10970215, 2006, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ijc.21852 by Readcube (Labtiva Inc.), Wiley Online Library on [07/12/2022].
See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Int. J. Cancer: 119, 1513–1518 (2006)
' 2006 Wiley-Liss, Inc.
MINI REVIEW
Understanding urothelial carcinoma through cancer pathways
Wolfgang A. Schulz*
Department of Urology, Heinrich-Heine-University D€
usseldorf, Germany
Urothelial carcinoma (UC), the common histological subtype of
bladder cancer, presents as a papillary tumor or as an invasive, of-
ten lethal form. To study UC molecular biology, candidate gene
and genome-wide approaches have been followed. Here, it is argued
that a Ôcancer pathwayÕ perspective is useful to integrate findings
from both approaches. According to this view, papillary cancers
typically exhibit activation of the MAPK pathway, as a conse-
quence of oncogenic mutations in FGFR3 or HRAS, with increased
Cyclin D1 expression. In contrast, invasive UC are characterized by
severe disturbances in proximate cell cycle regulators, e.g. RB1 and
CDKN2A/p16INK4A, which decrease dependency on mitogenic sig-
naling. In addition, these disturbances permit, promote and are in
turn exacerbated by chromosomal instability, which is further en-
hanced by loss of TP53 function. In another vicious cycle, defective
cell cycle regulation interacts with DNA methylation alterations.
The transition toward invasive UC may require concomitant and
interacting defects in cell cycle regulation and the control of
genomic stability. Intriguingly, neither canonical WNT/b-Catenin
nor hedgehog signaling appear to play major roles in UC. This
may reflect its origin from more differentiated urothelial cells
possessing a high regenerative potential rather than a stem cell
population.
' 2006 Wiley-Liss, Inc.
Key words: cell cycle regulation; TP53; DNA
chromosomal instability; cancer stem cells
Bladder cancer is the fifth most frequent cancer in industrialized
countries, accounting for up to 5% of all cancers. Histologically,
urothelial carcinoma (UC), formerly designated Ôtransitional cell
carcinoma,Õ is distinguished from rarer types like squamous cell
carcinoma (SCC) and adenocarcinoma. UC are derived from the
urothelium, the epithelium lining the urinary tract from the renal
pelvis to the urethra. Accordingly, some cases occur outside the
bladder in other segments of the urinary tract. Urothelial carcino-
mas express markers of urothelial differentiation, including CK7
and CK13 cytokeratins, and even uroplakins, proteins of the apical
membrane of terminally differentiated umbrella cells in the top
urothelial layer. Clinically, the most important distinction is that
between papillary and invasive UC. Papillary UC (pTa) are char-
acterized by hyperproliferation leading to lateral and vertical ex-
tension of the urothelial layer. As a consequence, the epithelium
folds up around connective tissue papillae into a polypous struc-
ture protruding into the lumen of the urinary tract. Papillary UC
can be treated by local transurethral resection (TUR). Unfortu-
nately, the disease tends to recur in up to 75% of the patients,
necessitating long-term monitoring by cystoscopy and repeated
treatments. Recurrences can be partly prevented by optimized re-
section and by adjuvant intravesical instillation therapy with cyto-
toxic drugs or with BCG, a mycobacterium vaccine preparation.
Thus, while papillary UC is seldom lethal, it causes considerable
suffering and high expenses. Moreover, a small but significant per-
centage of papillary UC, usually characterized by poor differentia-
tion, progress to invasive stages. Invasive bladder cancers are
mostly derived from another precursor, carcinoma in situ (pTIS),
flat highly dysplastic lesions. They are lethal by local progression
or metastasis, unless completely removed by cystectomy at early
invasive stages. Chemo- and radiotherapy are efficacious, but
rarely curative. In the older literature, papillary tumors of various
grades, carcinoma in situ, and carcinomas with invasion of the
lamina propria only (pT1), were often summarized as Ôsuperficial.Õ
Publication of the International Union Against Cancer
This designation is now obsolete, as they are recognized to exhibit
considerable clinical and molecular differences.
Molecular biological research on bladder cancer is challenged
to yield advances for its prevention, diagnostics and therapy. In
fact, several ideas referred in this article have already influenced
the latest WHO classification of urothelial cancers and thus con-
tributed to practical improvements.1 In the future, most urgently
needed are (1) markers identifying those patients with papillary
and early stage invasive tumors at high risk for recurrence and
progression, (2) reliable and affordable non-invasive techniques
for monitoring, (3) markers distinguishing between metastatic and
localized invasive disease and (4) more efficacious therapies for
patients with locally progressed and metastatic cancer.
Approaches to the molecular biology of UC
Two major approaches have been followed to analyze UC mo-
lecular biology: candidate gene analysis and genome-wide screens.
In 1981, the first human oncogene, a mutated HRAS, was isolated
from UC cell lines.2 Since then, almost every newly emerging
oncogene or tumor suppressor was investigated for a role in UC.
methylation; Several tumor suppressors were indeed found to be frequently af-
fected, including TP53, RB1 and CDKN2A. Several oncogenes were
observed to be activated in a fraction of cases, including HRAS, MYC
and CCND1.3,4 More recently, FGFR35–7 and E2F38,9 were identified
as oncogenes activated by point mutations and amplification, respec-
tively.
Familial cases of bladder cancer are very rare. In the case of an
exceptionally young index patient, an inherited translocation led
to the identification of a novel oncogene, CDC91L1 at 20q.10 Its
significance in sporadic cases remains to be ascertained.11 In gen-
eral, identification of genes important in bladder cancer must rely
on analyses of somatic alterations in tumor tissues and cell lines.
