REPRIMO Candidate mediator of the p53-dependent G2 arrest 0.9 0.9 1.1 0.8 0.4 0.8 0.1 0.3 0.7
TNFRSF6 Tumor necrosis factor receptor superfamily, member 6 1.1 0.7 0.7 1.4 1.2 1.5 1.1 0.9 1.0
TP53 Tumor protein p53 (Li–Fraumeni syndrome) 1.6 1.5 1.3 0.7 0.6 0.5 0.9 0.9 0.8
TP53BP1 Tumor protein p53 binding protein, 1 2.0 1.7 1.3 1.2 0.8 0.1 1.3 1.1 1.4
TP53BP2 Tumor protein p53-binding protein, 2 0.4 0.3 0.6 1.1 1.2 1.1 0.7 0.8 0.7
WIG1 p53 target zinc finger protein 1.1 0.8 0.8 1.1 1.5 1.0 0.9 0.7 0.9
ERCC3 Excision repair cross-complementing (xeroderma 1.5 1.6 0.9 1.1 1.1 0.9 0.9 1.1 0.9
(XPB) pigmentosum group B complementing)
XPA Excision repair cross-complementing (xeroderma 1.9 1.5 1.5 0.9 0.8 0.9 0.6 0.9 1.2
pigmentosum group A complementing)
XPC Excision repair cross-complementing (xeroderma 2.5 1.6 1.7 1.9 1.6 2.5 1.0 0.9 2.1
pigmentosum group C complementing)
ERCC5 Excision repair cross-complementing (xeroderma 1.2 2.1 1.7 0.5 0.7 0.5 0.7 1.4 1.4
(XPG) pigmentosum group G complementing)
Oncogene
UVB-induced gene expression profile of human epidermis
CD Enk et al
2608
underscoring the transient character of the apoptotic sigma (Hermeking et al., 1997). p53 target genes present
response. Supporting the picture of an early burst of on the Affymetrix Human Focus oligonucleotide arrays
proapoptotic activity, the antiapoptotic gene, Bcl-2, was are listed in Table 2. Surprisingly, applying the filter
downregulated at 24 h, whereas the apoptosis inhibitor, used throughout this paper, that is, twofold change in all
BIRC5 (Survivin), was strongly upregulated at 72 h. three volunteers, only GML was upregulated and BAI1
Evidence of involvement of the ROS pathway is found downregulated at 24 and 72 h, respectively. Applying a
in the downregulation of catalase, glutathione perox- less strict filter of 1.5-fold changes, additional p53 target
idase, oxidative-stress responsive 1, peroxiredoxin 2 and genes were changed, including BAX (up at 24 h), CASP6
prostaglandin-endoperoxide synthase. (up at 2 h), CCNG1 (up at 24 h), HERC1 (down at 24 h),
Recently, death-receptor-mediated, transcription-in- IGFBP3 (down at 2 h), TP53BP2 (down at 2 h), XPA
dependent signaling modulation has emerged as an (up at 2 h) and XPC (up at 2 and 24 h). However,
important determinant of cell survival during develop- even with these less restrictive cutoff criteria, key p53
ment and cellular homeostasis. These mechanisms targets such as WAF1/CIP1/p21, GADD45 and Mdm2
involve key post-translational modifications that affect remained unaffected following UVB exposure.
the activities of proteins at different levels in the To verify the validity of these profiling data, we first
signaling pathways (Tran et al., 2004). Thus, UVB- ascertained the accumulation of p53 protein in human
induced apoptosis, here evidenced by the presence of epidermis following exposure to 4 MED of UVB in vivo
sunburn cells (Figure 3a) and by a diffuse, epidermal (Figure 4). Our results confirmed that p53 has a very low
annexin V staining (Figure 3b), might be mediated by basal expression in human skin and accumulates upon
both gene activation and by transcription-independent UV irradiation (Hall et al., 1993; Liu et al., 1994;
mechanisms. Courtois et al., 1997). We next confirmed the microarray
data by semiquantitative RT–PCR using primers for key
p53 target. As can be seen in Figure 5, neither WAF1/
p53. The ‘guardian of the genome’, p53, is a tumor CIP1/p21, GADD45, Bax or Mdm2 gene expression was
suppressor protein that plays a key role in the protective upregulated in spite of p53 protein accumulation 24 h
responses against exogenous injuries such as UV by following UVB exposure, whereas SCCA1, included as a
initiating a cascade of fine-tuned events through positive control, was clearly increased. Thus, our data
transactivation of its target genes. Target genes of p53 show that in spite of p53 protein accumulation in
include genes involved in cell-cycle arrest such the epidermal cells following a single in vivo UVB exposure,
cyclin-dependent kinase inhibitor WAF1/CIP1/p21 (el p53 did not cause transactivation of its known target
Deiry et al., 1993), the proto-oncogene Mdm2 (Barak genes.
