ras signaling dysregulation in human embryonal rhabdomyosarcoma

GENES, CHROMOSOMES & CANCER 48:975–982 (2009)

RAS Signaling Dysregulation in Human EmbryonalRhabdomyosarcoma

Simone Martinelli,1* Heather P. McDowell,2,3 Silvia Delle Vigne,1 George Kokai,4

Stefania Uccini,5 Marco Tartaglia,1† and Carlo Dominici2,6,7†

1Departmentof Hematology,Oncologyand Molecular Medicine,Istituto Superiore di Sanita' ,Rome,Italy2Departmentof Pediatrics,‘‘Sapienza’’University,Rome,Italy3Departmentof Oncology,Alder Hey Children’s NHSFoundationTrust,Liverpool,UK4Departmentof Pathology,Alder Hey Children’s NHSFoundationTrust,Liverpool,UK5Departmentof Experimental Medicine/Pathology,‘‘Sapienza’’University,Rome,Italy6Laboratoryof Oncology,‘‘Bambino Gesu' ’’Children’s Hospital,Rome,Italy7School of Reproductive and Developmental Medicine,Liverpool University,Liverpool,UK

Rhabdomyosarcoma (RMS) is a common childhood solid tumor, resulting from dysregulation of the skeletal myogenesis

program. Two major histological subtypes occur in childhood RMS, embryonal and alveolar. While chromosomal rearrange-

ments account for the majority of alveolar tumors, the genetic defects underlying the pathogenesis of embryonal RMS

remain largely undetermined. A few studies performed on small series of embryonal tumors suggest that dysregulation of

RAS function may be relevant to disease pathogenesis. To explore further the biological and clinical relevance of mutations

with perturbing consequences on RAS signaling in embryonal RMS, we investigated the prevalence of PTPN11, HRAS, KRAS,

NRAS, BRAF, MEK1, and MEK2 mutations in a relatively large cohort of primary tumors. While HRAS and KRAS were found

to be rarely mutated, we identified somatic NRAS lesions in 20% of cases. All mutations were missense and affected codon

61, with the introduction of a positive charged amino acid residue representing the most common event. PTPN11 was

found mutated in one tumor specimen, confirming that somatic defects in this gene are relatively uncommon in RMS, while

no mutation was observed in BRAF and MEK genes. Although no clear association of mutations with any clinical variable

was observed, comparison of the outcome between mutation-positive and mutation-negative cases indicated a trend for a

higher percentage of patients exhibiting a better outcome in the former. Our findings provide evidence that dysregulation

of RAS signaling is a major event contributing to embryonal RMS pathogenesis. VVC 2009 Wiley-Liss, Inc.

INTRODUCTION

Rhabdomyosarcoma (RMS) represents the most

common malignant soft tissue tumor in children

aged 0–14 years, accounting for nearly 50% of all

soft tissue sarcomas in childhood (Gurney et al.,

1999). It is a highly malignant tumor that

originates from mesenchymal precursor cells

committed to undergo myogenesis, and closely

resembles early stages of prenatal skeletal muscle

differentiation (Parham and Ellison, 2006). Child-

hood RMS is subdivided into two major subtypes,

embryonal and alveolar, which are characterized

by distinct histological features (Patton and Horn,

1962). Age at diagnosis, distribution of primary

site and outcome are rather different in the two

subtypes because embryonal tumors are more fre-

quent among children aged 0–4 years, occur at

sites throughout the body, with head, neck, and

genitourinary tract being the most common, and

are associated with a better outcome, while alveo-

lar tumors tend to be equally frequent throughout

childhood, occur most commonly at extremities

and trunk, and carry a worse outcome (Gurney

et al., 1999; McDowell, 2003). These quite

distinctive morphological and clinical features are

thought to be related to the different genetic

alterations that take place in mesenchymal

precursors during tumorigenesis of each subtype.

Alveolar RMS frequently harbor either the

recurrent t(2;13)(q35;q14) or the less common

t(1;13)(p36;q14) chromosomal translocation,

resulting in overexpression of the PAX3-FOXO1

and PAX7-FOXO1 chimeric transcription factors,

Additional Supporting Information may be found in the onlineversion of this article.

yMarco Tartaglia and Carlo Dominici contributed equally tothis work.

Supported by: Telethon-Italy grant, Grant number: GGP07115.

*Correspondence to: Simone Martinelli, Department of Hema-tology, Oncology and Molecular Medicine, Istituto Superiore diSanita, Viale Regina Elena, 299, Rome 00161, Italy. E-mail:[email protected]

Received 21 April 2009; Accepted 20 July 2009

DOI 10.1002/gcc.20702

Published online 13 August 2009 inWiley InterScience (www.interscience.wiley.com).

