mesothelioma cancer

Malignant mesothelioma is a disease strongly associated with carcino¬gen
exposure (1). As has been observed in a wide variety of other
carcinogen-associated solid cancers, mesothelioma tumors accumulate a spectrum
of acquired genetic lesions during the molecular patho-genesis leading to overt
cancer. Perhaps reflecting the unique history of carcinogen exposure routinely
seen in mesothelioma, many of the well-characterized mutations found in other
cancers such as p53 and ras family alterations are not a common feature in
malignant mesothe-lioma (1). Nonetheless, a variety of well-defined molecular
abnormal¬ities have been identified in the majority of cases of mesothelioma. As
has often been the case in cancer genetics, the first information regard¬ing
genetic alterations in mesothelioma came from tumor karyotypic or family
studies.

Although generally observed to be a disease strongly associated with asbestos
exposure, familial clustering of mesothelioma independent of asbestos exposure
has been reported (2–6). Epidemiologic data suggest that there may be a possible
familial predisposition to mesothelioma, but the molecular basis for this
remains unclear (7). Examples of chro¬mosome 9p or 22 abnormalities have been
reported in single cases or families with early-onset mesothelioma, but
observations such as these have occurred infrequently and largely have confirmed
known genetic loci that are involved in mesothelioma pathogenesis (2,6). The
lack of a heritable model of mesothelioma, such as defined in breast cancer or
colon cancer, has concentrated studies on the genetics of asbestos-induced
mesothelioma. In these more common cases of sporadic mesothelioma with no
obvious familial clustering or early onset, both of which are ultimately rare in
mesothelioma, the most frequent cyto-genetic abnormality in tumors reported has
been the loss of chromo¬some 22 (8). As will be seen, this marks one of the most
common somatic genetic targets, NF2, identified to date in mesothelioma. Other
frequently observed karyotypic abnormalities include loss of the short arm (p)
of chromosomes 1, 3, and 9 (8–10). Again, one of these recur¬ring molecular
lesions, loss of the 9p21 locus, has been correlated with corresponding loss of
multiple well-characterized tumor suppressor


gene function. In addition, loss of the genetic material from the long arm of
chromosomes 6, 13, and 15 has also been commonly identified in mesothelioma
tumors (1). Reminiscent of investigations in a spec¬trum of cancers, gross
karyotypic abnormalities have given useful guidance as to the location of
somatic mutations in mesothelioma asso¬ciated with loss of function and
malignant transformation. These loss of function genes, or tumor suppressor
genes, regulate a variety of mo¬lecular phenomena ranging from regulating the
cell cycle, to cellular homeostasis and repair, to programmed cell death
mechanisms. In a similar way, reduplication, rearrangement, or amplification of
specific sequences can likewise give an indication of gain of function genetic
abnormalities. However, few examples exist of such gain of function genetic
abnormalities phenomenon in mesothelioma. For example, gain of genetic material
has been reported for 5p; however, no specific recurrent gene rearrangements
that might indicate a gain of function fusion gene product have been identified
(1).

This chapter discusses the known acquired genetic alteration and when possible
correlates them with the previously documented cyto-genetic data. However,
unlike general models in other solid tumors, little is known about the relative
timing or order of these genetic and epigenetic lesions. Nonetheless, recent
data from animal models can give us an indication of which molecular events are
directly correlated with potential carcinogenic stimuli such as asbestos or
simian virus 40 (SV40) infection.

Loss of Function from Cyclin-Dependent Kinase Inhibitors at the 9p21 Locus in
Mesothelioma

Multiple lines of research have revealed that among the most common acquired
genetic abnormalities in cancer is loss of G1 to S checkpoint control (Table
8.1) (11). Among the most critical elements in maintain¬ing control at this
checkpoint is the presence of hypophosphorylated pRb in maintaining the cell in
a quiescent state (12). Processive phos-phorylation of the 105-kd Rb
(retinoblastoma susceptibility gene) gene product (pRb) is mediated by the
family of cyclin-dependent kinases (CDKs) in conjunction with their
corresponding S phase cyclins con¬sisting primarily of cyclin E and the members
of the cyclin D family (11). The D family of cyclins, among the earliest
expressed proteins in the cell cycle, can physically interact with both pRb (13)
and their asso¬ciated CDKs, resulting in the critical phosphorylation of pRb.

The beginning of the understanding of the G1 to S molecular check¬point can be
traced to the cloning and characterization of the retinoblas-toma susceptibility
gene (Rb). The identification of the Rb gene in 1986 as being the genetic
element that is mutated or deleted in the germline of kindreds affected with
inherited retinoblastoma was the first proof of the tumor suppressor gene
paradigm (14) as put forth by Knutson 15 years earlier. This model, which
largely defines the tumor suppres¬sor gene, is currently understood to postulate
that loss of both alleles of a normal genetic element, in the case of familial
retinoblastoma this

being the two alleles of the Rb gene, leads to loss of cellular homeostasis
and transformation. This “two-hit” model applies to most of the somatic genetic
lesions identified in mesothelioma and discussed in this chapter. In fulfillment
of this model, the Rb gene locus of 13q14 is commonly deleted in a variety of
other cancers, most notably lung and bladder cancer, diseases in which
consequent loss of the Rb gene product, pRb, is found to be quite common (15).
The irony of somatic mutations acquired in mesothelioma, however, is that
although abnor¬malities of 13q are also common in mesothelioma, somewhat
surpris¬ingly, loss of function of pRb is rarely if ever observed in
mesothelioma (12). Therefore, although loss of function at the G1 to S
checkpoint occurs in mesothelioma, it is not accounted for by deletion or
mutation the Rb gene.

As previously mentioned, processive phosphorylation of pRb leads to inactivation
of pRb and the consequent release of the cell from the block into S phase.
Inhibition of this CDK-mediated phosphorylation occurs via a family of proteins
known appropriately as CDK inhibitors (CDKIs) (11). The first described CDKI,
p16INK4a (CDKN2, MTS) was first found to be commonly deleted in a form of
familial melanoma associated with loss of genetic material at 9p21 (16). It had
been known that 9p21 is a frequently targeted region of the genome in a variety
of cancers including lung, bladder, and brain tumors. Therefore, it was not
surprising to find that p16INK4a expression was also lost in many of these same
cancers. What was somewhat unexpected, though, was the tight correlation between
loss of p16INK4a function and the presence of wild-type pRb in many cancers, and
vice versa (17–19). Since, as pre¬viously discussed, pRb was not found to be a
common site of loss of function in mesothelioma, it might be anticipated that an
alternative but equally critical aspect of the G1 to S checkpoint, such as
p16INK4a, would be affected in mesothelioma.

