Mutant Ras proteins play a direct causal role in human
cancer. Oncogenic mutant Ras proteins are resistant to
down-regulation by GTPase activating proteins, and
therefore remain in their active, GTP-bound state persis-
tently (Bourne et al. 1991). Table 1 summarizes the fre-
quency with which Ras mutations have been detected in
primary human tumors and cancer-derived cell lines, us-
ing data from the Sanger Center COSMIC database
Although all three Ras genes, H-Ras, N-Ras, and K-
Ras, occur in varying frequencies in different tumor types,
K-Ras is the form that is causally involved in the major
cancers that afflict humans most frequently. The reasons
for these differences have not been determined. The sim-
plest interpretation is that these frequencies reflect levels
of expression of proto-oncogenic forms and therefore re-
flect the relative contributions each protein makes to sig-
nal output. In support of this, mice lacking both H-Ras
Cancer Targets in the Ras Pathway
P. RODRIGUEZ-VICIANA, O. TETSU, K. ODA, J. OKADA, K. RAUEN,
AND F. MCCORMICK
Cancer Research Institute, University of California San Francisco Comprehensive Cancer Center,
San Francisco, California 94115
Cold Spring Harbor Symposia on Quantitative Biology, Volume LXX. © 2005 Cold Spring Harbor Laboratory Press 0-87969-773-3.
Ras proteins play a direct causal role in human cancer and in other diseases. Mutant H-Ras, N-Ras, and K-Ras occur in vary-
ing frequencies in different tumor types, for reasons that are not known. Other members of the Ras superfamily may also con-
tribute to cancer. Mutations also occur in downstream pathways, notably B-Raf, PTEN, and PI 3´ kinase: These pathways in-
teract at multiple points, including cyclin D1, and act synergistically. In some cases mutations in Ras and effectors are
mutually exclusive; in other cases, they coexist. Drugs blocking elements of the pathway are in different stages of clinical de-
velopment. One of these, the Raf kinase/VEGF-R2 inhibitor Sorafenib, has already been approved for treatment of renal can-
cer and is being tested in other indications. However, therapeutic targets in the Ras pathway have not yet been fully validated
as bona fide targets.
Table 1. Ras Pathway Mutations in Human Cancer
% Positive samples
These data were collected from the Sanger Center COSMIC database, December 2005 release. Fre-
quencies of mutation in the four most common cancers are shown in the first four rows.
and N-Ras are viable (Esteban et al. 2001), whereas mice
lacking K-Ras are not (Johnson et al. 1997). However, we
have found that all three Ras proteins are usually ex-
pressed within the same cell, at approximately the same
level, suggesting that the three Ras proteins may have dif-
ferent functions which may be selected for during tumori-
genesis in a tissue-specific fashion.
The first half of H-, K-, and N-Ras, which includes the
switch 1 and 2 domains that are involved in interaction
with effectors, are identical. Consistent with this, we have
not found any difference in the ability of the three Ras pro-
teins to interact with a comprehensive list of effectors. On
the other hand, Ras proteins differ at the carboxyl terminus,
a region involved in membrane localization, and indeed,
H-, K-, and N-Ras have been found in different mi-
crodomains of the plasma membrane, as well as in differ-
ent endomembrane compartments. Although the biological
consequences of this differential localization are not yet
clear, it is possible that the Ras proteins recruit their effec-
tors and signal from different membrane compartments,
and differences may be related to their characteristic muta-
tional spectrum (Rodriguez-Viciana et al. 2004).