VVC 2009 Wiley-Liss, Inc.

respectively, which, in turn, inappropriately acti-

vate transcription of PAX3/PAX7 target genes

(Xia et al., 2002). Differently, in embryonal RMS,

no consistent chromosomal rearrangement has

been identified, even although genomic instabil-

ity and allelic imbalances at 11p15.5 have consis-

tently been reported (Xia et al., 2002).

Although a growing number of reports recently

documented that molecular defects affecting

genes with role in the TP53, RB1, and MYC

pathways, as well as telomere length abnormal-

ities and aberrant microRNA expression occur in

embryonal RMS (Xia et al., 2002; Ohali et al.,

2008; Subramanian et al., 2008; Ciarapica et al.,

2009), a number of reports consistently suggested

that dysregulation of RAS signaling is relevant to

the pathogenesis of this cancer (Stratton et al.,

1989; Wilke et al., 1993; Yoo et al., 1999; Reed

and Gutmann, 2001; Chen et al., 2006; Kratz

et al., 2007), even although most of these studies

were based on relatively small cohorts of patients.

RAS proteins participate in multiple signal trans-

duction pathways controlling cell proliferation,

differentiation and survival, and function as

GDP/GTP-regulated molecular switches to con-

trol intracellular signal flow (Mitin et al., 2005).

GDP/GTP cycling is controlled by GTPase acti-

vating proteins (GAPs) that accelerate the intrin-

sic GTPase activity, and guanylyl exchanging

factors (GEFs), which promote release of GDP

(Herrmann, 2003). Activating mutations of RASgenes occur in more than one-third of human

cancers (Bos, 1989). These substitutions lock

RAS in its GTP-bound active state by impairing

the intrinsic GTPase activity and/or conferring re-

sistance to GAPs (Trahey and McCormick, 1987).

In this study, we explored the molecular spec-

trum and clinical relevance of somatic RAS,

PTPN11, BRAF, and MEK gene mutations in a

relatively large cohort of embryonal RMS tumors.

Our findings indicate that a distinctive spectrum

of gain-of-function RAS defects represents a com-

mon molecular event during embryonal RMS de-

velopment, and that mutations tend to occur

more frequently among patients exhibiting a bet-

ter outcome.

MATERIALS AND METHODS

Patients

Thirty-one patients, 19 males and 12 females,

aged 3–183 months (median ¼ 56), with previ-

ously untreated embryonal RMS, admitted

between 1996 and 2005 at the Department of

Pediatrics, ‘‘Sapienza’’ University (Rome, Italy)

and Department of Oncology, Alder Hey Child-

ren’s NHS Foundation Trust (Liverpool, UK)

were retrospectively included in the study. The

DNA samples were obtained from tumors with a

histopathological diagnosis of embryonal subtype

and negative for expression of PAX3-FOXO1 and

PAX7-FOXO1 chimeric transcripts, evaluated by

RT-PCR screening (Barr et al., 1995). Primary

sites included head and neck parameningeal (N¼ 3); genitourinary, bladder or prostate (N ¼ 7);

genitourinary, nonbladder or nonprostate (N ¼ 9);

extremities (N ¼ 3); and others (N ¼ 9). Patients

were grouped according to the Intergroup Rhab-

domyosarcoma Study (IRS) postsurgical grouping

system (Maurer et al., 1988), using conventional

techniques of imaging and marrow examination,

and assigned as group I (N ¼ 4), II (N ¼ 12), III

(N ¼ 13), or IV (N ¼ 2). In patients with local-

ized disease (Groups I–III), treatment was admin-

istered according to the MMT-89 protocol of the

International Society of Pediatric Oncology

(SIOP) (Stevens et al., 2005), or to the RMS88 or

RMS96 protocol of the Italian Association of Pe-

diatric Hematology/Oncology (AIEOP) (Ferrari

et al., 2002), while in patients with metastatic dis-

ease (group IV) treatment was given according to

the SIOP MMT4-89 and MMT4-91 protocols

(Carli et al., 2004). Institutional written informed

consent was obtained from the patient’s parents

or legal guardians. The study underwent ethical

review and approval according to local institu-

tional guidelines. As of December 2008,

20 patients were disease-free (DF), 1 alive with

disease (AWD), and 10 dead of disease (DOD).

In the 21 DF or patients with AWD, median fol-

low-up was 75 months (range, 36–147); in the

10 patients with DOD, median follow-up was

24.5 months (range, 13–54). The principal clinical

features of all of the patients included in the

study have been summarized in Supporting

Information Table 1.