A variety of studies prior to the localization of the p16INK4a gene to the 9p21
locus had identified loss of 9p as a common event in mesothe-lioma tumors and
cell lines (9,10). Both the aforementioned retention of wild-type pRb expression
in mesothelioma and the subsequent iden¬tification of the 9p locus as being the
location of the CDK inhibitor p16INK4a, made this gene product a likely
candidate tumor suppressor gene deleted in mesothelioma. When mesothelioma cell
lines and tumors were examined for presence of the p16INK4a gene product, it was
found to be absent in all cases (12,20,21). Although this was not unex¬pected,
since many of the cell lines examined possess disruption of the 9p21 locus, the
extremely high frequency of loss of p16INK4a gene product expression was
surprising. Virtually all mesothelioma tumors and cell lines examined to date
have lost detectable expression of the p16INK4a gene product.

Tightly linked at the 9p21 locus to the p16INK4a gene is the highly homologous
p15INK4b gene (11). This 15-kd protein CDKI possesses much of the same
biochemical activity as p16INK4a, but its mutational spectrum is somewhat
different. While p16INK4a appears to be one of the most common genetic lesions
acquired in solid tumors, p15INK4b is affected at a lower frequency in many of
the common cancers. For example, in lung cancer, where the 9p21 locus is intact
in a large number of cases, the inactivation of the two genes via epigenetic
mech¬anisms, such as DNA hypermethylation, is not concordant (22). This pattern
of differential inactivation is similar to that observed between p16INK4a and
the third gene encoded at this locus, p14ARF, yet a third reg¬ulatory gene
located at the 9p21 locus (see below). In the case of the homologous CDKI,
p15INK4b, the gene is differentially inactivated in a significant percentage of
cases of both leukemia and myelodysplastic syndrome, where expression of the
other 9p21 regulatory genes are maintained (23,24). Nonetheless, in mesothelioma
many tumors have lost gross genetic material at the 9p21 locus, and have
consequently co-deleted all the genes encoded there (20). As will be discussed,
nondeletional loss of gene expression via epigenetic mechanisms of the genes at
the 9p21 locus is less common in mesothelioma than in other cancers (25–27).
With the high rate of deletion of the 9p21 locus, it is not surprising that at
least one study has determined that 72% of mesothelioma tumors have co-deleted
the genes for both p15INK4b and p16INK4a, and presumably for the other
regulatory gene encoded at this locus, p14ARF20.

In light of the high frequency of loss of p16INK4a product in mesothe-lioma, it
has been suggested that this may be a good target for gene replacement therapy
(28,29). Reexpression of p16INK4a protein is associ¬ated with cell cycle arrest,
apoptosis, and cell death in mesothelioma, and prolonged survival in
mesothelioma xenograft models (28,29). Alternative methods of mediating the
p16INK4a gene reexpression may also serve as potential novel therapies for
mesothelioma. As in other cancers that lack p16INK4a expression but retain
normal 9p alleles, DNA hypermethylation of the p16INK4a gene accounts for a
small percentage of p16INK4a gene silencing in mesothelioma (25–27). It is
interesting to note that a small percentage of mesothelioma tumors have demon-

strated response to methylation inhibiting cytidine analogues in phase II
clinical trials.

Loss of genetic material and consequent gene expression at the 9p21 locus
results in at least one additional well-characterized molecular defect in
mesothelioma. In one of the more surprising findings in cell cycle and cancer
molecular genetics, the DNA sequence encoding the p16INK4a gene also encodes an
alternative cell cycle regulatory gene in a unique reading frame, thus coding
for a second protein with an entirely different amino acid sequence and
molecular weight (11). This 14-kd protein, known as p14ARF (alternative reading
frame), inhibits mdm2-mediated degradation of the well-characterized p53 tumor
suppressor gene (11). This model predicts that in the absence of p14ARF,
wild-type p53 will be highly unstable due to a very short protein half-life.
Poten¬tially, then, loss of genetic material at the 9p21 locus leads to loss of
cell cycle regulation through both the pRb and p53 pathways. Data from murine
knockout models suggest that loss of p14ARF may be a stronger cancer initiating
event than loss of p16INK4a, although similar data from human cancer is not as
convincing with differential loss of one or the other gene product being
observed via methylation in tumors with intact 9p21 locus (30).

Loss of p14ARF gene expression appears to be as equally common in mesothelioma
as the loss of p16INK4a protein (1,27). Studies on protein expression have been
limited to this point due to the lack of appro¬priate immunologic reagents, but
reexpression of p14ARF appears to mediate many of the same effects in
mesothelioma cells and xenografts that p16INKa reexpression does (31). In a
correlation that strongly mimics the previously described situation in the
p16INK4a/pRb genetic switch in mesothelioma, p53 mutations are rarely found in
mesothelioma (1,32). Rather, the prevailing model suggests that loss of p53
check¬point is mediated through loss of p14ARF expression and consequent
enhanced mdm2-mediated p53 degradation (11,33). This fascinating picture of dual
loss of control at two of the best characterized cell cycle checkpoints by
disruption of a single genetic locus has proven to be a surprising twist in the
study of cancer biology. Although it may prove that only one of these pathways
is the critical step in the development of mesothelioma, at this time the
evidence suggests that loss of both of the proteins encoded for at the 9p21
locus is important in the patho-genesis of mesothelioma.

Additional Abnormalities of Cyclin-Dependent Kinase Inhibitors in Mesothelioma

Gross abnormalities in expression or mutations of p15INK4b and p16INK4a are
likely the most common genetic abnormalities of the CDK inhibitor family of
proteins not only in mesothelioma but in other solid tumors as well.
Nonetheless, differential expression of some of the other well-characterized CDK
inhibitors such as p27 (p27kip1) and p21 (WAF1/CIP1) has been documented. In
mesothelioma, studies have identified either elevated p27 expression by
immunohistochemistry to

be a positive prognostic sign or the converse, low expression of p27, to be a
poor prognostic sign (34–36). For example, in one study low expression was
associated with a median survival of 4 to 5 months, while normal or elevated
expression of p27 in mesothelioma portends a more favorable prognosis with a
median survival of 10 to 11 months (34). Similar prognostic significance of low
p27 expression has been identified in other solid tumors such as breast cancer
(37–40). In a like manner, the p21 CDK inhibitor has been identified as
differentially expressed in mesothelioma tumors by immunohistochemistry (41,42).
The original identification of p21 was as a factor upregulated by the expression
of wild-type p53, and thus it might be anticipated that most mesothelioma tumors
express detectable p53 (43). Alternatively, as pre¬viously discussed, p53 may be
destabilized or inactivated by secondary means, and thus any factor potentially
dependent on the presence of wild-type p53 for expression, such as p21, may be
relatively absent. Along these lines, in studies only about 35% of mesothelioma
tumors possess easily detectable p21 expression (41,42). However, the
rela¬tively few studies published have varied in the conclusion of whether there
is any clinical significance in the p21 expression pattern in mesothelioma.

