Oncogene (2001) 20, 2413 ± 2423
ã 2001 Nature Publishing Group All rights reserved 0950 ± 9232/01 $15.00
www.nature.com/onc
Function and regulation of AP-1 subunits in skin physiology and pathology
Peter Angel*,1, Axel Szabowski1 and Marina Schorpp-Kistner1
1
Deutsches Krebsforschungszentrum, Division of Signal Transduction and Growth Control, Im Neuenheimer Feld 280, 69120,
Heidelberg, Germany
The mouse skin has become the model of choice to study
the regulation and function of AP-1 subunits in many
physiological and pathological processes in vivo and in
vitro. Genetically modi®ed mice, in vitro reconstituted
skin equivalents and epidermal cell lines were established,
in which AP-1-regulated genetic programs of cell
proliferation, di€erentiation and tumorigenesis can be
analysed. Since the epidermis, as our interface with the
environment, is subjected to radiation and injury, signal
transduction pathways and critical AP-1 members
regulating the mammalian stress response could be
identi®ed. Regulated expression of important components
of the cytokine network, cell surface receptors and
proteases, which orchestrate the process of wound
healing has been found to rely on AP-1 activity. Here
we review our current knowledge on the function of AP-1
subunits and AP-1 target genes in these fascinating ®elds
of skin physiology and pathology. Oncogene (2001) 20,
2413 ± 2423.
Keywords: epidermis; Fos; Jun; proliferation; di€erentiation; target genes
functions of Fos, Jun and ATF proteins (see Jochum
et al. this issue).
The skin represents a very elegant, easily accessible
and highly informative system to study the regulation
and function of AP-1 and AP-1-regulated genes in
multiple biological processes such as cell proliferation
and di€erentiation. It allows elucidating cell-autonomous functions of AP-1 proteins as well as AP-1dependent genetic programs regulated by cell-cell and
cell-matrix interactions, or by soluble factors acting in
an autocrine or paracrine fashion. Moreover, both
tissue culture and animal models have been established
to determine the role of AP-1 subunits in tumour
formation and progression. Since the skin functions as
the outer protective barrier of a body exposed to a
variety of environmental stress factors (e.g. radiation
and chemical xenobiotics, injury by mechanical stress)
genetic programs controlling the cellular stress response and tissue remodelling and regeneration during
wound healing can be determined. This review
summarises our current knowledge on the function of
AP-1 members in these aspects of skin biology.
Introduction
Expression of AP-1 subunits in the skin
The transcription factor AP-1 mediates gene regulation
in response to a variety of extracellular stimuli
including growth factors, cytokines, oncogenes, tumour
promoters and chemical carcinogens (Angel and Karin,
1991; Karin et al., 1997). AP-1 represents a heterogenous set of dimeric proteins consisting of members of
the Jun, Fos and ATF families. Small di€erences in
sequence speci®city for individual AP-1 dimers have
been observed. Moreover, once bound to DNA, the
transactivation function of subunits of a given dimeric
complex may vary which can be explained by
di€erences in phosphorylation by upstream protein
kinases (see Rincon et al. this issue) as well as by
protein/protein interaction with other cellular factors.
In consequence, distinct sets of genes are targeted,
suggesting that individual members of the AP-1 family
have speci®c functions in AP-1-regulated cellular
processes. Loss of function approaches in mice and
analysis of cells derived from such mutants already led
to the identi®cation of speci®c, non-overlapping
The skin can be dissected into two major compartments, the epidermis representing the outermost skin
layer, and the dermis, both of which are separated by
the basal lamina. Homeostasis of the epidermis relies
on a tightly regulated balance between proliferation
and di€erentiation. Continuous proliferation of epidermal stem cells in the stratum basale and a shift in
the balance between proliferation and di€erentiation in
cells of the adjacent layers are a prerequisite to replace
the shed material of the stratum corneum (Figure 1).
The di€erent cell layers can be classi®ed by histological
means, or e.g. by abundant desmosomes as cells
belonging to the spinous layer and by the presence of
hyaline granula in the stratum granulosum. Importantly, typical marker genes indicative for a speci®c
phase of the di€erentiation process have been
identi®ed. Certain members of the cytokeratin protein
family, K5 and K14, are indicative for the basal layer,
whereas K1 and K10 are representative of the spinous
layer. Filaggrin, type I transglutaminase as well as the
envelope precursor proteins involucrin and loricrin are
other markers speci®cally expressed in the suprabasal
layers of the skin (for review Eckert and Welter, 1996).
*Correspondence: P Angel
AP-1 in skin biology
P Angel et al
2414
Figure 1 Expression pattern of AP-1 subunits in skin. A schematic illustration of the skin, which can be divided in the two major
compartments epidermis and dermis, is shown on the left. The multi-layered structure of the epidermis forming de®ned strata
re¯ects speci®c di€erentiation states of epidermal keratinocytes. The pattern of expression and abundance of human and mouse cJun, JunB, JunD, c-Fos, FosB, Fra-1 and Fra-2 proteins in these layers are shown on the right. Low but signi®cant expression of
most of these proteins is found in ®broblasts, which are located in the dermal compartment
While the expression pattern of AP-1 subunits in the
dermis, in particular in dermal ®broblasts, has not been
analysed in great detail, expression of a few family
members was studied in human and mouse epidermis
by both in situ hybridization and immunohistochemical
analysis (Welter and Eckert, 1995; Rutberg et al., 1996;
Schorpp-Kistner and Angel, unpublished). During
mouse embryonic development expression of c-Jun
and JunB in skin becomes ®rst visible at E17.5 showing
an ill-de®ned pattern of expression of c-jun in the
epidermis. JunB expression was found in the super®cial
stratum corneum, but not in basal layers of proliferating epidermal cells (Wilkinson et al., 1989). In contrast,
Fra-2-expressing cells were detected at E16.5 in the
basal layer and the upper granular layer, but not in the
stratum corneum (Carrasco and Bravo, 1995). These
data suggest that during the process of skin development (e.g. the stratum corneum ®rst arises at E16)
neither c-Jun, nor JunB play a decisive role. In the
epidermis of newborn mice a distinct expression
pattern of Fos and Jun proteins has been described
(Rutberg et al., 1996). Expression of c-jun, junD and
Fra-1 are co-ordinately expressed in the basal and
spinous layers of the epidermis. Basal cells also express
c-fos, junB and Fra-1, but not fosB. JunB and Fra-2 are
the only AP-1 members, which are also found in the
granular layer (Carrasco and Bravo, 1995; Rutberg et
al., 1996; Figure 1), suggesting the existence of
subgroups of AP-1 target genes involved in keratinocyte proliferation and di€erentiation, whose expression
is modulated by distinct AP-1 subunits. Interestingly,
di€erences in the expression pattern of Fos and Jun
proteins in mouse (newborn) and human (neonatal)
epidermis (Eckert and Welter, 1996) have been
detected. In human, JunB was, similar to JunD,
ubiquitously expressed in all layers of the epidermis
Oncogene
showing highest levels in the stratum basale and
stratum spinosum. In contrast, c-Jun was found to be
restricted to the granular layer (Figure 1). A somewhat
broader pattern was observed for c-Fos which was
present in both the spinous and granular layers but
seemed to be completely absent in the basal layer and
the lower zone of the stratum spinosum.
Interestingly, there was only a very low degree of
overlap in the expression of c-Fos and the other
members of the Fos protein family, particularly FosB,
suggesting subunit-speci®c regulation of target genes.
FosB, Fra-1 and Fra-2 proteins were co-ordinately
expressed in the spinous layer and expression of Fra-1
and Fra-2, but not FosB, was even further increased in
cells of the lower zone of the stratum granulosum.
(Figure 1; for review, Eckert and Welter, 1996). A
detailed analysis of expression of ATF proteins in the
skin has yet to be performed.