Identification of tumor suppressors is usually performed through
delineation of common regions of deletion followed by mutational
and methylation analysis of candidate genes. Genome-wide
screens using LOH analyses or cytogenetical techniques revealed
several consistently altered chromosomal regions in bladder can-
cer, but also a high background of chromosomal instability, with
essentially each chromosome affected in some cases.12,13 Some
recurrent chromosomal gains or losses reflect alterations in known
oncogenes or tumor suppressors. Losses at 9p21, 13q14 and 17p
reflect the established inactivation of CDKN2A, RB1 and TP53,
respectively. Some targets of less common losses can also be
assigned, e.g., 10q losses may usually reflect PTEN inactivation.
Gains at 7p12, 8q24 and 11q13 can result in overexpression of
EGFR, MYC and Cyclin D1, respectively.3,4 On a note of caution,
additional or alternative genes may be targeted by losses and am-
plifications in some cases. Recurrent amplifications at 6p22.3 in
advanced stage cancers are even more difficult to interpret because
Grant sponsor: Deutsche Forschungsgemeinschaft; Grant number: LI
1038/3-1.
*Correspondence to: Urologische Klinik, Heinrich-Heine-Universit€at,
Moorenstr. 5, 40225 D€usseldorf, Germany. Fax: 149-211-81-15846.
E-mail: wolfgang.schulz@uni-duesseldorf.de
Received 15 September 2005; Accepted 23 December 2005
DOI 10.1002/ijc.21852
Published online 23 March 2006 in Wiley InterScience (www.interscience.
wiley.com).

1514 SCHULZ
they are highly heterogeneous. Their most likely target is E2F3,
but several other genes in the region are over-expressed to a com-
parable or greater extent in individual cases.8,9,14,15 This example
illustrates the limitations to identifying targets of chromosomal
aberrations in cancers with pronounced genomic instability. Con-
sequently, the significance of many chromosomal gains in UC is
uncertain.
This predicament extends to chromosome losses, even highly
prevalent ones at 3p and 8p. The most exasperating problem con-
cerns chromosome 9q. Loss of (or LOH at) chromosome 9 is the
most frequent chromosomal alteration in bladder cancer through-
out all stages and grades, including well-differentiated papillary
cancers and occasionally preneoplastic hyperplasia.3,4,12,13 On 9p,
loss and recombination are centered around CDKN2A at 9p21.
This gene encodes p16INK4A, a CDK inhibitor and thereby activa-
tor of RB1, and p14ARF, an indirect activator of TP53, in different
reading frames. In UC, this general tumor suppressor is most fre-
quently inactivated by homozygous deletion.16,17 Point mutations,
typically inactivating both reading frames, and hypermethylation
of either or both alternative promoters contribute in a smaller frac-
tion of the cases. While often the entire chromosome is affected,
allelic losses at 9q occur also independently of changes at 9p. One
or several tumor suppressors are therefore assumed to be located
on 9q. Candidates comprise PTCH1 and TSC1 already implicated
in other cancers, and the novel, cautiously named DBCCR1
(deleted in bladder cancer candidate region 1). Mutations inacti-
vating the second allele, meeting standard expectations for tumor
suppressors, have been detected only in TSC1, at a low fre-
quency.18,19 Inherited TSC1 mutations predispose to several tumor
types, but not obviously to bladder cancer. Methylation of DBCCR1
increases in UC, but does not seem consistently associated with
gene silencing.20 Nevertheless, DBCCR1 overexpression inhibits
cell proliferation.21 The mechanism involved and the effects of
physiological expression levels remain to be elucidated.
PTCH1 is a tumor suppressor in the (sonic) hedgehog pathway
and is often inactivated in basal cell carcinoma of the skin and in
medulloblastoma.22 These cancers are generally characterized by
activation of the hedgehog pathway, which can be alternatively
brought about by oncogenic mutations in SMO, a membrane sig-
naling protein normally inhibited by PTCH1. Activation of intra-
cellular hedgehog signaling by either alteration results in increased
expression of several pathway components including the transcription
activators GLI1 and GLI2, and the inhibitory membrane proteins
HIP1 and PTCH1, through the increased activity of GLI-dependent
promoters. No evidence of such a response was found in UC cell
lines.23 In cancer tissues, PTCH1 expression is approximately halved
in cases with 9q loss in accord with a pure dosage effect. PTCH1
mutations are extremely rare in UC and none have been reported in
other pathway components.24 Taken together, the evidence indicates
that PTCH1 hemizygosity is not pathogenetic in UC.
Increasingly, global approaches are being used to investigate mo-
lecular changes in UC. These include array CGH and SNP arrays
at the DNA level25,26 microarray expression profiling at the RNA
level,27–30 and proteomics approaches.31 Early studies addressed dif-
ferences among cell lines, cancer versus normal tissues and papillary
versus invasive cancers. More recently, more sophisticated issues
have been tackled, such as response to therapy or differential progno-
sis in cases with similar histopathological parameters, e.g. the clini-
cally problematic pTaG3 tumors. Reassuringly, cluster analyses of
expression profiles separate most cases of papillary from muscle-
invasive cancers and cancers from normal tissue, with interesting
exceptions that deserve further study.
Unfortunately, microarray expression studies have yielded few
overlapping results with regard to the expression of individual
genes, which makes it difficult to draw any conclusions on molec-
ular mechanisms in UC. Some of the inconsistencies may derive
from technical reasons and can presumably be resolved by im-
proved standardization of arrays and of sample preparation, e.g.
through microdissection. Furthermore, a significant proportion of
10970215, 2006, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ijc.21852 by Readcube (Labtiva Inc.), Wiley Online Library on [07/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
the variability may have biological causes. Rampant chromosomal
instability in UC ought to lead to variable dose changes and
according differences in the expression of many genes, independ-
ently of whether they are causally involved in cancer development.