Mutations in the NF2 Gene Are a Common Feature of Mesothelioma

Neurofibromatosis type 2 (NF2) is an autosomal-dominant disease characterized by
development of brain tumors and schwannomas, particularly involving the eighth
cranial nerve (44). Epidemiologic studies have demonstrated a linkage to the
long arm of chromosome 22 (22q), and subsequent positional cloning isolated the
NF2 gene as the targeted gene whose loss of function accounts for this clinical
syn¬drome (45,46). The 70-kd NF2 gene product is a moesin ezrin radixin-like
protein that maps to 22q11-q13.1 and has also has been named both merlin (moesin
ezrin radixin-like protein) and schwannomin. As dis¬cussed previously, loss of
chromosome 22 appears to be the most com¬monly detected cytogenetic reported in
malignant mesothelioma (8). Following the identification of the NF2 gene on
chromosome 22 as the affected gene in familial neurofibromatosis, a variety of
investigators examined NF2 gene expression in mesothelioma (47,48). An initial
study of 15 mesothelioma cell lines revealed abnormalities in single-strand
conformation polymorphism (SSCP) in 53% of cDNAs from mesothelioma. Subsequent
sequencing revealed a high frequency of mutations in the NF2 protein coding
sequence, resulting in truncated merlin protein in all eight of the samples with
abnormal SSCP migra¬tion (48). When primary tumors were examined, all but one of
the matching primary tumors possessed the identical mutation found in the paired
cell line. Similar studies have reported 41% of tumors or cell lines with
detectable mutations or deletions in NF2 transcripts in mesothelioma, while no
such abnormalities were observed in lung cancer (47). Similar to the situation
with the genes at the 9p21 locus,

the NF2 gene appears to function as a classic tumor suppressor gene in
mesothelioma and likely is the second most common somatic muta¬tion in this
disease.

Although the high rate of mutation of NF2 transcripts in mesothe-lioma cell
lines and tumors indicates it is a frequent target of inactiva-tion in the
development of mesothelioma, it is interesting to note that mesothelioma is not
part of the well-described clinical syndrome of neurofibromatosis 2 (44).
However, this is not unlike what has been observed in other familial cancer
syndromes, including the rare inher¬ited abnormalities of 9p21 associated with
melanoma or pancreatic cancer. The clinical syndrome of neurofibromatosis is
exceedingly rare, and presumably the carrier rate of germline mutations of NF2
is as rare as 1 in 40,000 (44). It has been theorized that if germline mutations
in NF2 predispose patients to mesothelioma in the face of asbestos exposure, the
low frequency of both the incidence of NF2 germline mutations and asbestos
exposure would make the predisposition very difficult to detect. Along these
lines, a recent case was reported of an asbestos-exposed NF2 patient who
developed mesothelioma within several years of his exposure, rather than the
common prolonged latency period of 20 years or more (2). It seems likely that
NF2 patients are at increased risk of mesothelioma following additional exposure
to asbestos, or perhaps following infection with SV40 or other putative
mesothelioma promoting agents. However, the coincident low rate of germline NF2
mutation carriers and intense asbestos exposure makes detection of this
potential heritable predisposition to mesothelioma uncommon.

Alterations of p53 Appear Infrequently in Malignant Mesothelioma

Alterations of p53 Appear Infrequently in Malignant Mesothelioma

Mutations of the p53 gene are among the most frequent and best doc¬umented
acquired genetic abnormalities in solid tumors (49). The p53 locus is located at
17p13, a common hot spot for karyotypic abnor¬malities in a wide spectrum of
cancers. As its name indicates, the p53 gene encodes a 53-kd phosphoprotein that
appears to play a key role in maintenance of the integrity of genetic
information. Mutations of p53 generally are missense mutations resulting in
expression of a full-length inactive gene product that is preferentially
stabilized when compared to the relatively short-lived wild-type p53 protein. As
such, cancers that stain strongly for p53 protein on immunohistochemistry often,
but not invariably, possess mutant p53 protein, whereas tumors that stain weakly
are more often than not wild type. However, because of some variability between
the correlation of strong immunostaining and mutation of the p53 protein,
studies on the frequency of mutation of p53 in cancers can come to divergent
opinions (50).

Abnormalities of 17p have been noted in mesothelioma, although at a lower
frequency than in many other tumors (1). When mesothelioma cell lines or tumors
have been examined, generally few mutations in p53 have been found (51–54). One
explanation for this divergence from

the mutation patterns seen in other cancers may be the presence of SV40 tumor
antigen (Tag) as a putative co-carcinogen. It has been known that SV40 Tag can
bind to and inactivate wild-type p53, impli¬cating that infection with SV40 may
serve as a method for inactivating p53 in the absence of overt mutation. In
recent years, the detection of SV40 viral DNA in mesothelioma tumors has
generated much interest in this hypothesis, and this issue is discussed in
detail elsewhere in this text (1). Alternatively, as discussed previously, the
absence of p14ARF in mesothelioma would be expected to result in loss of the
inhibition of mdm2-mediated degradation of p53, leading to a shorter protein
half-life. Of interest, p53 mutations have rarely been observed in experi¬mental
models of asbestos-derived mesothelioma in rodents (55,56), but p53 deficient
mice are more susceptible to asbestos-induced mesothelioma (57,58). These
findings strongly argue that loss of p53 function plays a direct role in the
development of mesothelioma, but more likely through alternative mechanisms than
direct mutation or deletion of the p53 gene product.

Several investigators have reported increased levels of p53 protein in
mesothelioma tumors as detected by immunohistochemistry (51,52). Rates of
overexpression have been reported to be as high as 35% in resected tumors. In
addition, one report identified two of four mesothe-lioma cell lines with
missense mutations by DNA sequencing (59). However, these reports should
probably be regarded as the exception rather than the rule. Circulating
autoantibodies to p53 are often seen in solid tumors and are thought to
represent an immunologic response to the presentation of mutant p53 antigen, but
these autoantibodies are rarely detected in patients with mesothelioma, although
perhaps only in a small percentage of patients (less than 10%) (60). Moreover,
the p53 antisera detected in these patients in this study had relatively low
titers that did not vary with treatment, indicating that perhaps they bear
little relevance to the p53 status of the corresponding tumor.