et al., 1993), a negative regulator of p53, the growth Our current understanding of the regulation of p53 in
arrest and DNA damage-inducible GADD45 (Kastan human skin following UV damage is mostly based on
et al., 1992), Bax (Miyashita et al., 1994), Cyclin G keratinocyte cultures. To which degree such models
(Okamoto and Beach, 1994), PCNA (Shivakumar et al., reflect genuine physiological processes is largely un-
1995), IGF-BP3 (Buckbinder et al., 1995), B99 (Utrera known. Thus, the p53 repertoire and signaling pathways
et al., 1998), p53R2 (Tanaka et al., 2000a) and 14-3-3 of the cell is determined by several biological parameters
such as cell type, oncogenic composition, the nature and
the intensity of the extracellular stimulus, and the level
of p53 (Lassus et al., 1996; Li and Ho, 1998; Sionov and
Haupt, 1999; Zhao et al., 2000). Furthermore, the
resistance of keratinocytes to apoptosis depends on the
differentiation status and the location of the cells within
the epidermal layers (Qin et al., 2002). Many of these
parameters are not well controlled in the available cell
culture models. Our unexpected finding that only few of
the p53 target genes were activated following a single
in vivo UVB exposure might reflect crucial differences in
Oncogene
UVB-induced gene expression profile of human epidermis
CD Enk et al
2609
Cell cycle
NM_003914.1 CCNA1 Cyclin A1 ¼ m
NM_001759.1 CCND2 Cyclin D2 m k
NM_004701.2 CCNB2 Cyclin B2 m k
AW134535 CCNG2 Cyclin G2 m k
NM_000946 PRIM1 DNA primase, G1 to S cell-cycle reactome m ¼
NM_001269 CHC1 Chromosome condensation 1, (G1/S transition of m ¼
mitotic cell cycle DNA packaging)
N33167 CDKN1C Cyclin-dependent kinase inhibitor 1C (p57, Kip2) k m
NM_005197 CHES1 Checkpoint suppressor 1 k ¼
M68520 CDK2 Cyclin-dependent kinase 2 k ¼
U49844 ATR Ataxia telangiectasia and Rad3 related k k
Repair
NM_000380.1 XPA Xeroderma pigmentosum, complementation group A ¼ k
D21089.1 XPC Xeroderma pigmentosum, complementation group C m ¼
NM_000107 DDB2 Damage-specific DNA-binding protein 2, nucleotide-excision repair m ¼
NM_002691 POLD1 Polymerase delta 1, DNA replication DNA repair response to UV m ¼
NM_006297 XRCC1 X-ray repair complementing defective repair m ¼
AL080203 POLE Polymerase (DNA directed), epsilon; DNA replication and DNA repair m ¼
AJ243797 TREX1 Three prime repair exonuclease 1 m
NM_002690 POLB Polymerase (DNA directed), beta (DNA replication and DNA repair) ¼ m
NM_001983 ERCC1 Excision repair cross-complementing rodent repair deficiency, ¼ m
complementation group 1
NM_002412 MGMT O-6-methylguanine-DNA methyltransferase; DNA ligation, DNA repair ¼ m
Apoptosis
M14745 Bcl-2 ¼ m
NM_004052 BNIP3 BCL2/adenovirus E1B-interacting protein 3 (antiapoptosis) m ¼
AL117381 BCL2L1 BCL2-like 1 (antiapoptosis) m ¼
M59465 TNFAIP3 TNFa- induced protein 3 k m
NM_001066 TNFRSF1B Tumor necrosis factor receptor superfamily, member 1B ¼ m
BF439983 CASP8 Caspase 8, apoptosis-related cysteine protease m k
U13699 CASP1 Caspase 1, apoptosis-related cysteine protease ¼ m
AK024029 MOAP1 Modulator of apoptosis 1 ¼ m
U83981 PPP1R15A Protein phosphatase 1, apoptosis response to DNA damage stimulus ¼ m
BF686824 DAPK3 Death-associated protein kinase 3 k ¼
NM_001166 BIRC2 Baculoviral IAP repeat-containing 2 (antiapoptosis) ¼ k
Histones
NM_017445.1 H2BFT H2B histone family, member T m m
NM_003528.1 H2BFQ H2B histone family, member Q m m
AA451996 H2AFO H2A histone family, member O m m
NM_003522.1 H2BFG H2B histone family, member G m m
BC000893.1 H2BFT H2B histone family, member T m m
NM_021052.1 H2AFA H2A histone family, member A m m
BC002649.1 H1F2 H1 histone family, member 2 m m
NM_006026.1 H1FX H1 histone family, member X m m
Oncogene
UVB-induced gene expression profile of human epidermis
CD Enk et al
2611
Table 3 (continued )
GenBank Symbol Gene/protein Epidermis NHEK
AI554300 SERPINB1 Serine (or cysteine) proteinase inhibitor, clade B (ovalbumin), member 1 m ¼