Recently, constitutional activating mutations in H-ras
were detected in patients suffering from Costello syn-
drome, a complex developmental disorder characterized
by craniofacial abnormalities, developmental delays, and
a predisposition to neoplasia (Aoki et al. 2005; Estep et
al. 2006; Gripp et al. 2006; Kerr et al. 2006). 34G to A
transitions occur in 91% of affected individuals: This mu-
tation was characterized previously as an activating allele
of moderate transforming activity in focus-forming as-
says and occurs in human cancers. Less frequent muta-
tions include 35G to C also in codon 12, and 37G to T at
codon 13. In one case, it was possible to demonstrate loss
of the normal H-Ras allele in a tumor that developed in a
child suffering from this syndrome. Further analysis of
biological and genetic attributes of this syndrome and re-
lated disorders may be informative in understanding spe-
cific functions of H-Ras in human development. Consti-
tutional activating mutations have also been reported in
the K-Ras gene in Noonan syndrome, a condition related
to Costello syndrome (Schubbert et al. 2006). In these
cases, K-Ras is activated by a mutation that is not fully
transforming and is, in fact, partially sensitive to down-
regulation by GAP. Thus, individuals with fully activated
H-Ras show a similar (although not identical) phenotype
to those with a weak activated allele of K-Ras, confirm-
ing the suggestion that K-Ras is a more potent oncogene
in vivo. Interestingly, part of the Ras pathway is activated
by mutation in another syndrome, cardio-facial-cuta-
neous syndrome: In these individuals, activating muta-
tions in B-Raf, MEK1, and MEK2 appear to be the cause
of the disease (Rodriguez-Viciana et al. 2006). Interest-
ingly, the mutations that activate B-Raf are generally not
found in cancers, and indeed, these affected individuals
are not cancer-prone. In addition, activating mutations in
MEK1 and MEK2 have not been found in human cancers,
although they appear to be fully active by preliminary
Structures of Ras proteins have been solved in several
states: inactive, GDP-bound; active, bound to GppNHp
(nonhydrolyzable GTP analog), position-12 mutant ver-
sions, complexes with exchange factors, GAPs, and ef-
fectors (Wittinghofer and Pai 1991). Despite this wealth
of information, strategies for identifying Ras inhibitors
have not been forthcoming. Considering the extremely
high affinity of Ras proteins for GTP (Kd, 1 pM) it seems
unlikely that competitive inhibitors could be identified.
Screens for small molecules that restore GTP hydrolysis
of GTP bound to Ras failed to identify lead compounds,
and analysis of high-resolution structures of mutant pro-
teins suggested that this approach was doomed to fail.
Activating substitutions were thought to present a steric
block to GTP hydrolysis by preventing attack of γ-phos-
phate by a water molecule. This view has been revised re-
cently: The structure of Ras bound in a transition state
complexed with GDP.AlF4 and GAP suggests failure to
hydrolyze GTP is the result of displacement of critical
catalytic residues from GAP (the “arginine finger”) rather
than steric block. Consistent with this, a GTP analog in
which an amino group is covalently attached to the γ-
phosphate is hydrolyzed efficiently by mutant Ras pro-
teins (Ahmadian et al. 1999). Whether this presents an
opportunity for therapeutic intervention remains to be
seen. Meanwhile, most approaches to blocking Ras activ-
ity depend on inhibition of downstream effectors rather
than Ras itself. Considerable efforts expended on block-
ing Ras processing through farnesyl transferase were
thwarted by the existence of a “back-up” modification
that enables K-Ras to remain active through geranylger-
anyl modification instead of farnesylation. However, sev-
eral inhibitors have been developed extensively and
tested in clinical trials. Since H-Ras is not geranylger-
anylated in the absence of a farnesyl group, it is inhibited
effectively by these compounds. Furthermore, H-Ras ap-
pears to be dispensable in normal tissues, at least in mice,
suggesting that diseases driven by activated H-Ras would
likely respond to these inhibitors without significant side
effects on normal tissue.
INCREASING COMPLEXITY OF Ras
It was established in 1984 that Raf function was re-
quired for Ras transformation, and a direct interaction of
Ras and Raf and mechanisms of Ras-dependent Raf acti-
vation were described in 1992 and thereafter. Very soon,
however, it became apparent that Ras can also interact
with and activate other effectors, including class I PI 3´-
kinase (PI3´K) and RalGEFs. Furthermore, Ras effectors
were shown to act synergistically, implying that the full
transforming potential of Ras depends on simultaneous
activation of interacting downstream effector pathways
(Fig. 1). Other Ras effectors identified from two-hybrid
screens and other approaches include AF-6, RIN, PLCε,
Nore1/Rassf5, IMP, and many others that await charac-
terization (Rodriguez-Viciana et al. 2004). Analysis of
mutations in human tumors has validated both the Raf
and PI3´K pathways as crucial Ras effectors in human tu-
morigenesis (see below). However, data from a variety of
experimental systems, including knock-out mice defi-
cient for other effectors such as RalGDS, PLCε, and
462RODRIGUEZ-VICIANA ET AL.