For instance, individual bladder cancer cell lines over-express dif-
ferent genes from the 6p22.3 amplification unit up to a 100 fold.15
Moreover, gene expression profiling tends to identify those genes
reacting most strongly to primary genetic or epigenetic alterations
and not the causative change itself. Obviously, in a heterogeneous
cancer like UC, data interpretation needs to be anchored in a con-
ceptual framework to gain the most from genome-wide analyses.
TP53 as a biomarker in UC
An important goal of array investigations of UC is the develop-
ment of biomarkers for prognostic purposes. This aim has so also
been pursued by investigations focused on individual genes, the
most prominent one being TP53.32–34 The TP53 tumor suppressor
gene harbors missense mutations in up to 50% of bladder cancers.
The frequency of mutations increases with tumor stage and partic-
ularly with tumor grade. Specifically, their frequency is enhanced
in early stage cancers considered as high-risk from clinical experi-
ence, such as carcinoma in situ and poorly differentiated papillary
and early stage invasive tumors (pTaG3 and pT1G3). Moreover,
LOH at 17p is preferentially found in advanced cases. Evidently,
TP53 mutations are associated with an increased risk of progres-
sion in bladder cancer.3,4,32–34 The moot point is whether detection
of TP53 mutations can predict the natural course of the disease or
responses to therapy better than histopathological parameters. This
issue, too, is confounded by technical as well as biological factors.
Many studies have used techniques incapable of detecting all
TP53 mutations, e.g. single-strand conformation polymorphism
analysis plus sequencing starting from total tumor tissue. More-
over, accumulation of TP53 protein has often been used as a surro-
gate parameter for mutation. The correlation between mutations
and nuclear accumulation of TP53 is very good in bladder cancer,
but not perfect. In addition, TP53 immunohistochemistry poses its
own vagaries, although standardized techniques have meanwhile
been developed.32–34
These technical difficulties are exacerbated by the fact that
TP53 acts within a molecular network. Loss of TP53 function can
therefore not only be caused by mutations and deletions of the
gene itself, but also by alterations in the mechanisms acting
ÔupstreamÕ or ÔdownstreamÕ of TP53 in the network. Such altera-
tions may not have as severe effects as loss of TP53 itself, but
may suffice to impede crucial functions of the nodal protein in the
network. ÔUpstreamÕ alterations in the network diminish TP53 acti-
vation and include loss of p14ARF (encoded by CDKN2A) in a
large fraction of cases and amplification of HDM2 in a few.3,4,34
Defects in the function of the protein kinases ATM, CHK2 and
DNA-PK, which signal DNA damage to TP53, are implicated
too.35 Activation of these kinases in pTa and pT1 tumors may pre-
vent or delay their progression, but subsides in high-stage cancers
in spite of increased chromosomal instability. ÔDownstreamÕ alter-
ations diminish the efficacy of TP53 action through cell cycle
inhibitors like p21CIP1 (see later) and pro-apoptotic proteins.
Arguments for a ‘‘cancer pathway’’ approach to UC
The example of TP53, like that of PTCH1 and the hedgehog
pathway discussed earlier, illustrates how productive it is to con-
sider mutational and epigenetic gene alterations in the context of
relevant networks or pathways. This approach may be particularly
useful in a cancer with pronounced genomic instability such as
UC. Hanahan and Weinberg36 have postulated that the distinctive
properties of cancers are brought about by the activation or inacti-
vation of a limited number of regulatory systems termed Ôcancer
pathways.Õ In addition to the TP53 network and the hedgehog
pathway, these include the cell cycle regulatory system and the
(canonical) MAPK pathway discussed later, the PI3K, STAT and
 10970215, 2006, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ijc.21852 by Readcube (Labtiva Inc.), Wiley Online Library on [07/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
UNDERSTANDING UROTHELIAL CARCINOMA THROUGH CANCER PATHWAYS 1515
NFjB pathways and the TGFb response not discussed here
because of space limitations, and the WNT/b-Catenin pathway,
which can serve as another example for the value of this approach.
Constitutive activation of this pathway is the ÔgatekeeperÕ
change in several cancers, notably colorectal carcinoma. Activa-
tion is caused by loss of function of the tumor suppressors APC or
AXIN1, or by oncogenic mutations in b-Catenin. Either results in
the increased activity of promoters regulated by TCF4/b-Cate-
nin.22 The few published studies on this pathway in UC report no
or very few genetic alterations in pathway components.37,38 In-
deed, promoter-reporter constructs monitoring TCF/b-Catenin de-
pendent transcription were found to be inactive in all UC lines
tested and in normal urothelial cells, indicating lack of pathway
activity.39 Furthermore, in most UC lines and in normal cells, tran-
scription could not even be induced by overexpressed oncogenic
b-Catenin, suggesting that it is actively repressed. Interestingly,
exceptional inducible cell lines lacked expression of E-Cadherin,
a modulator of the pathway.39 Thus, some bladder cancers may ac-
quire the ability to respond to strong WNT signaling through loss
of E-Cadherin, but constitutive activation is apparently very rare.
Importantly, many cancer pathways interact with each other and
the effects of further regulators, e.g. cell adhesion molecules and
extracellular proteases, are partly mediated through these pathways.
The level of regulatory pathways may thus be optimal to integrate
results from single candidate gene analyses and large-scale ge-
nomics approaches. Not least, the identification of relevant cancer
pathways yields immediate options for pharmacological therapy.

Cell cycle regulation as a ‘‘cancer pathway’’

Many cancer pathways converge towards cell cycle regulation,
a regulatory network in itself (Fig. 1). Since enhanced cell prolif-
eration is a property of human cancers by definition, altered cell
cycle regulation seems a prerequisite for cancer development.