Similar to the reports on reexpression of p16INK4a and p14ARF, virally mediated
reexpression of p53 can result in cell growth inhibition and xenograft
inhibition in mesothelioma cells (61). Although this may seem to argue for the
presence of directly inactivated p53 in mesothe-lioma that can then be
ameliorated by the addition of exogenous wild-type p53, it is also consistent
with a model of secondary inactivation of p53 from the presence of SV40 Tag or
from protein destabilization in the absence of p14ARF. In the latter two
examples it may be predicted that overexpression of exogenous p53 will overcome
these mechanisms of p53 inactivation and thus reestablish normal cell
homeostasis. Con¬sistent with this interpretation, the replication sufficient
ONYX-015 adenovirus, which has a selective cytolytic effect in p53-deficient
cells, demonstrated cell killing in a mesothelioma cell line (MS-1) that
retained both normal p53 and p14ARF, while three cell lines (NCI-H28, NCI-H513,
211H) that possessed wild-type p53 but absent p14ARF were killed by the ONYX-015
virus (62). When p14ARF was reintroduced into these three cell lines, they
became resistant to ONYX-015–mediated killing as well, strongly arguing that p53
inactivation in these mesothe-lioma cell lines is mediated in part by the
absence of p14ARF protein.

However, these findings cannot eliminate the probable important role played by
the presence of possible SV40 Tag in moderating or inacti¬vating wild-type p53
function in mesothelioma.

Apoptosis-Mediating Gene Defects in Mesothelioma

Apoptosis-Mediating Gene Defects in Mesothelioma

Both the death receptor and mitochondrial-mediated pathways play significant
roles in cell death and cancer progression (63). Although much recent interest
has focused on abnormalities of caspase activity in cell immortalization and
cancer pathogenesis, little is known regard¬ing this pathway and the
pathogenesis of mesothelioma (64,65). In contrast, abnormalities of the Bcl-2
pathway (an inhibitor of the mitochondrial pathway) have been described in
mesothelioma, but to a limited degree. In one of the largest studies conducted
on this subject, researchers in Finland identified Bcl-2 positivity in seven out
of 35 (20%) mesothelioma tumors as analyzed by immunohistochemistry (66). Strong
expression of the related antiapoptotic proteins Bcl-X was found in all cases.
The absence of Bcl-2 expression in mesothelioma was strongly correlated, in this
relatively small series, with a higher apo-ptotic index and, paradoxically,
statistically significant poorer survival. Although a similar paradoxical result
has been reported in other cancers, these findings are surprising unless they
are indicative of a high tumor burden turnover in aggressive disease. No similar
correla¬tions were found in apoptotic index with Bcl-X expression, but given the
universally elevated levels detected it is understandable if no vari¬ation can
be detected.

In the face of elevated expression of the antiapoptotic Bcl-X gene product,
investigators have designed systems to attempt to down-regulate its expression
and render the mesothelioma cells more sus¬ceptible to apoptosis. Expression of
antisense Bcl-X has been reported to engender apoptosis in two mesothelioma cell
lines following dimin¬ished transcription of Bcl-X transcripts (67). In a like
manner, treatment of mesothelioma cells with sodium phenylbutyrate has been
described as leading to Bcl-X downregulation and cell death in mesothelioma
(68). In a somewhat different pharmacologic maneuver, these same investigators
introduced the proapoptotic Bak gene into mesothelioma cells to counteract the
overexpressed Bcl-X protein and found apopto-sis occurred in two cell lines
(69). It is hard to generalize from these experiments which, if any, of these
potential therapeutic modalities could prove of value in treating mesothelioma,
but it seems clear that as in many other solid tumors, overexpression of the
antiapoptotic Bcl family of proteins plays a significant role in the
pathogenesis of mesothelioma, and downregulation of these proteins can mediate
cell death in mesothelioma cell lines and tumors.

The death receptor pathway is initiated by death inducing ligands such as TRAIL
binding to their specific cell surface receptors and trig¬gering a cascade
involving caspase 8 and possibly caspase 10. There are four known TRAIL
receptors: DR4 and DR5, which initiate the apoptotic pathway upon activation,
and two decoy receptors, DcR1

and DcR2, which lack a death domain and are presumed to be anti-apoptotic.
However, recent data indicate that the decoy receptors are methylated and
silenced in pediatric tumors (70). Our unpublished data indicate that
methylation of the decoy receptors is one of the most frequent molecular changes
present in virtually all cancer types, with very high frequencies in
mesotheliomas. These observations indicate that the role of decoy receptors need
to be reevaluated, and that silenc¬ing of the decoy receptors may aid cell
survival rather than prevent apoptosis.

Epigenetic Inactivation of RASSF1A Occurs in Mesothelioma and After Simian Virus
40 Infection in Mesothelial Cells

Epigenetic Inactivation of RASSF1A Occurs in Mesothelioma and After Simian Virus
40 Infection in Mesothelial Cells

Many of the classic studies of the past decade in somatic mutations in solid
tumors have revolved around the central tenet that cancer is a disease of
acquired genetic damage. However, it is now clear that epigenetic mechanisms,
including DNA hypermethylation, play a sig¬nificant role in gene silencing in
cancer. As an example, as has been pre¬viously discussed, up to 10% of
mesothelioma tumors inactivate p16INK4a gene expression following acquired DNA
hypermethylation, although it must be noted that this rate of methylation is
much lower than that found in many other common cancers (26,27). Until recently,
however, the methylation pattern of other genes in mesothelioma has not been
extensively studied. A series profiling 66 mesothelioma and 40 lung
adenocarcinomas for methylation has yielded fascinating and com¬pelling results
(25). Methylation profiling of seven genes commonly inactivated by epigenetic
mechanisms in cancer was carried out in 66 mesothelioma samples. In summary, it
was found that the overall rate of gene methylation (“methylation index”) was
significantly lower than that found in adenocarcinoma of the lung. However, one
gene, the can¬didate tumor suppressor gene RASSF1A, was methylated at a
remark¬ably high frequency in this large cohort of mesothelioma tumors (32%).
The RASSF1A gene is a 39-kd ras-associated protein that fulfills the cri¬terion
of a classic tumor suppressor gene in lung cancer (71–73). The gene maps to the
3p21.3 locus, a common sight for loss of heterozygos-ity and cytogenetic
abnormalities in a wide variety of cancers, includ¬ing mesothelioma. An
additional intriguing finding arising from this methylation survey is that the
mesothelioma tumors analyzed that pos¬sessed evidence of SV40 Tag sequences (52%
of the tumors examined) were the subset that also possessed, significantly, a
higher methylation index (25). If SV40 infection mediates or is associated with
epigenetic cellular events, such as DNA hypermethylation, it may give insight
into the relatively fewer somatic genetic lesions seen in mesothelioma as
compared to other carcinogen associated malignancies.