BC005224.1 SERPINB3 SCCA1 Serine (or cysteine) proteinase inhibitor, clade B (ovalbumin), member 3 m ¼
Heat shock
X51757 HSPA6 Heat shock 70 kDa protein 6 (HSP70B0 ) k k
cell-cycle progression markers such as cyclins B2, D2 in the present study was performed with 150–250 mJ/cm2,
and G2 were all downregulated and p57/KIP2 upregu- whereas the keratinocyte cultures were exposed to
lated in NHEK but not in intact epidermis, where 40 mJ/cm2, using light sources with identical UVB spectra.
an opposite picture indicating cell-cycle progression Keeping in mind that approximately 100 mJ/cm2 of
was found. Also among apoptosis-related genes, a UVB can be encountered upon exposure to 60–90 min
functional dichotomy was evident, since the proapop- of midday Summer sun, it is worth noticing that the
titic genes, TNFAIP3, TNFRSF1B, CASP1, MOAP1 UVB exposure delivered to both intact epidermis and
and PPP1R15A, were upregulated only in NHEK, to the keratinocyte cultures is environmentally relevant;
whereas antiapoptotic genes such as BNIP3 and (3) The hypothesis-generating nature of microarray
BCL2L1 were upregulated only in intact epidermis. profiling appears to justify our approach of comparing
Likewise, in the DNA repair group, XPC, DDB2, two bioinformatic analyses despite the fact that they
POLD1, XRCC1, POLE and TREX1 were only were not performed in parallel. However, future
upregulated in intact epidermis. And in the p53 group parallel, hypothesis-driven studies are required to
where key target genes such as GADD45A, GADD45B compare selected candidate genes for their function
and WAF1/CIP1/p21 were unaffected in intact epider- and relevance under in vivo and in vitro conditions.
mis, the same genes were upregulated in NHEK.
The fact that the comparison between intact epidermis
and cultured keratinocytes was not performed in parallel
Methods
raises a number of issues: (1) Although both microarray
profilings were performed in the same laboratory, Array processing
different Affymetrix microarrays were used in the two All experiments were performed using Affymetrix Human
experiments. To allow a direct comparison between the Focus oligonucleotide arrays, as described (http://www.
two experiments, a series of compensatory bioinformatic affymetrix.com/support/technical/datasheets/human datasheet.
procedures were performed as detailed in Methods to pdf). Total RNA from each sample was used to prepare
compensate for potential pitfalls due to differences in biotinylated target RNA, with minor modifications from
construction of the microarrays,; (2) UV light impinging the manufacturer’s recommendations (http://www.affymetrix.
on intact human epidermis is absorbed or scattered by com/support/technical/manual/expression_manual.affx).
pigment cells and by chromophores in the stratum Briefly, 5 mg of mRNA was used to generate first-strand
cDNA by using a T7-linked oligo(dT) primer. After second-
corneum, and only 20–30% of the incoming light
strand synthesis, in vitro transcription was performed with
reaches the keratinocytes of the deeper layers of the biotinylated UTP and CTP (Enzo Diagnostics), resulting in
epidermis. None of these physiological structures approximately 300-fold amplification of RNA. The target
diminish the amount of UV light reaching the mono- cDNA generated from each sample was processed as per the
layer keratinocytes in culture. To compensate for these manufacturer’s recommendation using an Affymetrix Gene-
differences in UV accessibility, the in vivo UVB exposure Chip Instrument System (http://www.affymetrix.com/support/
Oncogene
UVB-induced gene expression profile of human epidermis
CD Enk et al
2612
technical/manual/expression_manual.affx). Briefly, spike con- that were compared to the list of 642 downregulated genes
trols were added to 15 mg fragmented cRNA before overnight from the in vivo study.