Cell Lines

Human RD, RD18, and CCA embryonal RMS

cell lines were kindly provided by Pierluigi Lol-

lini (McDowell et al., 2007). All cell lines were

grown in DMEM medium supplemented with

10% FCS (FCS) (Sigma, Dorset, UK), 1% L-glu-

tamine, and 1% streptomycin/penicillin. CCA cell

line was incubated in 7% CO2, while the remain-

ing cell lines in 5% CO2.

976 MARTINELLI ET AL.

Genes, Chromosomes & Cancer DOI 10.1002/gcc

Tumor Tissue Handling

Tumor tissue was obtained at diagnosis by

open surgery. A fraction of biopsy was formalin-

fixed and paraffin-embedded for histological anal-

ysis, while the remaining was immediately snap

frozen in liquid nitrogen and stored at �80�C

until molecular analyses were carried out. Cryo-

stat sections of frozen tissue samples used in this

study were characterized by tumor cells repre-

senting �90% of the total cell content.

Molecular Analysis

The entire HRAS, KRAS, and NRAS coding

exons and flanking intronic stretches were

screened for mutations. Primer sequences and po-

lymerase chain reactions (PCR) conditions are

available on request. On the basis of our previ-

ously generated data, mutational analysis was also

performed on exons 2, 3, 4, and 13 of PTPN11,

exons 6, 11, 12, 14, and 15 of BRAF, and exons 2

and 3 of MEK1 and MEK2, as previously

described (Tartaglia et al., 2002; Dentici et al.,

2009; Sarkozy et al., 2009). Unpurified PCR prod-

ucts were analyzed by denaturing high-perform-

ance liquid chromatography (DHPLC), using the

Wave 2100 DNA fragment analysis system

(Transgenomic, Omaha, NE) at column tempera-

tures recommended by the Navigator version

1.5.4.23 software (Transgenomic). Heterozygous

templates with previously identified mutations or

single nucleotide polymorphisms were used as

positive controls in each DHPLC run. Amplimers

having abnormal denaturing profiles were ream-

plified, purified (Microcon PCR; Millipore, Bed-

ford, MA) and sequenced bidirectionally using

the ABI BigDye Terminator Sequencing Kit ver-

sion 1.1 (Applied Biosystems, Carlsbad, CA) and

an ABI Prism 310 Genetic Analyzer (Applied Bio-

systems). Sequencing results were analyzed using

the Sequencing Analysis version 3.6.1 and

AutoAssembler version 2.1 software packages

(both from Applied Biosystems). Loss of hetero-

zygosity (LOH), uniparental disomy (UPD), and

possible homozygous condition for sequence vari-

ation at 11p15.5 was explored by DHPLC analy-

sis performed on the entire HRAS coding

sequence by using pooled DNAs.

Categorization of Clinical Variables

and Statistical Analysis

Clinical variables were categorized as follows:

age at diagnosis, favorable (12–120 months) ver-

sus unfavorable (<12 months or >120 months);

primary site, favorable (orbit and genitourinary

nonbladder or nonprostate) versus unfavorable

(head and neck parameningeal and nonparame-

ningeal, genitourinary bladder or prostate,

extremities and others); size of primary tumor,

�5 cm versus >5 cm; Groups I and II versus III

and IV. Clinical end point was overall survival

(OS), which was calculated from the date of his-

tological diagnosis until time of last follow-up or

death. Associations between RAS/PTPN11 muta-

tions and clinical variables were analyzed using

two-tailed Fisher’s exact test. The impact of

RAS/PTPN11 mutations on OS was evaluated

using Kaplan–Meier method, and log-rank test

was used for comparing survivals. Exact confi-

dence intervals of proportions (at 95% level) were

calculated based on binomial distribution. Statisti-

cal analyses were performed using the SPSS soft-

ware package (SPSS, Chicago, IL).

RESULTS

Prevalence of HRAS, KRAS, and NRAS muta-

tions in a pediatric cohort of 31 primary embry-

onal RMS was explored by DHPLC analysis and

direct sequencing. HRAS and NRAS lesions were

identified in 1 (3.2% of cases, 95% CI: 0.08–

16.7%) and 6 (19.4% of cases, 95% CI: 7.5–

37.5%) tumors, respectively (Table 1), while no

mutation was observed to affect the KRAS gene

(95% CI: 0–11.2%). Most cases were found to be

heterozygous for exonic silent changes or intronic

nucleotide substitutions known to occur as dis-

ease-unrelated variants in the population (Supp.