As has been discussed in this chapter and elsewhere in this text, there appears
to be significant evidence that expression of SV40 Tag plays a role in the
pathogenesis of mesothelioma (1). Identification of the pres¬ence of distinct
methylation patterns in SV40 Tag-positive mesothe-

lioma tumors raises the question of cause and effect. Recent investiga¬tions on
the acquired genetic and epigenetic abnormalities that occur following SV40
infection of mesothelial cells have revealed an inter¬esting pattern of DNA
hypermethylation in these cells. In early passage following infection, no
methylation of seven genes previously identi¬fied as being targets of
methylation in mesothelioma was found. However, later passages of SV40-infected
mesothelial cells have been found to possess progressive methylation of the
RASSF1A gene and a consequent decrease in transcript (74). Consistent with
methylation at this locus, treatment with the cytosine analogue and methylation
inhibitor 5-aza-2¢-deoxycytidine led to reexpression of the RASSF1A transcripts.
These findings provide a true mechanistic link between known tumor suppressor
gene inactivation and SV40 infection.

Dominant-Negative WT1 Gene Abnormalities Are Detected in Mesothelioma

Identification of karyotypic abnormalities at 11p13 in pediatric nephro-blastoma
(Wilms’ tumor) led to the subsequent cloning in 1990 of the Wilms’ tumor
susceptibility gene 1 (WT1) (75). The WT1 locus had already been shown to
fulfill one of the main criteria of a classic tumor suppressor gene even prior
to the identification of the coding sequence when it was demonstrated in 1987
that reconstitution of 11p to nephroblastoma cells reversed the malignant
phenotype (76). Sub¬sequent analyses have demonstrated loss or inactivation of
WT1 in most cases of pediatric nephroblastoma or Wilms’ tumor. The WT1 gene
product is a 52- to 54-kd protein with a complex expression pattern involving up
to 24 isoforms (77). The protein possesses a series of zinc finger motifs
consistent with a DNA-binding protein, and dis¬ruption of WT1 in murine models
leads to severe developmental abnor¬malities in the urogenital system. Similar
to the pattern seen with other tumor suppressor genes identified first in
germline cancer syndromes, acquired mutations of WT1 have also been noted in a
variety of other malignancies as well (78).

Mutations of the WT1 gene in mesothelioma were noted after the rather striking
finding that WT1 protein is routinely expressed in normal mesothelium just as it
is expressed in normal urogenital tissues (79). In addition, mesothelioma tumors
were found to generally express elevated levels of WT1 protein as detected by
immunohistochemistry (80,81). The presence of nuclear staining for WT1 has been
reported in 75% to 100% of mesothelioma tumors and cell lines examined to date.
This has led to WT1 expression being developed as a potentially useful
diagnostic test in differentiating mesothelioma tumors from lung cancer, since
normal WT1 expression is rarely detected in lung or lung cancers (81).
Mutational analyses, however, have yielded only sporadic missense mutation
identified in mesothelioma samples. Mutations of WT1 in mesothelioma, as well as
in other cancers, generally target the DNA binding motif of exons 7 through 10
containing the zinc finger motifs. By way of interaction with DNA through the
zinc finger motifs,

the WT1 gene product is thought to function as a transcriptional sup¬pressor
(82). In an intriguing alteration of the standard loss of function abnormalities
of tumor suppressor genes, mutations of WT1 in the zinc finger motif have been
demonstrated not only to abrogate the tran-scriptional repression activity of
wild-type WT1 but also to result in a dominant-negative transcriptional
activator phenotype (82). Such dominant-negative or activating mutations of WT1
occur in both germline nephroblastoma and acquired mutations in cases such as
mesothelioma (79), but probably at a low frequency. Additional reports indicate
DNA hypermethylation of the WT1 may occur with a high frequency in mesothelioma
at a CpG island with the 5¢ end of the gene, but the correlation of this finding
with gene product expression remains unclear (83).

Relationship to Simian Virus 40

While the relationship of the SV40 virus and the pathogenesis of malig¬nant
mesothelioma (MM) is discussed elsewhere in this book, the presence of SV40 is
associated with certain molecular changes that are discussed briefly here for
the sake of completeness. These changes include activation or upregulation of
the Met and Notch1 oncogenes and of insulin-like growth factor I (84). In
addition, methylation and silencing of the RASSF1A gene is significantly higher
in virus contain¬ing tumors (25,74). The Tag present in SV40-positive MMs has
been demonstrated to be capable of binding to and inhibiting cellular p53 and
retinoblastoma family proteins (84). In an in vitro model, both SV40 and
asbestos acted as co-carcinogens (85), suggesting that both of these etiologic
factors contribute to the molecular pathogenesis of MM in their own unique
manner. Of notable interest, both the Tag protein of SV40 and deletions of 9p
induced by asbestos exposure may contribute to inhibition of the cell cycle and
of the p53 protein.

Conclusion

Mesothelioma is a cancer marked by a distinct pattern of mutations unlike most
other solid tumors. Direct mutation of genes such as p53 or ras family members
may be rare and distinctly different from the well-characterized spectrum of
mutations seen in the more common epithelial malignancies such as lung or
bladder cancer. This may be partially a result of the unique contributions of
asbestos and SV40 infec¬tion as co-carcinogens in mesothelioma. Nonetheless,
aberrations of the regulatory genes expressed at the 9p21 locus (p16INK4a,
p14ARF) and chro¬mosome 22 (NF2) are clearly among the highest frequencies
observed in human cancers. In addition, recent investigations into epigenetic
inactivation following SV40 infection may yield new gene targets such as RASSF1A
that are inactivated by pathways that have not been suf¬ficiently explored in
the past. The presence of normal DNA within mesothelioma tumors is likely not to
be equated with the presence of a normal protein phenotype in the future. This
holds the promise

of potentially reversible epigenetic or infectious molecular defects accounting
for much of the transformed phenotype in mesothelioma, and the distinct
possibility of novel therapeutic modalities that may reverse these acquired
defects in gene expression patterns.