However, cancer types differ in the mechanisms that lead to
altered cell cycle regulation and it its degree. In bladder cancer,
genetic and epigenetic alterations of proximate regulators of the
cell cycle abound.3,4 Deletion and mutational inactivation of RB1
is found in a significant fraction of invasive cancers.33,40 A large
fraction of cancers of all stages harbor defects in CDKN2A, with
loss of the CDK4 inhibitor p16INK4A.16,17 In exceptional cases
CDK4 is overexpressed due to amplification of the gene at 12q1341
or mutations interfere with p16INK4A binding. In contrast, overex-
pression of Cyclin D1 is widely found. Overexpression of MYC
could also be counted as a cell-cycle related alteration. Like overex-
pression of Cyclin D1, it is caused by chromosomal gains or gene
amplification only in a small number of tumors, but is more regu-
larly a consequence of deregulation.37,42 The fact that overexpres-
sion of MYC is caused by amplification in some cases and by dereg-
ulation in others could explain why overexpression is not entirely
consistently associated with tumor stage and grade. These altera-
tions are compounded by additional changes in the CIP/KIP family
of CDK inhibitors. The p21CIP1 and p27KIP1 inhibitory proteins are
down-regulated in the majority of invasive cancers18,33,43,44 and ex-
pression of p57KIP2 is diminished or ablated by a combination of FIGURE 1 – Signaling pathways converging on cell cycle regulation
genetic and epigenetic mechanisms.45 in normal urothelial cells (a), typical papillary superficial UC (b) and
While disturbances of cell cycle regulation are doubtless crucial typical invasive UC (c). Inactive pathways: white, properly regulated
active pathways: plain grey; deregulated pathways or proteins: patchy
in all UC, closer analyses reveal interesting variations. Alterations grey; aberrantly inactivated pathways or proteins: hatched.
of RB1 are almost exclusively observed in muscle-invasive can-
cers and are associated with a significantly worse prognosis.33,39
In contrast, Cyclin D1 overexpression is characteristic of early There is thus convincing evidence that ÔprimaryÕ alterations in
stage and papillary cancers and rather associated with less malig- cell cycle regulators such as inactivation of RB1 and p16INK4A,
nant behavior.3,4 CDKN2A alterations are overall independent of usually by genetic mechanisms, together with down-regulation of
stage, grade and clinical course.16,17 Obviously, the degree of dis- CIP/KIP CDK inhibitors, predominantly by epigenetic mecha-
turbance of cell cycle regulation parallels the biological and clini- nisms, are crucial changes in invasive UC. It is not quite as clear
cal aggressiveness of bladder cancers. In line with this supposi- whether this conclusion extends to papillary cancers. A large pro-
tion, losses of p21CIP1, p27KIP1, and p57KIP2 are more frequent in portion of well-differentiated papillary cancers harbor oncogenic
muscle-invasive tumors,18,33,43–45 although their independent prog- mutations in FGFR3 which encodes a receptor for fibroblast
nostic value is debated. In contrast, many papillary tumors over- growth factors like FGF1 and FGF8. As a rule, tumors exhibiting
express p21CIP1. this change show fewer recurrences and their proportion decreases

1516 SCHULZ
5–7,46
strongly at higher stages. FGFR3 mutations are inversely
associated with TP53 mutations, but positively with Cyclin D1
overexpression.46 Overactivity of FGFR3 would be expected to
lead to activation of the MAPK pathway through RAS and RAF
and hence to induction of Cyclin D1 and cell cycle activation.
Indeed, HRAS mutations were reported to occur in a complemen-
tary pattern to those in FGFR3 suggesting they belong to the same
pathway.47 Thus, the growth of typical papillary cancers could be
driven mainly by activation of the MAPK pathway, either through
FGFR3 mutations, HRAS mutations, or through activation of
other tyrosine kinase receptors, e.g. EGFR (Fig. 1).
The proliferation of normal urothelium, e.g. in response to tissue
damage, is likewise thought to be stimulated by growth factors act-
ing on tyrosine kinase receptors, in particular by EGF family mem-
bers acting in an autocrine or paracrine fashion.48 These mecha-
nisms can be studied in primary cultures of urothelial cells which
proliferate spontaneously when placed in appropriate serum-free
culture media. Their proliferation is elicited by autocrine growth
factors such as HB-EGF and further enhanced by addition of exoge-
neous EGF. Inhibition of MAPK signaling decreases proliferation
of normal urothelial cells, but cell lines from invasive UC are much
less sensitive to MAP kinase inhibitors.49 Accordingly, while phos-
phorylated ERK and MEK can be detected in some UC lines, their
transcriptional response to MAPK activation is muted.49,50 These
findings suggest several interesting conclusions. Firstly, the more
massive alterations in cell cycle regulators found in invasive bladder
cancers apparently diminish the dependence on exogenous growth
factors and MAPK signaling, at least with regard to cell prolifera-
tion. Secondly, MAPK signaling is known to not only stimulate cell
proliferation, but also cell differentiation and to induce inhibitors
such as p21CIP1 which can induce cell cycle arrest and senescence.
Therefore, activation of the pathway in papillary cancers could lead
to a self-limiting tumor growth. This would explain the generally
lower malignancy of papillary cancers with FGFR3 mutations. Con-
tinuous tumor growth would require additional alterations that pre-
vent terminal differentiation and replicative senescence induced
eventually by activation of G1 checkpoints through TP53 and RB1.