TARGETS IN THE Ras PATHWAY463
With their ability to activate both the Raf and PI3K
pathways, the R-Ras subgroup of Ras family GTPases
may play a role in human cancer. In our hands, an acti-
vated L81 M-Ras mutant is as potent an oncogene as ac-
tive Ras in a variety of classic transformation assays, and
mutations in TC21 have already been found in a handful
of human tumors (Rodriguez-Viciana et al. 2004). De-
termination of whether mutational activation of these
GTPases is a rare or frequent way of deregulating the Ras
pathway in tumors with wild-type Ras will have to await
a comprehensive genetic analysis of human tumors, as
has already been performed for other gene families. We
have analyzed M-Ras sequences in a panel of 60 breast
cancer cell lines and have so far failed to detect activating
mutations in this gene (J. Gray and F. McCormick, un-
publ.). A further degree of complexity is illustrated by our
observation that, in some cases, Ras family GTPases can
cross-talk to each other and cooperate in the activation of
the same effector pathway. M-Ras, when activated, can
target a phosphatase holoenzyme complex made up of
Shoc2 and the catalytic subunit of protein phosphatase 1
to remove a negative regulatory phosphate group from
Raf kinase molecules complexed with active Ras pro-
teins, thereby further stimulating Raf-specific activity
(Fig. 3). Importantly, Shoc2 function is essential for ERK
activity in tumor cells with mutant Ras, and therefore, the
Shoc2-PP1C holoenzyme represents an additional target
for pharmacological inhibition of the ERK pathway (P.
Tiam1, suggest that other effectors will likely make an
important contribution to Ras-induced tumor formation.
Activated Ras has been implicated in many of the proper-
ties of the malignant phenotype (i.e., uncontrolled prolif-
eration, survival, invasion, and metastasis). It is likely
that different effector pathways (or combinations of
them) will contribute differentially to the various aspects
of tumor biology in different cell types as well as during
different stages of tumor progression.
In addition to the prototypic Ras proteins (H-, K-, and
N-Ras), the Ras family includes other closely related
GTPases that share many of the biochemical and biolog-
ical properties of Ras, including the ability to behave as
oncogenes (Fig. 2). In particular, members of the R-Ras
subgroup of the Ras family, which includes R-Ras,
TC21/R-Ras2, and M-Ras/R-Ras3, have a high degree of
overlap in their ability to interact with and regulate many
of Ras known effectors. For example, H-, N-, and K-Ras,
R-Ras, TC21, and M-Ras have similar abilities to activate
the RalGEF family of effectors and the p110α and γ iso-
forms of class I PI3K. H-, N-, and K-Ras, on the other
hand, are the strongest at interacting with and activating
Raf kinases. R-Ras and TC21, however, are the only
GTPases that can activate the p110δ PI3K (Rodriguez-
Viciana et al. 2004). Individual GTPases have specific
blueprints of effector interactions, and their signaling and
biological properties should be considered in the context
of the full spectrum of their many effector interactions.
Figure 1. Signaling pathways downstream of Ras.
Rodriguez-Viciana and F. McCormick, in prep.). Clearly,
a better understanding of the many effector pathways reg-
ulated by Ras and its closely related GTPases will likely
lead to the identification of novel targets of therapeutic
intervention in human cancer.
CANCER GENETICS AND RAS EFFECTORS
Until recently, the PI 3´K pathway appeared to be the
more critical Ras effector pathway in human cancer,
based on the observation that mutations occur fre-
quently in major regulators of this pathway, including
PI3´K itself and the lipid phosphate PTEN. Activating
mutations in Akt2, PDK1, and PAK4 have also been re-
ported at lower frequency. In 2002, frequent activation
of B-Raf was reported in many types of human cancer,
most notably malignant melanoma (Davies et al. 2002).
This discovery confirmed the importance of the MAPK
pathway in cancer and encouraged development of
drugs that target enzymes in the pathway. In addition,
the coexistence of mutations in the Ras pathway in hu-
man tumors has led to new insights relating to signal
cross-talk and codependence. For example, activating
mutations in N-Ras are mutually exclusive with muta-
464 RODRIGUEZ-VICIANA ET AL.
Figure 2. The Ras superfamily of small GTPases.