hybridization. Arrays were then washed and stained with
streptavidin–phycoerythrin, before being scanned on an Volunteers
Affymetrix GeneChip scanner. Additionally, quality and Three healthy volunteers were recruited after informed consent
amount of starting RNA was confirmed using an agarose and prior approval from the Ethical Committee on Experi-
gel. After scanning, array images were assessed by eye to ments on Humans (Helsinki Committee). Inclusion criteria
confirm scanner alignment and the absence of significant were skin type II–III, age 18-30 years, non-smokers, no regular
bubbles or scratches on the chip surface. 30 /50 ratios for or concomitant medication.
GAPDH and beta-actin were confirmed to be within
acceptable limits (1.01–1.59 and 1.04–1.82), and BioB spike UVB irradiation
controls were found to be present on all chips, with BioC, Areas of 9 cm2 on previously unexposed skin of the inner
BioD and CreX also present in increasing intensity. When forearms were irradiated with UVB from a bank of four FS40
scaled to a target intensity of 150 (using Affymetrix MAS 5.0 fluorescent lamps (Philips) that emit wavelengths between
array analysis software), scaling factors for all arrays were 280 and 320 nm with a peak at 313 nm. Light intensity, deter-
within acceptable limits (1.07–2.08), as were background, mined using a Waldmann (Waldmann GBH, Schwenningen,
Q values and mean intensities. Details of quality control Germany) UV radiometer, was 0.75 mW/cm2 at a distance of
measures can be found at http://www.ncbi.nlm.nih.gov/geo/ 20 cm from the light source. Single doses of 150–250 mJ/cm2,
and at http://eng.sheba.co.il/genomics. equivalent to 4 MED, were used.
References
Abts HF, Welss T, Mirmohammadsadegh A, Kohrer K, Cataltepe S, Gornstein ER, Schick C, Kamachi Y, Chatson K,
Michel G, Ruzicka T. (1999). J Mol Biol 293: 29–39. Fries J et al. (2000). J Histochem Cytochem 48: 113–122.
Alowami S, Qing G, Emberley E, Snell L, Watson PH. (2003). Celis JE, Madsen P, Celis A, Nielsen HV, Gesser B. (1987).
BMC Dermatol 3: 1. FEBS Lett 220: 1–7.
Aragane Y, Kulms D, Metze D, Wilkes G, Poppelmann B, Chaturvedi V, Bacon P, Bodner B, Nickoloff BJ. (2004).
Luger TA et al. (1998). J Cell Biol 140: 171–182. J Invest Dermatol 123: 1200–1203.
Baadsgaard O. (1991). Arch Dermatol 127: 99–109. Courtois SJ, Woodworth CD, Degreef H, Garmyn M. (1997).
Barak Y, Juven T, Haffner R, Oren M. (1993). EMBO J 12: Exp Cell Res 233: 135–144.
461–468. Crombach G, Scharl A, Vierbuchen M, Wurz H, Bolte A.
Becker B, Vogt T, Landthaler M, Stolz W. (2001). J Invest (1989). Cancer 63: 1337–1342.
Dermatol 116: 983–988. Dazard JE, Gal H, Amariglio N, Rechavi G, Domany E,
Birkedal-Hansen H, Moore WG, Bodden MK, Windsor LJ, Givol D. (2003). Oncogene 22: 2993–3006.
Birkedal-Hansen B, DeCarlo A et al. (1993). Crit Rev Oral Decraene D, Gan D, Mammone T, Maes D, Declereq L,
Biol Med 4: 197–250. Germyn M. (2004). J Investig Dermatol 122: A140.
Boni R, Heizmann CW, Doguoglu A, Ilg EC, Schafer BW, Dennis Jr G, Sherman BT, Hosack DA, Yang J, Gao W, Lane
Dummer R et al. (1997). J Cutan Pathol 24: 76–80. HC et al. (2003). Genome Biol 4: 3.