Alterations compromising these networks in UC may therefore also
prevent replicative senescence. Specifically, complete or even par-
tial inactivation of CDKN2A also common in pTa cases17 may com-
plement activation of the MAPK pathway. Interestingly, mutant
RAS proteins may more efficiently induce premature senescence
through a different pathway dependent on p38MAPK.51 Tumors with
mutant HRAS could therefore behave differently from those with
FGFR mutations, in spite of similar effects on canonical MAPK
pathway activity. Thirdly and importantly, papillary and invasive
UC may respond quite differently to therapeutic MAPK pathway in-
hibition, with additional differences within each group.
The notion that papillary UC are typically characterized by
MAPK activation, and invasive UC by defects in the TP53 and
RB1 networks is corroborated by animal models. Transgenic mice
that overexpress a mutated Ha-Ras in urothelial cells develop uro-
thelial hyperplasia and papillary UC,52 whereas animals express-
ing the SV40 large-T antigen, which inactivates Rb1 and Tp53,
present with carcinoma in situ or invasive cancers.53
Consequences of cell cycle disturbances in UC
The almost universal disturbance of the cell cycle regulatory
network in bladder cancers has consequences beyond increased
cell proliferation (Fig. 2). Hyperproliferative signals elicited by
activated oncogenes induce cellular senescence through CDKN2A.
Inactivation of this locus and down-regulation of other CDK in-
hibitors that can mediate cellular senescence, p21CIP1 and p57KIP2,
would thus allow bladder cancers to tolerate persistently high pro-
liferation signals, emerging from mutated FGFR3, mutated RAS,
over-expressed EGFR, or amplified MYC.
Proper cell cycle control is also required for the function of cel-
lular checkpoints responding to DNA damage and mitotic distur-
10970215, 2006, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ijc.21852 by Readcube (Labtiva Inc.), Wiley Online Library on [07/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
FIGURE 2 – Mutual interaction of defective cell cycle regulation
with mitogenic signaling, chromosomal stability and maintenance of
DNA methylation patterns in UC.
bances. Chromosomal instability in invasive bladder cancers indi-
cates a dysfunction of these checkpoints. The prevalent loss of
TP53 function in these carcinomas certainly contributes to this
dysfunction. In cancers retaining intact TP53, down-regulation of
p21CIP1 or loss of RB1 should compromise its ability to arrest the
cell cycle upon check point activation. In fact, wild-type TP53
transfected into UC lines elicits apoptosis rather than cell cycle
arrest.54 Accordingly, recent evidence suggests that disturbances
of checkpoint regulation are not due to TP53 loss only. The check-
point kinases ATM, CHK2 and CHK1 were observed in activated
states in papillary UC, but more weakly active in invasive can-
cers.35 Apparently, dysregulation of DNA replication caused by
oncogenic FGFR3, RAS or MYC or by defects in the RB1 cell
cycle regulatory network elicit activation of S-phase and G2/M
checkpoints. Overcoming these checkpoints through severe
defects in cell cycle regulation or inactivation of checkpoint effec-
tors is therefore a prerequisite for continued tumor growth and
progression. Indeed, checkpoint responses become increasingly
compromised as bladder cancers progress. While the underlying
mechanisms are incompletely understood, loss of RB1 in particu-
lar has been linked to chromosomal instability.40 This relationship
may have several reasons. Firstly, checkpoint signaling may not
elicit G1 arrest in cells lacking functional RB1. Secondly, defects
in replication and mitosis may occur more frequently during shorter
and less coordinated cell cycles. Thirdly, RB1 is implied directly in
the control of mitosis.40 In summary, therefore, the chromosomal
instability characteristic of invasive bladder cancers may partly rep-
resent a consequence of severe disturbances of cell cycle regulation.
Conversely, chromosomal instability causes genetic alterations that
exacerbate cell cycle deregulation (Fig. 2).
Another vicious cycle in UC links cell cycle deregulation to
alterations of DNA methylation, which promote tumor growth and
genomic instability (Fig. 2). As in other cancers, 2 kinds of meth-
ylation alterations are observed, hypermethylation at CpG-rich
promoter regions of selected genes and a genome-wide decrease
in overall methylation, which is most evident at retrotransposons
and satellites sequences densely methylated in normal cells. Hyper-
methylation events are more frequent in invasive than in papillary
UC,55 while hypomethylation is essentially universal in invasive
cancers.56 The mechanisms leading to these changes are complex
and far from understood, but clearly proper coordination between
the activity of DNA methyltransferases and DNA replication is cru-
cial for the maintenance of methylation patterns in normal cells.
Specifically, the major enzyme in somatic cells, DNMT1, is coordi-
nated with the cell cycle by RB1 via E2F transcription factors. Loss

UNDERSTANDING UROTHELIAL CARCINOMA THROUGH CANCER PATHWAYS
of proper RB1 function would be expected to disturb this coordi-
nation. Indeed, DNMT1 expression was found to become un-
coupled from increased cell proliferation in UC.57 Thus, while
DNMT1 expression even increases in some UC, the enzyme can-
not maintain normal methylation patterns. In turn, methylation
alterations affect cell cycle regulators. The CDK inhibitor p57KIP2,
e.g., is down-regulated during UC progression by promoter hyper-
methylation as well as hypomethylation of a more distant region
controlling its imprinting.45
Conclusion and perspective
In conclusion, 2 major ‘‘cancer pathways,’’ cell cycle regulation
and the TP53 network, appear to be affected in UC and the extent
of their respective deregulation in each tumor significantly deter-
mines its biological and clinical characteristics (Fig. 1). Activation
of the canonical MAPK pathway may characterize an important
subclass of papillary tumors.1,4,58 If so, this subclass may be more
responsive to appropriate drug treatment than UC in general. Intri-
guingly, neither hedgehog nor WNT/b-Catenin signaling appear
to play major parts in UC, in spite of their fundamental importance
in other carcinomas. Apparently, stimulation of cell proliferation
in UC is accomplished by other alterations, notably severe defects
in proximate cell cycle regulation. In addition, the lack of impor-
References
1. Eple JN, Sauter G, Epstein JI, Sesterhenn IA, eds. Pathology and
genetics: tumours of the urinary system and male genital organs.