Figure 3. Model for the action of R-Ras3 (M-Ras) on Raf kinase, through recruitment of SHOC-2/PP1C complexes to the plasma
tions in PTEN in this disease, and also with mutations in
B-Raf. This suggests that both major Ras effector arms
need to be activated to sustain the malignant phenotype,
and that the combined effect of activated MAPK and
PI3´K can be achieved by a single mutation in N-Ras, or
two independent mutations in B-Raf and PTEN. These
two pathways may converge at several key points: For
example, p27 expression is increased by hyperactiva-
tion of MAPK, potentially leading to growth arrest.
However, the PI3´K pathway targets p27 for degrada-
tion. These observations suggest an obvious molecular
basis for coactivation of these pathways. Another point
of intersection may be cyclin D1. Transcription of cy-
clin D1 is regulated by the MAPK pathway, but cyclin
D1 protein is stabilized by the PI3´K pathway; high lev-
els of cyclin D1 protein obviously need inputs from both
signaling pathways. Consistent with a role for cyclin D1
in malignant melanoma, this gene is amplified in rare
forms of melanoma in which Ras, B-Raf, and PTEN are
wild type (Bastian et al. 2001).
Transcriptional regulation of cyclin D1 involves several
different types of transcription factors such as AP-1/c-Jun
(Albanese et al. 1995; Bakiri et al. 2000), CREB (Tetsu
and McCormick 1999), NF-κB (Guttridge et al. 1999;
Hinz et al. 1999), and c-Ets (Albanese et al. 1995; Tetsu
and McCormick 1999). In addition, we showed that cyclin
D1 is a major transcriptional target of the APC/β-
catenin/TCF signaling pathway (Tetsu and McCormick
1999; Hulit et al. 2004; Lepourcelet et al. 2004). Increased
expression of cyclin D1 protein therefore depends on hy-
peractive signals from a number of pathways. Stabiliza-
tion of the protein through activation of the PI3´K path-
way has also been reported, suggesting that signals both
from the RAF/MEK/ERK pathway, leading to transcrip-
tional activation, and from the PI3´K pathway, leading to
posttranscriptional activation, may be necessary: A simi-
lar situation has been described for Mdm2, in which the
RAF/MEK/ERK pathway increases transcription, and the
PI3´K pathway leads to stabilization.
In contrast to the relatively simple relationship be-
tween Ras and its major effectors in melanoma, muta-
tions in K-Ras, PTEN, and PI3´K coexist in other types of
cancer: Mutations in K-Ras and PI3´K often coexist in
colon cancer, for example. We have analyzed mutations
in the PIK3CA gene in 66 endometrial carcinoma pa-
tients. We identified coexistence of PTEN and PIK3CA
mutations in 26% (17/66) of patients (Oda et al. 2005).
Tumors with PTEN mutation showed a tendency to carry
PIK3CA mutation more frequently (17/37 = 46%) than
tumors without PTEN mutation (7/29 = 24%), although
statistical significance was not reached (p = 0.078 in
Fisher’s exact test). Subsequently, we evaluated the rela-
tionship between PIK3CA mutation and other clinico-
pathological factors. There was no evidence of an associ-
ation of PIK3CAmutations with histological grade, FIGO
stage, lymph node metastasis, and ER/PgR status. These
data are in striking contrast to those of breast carcinoma,
showing that PIK3CA mutations correlate with expres-
sion of hormone receptors and node metastasis, and are
mutually exclusive with loss of PTEN expression. How-
ever, 2/8 (25%) PTEN mutant breast carcinomas also
possessed PIK3CA mutations, suggesting that coexis-
tence of PTEN/PIK3CA mutations could occur in other
tumors as well (Oda et al. 2005).
Coexistence of mutations in both PTEN and PIK3CA
may imply that more than one input activating the
PI3K/Akt pathway is required to completely activate this
pathway in endometrial cancers, but not in melanoma. In
breast carcinoma, PIK3CA mutations correlate with
ErbB2 overexpression, suggesting that another activating
event may be necessary to fully activate the PI3K path-
way. Alternatively, either PTEN or p110α may possess
additional function(s) distinct from the PI3K pathway.