References

1. Carbone M, Kratzke RA, Testa JR. The pathogenesis of mesothelioma. Semin
Oncol 2002;29:2–17.

2. Baser ME, De Rienzo A, Altomare D, et al. Neurofibromatosis 2 and malig¬nant
mesothelioma. Neurology 2002;59:290–291.

3. Kane MJ, Chahinian AP, Holland JF. Malignant mesothelioma in young adults.
Cancer 1990;65:1449–1455.

4. Risberg B, Nickels J, Wagermark J. Familial clustering of malignant
mesothelioma. Cancer 1980;45:2422–2427.

5. Hammar SP, Bockus D, Remington F, Freidman S, LaZerte G. Familial
mesothelioma: a report of two families. Hum Pathol 1989;20:107–112.

6. Musti M, Cavone D, Aalto Y, Scattone A, Serio G, Knuutila S. A cluster of
familial malignant mesothelioma with del(9p) as the sole chromosomal anomaly.
Cancer Genet Cytogenet 2002;138:73–76.

7. Roushdy-Hammady I, Siegel J, Emri S, Testa JR, Carbone M.
Genetic-susceptibility factor and malignant mesothelioma in the Cappadocian
region of Turkey. Lancet 2001;357:444–445.

8. Flejter WL, Li FP, Antman KH, Testa JR. Recurring loss involving chromo¬somes
1, 3, and 22 in malignant mesothelioma: possible sites of tumor sup¬pressor
genes. Genes Chromosomes Cancer 1989;1:148–154.

9. Taguchi T, Jhanwar SC, Siegfried JM, Keller SM, Testa JR. Recurrent
dele¬tions of specific chromosomal sites in 1p, 3p, 6q, and 9p in human
malig¬nant mesothelioma. Cancer Res 1993;53:4349–4355.

10. Cheng JQ, Jhanwar SC, Lu YY, Testa JR. Homozygous deletions within 9p21-p22
identify a small critical region of chromosomal loss in human malignant
mesotheliomas. Cancer Res 1993;53:4761–4763.

11. Sherr CJ. Cancer cell cycles. Science 1996;274:1672–1677.

12. Kratzke RA, Otterson GA, Lincoln CE, et al. Immunohistochemical analysis of
the p16INK4 cyclin-dependent kinase inhibitor in malignant mesothelioma. J Natl
Cancer Inst 1995;87:1870–1875.

13. Dowdy SF, Hinds PW, Louie K, Reed SI, Arnold A, Weinberg RA. Physical
interaction of the retinoblastoma protein with human D cyclins. Cell
1993;73:499–511.

14. Friend SH, Horowitz JM, Gerber MR, et al. Deletions of a DNA sequence in
retinoblastomas and mesenchymal tumors: organization of the sequence and its
encdoded protein. Proc Natl Acad Sci USA 1987;84:9059–9063.

15. Kratzke RA, Shimizu E, Kaye FJ. Oncogenes in human lung cancer. Cancer Treat
Res 1992;63:61–85.

16. Kamb A, Gruis NA, Weaver-Feldhaus J, et al. A cell cycle regulator
poten¬tially involved in genesis of many tumor types. Science 1994;264:436– 440.

17. Kelley MJ, Nakagawa K, Steinberg SM, Mulshine JL, Kamb A, Johnson BE.
Differential inactivation of CDKN2 and Rb protein in non-small-cell and
small-cell lung cancer cell lines. J Natl Cancer Inst 1995;87:756–761.

18. Kratzke RA, Greatens TM, Rubins JB, et al. Rb and p16INK4a expression in
resected non-small cell lung tumors. Cancer Res 1996;56:3415–3420.

19. Otterson GA, Kratzke RA, Coxon A, Kim YW, Kaye FJ. Absence of p16INK4
protein is restricted to the subset of lung cancer lines that retains wild-type
RB. Oncogene 1994;9:3375–3378.

20. Xio S, Li D, Vijg J, Sugarbaker DJ, Corson JM, Fletcher JA. Codeletion of
p15 and p16 in primary malignant mesothelioma. Oncogene 1995;11: 511–515.

21. Cheng JQ, Jhanwar SC, Klein WM, et al. p16 alterations and deletion mapping
of 9p21-p22 in malignant mesothelioma. Cancer Res 1994;54: 5547–5551.

22. Zhou JX, Niehans GA, Shar A, Rubins JB, Frizelle SP, Kratzke RA. Mecha¬nisms
of G1 checkpoint loss in resected early stage non-small cell lung cancer. Lung
Cancer 2001;32:27–38.

23. Daskalakis M, Nguyen TT, Nguyen C, et al. Demethylation of a
hypermeth-ylated P15/INK4B gene in patients with myelodysplastic syndrome by
5-Aza-2¢-deoxycytidine (decitabine) treatment. Blood 2002;100:2957–2964.

24. Quesnel B, Guillerm G, Vereecque R, et al. Methylation of the p15(INK4b)
gene in myelodysplastic syndromes is frequent and acquired during disease
progression. Blood 1998;91:2985–2990.

25. Toyooka S, Pass HI, Shivapurkar N, et al. Aberrant methylation and simian
virus 40 tag sequences in malignant mesothelioma. Cancer Res 2001;61: 5727–5730.

26. Wong L, Zhou J, Anderson D, Kratzke RA. Inactivation of p16(INK4a)
expression in malignant mesothelioma by methylation. Lung Cancer 2002;
38:131–136.

27. Hirao T, Bueno R, Chen CJ, Gordon GJ, Heilig E, Kelsey KT. Alterations of
the p16(INK4) locus in human malignant mesothelial tumors. Carcinogen-esis
2002;23:1127–1130.

28. Frizelle SP, Grim J, Zhou J, et al. Re-expression of p16INK4a in
mesothe-lioma cells results in cell cycle arrest, cell death, tumor suppression
and tumor regression. Oncogene 1998;16:3087–3095.

29. Frizelle SP, Rubins JB, Zhou JX, Curiel DT, Kratzke RA. Gene therapy of
established mesothelioma xenografts with recombinant p16INK4a ade-novirus.
Cancer Gene Ther 2000;7:1421–1425.

30. Kamijo T, Zindy F, Roussel MF, et al. Tumor suppression at the mouse INK4a
locus mediated by the alternative reading frame product p19ARF. Cell
1997;91:649–659.