Lyon: IARC Press, 2004. 359 p.
2. Parada LF, Tabin CJ, Shih C, Weinberg RA. Human EJ bladder carci-
noma oncogene is homologue of Harvey sarcoma virus ras gene.
Nature 1982;297:474–8.
3. Knowles MA. What we could do now: molecular pathology of blad-
der cancer. Mol Pathol 2001;54:215–21.
4. Dinney CP, McConkey DJ, Millikan RE, Wu X, Bar-Eli M, Adam L,
Kamat AM, Siefker-Radtke AO, Tuziak T, Sabichi AL, Grossman HB,
Benedict WF. Focus on bladder cancer. Cancer Cell 2004;6:111–6.
5. Cappellen D, De Oliveira C, Ricol D, de Medina S, Bourdin J, Sastre-
Garau X, Chopin D, Thiery JP, Radvanyi F. Frequent activating muta-
tions of FGFR3 in human bladder and cervix carcinomas. Nat Genet
1999;23:18–20.
6. Sibley K, Cuthbert-Heavens D, Knowles MA. Loss of heterozygosity
at 4p16.3 and mutation of FGFR3 in transitional cell carcinoma.
Oncogene 2001;20:686–91.
7. van Rhijn BW, Lurkin I, Radvanyi F, Kirkels WJ, van der Kwast TH,
Zwarthoff EC. The fibroblast growth factor receptor 3 (FGFR3) muta-
tion is a strong indicator of superficial bladder cancer with low recur-
rence rate. Cancer Res 2001;61:1265–8.
8. Feber A, Clark J, Goodwin G, Dodson AR, Smith PH, Fletcher A,
Edwards S, Flohr P, Falconer A, Roe T, Kovacs G, Dennis N.
Amplification and overexpression of E2F3 in human bladder cancer.
Oncogene 2004;23:1627–30.
9. Oeggerli M, Tomovska S, Schraml P, Calvano-Forte D, Schafroth S,
Simon R, Gasser T, Mihatsch MJ, Sauter G. E2F3 amplification and
overexpression is associated with invasive tumor growth and rapid
tumor cell proliferation in urinary bladder cancer. Oncogene 2004;23:
5616–23.
10. Guo Z, Linn JF, Wu G, Anzick SL, Eisenberger CF, Halachmi S,
Cohen Y, Fomenkov A, Hoque MO, Okami K, Steiner G, Engles JM,
et al. CDC91L1 is a newly discovered oncogene in human bladder
cancer. Nat Med 2004;10:374–81.
11. Schultz IJ, Kiemeney LA, Witjes JA, Schalken JA, Willems JL, Swin-
kels DW, de Kok JB. CDC91L1 (PIG-U) mRNA expression in urothe-
lial cell carcinomas. Int J Cancer 2005;116:282–4.
12. Knowles MA, Elder PA, Williamson M, Cairns JP, Shaw ME,
Law MG. Allelotype of human bladder cancer. Cancer Res 1994;54:
531–538.
13. Hovey RM, Chu L, Balazs M, DeVries S, Moore D, Sauter G, Carroll PR,
Waldman FM. Genetic alterations in primary bladder cancers and
their metastases. Cancer Res 1998;58:3555–60.
14. Bruch J, Schulz WA, Melzner I, Kemmerling R, Br€uderlein S, M€oller P,
Vogel W, Hameister H. Concurrent gain of chromosomes 5p, 6p, and
20q in bladder carcinoma cell lines: delineation of the 6p22 amplifica-
tion unit. Cancer Res 2000;60:4526–30.
15. Wu Q, Hoffmann MJ, Hartmann FH, Schulz WA. Amplification and
overexpression of the ID4 gene at 6p22.3 in bladder cancer. BMC
Mol Cancer 2005;4:16.
10970215, 2006, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ijc.21852 by Readcube (Labtiva Inc.), Wiley Online Library on [07/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
1517
tance of these pathways might relate to the urothelial origin of
UC. Urothelium is a quiescent tissue, but capable of rapid prolifer-
ation in response to injury. The proliferating cells come from the
basal and intermediate layers of the tissue. During acute injury,
stem cell recruitment does not seem to be required. Thus, the func-
tions of WNT and hedgehog signaling in maintaining stem cell
compartments may not be as important in normal urothelial tissue
as in continuously proliferating tissues like the skin or the gut epi-
thelium. Consequentially, UC may arise from more differentiated
cells, in which these pathways are suppressed. Indeed, UC usually
express proteins characteristic of differentiated urothelial cells.
According to this view, the property of unlimited growth, which
cancer cells share with stem cells, would be acquired secondarily
in UC, most likely by inactivation of TP53 and RB1 function.
Obviously, to even better understand bladder cancer, the cancer
pathway perspective sketched here should be widened to encom-
pass the context of urothelial tissue organization.
Acknowledgements
The author thanks Andrea Linnemann-Florl, Hans-Helge Seifert,
Sascha Raschke and Jiri Hatina for critical reading of the
manuscript.
16. Florl AR, Franke KH, Niederacher D, Gerharz CD, Seifert HH,
Schulz WA. DNA methylation and the mechanisms of CDKN2A inac-
tivation in transitional cell carcinoma of the urinary bladder. Lab
Invest 2000;80:1513–22.
17. Chapman EJ, Harnden P, Chambers P, Johnston C, Knowles MA.
Comprehensive analysis of CDKN2A status in microdissected urothelial
cell carcinoma reveals potential haploinsufficiency, a high frequency of
homozygous co-deletion and associations with clinical phenotype. Clin
Cancer Res 2005;11:5740–7.