Info. Table 2). All mutations were missense

changes, affected the glutamine residue at codon

61, and introduced a positively charged amino

acid residue. Since previous studies reported a

relatively high prevalence of mutations at codons

12 and 13, to confirm our DHPLC data, we per-

formed bidirectional sequencing of HRAS, KRAS,

and NRAS exon 1 in all RAS gene mutation-nega-

tive tumors. We failed in identifying any new

lesion, confirming the absence of RAS gene

defects at those codons in our cohort. Genotyping

of available DNAs from circulating leukocytes of

3 patients with a mutation-positive RMS, demon-

strated the absence of the mutated allele in each

tested case (Fig. 1A), providing evidence that the

mutations were somatic events acquired in cancer

cells.

In one primary tumor, DHPLC profiles and

electropherograms indicated that the mutant

RAS SIGNALING DYSREGULATION IN HUMAN EMBRYONAL RMS 977


allele was less represented compared to the wild

type allele. The histological evaluation indicated

that the tumor cell population was highly repre-

sented (>90% of total cells), supporting the idea

that this lesion likely occurred in a fraction of tu-

mor cells being acquired as a late event during

disease progression. A similar finding was docu-

mented for the KRAS 182A>T missense change

(Q61L) identified in the CCA tumor cell line. On

the other hand, genotyping of the RD (Chardin

et al., 1985) and RD-derived RD18 (Lollini et al.,

1991) RMS cell lines demonstrated the almost

exclusively occurrence of the NRAS 183A>T al-

lele (Q61H), indicating that either gene conver-

sion, deletion of the wild type allele or

amplification of the mutated allele occurred in

the ancestral tumor cell clone (Fig. 1B). On the

basis of this finding, as well as on the evidence

indicating that allelic imbalances at 11p15.5 fre-

quently occur during embryonal RMS develop-

ment (Xia et al., 2002), we systematically

investigated the possibility that HRAS mutations

might occur as hemizygous defects by DNA pool-

ing-based DHPLC analysis; however, no homozy-

gous/hemizygous lesion was identified.

Although our previous PTPN11 mutational

screening of an independent relatively small

RMS cohort did not reveal any cancer-associated

mutation (Martinelli et al., 2006), a recent report

documented a PTPN11 lesion in a 2-year-old girl

with stage IV embryonal RMS (Chen et al.,

2006). To explore further the prevalence of

PTPN11 defects in primary RMS tumors, muta-

tional analysis was carried out. DHPLC screening

of cancer-associated PTPN11 mutational hot-spots

allowed the identification of one missense muta-

tion (3.2% of cases, 95% CI: 0.08–16.7%) predict-

ing the E69K change in a 2-year-old boy with a

Group II embryonal RMS (Table 1), who did not

show any clinical features suggestive of Noonan

syndrome (NS), a genetic disease caused by

germline PTPN11 defects (Tartaglia et al., 2001).

Mutational analysis of genomic DNA from

TABLE 1. RAS and PTPN11 Gene Mutations in Primary Embryonal RMS Specimens and Cell Lines

Mutated gene Number of cases Nucleotide change Exon Amino acid substitution

Primary tumorsNRAS 5a CAA > AAA 2 Q61K

1 CAA > CGA 2 Q61RHRAS 1 CAG > AAG 2 Q61KPTPN11 1 GAG > AAG 3 E69K

Tumor cell linesNRAS 2

bCAA > CAT 2 Q61H

KRAS 1c CAA > CTA 2 Q61L

aThe mutation represents a subpopulation of RMS cells in one case.bThis nucleotide change was observed in the RD and RD-derived RD18 RMS cell lines. The mutant allele, which was originally reported by Chardin

et al. (1985), is overrepresented due to possible gene conversion, gene amplification or deletion of the wild type allele.cThis nucleotide change was observed in the CCA RMS cell line. The mutation represents a subpopulation of RMS cells.