DEVELOPMENT OF SORAFENIB/NEXAVAR,
A Raf KINASE INHIBITOR
In 1992, the drug discovery group at ONYX Pharma-
ceuticals screened a library of compounds for inhibitors
of Raf-1 kinase activity. As a source of active kinase for
this screen, Raf-1 was expressed in Sf9 cells using bac-
ulovirus vectors. The protein had been engineered with
an epitope tag to facilitate purification, and cells were
co-infected with a vector expressing v-src to activate
Raf-1 kinase. A compound that was selective for Raf-1
was identified and a medicinal chemistry program was
initiated to identify derivatives with improved proper-
ties. The compound shown in Figure 4, first referred to
as BAY43-9006, then Sorafenib, entered Phase I clini-
cal trials in 2000. At that time, it was assumed that inhi-
bition of Raf kinase in cancer cells would reverse as-
pects of the transformed phenotype. This assumption
was based on experimental systems in which Raf kinase
was inhibited using microinjected antibodies or domi-
nant negative constructs. These approaches suggested
that Raf kinase is an appropriate target for intervention,
but did not predict side effects through Raf inhibition in
normal tissue, and did not predict the consequences of
blocking Ras in cancers in vivo. Furthermore, it was not
clear whether Raf kinase is activated in tumors that do
not harbor oncogenic Ras alleles. For these reasons,
Phase I trials were launched for all types of cancers,
without bias toward cancers with high frequencies of
Ras mutation. During these trials, stabilization of dis-
ease was noted in patients suffering from renal cell car-
cinoma, a disease with no obvious association with acti-
vated Raf kinase. At this time, the specificity of
Sorafenib/Nexavar was scrutinized, and a potent effect
on VEGF-R2 was observed: This may well account for
clinical effects in renal cell carcinoma, since this disease
is associated with activated VEGF signaling through
loss of the VHL tumor suppressor (Wilhelm et al. 2004).
After completion in 2005 of a successful Phase III trial,
in which a significant improvement in time to progres-
sion was reported, Nexavar was approved in the U.S. for
treatment of metastatic renal cell carcinoma (Fig. 5).
In 2002, existence of BRAF mutations in human can-
cers was first reported. Activation frequencies approach-
ing 70% were detected in malignant melanoma, and sig-
nificant frequencies in several other human cancers
(Davies et al. 2002). On the basis of these discoveries,
TARGETS IN THE Ras PATHWAY465
clinical trials were initiated in which Nexavar was tested
in patients suffering from malignant melanoma. To date,
little clinical activity has been observed as a single
agent, however, even though the drug appeared to hit the
target effectively, i.e., MEK phosphorylation was
blocked significantly. Lack of efficacy could be because
(1) late-stage melanomas may no longer be Raf-depen-
dent; (2) Raf inhibition causes growth arrest, but not
apoptosis; or (3) the drug is not potent enough. Relating
perhaps to the second point, BRAF mutations frequently
coexist with PTEN mutations, as described above. Loss
of PTEN has been reported to render other targeted ther-
apies ineffective, presumably because cells in which
PI3´K has been up-regulated are more difficult to kill.
Nevertheless, Nexavar is currently being tested in com-
bination therapies and as a single agent in multiple dis-
ease indications. However, the clinical value of blocking
the Ras or Raf pathway remains uncertain.
466 RODRIGUEZ-VICIANA ET AL.
Figure 4. Structure of the lead compound identified in a screen for Raf kinase inhibitors, and structure of Sorafenib.
Figure 5. Increased survival in patients treated with Sorafenib/Nexavar as an orally active single agent, in patients suffering from
renal cell carcinoma.
Ras proteins and their downstream effectors play direct
causal roles in human cancer. Analysis of mutations in
these pathways reveals that the MAPK pathway and the
PI3K pathways are activated frequently in cancers, in ad-
dition to K-Ras, and, to a lesser degree, N-Ras and H-Ras.