Regulation and function of Fos and Jun proteins in
keratinocyte proliferation and di€erentiation
Signalling pathways and AP-1 target genes in skin
The distinct expression pattern of the Jun and Fos
proteins in the epidermis suggests a general role for
individual AP-1 subunits in cell proliferation and
di€erentiation. Protein interference experiments in
tissue culture cells, such as expression of antisense
RNA or injection of neutralizing antibodies, and
generation of subunit-speci®c knock out mice and cells
derived thereof have established a critical role for Fos
and Jun members in these processes. This includes the
negative regulatory function of c-Jun on p53, its
positive e€ect on cyclin D1 regulation and the positive
AP-1 in skin biology
P Angel et al
transcriptional regulation of p16 by JunB as reported
for ®broblasts (see Jochum et al.; Shaulian and Karin,
this issue). Moreover, binding of c-Jun to Rb has been
described in human keratinocyte cell lines (Nead et al.,
1998) suggesting protein/protein interactions between
AP-1 members and cell cycle regulators as an
additional level of cell cycle control by AP-1 subunits.
However, it is important to note that the general
concept of c-Jun being a positive regulator of cell
proliferation and of JunB (and possibly JunD) acting
as an attenuator of cell cycle progression to favour
di€erentiation may not account for cells of the
epidermis, at least in human. As described above, the
proliferating cells of the basal layer do neither express
c-Jun nor c-Fos, but exhibit signi®cant amounts of
JunB and JunD. Surprisingly, c-Jun (and c-Fos)
expression is restricted to proliferation incompetent,
terminally di€erentiated cells of the stratum granulosum, which express only low amounts of JunB and
JunD. These ®ndings suggest that in human AP-1
subunits expressed in keratinocytes are predominantly
involved in the regulation of the di€erentiation
program rather than in the proliferation control. On
the other hand, the expression pattern of Fos and Jun
proteins in mice may point to a critical role of c-Jun,
JunD and Fra-1 in keratinocyte proliferation or in the
early stages of di€erentiation.
Additional lines of evidence for a critical role of
AP-1 subunits in keratinocyte proliferation and/or
di€erentiation come from the increasing number of
genes, which have been found to harbour AP-1 binding
sites in their promoters, and which are expressed in
distinct layers of the epidermis. In most cases,
functional relevance of these sites for di€erentiationspeci®c basal expression or for induced expression
upon critical extracellular signals, e.g. levels of calcium,
was con®rmed in established keratinocyte cell lines.
The list of AP-1 target genes in the epidermis, which
are considered to encode critical players of skin
homeostasis includes transglutaminase and di€erent
members of the cytokeratin gene family such as K1,
K5, K6, K8, K14, K18 and K19 (for references, see
Eckert and Welter, 1996; Navarro et al., 1995; Hu and
Gudas, 1994). The requirement of the AP-1 site in the
human K14 and K18 promoters was con®rmed in
transgenic mice (Rhodes and Oshima, 1998; Sinha et
al., 2000). High basal level expression of the pro®laggrin gene in epidermal keratinocytes depends on c-Jun/
c-Fos heterodimers (Jang et al., 1996), whereas cooperative action of AP-1 and Sp1 is required for
maximal expression of the human involucrin promoter
(Lopez-Bayghen et al., 1996). JunD, Fra-1 and Fra-2
proteins, whose levels are a€ected by changes in the
concentration of extracellular calcium, preferentially
bind to the calcium response element in this promoter
(Ng et al., 2000). However, Fra-1, JunB and JunD
appear to be responsible for the regulation of
involucrin promoter activity by phorbol esters (Welter
et al., 1995). Functional synergism between AP-1 and
ETS proteins was found to mediate expression of the
keratinocyte terminal di€erentiation markers small
proline-rich protein (SPRR)-1A (Sark et al., 1998)
and SPRR3 (Fischer et al., 1999). In contrast,
expression of another member of the SPRR family
sprr2, is negatively regulated by c-Jun. Interestingly,
this e€ect is likely to be indirect, since repression is
mediated by a region, which di€ers from the AP-1 site
in the sprr promoter (Lohman et al., 1997). Transient
co-transfection experiments suggested that c-Fos/JunB
heterodimers are critically involved in the di€erentiation-speci®c expression of loricrin. However, analysis
of the loricrin promoter in transgenic mice showed that
an additional negatively acting element is needed to
restrict loricrin expression to the granular layer (Di
Sepio et al., 1999). Similarly, AP-1, ETS and members
of the AP-2 transcription factor families, binding to
two distinct promoter regions, are required to orchestrate keratinocyte-speci®c activity of the K14 promoter
in transgenic mice (Sinha et al., 2000).
In addition to this impressive and still increasing
number of AP-1-regulated genes in epidermal cells,
di€erences in the expression pattern of AP-1 subunits
during in vitro di€erentiation of keratinocytes have
pointed to a regulatory function of AP-1 in the
formation and maintenance of the epidermis. In vitro
multiple phenotypes of keratinocyte di€erentiation can
be recapitulated e.g. by raising the calcium concentration or treatment of cells with phorbol esters (for
references see Rutberg et al., 1997). Under such
conditions speci®c alterations in the expression of Fos,
Jun and AP-1 target genes (sprr1, pro®laggrin) can be
detected suggesting speci®c and partially opposing
functions of AP-1 subunits in keratinocyte di€erentiation (Gandarillas and Watt, 1995; Rutberg et al., 1997).
Moreover, there is strong evidence that signal transduction pathways modulating AP-1 activity (see Rincor et
al. this issue) are main regulators of keratinocyte
proliferation and di€erentiation. For example, calcium-induced terminal di€erentiation depends on speci®c isoforms of PKC, and PKC activation is required
for calcium-induced expression of Fra-2, JunB and
JunD expression and enhanced AP-1 binding activity
(Rutberg et al., 1996; Takahashi et al., 1998). On the
other hand, transdominant negative versions of
MEKK1, MEK1, MEK3, MEK7, p38 and c-jun, but
not mutants of the Raf/Erk or MEK4/JNK signalling
pathway, interfere with TPA-dependent induction of
involucrin expression (E®mova et al., 1998). These data
suggest that at least one route of the stress signal
transduction pathway may also be involved in the
regulation of keratinocyte di€erentiation.
An additional argument for the involvement of AP-1
in skin biology is the repression of involucrin and K1,
K5 and K14 gene expression by retinoids and steroid
hormones, which are the most commonly used
therapeutic agents in dermatology (Lu et al., 1994;
Monzon et al., 1996; Radoja et al., 2000). Although in
each case repression is mediated through an AP-1 site,
loss of K1 expression might be explained through
competitive binding of AP-1 and steroid hormone
receptors to a composite AP-1/steroid hormone
receptor element (Lu et al., 1994). Repression of the
2415
Oncogene
AP-1 in skin biology
P Angel et al
2416
other genes is likely to be mediated by physical
interaction of AP-1 with the retinoic acid or
glucocorticoid receptor resulting in mutual inactivation
of their transactivation functions (see Herrlich, this
issue).
Role of AP-1 in wound healing
Despite this large body of data on the expression
pattern of AP-1 members and potential AP-1 target
genes in tissue culture cells and in the mouse, the
phenotypes of mice lacking c-Fos, FosB or JunD in the
epidermis did not provide evidence for a critical role of
these proteins in normal skin (see Jochum et al. this
issue). Moreover, speci®c deletion of c-Jun in cells of
the stratum basale did not yield an obvious skin
phenotype (R Zenz and EF Wagner, personal communication). These data suggest that single subunits (e.g.
c-Jun, c-Fos) may not be required for skin development, or that other members of the protein family may
functionally compensate their loss. The process of skin
metabolism initiated by signals that trigger undi€erentiated, proliferative cells to undergo cell di€erentiation and to ®nally become shed material of the
corni®ed layer is fairly slow. Therefore, in light of
the rapid and transient kinetics of expression and
activity of AP-1 subunits in response to many
extracellular signals (for review Angel and Karin,
1991; Karin et al., 1997) it is tempting to speculate
that AP-1 activity may be required under conditions
where the balance of keratinocyte proliferation and
di€erentiation has to be rapidly and temporally altered.