Brash DE, Ziegler A, Jonason AS, Simon JA, Kunala S, Leffell Duk JM, Voorst Vader PC, ten Hoor KA, Hollema H,
DJ. (1996). J Investig Dermatol Symp Proc 1: 136–142. Doeglas HM, de Bruijn HW. (1989). Cancer 64: 1652–1656.
Broome AM, Ryan D, Eckert RL. (2003). J Histochem el Deiry WS, Tokino T, Velculescu VE, Levy DB, Parsons R,
Cytochem 51: 675–685. Trent JM et al. (1993). Cell 75: 817–825.
Brown DW, Libertini LJ, Suquet C, Small EW, Smerdon MJ. Enk CD, Katz SI. (1994). Arch Dermatol Res 287: 72–77.
(1993). Biochemistry 32: 10527–10531. Enk CD, Shahar I, Amariglio N, Rechavi G, Kaminski N,
Buckbinder L, Talbott R, Velasco-Miguel S, Takenaka I, Faha Hochberg M. (2004). Photodermatol Photoimmunol Photo-
B, Seizinger BR et al. (1995). Nature 377: 646–649. med 20: 129–137.
Oncogene
UVB-induced gene expression profile of human epidermis
CD Enk et al
2614
Fisher GJ, Datta SC, Talwar HS, Wang ZQ, Varani J, Kang S Pfundt R, Wingens M, Bergers M, Zweers M, Frenken M,
et al. (1996). Nature 379: 335–339. Schalkwijk J. (2000). Arch Dermatol Res 292: 180–187.
Gibbs S, Boelsma E, Kempenaar J, Ponec M. (1998). Cell Pisarchik A, Wortsman J, Slominski A. (2004). Gene 341:
Tissue Res 292: 107–114. 199–207.
Gniadecki R, Hansen M, Wulf HC. (1997). J Invest Dermatol Qin JZ, Chaturvedi V, Denning MF, Bacon P, Panella J,
109: 163–169. Choubey D et al. (2002). Oncogene 21: 2991–3002.
Grimbaldeston MA, Geczy CL, Tedla N, Finlay-Jones JJ, Sage E. (1993). Photochem Photobiol 57: 163–174.
Hart PH. (2003). J Invest Dermatol 121: 1168–1174. Scharffetter K, Wlaschek M, Hogg A, Bolsen K, Schothorst A,
Hahn WC, Weinberg RA. (2002). Nat Rev Cancer 2: 331–341. Goerz G et al. (1991). Arch Dermatol Res 283: 506–511.
Hall PA, McKee PH, Menage HD, Dover R, Lane DP. (1993). Sellheyer K, Belbin TJ. (2004). J Am Acad Dermatol 51:
Oncogene 8: 203–207. 681–692.
Hermeking H, Lengauer C, Polyak K, He TC, Zhang L, Sesto A, Navarro M, Burslem F, Jorcano JL. (2002). Proc Natl
Thiagalingam S et al. (1997). Mol Cell 1: 3–11. Acad Sci USA 99: 2965–2970.
Horiuchi Y, Tsukahara T, Otoyama K. (1994). J Dermatol 21: Shieh SY, Ikeda M, Taya Y, Prives C. (1997). Cell 91: 325–334.
67–72. Shivakumar CV, Brown DR, Deb S, Deb SP. (1995). Mol Cell
Hosack DA, Dennis Jr G, Sherman BT, Lane HC, Lempicki Biol 15: 6785–6793.
RA. (2003). Genome Biol 4: R70. Siliciano JD, Canman CE, Taya Y, Sakaguchi K, Appella E,
Joos L, Eryuksel E, Brutsche MH. (2003). Swiss Med Wkly Kastan MB. (1997). Genes Dev 11: 3471–3481.
133: 31–38. Sionov RV, Haupt Y. (1999). Oncogene 18: 6145–6157.
Kahari VM, Saarialho-Kere U. (1997). Exp Dermatol 6: Sudel KM, Venzke K, Knussmann-Hartig E, Moll I,
199–213. Stab F, Wenck H et al. (2003). Photochem Photobiol 78:
Kannan K, Amariglio N, Rechavi G, Jakob-Hirsch J, Kela I, 355–360.
Kaminski N et al. (2001). Oncogene 20: 2225–2234. Suminami Y, Nagashima S, Vujanovic NL, Hirabayashi K,
Kastan MB, Zhan Q, el Deiry WS, Carrier F, Jacks T, Walsh Kato H, Whiteside TL. (2000). Br J Cancer 82: 981–989.