To date, only one drug (Nexavar) targeting the Ras path-
way has been approved for cancer treatment, but a num-
ber of others are under development. The clinical value of
targeting these pathways remains unclear, since Nexavar
was approved for treatment of renal cell cancer based,
most likely, on its activity against VEGF-R2 rather than
Raf kinase. However, we expect that single agents and
combinations of agents targeting key enzymes in these
pathways will play a powerful role in cancer treatment in
the future, and that their successful clinical development
will depend on a clear understanding of the complexities
of the entire Ras pathway.
Ahmadian M.R., Zor T., Vogt D., Kabsch W., Selinger Z., Wit-
tinghofer A., and Scheffzek K. 1999. Guanosine triphos-
phatase stimulation of oncogenic Ras mutants. Proc. Natl.
Acad. Sci. 96: 7065.
Albanese C., Johnson J., Watanabe G., Eklund N., Vu D.,
Arnold A., and Pestell R.G. 1995. Transforming p21ras mu-
tants and c-Ets-2 activate the cyclin D1 promoter through dis-
tinguishable regions. J. Biol. Chem. 270: 23589.
Aoki Y., Niihori T., Kawame H., Kurosawa K., Ohashi H.,
Tanaka Y., Filocamo M., Kato K., Suzuki Y., Kure S., and
Matsubara Y. 2005. Germline mutations in HRAS proto-
oncogene cause Costello syndrome. Nat. Genet. 37: 1038.
Bakiri L., Lallemand D., Bossy-Wetzel E., and Yaniv M. 2000.
Cell cycle-dependent variations in c-Jun and JunB phos-
phorylation: A role in the control of cyclin D1 expression.
EMBO J. 19: 2056.
Bastian B.C., LeBoit P.E., and Pinkel D. 2001. Genomic ap-
proaches to skin cancer diagnosis. Arch. Dermatol. 137: 1507.
Bourne H.R., Sanders D.A., and McCormick F. 1991. The
GTPase superfamily: Conserved structure and molecular
mechanism. Nature 349: 117.
Davies H., Bignell G.R., Cox C., Stephens P., Edkins S., Clegg
S., Teague J., Woffendin H., Garnett M.J., Bottomley W.,
Davis N., Dicks E., Ewing R., Floyd Y., Gray K., Hall S.,
Hawes R., Hughes J., Kosmidou V., Menzies A., Mould C.,
Parker A., Stevens C., Watt S., Hooper S., Wilson R., Jay-
atilake H., Gusterson B.A., Cooper C., Shipley J., Hargrave
D., Pritchard-Jones K., Maitland N., Chevenix-Trench G.,
Riggins G.J., Bigner D.D., Palmieri G., Cossu A., Flanagan
A., Nicholson A., Ho J.W., Leung S.Y., Yuen S.T., Weber
B.L., Seigler H.F., Darrow T.L., Paterson H., Marais R.,
Marshall C.J., Wooster R., Stratton M.R., and Futreal P.A.
2002. Mutations of the BRAF gene in human cancer. Nature
Esteban L.M., Vicario-Abejon C., Fernandez-Salguero P., Fer-
nandez-Medarde A., Swaminathan N., Yienger K., Lopez E.,
Malumbres M., McKay R., Ward J.M., Pellicer A., and San-
tos E. 2001. Targeted genomic disruption of H-ras and N-ras,
individually or in combination, reveals the dispensability of
both loci for mouse growth and development. Mol. Cell. Biol.
Estep A.L., Tidyman W.E., Teitell M.A., Cotter P.D., and Rauen
K.A. 2006. HRAS mutations in Costello syndrome: Detection
of constitutional activating mutations in codon 12 and 13 and
loss of wild-type allele in malignancy. Am. J. Med. Genet. A
Gripp K.W., Lin A.E., Stabley D.L., Nicholson L., Scott C.I., Jr.,
Doyle D., Aoki Y., Matsubara Y., Zackai E.H., Lapunzina P.,
Gonzalez-Meneses A., Holbrook J., Agresta C.A., Gonzalez
I.L., and Sol-Church K. 2006. HRAS mutation analysis in
Costello syndrome: Genotype and phenotype correlation. Am.
J. Med. Genet. A 140: 1.
Guttridge D.C., Albanese C., Reuther J.Y., Pestell R.G., and
Baldwin A.S., Jr. 1999. NF-kappaB controls cell growth and
differentiation through transcriptional regulation of cyclin
D1. Mol. Cell. Biol. 19: 5785.