The best example for the requirement of such a switch
of genetic programs is the process of wound repair and
tissue regeneration, also known as cutaneous wound
healing. The healing of an adult wound is a very
complex, dynamic and interactive process involving
soluble mediators, blood cells, dermal cells (in
particular ®broblasts which are responsible for
synthesis of components of the extracellular matrix)
and keratinocytes of the epidermis (Figure 2; for
review; Martin, 1997; Singer and Clark, 1999).
The Ras/MAP/AP-1 pathway appears to play an
important role in cell proliferation control in response
to wounding since interference with Ras inhibits DNA
synthesis, cell migration and induction of AP-1
members including c-Fos in bovine endothelial cells
after mechanical wounding (Sosnowski et al., 1993).
On the other hand, in mice lacking c-Jun in
keratinocytes tissue regeneration after wounding is
not signi®cantly disturbed (R Zenz and EF Wagner,
personal communication), excluding the absolute
requirement of c-Jun in cell-autonomous regulation of
keratinocyte proliferation, migration and di€erentiation during wound healing in vivo. Nevertheless, there
is strong evidence that AP-1 is critically involved in this
process, since Jun, Fos and ATF proteins are at the
receiving end of signalling pathways initiated by
Figure 2 AP-1-dependent players of the cytokine network controlling the process of wound healing. Growth factors and cytokines
secreted by various cell types during wound healing are shown. Only those, whose expression is regulated by AP-1, or whose activity
in target cells depends, at least in part, on AP-1 activity, are considered. For a more detailed description of the wound healing
process, see Martin, (1997) and Singer and Clark (1999). On the left, the double paracrine network of epithelial-mesenchymal
interaction established in an in vitro organotypic coculture system is highlighted. After wounding keratinocytes secrete IL-1 which
induces expression of KGF and GM-CSF in ®broblasts. C-Jun and JunB regulate expression of both cytokines in ®broblasts in an
antagonistic manner. An appropriate balance of KGF and GM-CSF is required for co-ordinated keratinocyte proliferation and
di€erentiation during reepithelialization of the epidermis
Oncogene
AP-1 in skin biology
P Angel et al
cytokines and growth factors in appropriate target cells
and, furthermore, since AP-1 is involved in the
regulation of synthesis of the factors themselves.
In normal and wounded skin, several growth
factors and interleukins have been detected, such as
IL-1, IL-6, IL-8, GM-CSF, TFG-a and TGF-b,
NGF, PDGF as well as members of the FGF family
(for review see SchroÈder, 1995; Martin, 1997; Singer
and Clark, 1999). First evidence for an involvement
of AP-1 in wound healing was provided by the fact
that certain factors, such as EGF, FGF, members of
the TGFb, TNF-a and interleukin protein families
stimulate AP-1 activity through the activation of the
MAP kinases JNK and p38, which, in turn
hyperphosphorylate and thereby activate c-Jun and
ATF-2, the main mediators of c-jun expression in
response to cytokines and genotoxic agents. Upregulated gene expression resulting in enhanced levels of
Jun and Fos proteins triggers alterations in the
expression of AP-1 target genes, for example matrix
metalloproteinases and IL-2 (Muegge et al., 1989;
Borden and Heller, 1997).
Second, a number of cytokines have been suggested
to be potential AP-1 target genes. For example
regulation of the human TNF promoter by phorbol
esters depends in a variety of cell types on both an
AP-1 and a juxtaposed PEA3/ETS-1 site (Kramer et
al., 1995). In myelomonocytic cells, AP-1 synergizes
with C/EBPb in TNF-a expression. Interestingly, in
these cells co-operation does not require the transactivation domain of c-Jun but may be explained by
protein/protein interaction between c-Jun and c/EBPb
leading to enhanced DNA binding (Zagariya et al.,
1998). In vitro binding studies and mutational analysis
revealed the importance of Jun and Fos proteins in the
expression of mouse M-CSF in ®broblasts and
monocytes (Konicek et al., 1998). AP-1 sites have been
found in the promoter region of human and murine
KGF and GM-CSF genes (Finch et al., 1995; Wang et
al., 1994; Ye et al., 1996; Viebig, Szabowski and Angel,
unpublished). The importance of AP-1 for the
induction of KGF promoter activity during wound
healing and repression by glucocorticoids (Brauchle et
al., 1995) has not been analysed in detail. In the case of
the GM-CSF promoter, AP-1 exhibits its regulatory
activity through functional synergism with other
transcription factors including ETS-1, NF-kB (Wang
et al., 1994; Thomas et al., 1997), a Sp1-related
complex (Ye et al., 1996) and NFAT (Masuda et al.,
1993). By contrast, AP-1-depended GM-CSF promoter
activity can be repressed by the nuclear factor YY1 (Ye
et al., 1996). Functional synergism between AP-1, NFIL-6 and NF-kB has been observed for the regulation
of IL-8 expression (Roebuck, 1999). Two adjacent
AP-1 binding sites recognized by JunD and a Fosrelated protein mediate transcriptional activation of
tissue factor (TF), the cellular initiator of the protease
cascade leading to blood coagulation, by serum, as well
as of de®ned growth factors, such as PDGF and FGF
(Felts et al., 1995). Interestingly, overexpression of cFos and JunD abrogates the requirement for serum in
the stimulation of the TF promoter activity (Felts et
al., 1995). Yet, it remains to be determined whether in
addition to c-Fos and JunD other AP-1 subunits are
involved in TF expression. Finally, an ATF/CREBbinding site in the macrophage in¯ammatory protein 1
beta (MIP-1 beta) seems to control cell type- and LPSinduced expression of this gene in macrophages
(Prott et al., 1995).
AP-1 sites have been found in the promoter region
of the ®broblast growth factor receptor 1 (Saito et al.,
1992), and IL-1 receptor antagonist promoter (Smith et
al., 1992), suggesting that AP-1 is also involved in the
regulation of cytokine receptors thereby modulating
the susceptibility of target cells for cytokine action.
As illustrated in Figure 2 cytokine-mediated cross
talk involves a variety of cell types in the epidermis and
in the mesenchyme which makes it very dicult to
de®ne precisely the source of the cytokine of interest.
However, a simpli®ed in vitro tissue engineered skin
equivalent model mimicking tissue physiology has been
established, in which the regulatory cross talk between
keratinocytes and dermal ®broblasts can be investigated. In these organotypic cultures keratinocyte
proliferation and di€erentiation is maintained under
the control of co-cultured dermal ®broblasts from
human or mouse to form a three-dimensionally
organized epithelium (Bell et al., 1981; Fusenig, 1994;
Choi and Fuchs, 1990). Keratinocytes form a typical
epidermal tissue architecture expressing essentially all
characteristic di€erentiation markers including a basement membrane (Smola et al., 1998). Importantly, this
in vitro system mimics many aspects of the reepithelialization process during wound closure. This
mesenchymal/epithelial interplay via di€usible factors
is initiated by keratinocyte-derived IL-1 acting on
®broblasts to induce the expression of growth factors
such as KGF and GM-SCF in mouse ®broblasts acting
back on human keratinocytes in a paracrine fashion.
The speci®c contribution of AP-1-dependent cytokine
expression in ®broblasts regulating keratinocyte proliferation and di€erentiation was investigated using
®broblasts from c-jun or junB-de®cient mice (Szabowski et al., 2000). Whereas c-jun7/7 ®broblasts did not
support keratinocyte growth, junB7/7 ®broblasts
caused enhanced keratinocyte growth and di€erentiation. These phenotypes may be explained by the lack of
constitutive and Il-1-induced expression of KGF and
GM-CSF in c-jun7/7 cells and enhanced constitutive
expression of both genes in junB7/7 cells. Growthinhibited thin epithelia in c-jun7/7 ®broblasts containing cultures could be partially rescued by KGF to a
proliferative but less di€erentiating epithelium, whereas
addition of both KGF and GM-CSF completely
restored the wild-type phenotype. On the other hand,
blocking antibodies to GM-CSF and IL-1 reduced the
phenotype obtained by junB7/7 ®broblasts. These data
suggest that the relative abundance and activation state
of a given AP-1 subunit in a non-cell autonomous,
trans-regulatory fashion directs regeneration of the
epidermis and maintenance of tissue homeostasis in
skin (Szabowski et al., 2000).