31. Yang CT, You L, Yeh CC, et al. Adenovirus-mediated p14(ARF) gene trans¬fer
in human mesothelioma cells. J Natl Cancer Inst 2000;92:636–641.

32. Lee WC, Testa JR. Somatic genetic alterations in human malignant
meso-thelioma (review). Int J Oncol 1999;14:181–188.

33. Quelle DE, Zindy F, Ashmun RA, Sherr CJ. Alternative reading frames of the
INK4a tumor suppressor gene encode two unrelated proteins capable of inducing
cell cycle arrest. Cell 1995;83:993–1000.

34. Beer TW, Shepherd P, Pullinger NC. p27 immunostaining is related to
prog¬nosis in malignant mesothelioma. Histopathology 2001;38:535–541.

35. Bongiovanni M, Cassoni P, De Giuli P, et al. p27(kip1) immunoreactivity
correlates with long-term survival in pleural malignant mesothelioma. Cancer
2001;92:1245–1250.

36. Beer TW. Immunohistochemical MIB-1 and p27 as prognostic factors in pleural
mesothelioma. Pathol Res Pract 2001;197:859.

37. Catzavelos C, Tsao MS, DeBoer G, Bhattacharya N, Shepherd FA, Slingerland
JM. Reduced expression of the cell cycle inhibitor p27Kip1 in non-small cell
lung carcinoma: a prognostic factor independent of Ras. Cancer Res
1999;59:684–688.

38. Yang RM, Naitoh J, Murphy M, et al. Low p27 expression predicts poor
disease-free survival in patients with prostate cancer. J Urol 1998;159:941–
945.

39. Fredersdorf S, Burns J, Milne AM, et al. High level expression of p27(kip1)
and cyclin D1 in some human breast cancer cells: inverse correlation between the
expression of p27(kip1) and degree of malignancy in human breast and colorectal
cancers. Proc Natl Acad Sci USA 1997;94:6380–6385.

40. Tan P, Cady B, Wanner M, et al. The cell cycle inhibitor p27 is an
indepen¬dent prognostic marker in small (T1a,b) invasive breast carcinomas.
Cancer Res 1997;57:1259–1263.

41. Isik R, Metintas M, Gibbs AR, et al. p53, p21 and metallothionein
immunoreactivities in patients with malignant pleural mesothelioma: correlations
with the epidemiological features and prognosis of mesothe-liomas with
environmental asbestos exposure. Respir Med 2001;95:588– 593.

42. Baldi A, Groeger AM, Esposito V, et al. Expression of p21 in SV40 large T
antigen positive human pleural mesothelioma: relationship with survival. Thorax
2002;57:353–356.

43. el-Deiry WS, Tokino T, Velculescu VE, et al. WAF1, a potential mediator of
p53 tumor suppression. Cell 1993;75:817–825.

44. Evans DG, Sainio M, Baser ME. Neurofibromatosis type 2. J Med Genet
2000;37:897–904.

45. Rouleau GA, Merel P, Lutchman M, et al. Alteration in a new gene encod¬ing a
putative membrane-organizing protein causes neuro-fibromatosis type 2. Nature
1993;363:515–521.

46. Trofatter JA, MacCollin MM, Rutter JL, et al. A novel moesin-, ezrin-,
radixin-like gene is a candidate for the neurofibromatosis 2 tumor sup¬pressor.
Cell 1993;72:791–800.

47. Sekido Y, Pass HI, Bader S, et al. Neurofibromatosis type 2 (NF2) gene is
somatically mutated in mesothelioma but not in lung cancer. Cancer Res
1995;55:1227–1231.

48. Bianchi AB, Mitsunaga SI, Cheng JQ, et al. High frequency of inactivating
mutations in the neurofibromatosis type 2 gene (NF2) in primary malig¬nant
mesotheliomas. Proc Natl Acad Sci USA 1995;92:10854–10858.

49. Harris CC, Hollstein M. Clinical implications of the p53 tumor-suppressor
gene. N Engl J Med 1993;329:1318–1327.

50. Greatens TM, Niehans GA, Rubins JB, et al. Do molecular markers predict
survival in non-small-cell lung cancer? Am J Respir Crit Care Med 1998;
157:1093–1097.

51. Liu BC, Fu DC, Miao Q, Wang HH, You BR. p53 gene mutations in asbestos
associated cancers. Biomed Environ Sci 1998;11:226–232.

52. Mayall FG, Jacobson G, Wilkins R. Mutations of p53 gene and SV40 sequences
in asbestos associated and non-asbestos-associated mesothe-liomas. J Clin Pathol
1999;52:291–293.

53. Ni Z, Liu Y, Keshava N, Zhou G, Whong W, Ong T. Analysis of K-ras and p53
mutations in mesotheliomas from humans and rats exposed to asbestos. Mutat Res
2000;468:87–92.

54. Pass HI, Stevens EJ, Oie H, et al. Characteristics of nine newly derived
mesothelioma cell lines. Ann Thorac Surg 1995;59:835–844.

55. Unfried K, Roller M, Pott F, Friemann J, Dehnen W. Fiber-specific molecu¬lar
features of tumors induced in rat peritoneum. Environ Health Perspect
1997;105S:1103–1108.

56. Kleymenova EV, Horesovsky G, Pylev LN, Everitt J. Mesotheliomas in¬duced in
rats by the fibrous mineral erionite are independent from p53 alterations.
Cancer Lett 1999;147:55–61.

57. Vaslet CA, Messier NJ, Kane AB. Accelerated progression of asbestos-induced
mesotheliomas in heterozygous p53+/- mice. Toxicol Sci 2002;68: 331–338.

58. Marsella JM, Liu BL, Vaslet CA, Kane AB. Susceptibility of p53-deficient
mice to induction of mesothelioma by crocidolite asbestos fibers. Environ Health
Perspect 1997;105 (suppl 5):1069–1072.

59. Cote RJ, Jhanwar SC, Novick S, Pellicer A. Genetic alterations of the p53
gene are a feature of malignant mesotheliomas. Cancer Res 1991;51:5410– 5416.

60. Creaney J, McLaren BM, Stevenson S, et al. p53 autoantibodies in patients
with malignant mesothelioma: stability through disease progression. Br J Cancer
2001;84:52–56.

61. Giuliano M, Catalano A, Strizzi L, Vianale G, Capogrossi M, Procopio A.
Adenovirus-mediated wild-type p53 overexpression reverts tumouri-genicity of
human mesothelioma cells. Int J Mol Med 2000;5:591– 596.