18. Adachi H, Igawa M, Shiina H, Urakami S, Shigeno K, Hino O.
Human bladder tumors with 2-hit mutations of tumor suppressor gene
TSC1 and decreased expression of p27. J Urol 2003;170:601–4.
19. Knowles MA, Habuchi T, Kennedy W, Cuthbert-Heavens D. Muta-
tion spectrum of the 9q34 tuberous sclerosis gene TSC1 in transitional
cell carcinoma of the bladder. Cancer Res 2003;63:7652–6.
20. Habuchi T, Takahashi T, Kakinuma H, Wang L, Tsuchiya N, Satoh S,
Akao T, Sato K, Ogawa O, Knowles MA, Kato T. Hypermethylation
at 9q32-33 tumour suppressor region is age-related in normal urothe-
lium and an early and frequent alteration in bladder cancer. Oncogene
2001;20:531–7.
21. Wright KO, Messing EM, Reeder JE. DBCCR1 mediates death in cul-
tured bladder tumor cells. Oncogene 2004;23:82–90.
22. Beachy PA, Karhadkar SS, Berman DM. Tissue repair and stem cell
renewal in carcinogenesis. Nature 2004;432:324–31.
23. Thievessen I, Wolter M, Prior A, Seifert HH, Schulz WA. Hedgehog
signaling in normal urothelial cells and in urothelial carcinoma cell
lines. J Cell Physiol 2005;203:372–7.
24. Aboulkassim TO, LaRue H, Lemieux P, Rousseau F, Fradet Y. Alter-
ation of the PATCHED locus in superficial bladder cancer. Oncogene
2003;22:2967–71.
25. Veltman JA, Fridlyand J, Pejavar S, Olshen AB, Korkola JE, DeVries S,
Carroll P, Kuo WL, Pinkel D, Albertson D, Cordon-Cardo C, Jian AN.
Array-based comparative genomic hybridization for genome-wide
screening of DNA copy number in bladder tumors. Cancer Res 2003;63:
2872–80.
26. Primdahl H, Wikman FP, von der Maase H, Zhou XG, Wolf H,
Orntoft TF. Allelic imbalances in human bladder cancer: genome-wide
detection with high-density single-nucleotide polymorphism arrays.
J Natl Cancer Inst 2002;94:216–23.
27. Dyrskjot L, Thykjaer T, Kruhoffer M, Jensen JL, Marcussen N, Ham-
ilton-Dutoit S, Wolf H, Orntoft TF. Identifying distinct classes of
bladder carcinoma using microarrays. Nat Genet 2003;33:90–6.
28. Modlich O, Prisack HB, Pitschke G, Ramp U, Ackermann R, Bojar H,
V€ogeli TA, Grimm MO. Identifying superficial, muscle-invasive, and
metastasizing transitional cell carcinoma of the bladder: use of cDNA
array analysis of gene expression profiles. Clin Cancer Res 2004;10:
3410–21.
29. Blaveri E, Simko J, Korkola JE, Brewer JL, Baehner F, Mehta K,
deVries S, Koppie T, Pejavar S, Carroll P, Waldman FM. Bladder
cancer outcome and subtype classification by gene expression. Clin
Cancer Res 2005;11:4044–55.

1518 SCHULZ
30. Wild PJ, Herr A, Wissmann C, Stoehr R, Rosenthal A, Zaak D, Simon R,
Knuechel R, Pilarsky C, Hartmann A. Gene expression profiling of
progressive papillary noninvasive carcinomas of the urinary bladder.
Clin Cancer Res 2005;11:4415–29.
31. Celis JE, Gromova I, Moreira JM, Cabezon T, Gromov P. Impact of pro-
teomics on bladder cancer research. Pharmacogenomics 2004;5:381–94.
32. Helpap B, Schmitz-Drager BJ, Hamilton PW, Muzzonigro G, Galosi AB,
Kurth KH, Lubaroff D, Waters DJ, Droller MJ. Molecular pathology of
non-invasive urothelial carcinomas. Virchows Arch 2003;442:309–16.
33. Chatterjee SJ, Datar R, Youssefzadeh D, George B, Goebell PJ, Stein JP,
Young L, Shi SR, Gee C, Groshen S, Skinner DG, Cote RJ. Combined
effects of p53, p21, and pRb expression in the progression of bladder
transitional cell carcinoma. J Clin Oncol 2004;22:1007–13.
34. Malats N, Bustos A, Nascimento CM, Fernandez F, Rivas M, Puente D,
Kogevinas M, Real FX. P53 as a prognostic marker for bladder cancer:
a metaanalysis and review. Lancet Oncol 2005;6:678–686.
35. Bartkova J, Horejsi Z, Koed K, Kr€amer A, Tort F, Zieger K, Guldberg P,
Sehested M, Nesland JM, Lukas C, Orntoft T, Lukas J. DNA damage
response as a candidate anti-cancer barrier in early human tumorigenesis.
Nature 2005;434:864–70.
36. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000;100:57–70.
37. Shiina H, Igawa M, Shigeno K, Terashima M, Deguchi M, Yamanaka
M, Ribeiro-Filho L, Kane CJ, Dahiya R. Beta-catenin mutations cor-
relate with over expression of C-myc and cyclin D1 genes in bladder
cancer. J Urol 2002;168:2220–6.
38. Stoehr R, Krieg RC, Knuechel R, Hofstaedter R, Pilarsky C, Zaak D,
Schmitt R, Hartmann A. No evidence for involvement of beta-catenin
and APC in urothelial carcinomas. Int J Oncol 2002;20:905–11.