2417
Oncogene
AP-1 in skin biology
P Angel et al
2418
Oncogene
Fibroblast-derived KGF and GM-CSF, whose
function in skin homeostasis has been con®rmed in
mice (Rubin et al., 1989; Werner et al., 1992; Xing et
al., 1997) represent the ®rst members of a potentially
growing list of critical AP-1-regulated cytokines, which
are produced by the various cell types during the
di€erent phases of the wound healing process (for
review, Martin, 1997; Singer and Clark, 1999). These
factors very likely co-operate with ctyokines that are
expressed in an AP-1 independent manner, e.g.
members of the TGF-b family, their receptors (Frank
et al., 1996) and downstream targets of the activin A
pathway (Munz et al., 1999).
In addition to autocrine and paracrine cytokine and
growth factor action, there is convincing evidence that
cell-cell and cell matrix interactions are important
regulators of keratinocyte proliferation, migration and
di€erentiation. AP-1-regulated genes may also a€ect
this type of regulatory mechanism. AP-1 binding sites
have been identi®ed in the promoter regions of avian
beta-3 integrin (Cao et al., 1993), alpha2 (Zutter et al.,
1994), alpha5 (Corbi et al., 2000) alpha6 (Nishida et
al., 1997) and alpha7 integrins (Ziober and Kramer,
1996), laminin B1 (Matsui, 1996), laminin gamma2chain (Olsen et al., 2000), decorin (Mauviel et al., 1996)
and tenascin-C (Mettouchi et al., 1997). Only in some
cases, speci®c AP-1 subunit compositions have been
described, which may be responsible for regulation of
these genes, such as JunD regulating expression of
laminin gamma2 by HGF (Olsen et al., 2000), and
JunD/Fra-2 modulating basal and TGF-b-dependent
laminin alpha 3A transcription (Virolle et al., 1998).
Moreover, to allow tissue remodelling during wound
healing the composition of the extracellular matrix has
to be altered. This can be achieved by transcriptional
regulation of cellular genes encoding ECM components
as well as by proteolytic enzymes (particularly matrix
metalloproteinases) and their inhibitors that control the
ECM turnover on the post-translational level (for
review: Borden and Heller, 1997; Vu and Werb, 2000).
Over the past years, a critical role for AP-1 in the
regulation of ECM components has been established.
For example, an AP-1 binding site mediates expression
of human alpha2 (I) collagen (COL1A2) gene by TGFb and TNF-a (Chung et al., 1996). Expression of
syndecan-1, a cell surface heparan sulphate proteoglycan that binds growth factors, is activated during
cutaneous wound healing through a so-called AP-1driven FGF-inducible response element (FiRE). Interestingly, FiRE activity is selectively upregulated in
migrating keratinocytes at the edge of the wound and
is persistent until completion of the re-epithelialization
process. Although the exact composition of AP-1
complexes binding to this element has not yet been
identi®ed, modulation of AP-1 complexes is mediated
by members of the MAP kinase family including ERKs
and p38 (Jaakkola et al., 1998).
AP-1 plays a dominant role in the transcriptional
activation of the majority of matrix-metalloproteinases
(MMPs) that are required for cell migration, ECM
degradation and tissue reorganization during the
wound healing process. Basal transcription as well as
activation of MMP promoters in response to phorbol
esters, cytokines, growth factors and cell matrix
interactions and altered cell-cell contacts require the
speci®c interaction of AP-1 and ETS proteins (for
review see Benbow and Brinckerho€, 1997). In general,
MMPs are not constitutively expressed in skin but are
induced temporarily in response to the above mentioned exogenous signals to trigger the proteolytic
remodelling of the ECM in physiological and pathological situations e.g. tissue morphogenesis, tissue
repair but also dermal photoaging and tumour cell
invasion (for review KaÈhaÈri and Saarialho-Kere, 1997).
In acute murine excisional skin wounds interstitial
collagenase (MMP-1; MMP-13), MMP-9, MMP-3,
MMP-10 and their physiological inhibitor TIMP-1
are strongly induced showing a unique spatial and
temporal transcription pattern (Madlener et al., 1998).
Tissue culture as well as animal studies have provided
evidence for the importance of the AP-1 target genes
MMP-1 and MMP-3 during wound healing in
keratinocyte migration (MMP-1; Lund et al., 1999)
and wound contraction, respectively (MMP-3; Bullard
et al., 1999).
Involvement of AP-1 subunits in photoaging
According to the function of the skin to serve as a
protective barrier against the environment, such as
ultraviolet irradiation from sunlight, skin cells actively
respond to such signals by changing their repertoire of
gene expression to prevent radiation-induced damage.
Long-term exposure to UV causes premature skin
aging (photoaging), in the course of which the
symptoms include leathery texture, mottled pigmentation and wrinkles. There is strong evidence that this
process is driven by major changes in gene expression
of ®broblasts in the dermal compartment (Bernerd and
Asselineau, 1998). This includes induction of AP-1regulated matrix metalloproteinases, particularly interstitial collagenase (MMP-1), stromelysin (MMP-3) and
gelatinase B (MMP-9; Fisher et al., 1997). In
agreement with the critical role of AP-1 in the
transactivation of these genes by UV, transcription of
the c-jun, junB and c-fos genes is also enhanced upon
UV irradiation (Gillardon et al., 1994; Fisher et al.,
1999; Soriani et al., 2000; Shaulian and Karin, this
issue). However, there is evidence that, depending on
the wavelength of the UV light, the eciency of
induction di€ers among the Fos and Jun members
(Ariizumi et al., 1996). Importantly, pre-existing and
newly synthesized c-Jun protein becomes eciently
hyperphosphorylated, most likely by members of the
JNK/SAPK kinase family (Ramaswamy et al., 1998;
Fisher et al., 1999). Activation of c-Jun seems also to
depend on PKCzeta, because pre-treatment of mouse
skin with PKCzeta antisense oligonucleotides impaired
UV induced c-Jun-speci®c gene expression (Huang et
al., 2000). Interestingly, pre-treatment of cells with
retinoic acid, which diminishes the eciency of
photoaging inhibits UV-dependent transcriptional acti-
AP-1 in skin biology
P Angel et al
vation of MMPs. The repression seems to be mediated
by an accelerated breakdown of the c-Jun protein via
the ubiquitin-proteasome pathway (Fisher et al., 1999).
However, superfusion of rat skin with c-fos antisense
oligonucleotides did not signi®cantly reduce the
formation of sunburn cells in the UV-irradiated
epidermis (Gillardon et al., 1994).
Role of AP-1 in tumourigenesis and other skin diseases
The ability of speci®c members of the Jun and Fos
protein family, as well as their retroviral derivatives, to
transform tissue culture cells and to induce tumours in
the animal (avian, rat, mouse) is well established (see
van Dam and Castellazzi and Jochum et al., this issue).
There is evidence that changes in AP-1 expression and
activity correlate with di€erent stages of melanoma
development and progression (Rutberg et al., 1994). It
was, however, the analysis of AP-1 function in
keratinocyte cell lines and in epidermal tumours of
the mouse, which has contributed considerably to our
current knowledge on the function of AP-1 in
tumourigenesis.
Di€erent tumour model systems of mouse skin have
become available:
Formation of squamous cell carcinoma can be
induced in hairless mice upon chronical exposure to
UV-B irradiation. While a single exposure leads to a
transient upregulation of c-fos, c-jun and c-myc, high
constitutive expression of c-fos, c-myc and Ha-ras was
observed in the tumour cells (Sato et al., 1997).