WV et al. (1992). Cell 71: 587–597. Takao J, Ariizumi K, Dougherty II, Cruz Jr PD. (2002).
Kato H. (1996). Anticancer Res 16: 2149–2153. Photodermatol Photoimmunol Photomed 18: 5–13.
Kosmoski JV, Ackerman EJ, Smerdon MJ. (2001). Proc Natl Takeda A, Higuchi D, Takahashi T, Ogo M, Baciu P,
Acad Sci USA 98: 10113–10118. Goetinck PF et al. (2002). J Invest Dermatol 118: 147–154.
Lahmann C, Young AR, Wittern KP, Bergemann J. (2001). Tanaka H, Arakawa H, Yamaguchi T, Shiraishi K, Fukuda S,
Photochem Photobiol 73: 657–663. Matsui K et al. (2000a). Nature 404: 42–49.
Lassus P, Ferlin M, Piette J, Hibner U. (1996). EMBO J 15: Tanaka N, Fujioka A, Tajima S, Ishibashi A, Hirose S.
4566–4573. (2000b). Br J Dermatol 143: 728–732.
Lee JH, An HT, Chung JH, Kim KH, Eun HC, Cho KH. Thorey IS, Roth J, Regenbogen J, Halle JP, Bittner M, Vogl T
(2002). Photodermatol Photoimmunol Photomed 18: 253–261. et al. (2001). J Biol Chem 276: 35818–35825.
Lee KM, Lee JG, Seo EY, Lee WH, Nam YH, Yang JM et al. Tran SE, Meinander A, Eriksson JE. (2004). Trends Biochem
(2005). Br J Dermatol 152: 52–59. Sci 29: 601–608.
Li D, Turi TG, Schuck A, Freedberg IM, Khitrov G, Utrera R, Collavin L, Lazarevic D, Delia D, Schneider C.
Blumenberg M. (2001). FASEB J 15: 2533–2535. (1998). EMBO J 17: 5015–5025.
Li G, Ho VC. (1998). Br J Dermatol 139: 3–10. Wang X, Taplick J, Geva N, Oren M. (2004). FEBS Lett 561:
Liu M, Dhanwada KR, Birt DF, Hecht S, Pelling JC. (1994). 195–201.
Carcinogenesis 15: 1089–1092. Wehrli P, Viard I, Bullani R, Tschopp J, French LE. (2000).
Liu X, Mann DB, Suquet C, Springer DL, Smerdon MJ. J Invest Dermatol 115: 141–148.
(2000). Biochemistry 39: 557–566. Welss T, Sun J, Irving JA, Blum R, Smith AI, Whisstock JC
Maelandsmo GM, Florenes VA, Mellingsaeter T, Hovig E, et al. (2003). Biochemistry 42: 7381–7389.
Kerbel RS, Fodstad O. (1997). Int J Cancer 74: 464–469. Xirodimas DP, Saville MK, Bourdon JC, Hay RT, Lane DP.
Marionnet C, Bernerd F, Dumas A, Verrecchia F, Mollier K, (2004). Cell 118: 83–97.
Compan D et al. (2003). J Invest Dermatol 121: 1447–1458. Yang A, Kaghad M, Wang Y, Gillett E, Fleming MD, Dotsch
Miyashita T, Krajewski S, Krajewska M, Wang HG, Lin HK, V et al. (1998). Mol Cell 2: 305–316.
Liebermann DA et al. (1994). Oncogene 9: 1799–1805. Young AR, Chadwick CA, Harrison GI, Hawk JL, Nikaido
Murakami T, Fujimoto M, Ohtsuki M, Nakagawa H. (2001). O, Potten CS. (1996). J Invest Dermatol 106: 1307–1313.
J Dermatol Sci 27: 121–129. Zhang T, Woods TL, Elder JT. (2002). J Invest Dermatol 119:
Okamoto K, Beach D. (1994). EMBO J 13: 4816–4822. 1196–1201.
Park WY, Hwang CI, Im CN, Kang MJ, Woo JH, Kim JH Zhao R, Gish K, Murphy M, Yin Y, Notterman D, Hoffman
et al. (2002). Oncogene 21: 8521–8528. WH et al. (2000). Genes Dev 14: 981–993.
Oncogene