39. Thievessen I, Seifert HH, Swiatkowski S, Florl AR, Schulz WA.
E-cadherin involved in inactivation of WNT/b-catenin signalling in
urothelial carcinoma and normal urothelial cells. Br J Cancer 2003;
88:1932–8.
40. Hernando E, Nahle Z, Juan G, Diaz-Rodriguez E, Alaminos M,
Hemann M, Michel L, Mittal V, Gerald W, Benezra R, Lowe SW,
Cordon-Cardo C. Rb inactivation promotes genomic instability by
uncoupling cell cycle progression from mitotic control. Nature 2004;
430:797–802.
41. Simon R, Struckmann K, Schraml P, Wagner U, Forster T, Moch H,
Fijan A, Bruderer J, Wilber K, Mihatsch MJ, Gasser T, Sauter G.
Amplification pattern of 12q13-q15 genes (MDM2, CDK4, GLI) in
urinary bladder cancer. Oncogene 2002;21:2476–83.
42. Christoph F, Schmidt B, Schmitz-Dr€ager BJ, Schulz WA. Overexpression
and amplification of the c-myc gene in human urothelial carcinoma. Int
J Cancer 1999;84:169–73.
43. Clasen S, Schulz WA, Gerharz CD, Grimm MO, Christoph F,
Schmitz-Dr€ager BJ. Frequent and heterogenous expression of cyclin-
dependent kinase inhibitor WAF1/p21 protein and mRNA in urothe-
lial carcinoma. Br J Cancer 1998;77:515–21.
44. Franke KH, Miklosi M, Goebell P, Clasen S, Steinhoff C, Anastasiadis AG,
Gerharz CD, Schulz WA. Cyclin-dependent kinase inhibitor p27KIP1 is
expressed preferentially in early stages of urothelial carcinoma. Urology
2000;56:689–95.
10970215, 2006, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ijc.21852 by Readcube (Labtiva Inc.), Wiley Online Library on [07/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
45. Hoffmann MJ, Florl AR, Seifert HH, Schulz WA. Multiple mecha-
nisms inactivating CDKN1C in bladder cancer. Int J Cancer 2005;
114:406–13.
46. Bakkar AA, Wallerand H, Radvanyi F, Lahaye JB, Pissard S, Lecerf L,
Kouyoumdjian JC, Abbou CC, Pairon JC, Jaurand MC, Thiery JP,
Chopin OK. FGFR3 and TP53 gene mutations define two distinct path-
ways in urothelial cell carcinoma of the bladder. Cancer Res 2003;63:
8108–12.
47. Jebar AH, Hurst CD, Tomlinson DC, Johnston C, Taylor CF, Knowles MA.
FGFR3 and Ras gene mutations are mutually exclusive genetic events
in urothelial cell carcinoma. Oncogene 2005;24:5218–25.
48. Varley C, Hill G, Pellegrin S, Shaw NJ, Selby PJ, Trejdosiewicz LK,
Southgate J. Autocrine regulation of human urothelial proliferation
and migration during regenerative responses in vitro. Exp Cell Res
2005;306:216–29.
49. Swiatkowski S, Seifert HH, Steinhoff C, Prior A, Thievessen I, Schli-
ess F, Schulz WA. Activities of MAP-kinase pathways in normal
uroepithelial cells and urothelial carcinoma cell lines. Exp Cell Res
2003;282:48–57.
50. Hoshino R, Chatani Y, Yamori T, Tsuruo T, Oka H, Yoshida O,
Shimada Y, Ari-i S, Wada H, Fujimoto J, Kohno M. Constitutive acti-
vation of the 41-/43-kDa mitogen-activated protein kinase signaling
pathway in human tumors. Oncogene 1999;18:813–22.
51. Deng Q, Liao R, Wu BL, Sun P. High intensity RAS signaling induces
premature senescence by activating p38 pathway in primary human
fibroblasts. J Biol Chem 2004;279:1050–9.
52. Zhang ZT, Pak J, Huang HY, Shapiro E, Sun TT, Pellicer A, Wu XR.
Role of Ha-ras activation in superficial papillary pathway of urothelial
tumor formation. Oncogene 2001;20:1973–80.
53. Zhang ZT, Pak J, Shapiro E, Sun TT, Wu XR. Urothelium-specific
expression of an oncogene in transgenic mice induced the formation
of carcinoma in situ and invasive transitional cell carcinoma. Cancer
Res 1999;59:3512–7.
54. Makri D, Schulz WA, Grimm MO, Clasen S, Bojar H, Schmitz-
Dr€ager BJ. WAF1/p21 regulates proliferation but does not mediate
p53-dependent apoptosis in urothelial carcinoma cell lines. Int J Oncol
1998;12:621–8.
55. Catto JW, Azzouzi AR, Rehman I, Feeley KM, Cross SS, Amira N,
Fromont G, Sibony M, Cussenot O, Meuth M, Hamdy FC. Promoter
hypermethylation is associated with tumor location, stage, and subse-
quent progression in transitional cell carcinoma. J Clin Oncol 2005;23:
2903–10.
56. Florl AR, Loewer R, Schmitz-Dr€ager BJ, Schulz WA. DNA meth-
ylation and expression of L1 LINE and HERV-K provirus sequen-
ces in urothelial and renal cell carcinoma. Br J Cancer 1999;80:
1312–21.
57. Kimura F, Seifert HH, Florl AR, Santourlidis S, Steinhoff C, Swiat-
kowski S, Mahotka C, Gerharz CD, Schulz WA. Decreased DNA
methyltransferase 1 expression relative to cell proliferation in transi-
tional cell carcinoma. Int J Cancer 2003;104:568–78.
58. Wu XR. Urothelial tumorigenesis: a tale of divergent pathways. Nat
Rev Cancer 2005;5:713–25.