Interestingly, application of the monoterpene perillyl
alcohol (POH) inhibited tumour incidence and UV-Binduced AP-1 activity in mouse skin (Barthelman et al.,
1998). The speci®c contribution of these proteins in the
induction of skin photocarcinogenesis still remains to
be con®rmed by functional assays.
Skin papillomas can be induced with very high
eciency upon constitutive activation of the Ras
pathway in basal keratinocytes (Sibilia et al., 2000).
Tumour formation is impaired in mice expressing a cJun mutant, which lacks the phosphoacceptor sites for
JNK (Behrens et al., 2000), which strongly suggests
that activation of c-Jun and c-Jun-dependent target
genes is critically involved in the development of skin
papillomas.
Expression of bovine papilloma virus (BPV) in
mouse dermal ®broblasts induces the formation of
mild and aggressive ®bromatosis, and ®brosarcoma.
The latter two stages are characterized by strongly
enhanced expression of c-Jun and JunB, but not JunD
and Fos (Bossy-Wetzel et al., 1992). Overexpression of
c-Jun or JunB, but not JunD, in cells of mild
®bromatosis induced anchorage independence and
changes in the cell shape, suggesting that both proteins
may function as progression factors at intermediate
stages of tumorigenesis. Fibrosarcomas have been
observed also in v-Jun transgenic mice, which exhibit
high basal level expression of the transgene in
®broblasts (Schuh et al., 1990). However, tumour
formation depends on obligatory wounding, suggesting
that, in contrast to BPV, transformation by v-Jun
requires co-operative action with other yet unknown
factors.
When targeted to the basal cells of the epidermis,
expression of human papilloma virus 16 (HPV-16)
leads to progressive squamous epithelial neoplasia,
which is accompanied by alterations in the expression
of K5, K10, K14 and ®laggrin (Arbeit et al., 1994).
Products of HPV (E6, E7) may directly or indirectly
a€ect AP-1 activity, resulting in changes in gene
expression and keratinocyte proliferation and di€erentiation (Antinore et al., 1996; Shino et al., 1997). On
the other hand, transcriptional activity of HPV-16 and
other serotypes, e.g. HPV-18 and HPV-11 depends on
AP-1 binding sites (Cripe et al., 1990; Thierry et al.,
1992; Zhao et al., 1997). Exempli®ed by HPV-18 there
is increasing evidence that speci®c AP-1 subunits di€er
in their activity to either activate or trans-repress HPVlinked carcinogenesis of the uterine cervix (Thierry et
al., 1992; Soto et al., 1999).
The contribution of AP-1 subunits to premalignant
conversion and malignant progression of epidermal
cells is best documented by the initiation-promotion
protocol of mouse multistage skin carcinogenesis,
which ranks among the best studied in vivo approaches
in experimental cancer research (for review see Bowden
et al., 1994; Brown and Balmain, 1995; Yuspa, 1998).
Upon treatment of the skin with a single dose of
mutagen (e.g. DMBA), followed by repeated application of tumour promoters, such as the phorbol ester
12-0-tetradecanoylphorbol-13-acetate (TPA), benign
papillomas develop, some of which will enter malignant
progression to squamous cell carcinomas. In mice
lacking c-fos the development of malignant skin
tumours from papillomas induced by treatment with
TPA is blocked at the transition from benign to
malignancy (Saez et al., 1995). The eciency of
formation of papilloma did not di€er from wild-type
mice, but papillomas quickly hyperkeratinized. In
agreement with these data transgenic mice expressing
a transdominant-negative version of c-Jun (TAM67)
are resistant to DMBA-TPA induced skin tumorigenesis (Young et al., 1999). Surprisingly, transgenic mice
expressing v-Fos under the control of the human
keratin-1 promoter exhibit hyperplasia, hyperkeratosis
and squamous papillomas in wounded ears (Greenhalgh et al., 1993), suggesting that both overexpression
and loss of Fos activity interferes with skin homeostasis. An additional major advantage of this
carcinogenesis model is the availability of mouse
keratinocyte cell lines that represent each of the
intermediate stages in tumour development. In these
cells constitutive AP-1 binding and transactivating
ability of cells of the malignant, but not benign state
of the tumour was observed (Domann et al., 1994).
Moreover, high levels of phosphorylated c-Jun, Fra-1,
Fra-2 and ATF-2 proteins and a reduction in JunB
expression correlate with malignancy (Joselo€ and
Bowden, 1997; Zoumpourlis et al., 2000). Overexpression of c-Fos in the papilloma cell line seems to be
2419
Oncogene
AP-1 in skin biology
P Angel et al
2420
Oncogene
sucient for malignant conversion (Greenhalgh and
Yuspa, 1988), whereas expression of a transdominant
negative mutant version of c-Jun in malignant cells
inhibited expression of AP-1-dependent reporter genes
and the ability of the cells to form tumours in nude
mice (Bowden et al., 1994). These results are in good
agreement with data obtained in two variants of the
mouse keratinocyte cell line JB6, which are either
promotion-sensitive (P+) or promotion resistant (P7)
to TPA-induced transformation. TPA-induced expression of c-Jun is reduced in P7 cells and overexpression
of c-Jun increased the probability of progression in
P+, but not in P7 cells (Watts et al., 1995).
Overexpression of a dominant negative c-Jun mutant
blocked TPA-induced AP-1 activity, cell transformation and invasion (Dong et al., 1994, 1997). Block of
tumorigenesis is accompanied by a loss of TPAinduced expression of MMPs, such as interstitial
collagenase (MMP-13), stromelysin-1 (MMP-3, stromelysin-2 (MMP-10), gelatinase A (MMP-2) and MT1MMP (MMP-14; Dong et al., 1997). The TPA-induced
cellular response is mediated by MAP kinases, since
expression of transdominant negative version of ERK2
inhibits AP-1-dependent gene expression and neoplastic
transformation (Watts et al., 1998).
Co-treatment with retinoic acid (Huang et al., 1997)
or steroid hormones, particularly glucocorticoids (Belman and Troll, 1972), can eciently interfere with
chemically-induced tumour formation in the mouse.
TPA-dependent transcriptional upregulation of MMP3 and MMP-13 is similarly blocked by glucocorticoids
(Tuckermann et al., 1999). Induction of these genes is
abolished with similar eciency in wild-type and in
GRdim mice expressing a DNA-binding-defective
mutant version of GR, demonstrating that the DNA
binding-dependent function of GR is fully dispensable
for repression of AP-1 activity in vivo and that DNAbinding-independent functions of the GR are responsible for the antitumour promoting activity of
glucocorticoids (Tuckermann et al., 1999). Similarly,
the repression function of the retinoic acid receptor,
but not the activation of retinoic acid responsive genes
is required for the antitumour promotion e€ect of
retinoic acid (Huang et al., 1997). Remarkable results
of a recent study suggest that the eciency of antiin¯ammatory agents and cancer chemoprevention may
also be ascribed to an impaired recruitment and/or
function of in¯ammatory cells that supply the AP-1
target gene MMP-9. Analysis of the role of MMP-9
during carcinogenesis in K14-HPV16 transgenic mice
and in mice lacking MMP-9 revealed that MMP-9
expressed by in¯ammatory cells is functionally
involved in the regulation of oncogene-induced
keratinocyte hyperproliferation, progression to invasive cancer and end-stage malignant grade epithelial
carcinomas (Coussens et al., 2000). Thus, tumour
development and progression are complex multicellular processes, in which normal cells, such as
in¯ammatory cells, or normal constituents of wound
repair are conscripted to signi®cantly contribute to the
tumour phenotype (Boudreau and Bissell, 1998;
Hanahan and Weinberg, 2000). Moreover, there seems
to exist a quantitative requirement or threshold for an
acute neoplastic response and its progression to
malignant stages which in addition can depend on
the nature of the target cell as it has been
demonstrated for v-Jun-mediated wound tumorigenesis
(Shalaby et al., 1994).