Hinz M., Krappmann D., Eichten A., Heder A., Scheidereit C.,
and Strauss M. 1999. NF-kappaB function in growth control:
Regulation of cyclin D1 expression and G0/G1-to-S-phase
transition. Mol. Cell. Biol. 19: 2690.
Hulit J., Wang C., Li Z., Albanese C., Rao M., Di Vizio D., Shah
S., Byers S.W., Mahmood R., Augenlicht L.H., Russell R.,
and Pestell R.G. 2004. Cyclin D1 genetic heterozygosity reg-
ulates colonic epithelial cell differentiation and tumor number
in ApcMin mice. Mol. Cell. Biol. 24: 7598.
Johnson L., Greenbaum D., Cichowski K., Mercer K., Murphy
E., Schmitt E., Bronson R.T., Umanoff H., Edelmann W.,
Kucherlapati R., and Jacks T. 1997. K-ras is an essential gene
in the mouse with partial functional overlap with N-ras.
Genes. Dev. 11: 2468.
Kerr B., Delrue M.A., Sigaudy S., Perveen R., Marche M.,
Burgelin I., Stef M., Tang B., Eden T., O’Sullivan J., De San-
dre-Giovannoli A., Reardon W., Brewer C., Bennett C., Quar-
rell O., McCann E., Donnai D., Stewart F., Hennekam R.,
Cave H., Verloes A., Philip N., Lacombe D., Levy N.,
Arveiler B., and Black G. 2006. Genotype-phenotype correla-
tion in Costello syndrome; HRAS mutation analysis in 43
cases. J. Med. Genet. (in press).
Lepourcelet M., Chen Y.N., France D.S., Wang H., Crews P.,
Petersen F., Bruseo C., Wood A.W., and Shivdasani R.A.
2004. Small-molecule antagonists of the oncogenic Tcf/beta-
catenin protein complex. Cancer Cell 5: 91.
Oda K., Stokoe D., Taketani Y., and McCormick F. 2005. High
frequency of coexistent mutations of PIK3CA and PTEN
genes in endometrial carcinoma. Cancer Res. 65: 10669.
Rodriguez-Viciana P., Sabatier C., and McCormick F. 2004.
Signaling specificity by Ras family GTPases is determined by
the full spectrum of effectors they regulate. Mol. Cell. Biol.
Rodriguez-Viciana P., Tetsu O., Tidyman W.E., Estep A.L.,
Conger B.A., Santa Cruz M., McCormick F., and Rauen K.A.
2006. Germline mutations in genes within the MAPK pathway
cause cardio-facio-cutaneous syndrome. Science 311: 1287.
Schubbert S., Zenker M., Rowe S.L., Boll S., Klein C., Bollag G.,
van der Burgt I., Musante L., Kalscheuer V., Wehner L.E.,
Nguyen H., West B., Zhang K.Y., Sistermans E., Rauch A.,
Niemeyer C.M., Shannon K., and Kratz C.P. 2006. Germline
KRAS mutations cause Noonan syndrome. Nat. Genet. 38:
Tetsu O. and McCormick F. 1999. Beta-catenin regulates expres-
sion of cyclin D1 in colon carcinoma cells. Nature 398: 422.
Wilhelm S.M., Carter C., Tang L., Wilkie D., McNabola A.,
Rong H., Chen C., Zhang X., Vincent P., McHugh M., Cao
Y., Shujah J., Gawlak S., Eveleigh D., Rowley B., Liu L., Ad-
nane L., Lynch M., Auclair D., Taylor I., Gedrich R., Vozne-
sensky A., Riedl B., Post L.E., Bollag G., and Trail P.A. 2004.
BAY 43-9006 exhibits broad spectrum oral antitumor activity
and targets the RAF/MEK/ERK pathway and receptor tyro-
sine kinases involved in tumor progression and angiogenesis.
Cancer Res. 64: 7099.
Wittinghofer A. and Pai E.F. 1991. The structure of Ras protein:
A model for a universal molecular switch. Trends Biochem.
Sci. 16: 382.
TARGETS IN THE Ras PATHWAY 467