62. Yang CT, You L, Uematsu K, Yeh CC, McCormick F, Jablons DM. p14(ARF)
modulates the cytolytic effect of ONYX-015 in mesothelioma cells with wild-type
p53. Cancer Res 2001;61:5959–5963.

63. Rudin CM, Thompson CB. Apoptosis and disease: regulation and clinical
relevance of programmed cell death. Annu Rev Med 1997;48:267–281.

64. Harada K, Toyooka S, Shivapurkar N, et al. Deregulation of caspase 8 and 10
expression in pediatric tumors and cell lines. Cancer Res 2002;62: 5897–5901.

65. Shivapurkar N, Toyooka S, Eby MT, et al. Differential inactivation of
caspase-8 in lung cancers. Cancer Biol Ther 2002;1:65–69.

66. Soini Y, Kinnula V, Kaarteenaho-Wiik R, Kurttila E, Linnainmaa K, Paakko P.
Apoptosis and expression of apoptosis regulating proteins bcl-2, mcl-1, bcl-X,
and bax in malignant mesothelioma. Clin Cancer Res 1999;5:3508– 3515.

67. Smythe WR, Mohuiddin I, Ozveran M, Cao XX. Antisense therapy for malignant
mesothelioma with oligonucleotides targeting the bcl-xl gene product. J Thorac
Cardiovasc Surg 2002;123:1191–1198.

68. Mohiuddin I, Cao X, Fang B, Nishizaki M, Smythe WR. Significant
aug¬mentation of pro-apoptotic gene therapy by pharmacologic bcl-xl
down-regulation in mesothelioma. Cancer Gene Ther 2001;8:547–554.

69. Pataer A, Smythe WR, Yu R, et al. Adenovirus-mediated Bak gene transfer
induces apoptosis in mesothelioma cell lines. J Thorac Cardiovasc Surg
2001;121:61–67.

70. van Noesel MM, van Bezouw S, Salomons GS, et al. Tumor-specific
down-regulation of the tumor necrosis factor-related apoptosis-inducing ligand
decoy receptors DcR1 and DcR2 is associated with dense promoter
hyper-methylation. Cancer Res 2002;62:2157–2161.

71. Pfeifer GP, Yoon JH, Liu L, Tommasi S, Wilczynski SP, Dammann R.
Methy-lation of the RASSF1A gene in human cancers. Biol Chem 2002;383:907– 914.

72. Dreijerink K, Braga E, Kuzmin I, et al. The candidate tumor suppressor gene,
RASSF1A, from human chromosome 3p21.3 is involved in kidney tumorigenesis. Proc
Natl Acad Sci USA 2001;98:7504–7509.

73. Dammann R, Takahashi T, Pfeifer GP. The CpG island of the novel tumor
suppressor gene RASSF1A is intensely methylated in primary small cell lung
carcinomas. Oncogene 2001;20:3563–3567.

74. Toyooka S, Carbone M, Toyooka KO, et al. Progressive aberrant methyla-tion
of the RASSF1A gene in simian virus 40 infected human mesothelial cells.
Oncogene 2002;21:4340–4344.

75. Call KM, Glaser T, Ito CY, et al. Isolation and characterization of a zinc
finger polypeptide gene at the human chromosome 11 Wilms’ tumor locus. Cell
1990;60:509–520.

76. Weissman BE, Saxon PJ, Pasquale SR, Jones GR, Geiser AG, Stanbridge EJ.
Introduction of a normal human chromosome 11 into a Wilms’ tumor cell line
controls its tumorigenic expression. Science 1987;236:175–180.

77. Yang Y, Gubler MC, Beaufils H. Dysregulation of podocyte phenotype in
idiopathic collapsing glomerulopathy and HIV-associated nephropathy. Nephron
2002;91:416–423.

78. Little M, Wells C. A clinical overview of WT1 gene mutations. Hum Mutat
1997;9:209–225.

79. Park S, Schalling M, Bernard A, et al. The Wilms tumour gene WT1 is
expressed in murine mesoderm-derived tissues and mutated in a human
mesothelioma. Nat Genet 1993;4:415–420.

80. Amin KM, Litzky LA, Smythe WR, et al. Wilms’ tumor 1 susceptibility (WT1)
gene products are selectively expressed in malignant mesothelioma. Am J Pathol
1995;146:344–356.

81. Foster MR, Johnson JE, Olson SJ, Allred DC. Immunohistochemical analy¬sis of
nuclear versus cytoplasmic staining of WT1 in malignant mesothe-liomas and
primary pulmonary adenocarcinomas. Arch Pathol Lab Med 2001;125:1316–1320.

82. Gubler MC. WT1, a multiform protein. Contribution of genetic models to the
understanding of its various functions. J Am Soc Nephrol 2002;13: 2192–2194.

83. Kleymenova EV, Yuan X, LaBate ME, Walker CL. Identification of a
tumor-specific methylation site in the Wilms tumor suppressor gene. Oncogene
1998;16:713–720.

84. Gazdar AF, Butel JS, Carbone M. SV40 and human tumours: myth, associ¬ation
or causality? Nat Rev Cancer 2002;2:957–964.

85. Bocchetta M, Di Resta I, Powers A, et al. Human mesothelial cells are
unusually susceptible to simian virus 40- mediated transformation and asbestos
cocarcinogenicity. Proc Natl Acad Sci USA 2000;97:10214–10219.

86. Pipas JM, Levine AJ. Role of T antigen interactions with p53 in
tumorigen-esis. Semin Cancer Biol 2001;11:23–30.

87. Carbone M, Rizzo P, Grimley PM, et al. Simian virus-40 large-T antigen binds
p53 in human mesotheliomas. Nat Med 1997;3:908–912.

88. Foddis R, De Rienzo A, Broccoli D, et al. SV40 infection induces telomerase
activity in human mesothelial cells. Oncogene 2002;21:1434–1442.

89. Zanella CL, Timblin CR, Cummins A, et al. Asbestos-induced phosphory-lation
of epidermal growth factor receptor is linked to c-fos and apoptosis. Am J
Physiol 1999;277:L684–693.

90. Cacciotti P, Strizzi L, Vianale G, et al. The presence of simian-virus 40
sequences in mesothelioma and mesothelial cells is associated with high levels
of vascular endothelial growth factor. Am J Respir Cell Mol Biol
2002;26:189–193.

91. Bocchetta M, Miele L, Pass HI, Carbone M. Notch-1 induction, a novel
activity of SV40 required for growth of SV40-transformed human mesothe-lial
cells. Oncogene 2003;22:81–89.