Evidence for trans-regulatory functions of cellular
Jun proteins in the regeneration of the epidermis and
maintenance of tissue homeostasis comes from the
analysis of a reconstituted skin by combining primary
human keratinocytes and mouse wild-type, c-jun7/7 or
junB7/7 ®broblasts described above (Szabowski et al.,
2000; Figure 2). An aberrant or disturbed balance
between cytokines may also be implicated in pathological diseases in human, such as lichen planus,
psoriasis or dermato®brosis (Lever, 1983). Histologically cutaneous lichen planus is characterized by
epidermal hypergranulosis (a pronounced stratum
granulosum with high amounts of keratohyalin
granula), irregular acanthosis, hyperkeratosis, dermal
dibrosis and a dense in®ltrate (Stefanidou et al., 1999).
By contrast, absence of granular cells, hyperproliferation and parakeratosis is observed in psoriatic
epidermis (Lever, 1983). In vitro, normal keratinocytes
develop psoriatic-skin like epidermal phenotypes when
grown on ®broblasts from psoriatic lesions (Saiag et
al., 1985), providing evidence that the epidermal
alterations in psoriasis and dermato®broma are caused
by changes in the underlying dermal ®broblasts. The
critical interplay between ®broblasts and keratinocytes
can also be exempli®ed in dermato®bromas where a
signi®cant hyperplasia of the epidermis can be
observed overlying the benign ®broblastic tumour
nodule. Thereby, the ®brotic cells stimulate the
epidermis in a fashion similar to that of embryonic
mesenchyme (Lever, 1983).
Enhanced DNA binding activity of AP-1 was
observed in ®broblasts from recessive dystrophic
epodermolysis bullosa patients, which may be responsible for elevated mRNA and protein levels of AP-1
target genes, such as MMP-1 and MMP-3 (Unemoria
et al., 1994). In ®broblasts of the tight skin (TSK)
mouse, representing a model of ®brosis characterized
by exaggerated connective tissue accumulation in skin
and visceral organs, increased levels of alpha 1(I) and
alpha 1 (III) chain mRNA were found. This was
explained by a loss of the negatively acting AP-1
complex binding to the promoter of these genes
(Philips et al., 1995). The identi®cation of the speci®c
subunits of this complex, however, remains to be
determined. Similarly, the nature of the AP-1 subunits
exhibiting enhanced DNA binding to the collagenase
promoter in ®broblasts derived from patients with cutis
laxa (Hatamochi et al., 1996) is yet unknown.
Enhanced mRNA and protein levels of c-Jun and cFos were found in keloids and hypertrophic scars
representing a model of altered wound healing
characterized by hyperproliferation of dermal ®broblasts and accelerated production of components of the
extracellular matrix (Teofoli et al., 1999).
AP-1 in skin biology
P Angel et al
Summary and outlook
AP-1 participates in key physiological and pathophysiological processes due to its central role as a cellular
switch of genetic programs in response to extracellular
signals. The distinct expression pattern of AP-1
subunits in the skin as well as the large number of
AP-1 target genes in epidermal cells indicate that AP-1
plays a pivotal role in normal and pathological skin
physiology. Even though the loss of single AP-1
subunits in mice did not result in obvious skin
phenotypes, experiments with in vitro reconstituted
skin provided direct evidence for the orchestrated
antagonistic action of c-Jun and JunB to control
cytokine-regulated mesenchymal-epidermal interaction
in normal skin. Moreover, the regulatory function of
AP-1 in wound healing, photoaging and tumorigenesis
is demonstrated by numerous studies in tissue culture
cells and mouse models. Taken together, AP-1
transactivation of target genes contributes to both
protective and pathological responses in individual cells
or tissues, including the skin.
The analysis of transgenic mouse models and of in
vitro tissue engineered skin equivalent models using
genetically altered AP-1 subunit-de®cient cell types will
be the focus of future studies. As more data will be
generated, our current knowledge on the individual
functional involvement of AP-1 subunits in skin
homeostasis, wound healing and skin carcinogenesis
will be enhanced.
The design of novel therapeutic strategies and
potential clinical applications for skin diseases and
cancer will depend on the scienti®c progress made in
de®ning cell-type and stimulus-speci®c activation
modes of individual AP-1 subunits, on the character-
ization of AP-1 subunit-speci®c target genes and
genetic programs within a given cell type at a certain
di€erentiation state. The method of choice will be the
use of comprehensive array and bioinformatic tools to
compare the gene expression patterns for example of
keratinocytes located in the di€erent epidermal layers,
cells from non-treated and UV-treated skin, or cancer
cells of di€erent stages of tumour progression. The
major challenge, however, will be identi®cation of
those changes that are important, e.g. for cancer
development and progression. Clearly, the availability
of the broad spectrum of both in vivo and in vitro
experimental approaches described above represents a
powerful selection system to rapidly ®lter the enormous
number of di€erentially expressed genes and to focus
on the most promising tumour-associated genes. The
detailed analysis of known and newly identi®ed AP-1
target genes in such model systems o€ers the unique
opportunity to con®rm the importance of these genes
as potential drug targets and will proved a molecular
basis by which alterations in AP-1 subunit composition
and activity may act together in regulating key aspects
of normal or pathological phenotype development and
maintenance.
2421
Acknowledgments
We thank Erwin F Wagner for proof-reading of the
manuscript. Our laboratory is supported by grants from
the Deutsche Forschungsgemeinschaft, by the Cooperation
Program in Cancer Research of the DKFZ and Israeli's
Ministry of Science, and by the BioMed-2 and Training
and Mobility of Researchers (TNR) Programs of the
European Community.
References
Angel P and Karin M. (1991). Biochim. Biophys. Acta., 1072,
129 ± 157.
Antinore MJ, Birrer MJ, Patel D, Nader L and McCance DJ.
(1996). EMBO J., 15, 1950 ± 1960.
Arbeit JM, Munger K, Howley PM and Hanahan D. (1994).
J. Virol., 68, 4358 ± 4368.
Ariizumi K, Bergstresser PR and Takashima A. (1996).
J. Dermatol. Sci., 12, 147 ± 155.
Barthelman M, Chen W, Gensler HL, Huang C, Dong Z and
Bowden GT. (1998). Cancer Res., 58, 711 ± 716.
Behrens A, Jochum W, Sibilia M and Wagner EF. (2000).
Oncogene, 19, 2657 ± 2663.
Bell E, Ehrlich HP, Buttle DJ and Nakatsuji T. (1981).
Science, 211, 1052 ± 1054.
Belman S and Troll W. (1972). Cancer Res., 32, 450 ± 454.
Benbow U and Brinckerho€ CE. (1997). Matrix Biol., 15,
519 ± 526.
Bernerd F and Asselineau D. (1998). Cell Death Di€er., 5,
792 ± 802.
Borden P and Heller RA. (1997). Crit. Rev. Eukaryot. Gene
Expr., 7, 159 ± 178.
Bossy-Wetzel E, Bravo R and Hanahan D. (1992). Genes
Dev., 6, 2340 ± 2351.
Boudreau N and Bissell MJ. (1998). Curr. Opin. Cell Biol.,
10, 640 ± 646.
Bowden GT, Schneider B, Domann R and Kulesz-Martin M.
(1994). Cancer Res., 54, 1882s ± 1885s.
Brauchle M, Fassler R and Werner S. (1995). J. Invest.
Dermatol., 105, 579 ± 584.
Brown K and Balmain A. (1995). Cancer Metastasis Rev., 14,
113 ± 124.
Bullard KM, Lund L, Mudgett JS, Mellin TN, Hunt TK,
Murphy B, Ronan J, Werb Z and Banda MJ. (1999). Ann.
Surg., 230, 260 ± 265.
Cao X, Ross FP, Zhang L, MacDonald PN, Chappel J and
Teitelbaum SL. (1993). Biol. Chem., 268, 27371 ± 27380.
Carrasco D and Bravo R. (1995). Oncogene, 10, 1069 ± 1079.
Choi Y and Fuchs E. (1990). Cell Regul., 1, 791 ± 809.
Chung KY, Agarwal A, Uitto J and Mauviel A. (1996). J.
Biol. Chem., 271, 3272 ± 3278.
Corbi AL, Jensen UB and Watt FM. (2000). FEBS Lett., 474,
201 ± 207.
Coussens LM, Tinkle CL, Hanahan D and Werb Z. (2000).
Cell, 103, 481 ± 490.
Oncogene
AP-1 in skin biology
P Angel et al
2422
Oncogene
Cripe TP, Alderborn A, Anderson RD, Parkkinen S,
Bergman P, Haugen TH, Pettersson U and Turek LP.
(1990). New Biol., 2, 450 ± 463.
Di Sepio D, Bickenbach JR, Longley MA, Bundman DS,
Rothnagel JA and Roop DR. (1999). Di€erentiation, 64,
225 ± 235.
Domann Jr FE, Levy JP, Finch JS and Bowden GT. (1994).
Mol. Carcinog., 9, 61 ± 66.
Dong Z, Birrer MJ, Watts RG, Matrisian LM and Colburn
NH. (1994). Proc. Natl. Acad. Sci. USA, 91, 609 ± 613.
Dong Z, Crawford HC, Lavrosky V, Taub D, Watts R,
Matrisian LM and Colburn NH. (1997). Mol. Carcinog.,
19, 204 ± 212.
Eckert RL and Welter JF. (1996). Mol. Biol. Rep., 23, 59 ± 70.
E®mova T, LaCelle P, Welter JF and Eckert RL. (1998).
J. Biol. Chem., 273, 24387 ± 24395.
Felts SJ, Sto¯et ES, Eggers CT and Getz MJ. (1995).
Biochemistry, 34, 12355 ± 12362.
Finch PW, Lengel C and Chedid M. (1995). J. Biol. Chem.,
270, 11230 ± 11237.
Fischer DF, Sark MW, Lehtola MM, Gibbs S, van de Putte P
and Backendorf C. (1999). Genomics, 55, 88 ± 99.
Fisher GJ, Talwar HS, Lin J and Voorhees JJ. (1999).
Photochem. Photobiol., 69, 154 ± 157.
Fisher GJ, Wang ZQ, Datta SC, Varani J, Kang S and
Voorhees JJ. (1997). Engl. J. Med., 337, 1419 ± 1428.
Frank S, Madlener M and Werner S. (1996). J. Biol. Chem.,
271, 10188 ± 10193.
Fusenig NE. (1994). In: The Keratinocyte Handbook, Leigh,
B Lane and F Watt, eds. (Cambridge: Cambridge
University Press), pp. 71 ± 94.
Gandarillas A and Watt FM. (1995). Oncogene, 11, 1403 ±
1407.
Gillardon F, Eschenfelder C, Uhlmann E, Hartschuh W and
Zimmermann M. (1994). Oncogene, 9, 3219 ± 3225.
Greenhalgh DA, Rothnagel JA, Wang XJ, Quintanilla MI,
Orengo CC, Gagne TA, Bundman DS, Longley MA,
Fisher C and Roop DR. (1993). Oncogene, 8, 2145 ± 2157.
Greenhalgh DA and Yuspa SH. (1988). Mol. Carcinog., 1,
134 ± 143.
Hanahan D and Weinberg RA. (2000). Cell, 100, 57 ± 70.
Hatamochi A, Arakawa M, Mori K, Mori Y, Ueki H and
Yoshioka H. (1996). Dermatol. Sci., 11, 97 ± 103.
Hu L and Gudas LJ. (1994). J. Biol. Chem., 269, 183 ± 191.
Huang C, Li J, Chen N, Ma W, Bowden GT and Dong Z.
(2000). Mol. Carcinog., 27, 65 ± 75.
Huang C, Ma WY, Dawson MI, Rincon M, Flavell RA and
Dong Z. (1997). Proc. Natl. Acad. Sci. USA, 94, 5826 ±
5830.
Jaakkola P, Maatta A and Jalkanen M. (1998). Oncogene,
17, 1279 ± 1286.
Jang SI, Steinert PM and Markova NG. (1996). J. Biol.
Chem., 271, 24105 ± 24114.
Joselo€ E and Bowden GT. (1997). Mol. Carcinog., 18, 26 ±
36.
Kahari VM and Saarialho-Kere U. (1997). Exp. Dermatol.,
6, 199 ± 213.
Karin M, Liu Z and Zandi E. (1997). Curr. Opin. Cell Biol., 9,
240 ± 246.
Konicek BW, Xia X, Rajavashisth T and Harrington MA.
(1998). DNA Cell Biol., 17, 799 ± 809.
Kramer B, Wiegmann K and Kronke M. (1995). J. Biol.
Chem., 270, 6577 ± 6583.
Lever WF. (1983). In: Histopathology of the skin, 6th edition,
Lever WF and Schaumberger-Lever G (eds). (Philadelphia: JB Lippincott), pp 139 ± 622.
Lohman FP, Gibbs S, Fischer DF, Borgstein AM, van de
Putte P and Backendorf C. (1997). Oncogene, 14, 1623 ±
1627.
Lopez-Bayghen E, Vega A, Cadena A, Granados SE, Jave
LF, Gariglio P and Alvarez-Salas LM. (1996). J. Biol.
Chem., 271, 512 ± 520.
Lu B, Rothnagel JA, Longley MA, Tsai SY and Roop DR.
(1994). J. Biol. Chem., 269, 7443 ± 7449.
Lund LR, Romer J, Bugge TH, Nielsen BS, Frandsen TL,
Degen JL, Stephens RW and Dano K. (1999). EMBO J.,
18, 4645 ± 4656.
Madlener M, Parks WC and Werner S. (1998). Exp. Cell
Res., 242, 201 ± 210.
Martin P. (1997). Science, 276, 75 ± 81.
Masuda ES, Tokumitsu H, Tsuboi A, Shlomai J, Hung P,
Arai K and Arai N. (1993). Mol. Cell. Biol., 13, 7399 ±
7407.
Matsui T. (1996). Biochem. Biophys. Res. Commun., 220,
405 ± 410.
Mauviel A, Korang K, Santra M, Tewari D, Uitto J and
Iozzo RV. (1996), Biol. Chem., 271, 24824 ± 24829.
Mettouchi A, Cabon F, Montreau N, Dejong V, Vernier P,
Gherzi R, Mercier G and Binetruy B. (1997). Mol. Cell
Biol., 17, 3202 ± 3209.
Monzon RI, LaPres JJ and Hudson LG. (1996). Cell Growth
Di€er., 7, 1751 ± 1759.
Muegge K, Williams TM, Kant J, Karin M, Chiu R, Schmidt
A, Siebenlist U, Young HA and Durum SK. (1989).
Science, 246, 249 ± 251.
Munz B, Smola H, Engelhardt F, Bleuel K, Brauchle M, Lein
I, Evans LW, Huylebroeck D, Balling R and Werner S.
(1999). EMBO J., 18, 5205 ± 5215.
Navarro JM, Casatorres J and Jorcano JL. (1995). J. Biol.
Chem., 270, 21362 ± 21367.
Nead MA, Baglia LA, Antinore MJ, Ludlow JW and
McCance DJ. (1998). EMBO J., 17, 2342 ± 2352.
Ng DC, Shafaee S, Lee D and Bikle DD. (2000). J. Biol.
Chem., 275, 24080 ± 24088.
Nishida K, Kitazawa R, Mizuno K, Maeda S and Kitazawa
S. (1997). Biochem. Biophys. Res. Commun., 241, 258 ±
263.
Olsen J, Lefebvre O, Fritsch C, Troelsen JT, Orian-Rousseau
V, Kedinger M and Simon-Assmann P. (2000). Biochem.
J., 347, 407 ± 417.
Philips N, Bashey RI and Jimenez SA. (1995). J. Biol. Chem.,
270, 9313 ± 9321.
Prott J, Crabtree G, Grove M, Daubersies P, Bailleul B,
Wright E and Plumb M. (1995). Gene., 152, 173 ± 179.
Radoja N, Komine M, Jho SH, Blumenberg M and TomicCanic M, (2000). Mol. Cell Biol., 20, 4328 ± 4339.
Ramaswamy NT, Ronai Z and Pelling JC. (1998). Oncogene,
16, 1501 ± 1505.
Rhodes K and Oshima RG. (1998). J. Biol. Chem., 273,
26534 ± 26542.
Roebuck KA. (1999). J. Interferon Cytokine Res., 19, 429 ±
438.
Rubin JS, Osada H, Finch PW, Taylor WG, Rudiko€ S and
Aaronson SA. (1989). Proc. Natl. Acad. Sci. USA, 86,
802 ± 806.
Rutberg SE, Goldstein IM, Yang YM, Stackpole CW and
Ronai Z. (1994). Mol. Carcinog., 10, 82 ± 87.
Rutberg SE, Saez E, Glick A, Dlugosz AA, Spiegelman BM
and Yuspa SH. (1996). Oncogene, 13, 167 ± 176.
Rutberg SE, Saez E, Lo S, Jang SI, Markova N, Spiegelman
BM and Yuspa SH. (1997). Oncogene, 15, 1337 ± 1346.
AP-1 in skin biology
P Angel et al
Saez E, Rutberg SE, Mueller E, Oppenheim H, Smoluk J,
Yuspa SH and Spiegelman BM. (1995). Cell, 82, 721 ± 732.
Saiag P, Coulomb B, Lebreton C, Bell E and Dubertret L.
(1985). Science, 230, 669 ± 672.
Saito H, Kouhara H, Kasayama S, Kishimoto T and Sato B.
(1992). Biochem. Biophys. Res. Commun., 183, 688 ± 693.
Sark MW, Fischer DF, de Meijer E, van de Putte P and
Backendorf C. (1998). J. Biol. Chem., 273, 24683 ± 24692.
Sato H, Suzuki JS, Tanaka M, Ogiso M, Tohyama C and
Kobayashi S. (1997). Photochem. Photobiol., 65, 908 ± 914.
SchroÈder JM. (1995). J. Invest. Dermatol., 105, 20S ± 24S.
Schuh AC, Keating SJ, Monteclaro FS, Vogt PK and
Breitman ML. (1990). Nature, 346, 756 ± 760.
Shalaby F, Schuh AC and Breitman ML. (1994). Oncogene,
9, 2579 ± 2588.
Shino Y, Shirasawa H, Kinoshita T and Simizu B. (1997).
J. Virol., 71, 4310 ± 4318.
Sibilia M, Fleischmann A, Behrens A, Stingl L, Carroll J,
Watt FM, Schlessinger J and Wagner EF. (2000). Cell,
102, 211 ± 220.
Singer AJ and Clark RA. (1999). N. Engl. J. Med., 341, 738 ±
746.
Sinha S, Degenstein L, Copenhaver C and Fuchs E. (2000).
Mol. Cell. Biol., 20, 2543 ± 2555.
Smith Jr MF, Eidlen D, Brewer MT, Eisenberg SP, Arend
WP and Gutierrez-Hartmann A. (1992). J. Immunol., 149,
2000 ± 2007.
Smola H, Stark HJ, Thiekotter G, Mirancea N, Krieg T and
Fusenig NE. (1998). Exp. Cell. Res., 239, 399 ± 410.
Soriani M, Hejmadi V and Tyrrell RM. (2000). Photochem.
Photobiol., 71, 551 ± 558.
Sosnowski RG, Feldman S and Feramisco JR. (1993).
J. Cell. Biol., 121, 113 ± 119.
Soto U, Das BC, Lengert M, Finzer P, zur Hausen H and
Rosl F. (1999). Oncogene, 18, 3187 ± 3198.
Stefanidou MP, Ionnidou DJ, Panayiotides JG and Tosca
AD. (1999). Br. J. Dermatol., 141, 1040 ± 1045.
Szabowski A, Maas-Szabowski N, Andrecht S, Kolbus A,
Schorpp-Kistner M, Fusenig NE and Angel P. (2000).
Cell, 103, 745 ± 755.
Takahashi H, Asano K, Manabe A, Kinouchi M, IshidaYamamoto A and Iizuka H. (1998). J. Invest. Dermatol.,
110, 218 ± 223.
Teofoli P, Bardaugni S, Ribu€o M, Campanella A, De Pita O
and Puddu P. (1999). J. Dermatol. Sci., 22, 31 ± 37.
Thierry F, Spyrou G, Yaniv M and Howley P. (1992).
J. Virol., 66, 3740 ± 3748.
Thomas RS, Tymms MJ, McKinlay LH, Shannon MF, Seth
A and Kola I. (1997). Oncogene, 14, 2845 ± 2855.
Tuckermann JP, Reichardt HM, Arribas R, Richter KH,
Schutz G and Angel P. (1999). J. Cell. Biol., 147, 1365 ±
1370.
Unemoria EN, Mauch C, Hoe‚er W, Kim Y, Amento EP
and Bauer EA. (1994). Exp. Cell. Res., 211, 212 ± 218.
Virolle T, Monthouel MN, Djabari Z, Ortonne JP,
Meneguzzi G and Aberdam D. (1998). J. Biol. Chem.,
273, 17318 ± 17325.
Vu TH and Werb Z. (2000). Genes Dev., 14, 2123 ± 2133.
Wang CY, Bassuk AG, Boise LH, Thompson CB, Bravo R
and Leiden JM. (1994). Mol. Cell. Biol., 14, 1153 ± 1159.
Watts RG, Ben-Ari ET, Bernstein LR, Birrer MJ, Winterstein D, Wendel E and Colburn NH. (1995). Mo.
Carcinog., 13, 27 ± 36.
Watts RG, Huang C, Young MR, Li JJ, Dong Z, Pennie WD
and Colburn NH. (1998). Oncogene, 17, 3493 ± 3498.
Welter JF, Crish JF, Agarwal C and Eckert RL. (1995).
J. Biol. Chem., 270, 12614 ± 12622.
Welter JF and Eckert RL. (1995). Oncogene, 11, 2681 ± 2687.
Werner S, Peters KG, Longaker MT, Fuller-Pace F, Banda
MJ and Williams LT. (1992). Proc. Natl. Acad. Sci. USA,
89, 6896 ± 6900.
Wilkinson DG, Bhatt S, Ryseck RP and Bravo R. (1989).
Development, 106, 465 ± 471.
Xing Z, Gauldie J, Tremblay GM, Hewlett BR and Addison
C. (1997). Lab. Invest., 77, 615 ± 622.
Ye J, Zhang X and Dong Z. (1996). Mol. Cell. Biol., 16, 157 ±
167.
Young MR, Li JJ, Rincon M, Flavell RA, Sathyanarayana
BK, Hunziker R and Colburn N. (1999). Proc. Natl. Acad.
Sci. USA, 96, 9827 ± 9832.
Yuspa SH. (1998). J. Dermatol. Sci., 17, 1 ± 7.
Zagariya A, Mungre S, Lovis R, Birrer M, Ness S,
Thimmapaya B and Pope R. (1998). Mol. Cell. Biol., 18,
2815 ± 2824.
Zhao W, Chow LT and Broker TR. (1997). J. Virol., 71,
8832 ± 8840.
Ziober BL and Kramer RH. (1996). J. Biol. Chem., 271,
22915 ± 22922.
Zoumpourlis V, Papassava P, Linardopoulos S, Gillespie D,
Balmain A and Pintzas A. (2000). Oncogene, 19, 4011 ±
4021.
Zutter MM, Santoro SA, Painter AS, Tsung YL and Ga€ord
A. (1994). J. Biol. Chem., 269, 463 ± 469.
2423
Oncogene