Figure 1. Somatic RAS and PTPN11 mutations in childhood embry-onal RMS. (A) Representative DHPLC profiles showing the occur-rence of an NRAS (left) or PTPN11 (right) missense mutation in twoembryonal RMS specimens; in both cases, mutations were observedin tumor cells (above), but were undetectable in peripheral blood(below). Electropherograms documenting the somatic changes arealso shown. (B) Electropherograms showing the occurrence of theNRAS 181C>A mutation in a subpopulation of cancer cells (left), andthe loss of the wild type allele in the RD18 RMS cell line (right).[Color figure can be viewed in the online issue, which is available atwww.interscience.wiley.com.]



normal tissue demonstrated absence of the

mutated allele, confirming the somatic origin of

the defect (Fig. 1A). On the basis of the available

crystallographic data on SHP2, the identified

mutation is predicted to have an activating effect

on protein function by perturbing the stability of

SHP20s catalytically inactive conformation (Hof

et al., 1998). Consistent with this hypothesis, the

same amino acid substitution had previously been

observed in children with juvenile myelomono-

cytic leukemia (JMML) and B-cell precursor

acute lymphoblastic leukemia (ALL) (Tartaglia

et al., 2003, 2004, 2006). We failed in identifying

any BRAF, MEK1, and MEK2 mutation in 23

RMS samples for which DNAs were available.

Possible associations between RAS/PTPN11mutations and clinical features were investigated.

Patients were grouped in subsets according to

each variable of interest, and distributions were

compared using two-tailed Fisher’s exact test.

Mutations were equally distributed in the differ-

ent subsets and no significant association was

found (Table 2). The impact of RAS and

PTPN11 mutations on OS was analyzed using

Kaplan–Meier method, and log-rank test was

used for comparing survival between groups.

Interestingly, patients carrying a RAS/PTPN11gene mutation tended to have a better outcome,

with a higher OS probability compared to

patients without mutations (88% vs. 57%),

although this difference did not reach statistical

significance (P ¼ 0.17), possibly due to the rela-

tively small size of the study cohort (Fig. 2).

DISCUSSION

The molecular mechanisms underlying the de-

velopment of human RMS, particularly of the

embryonal subtype, are largely unknown. Experi-

mental evidence suggests that dysregulation of

the RAS cascade may be relevant to the pathoge-

nesis and/or progression of these tumors. First,

constitutive RAS-mediated signal flow abrogates

myogenic differentiation by downregulating myo-

genic factors such as MyoD1 and myogenin

(Konieczny et al., 1989; Lassar et al., 1989), and

inhibits differentiation-associated apoptosis (Dee

et al., 2002). Second, a RAS gain-of-function/

TP53 loss-of-function RMS mouse model

recently established a clear contributing role for

RAS activation in the genesis of translocation-

negative RMS (Tsumura et al., 2006). In addi-

tion, a zebrafish model of RAS-induced RMS

demonstrated that tumors express markers typical

of human RMS, and are morphologically similar

to the embryonal subtype, supporting the idea

that RAS activation is sufficient to initiate tumor-

igenesis in vivo (Langenau et al., 2007). While

substantial genetic evidence appears to support

Figure 2. Kaplan–Meier survival curves for RAS/PTPN11 mutation-negative and mutation-positive patients with childhood embryonalRMS. Statistical differences between groups were calculated using thelog-rank test. Vertical bars indicate censored observations.

TABLE 2. Clinical Relevance of Somatic RAS/PTPN11 Mutations in Primary Embryonal RMS

Variable Classes Number of patients

RAS/PTPN11 mutational status

PaWild type Mutated

Age (months) 12–120 25 19 (76%) 6 (24%) 0.63<12 or >120 6 4 (67%) 2 (33%)

Sex Male 19 13 (68%) 6 (32%) 0.43Female 12 10 (83%) 2 (17%)

Primary site Favorableb 9 6 (67%) 3 (33%) 0.66Unfavorablec 22 17 (77%) 6 (23%)

Primary size �5 cm 8 6 (75%) 2 (25%) 0.99>5 cm 23 17 (74%) 6 (26%)

Group I and II 16 11 (69%) 5 (31%) 0.68III and IV 15 12 (80%) 3 (20%)

aTwo-tailed Fisher’s exact test.bOrbit and genitourinary nonbladder/prostate.cHead and neck parameningeal and nonparameningeal, genitourinary bladder/prostate, extremities and others.



this hypothesis (Stratton et al., 1989; Wilke et al.,

1993; Yoo et al., 1999; Chen et al., 2006; Kratz

et al., 2006), with the exception of the report by

Chen et al., who performed RAS genotyping in

20 embryonal RMS samples, previous studies

included rather small cohorts, not allowing a reli-

able estimate of the actual prevalence of RASgene lesions in these tumors. In this report, we

documented that somatic missense mutations in

RAS genes represent a recurrent event in pediat-

ric embryonal RMS, accounting for approximately

one-fourth of cases (95% CI: 9.6–41.1%).

Activating RAS mutations occur in approxi-

mately 30% of human cancers (Bos, 1989). Differ-

ent tumor types often display specificity in RASgene involvement. For example, HRAS mutations

are prevalent in bladder and kidney carcinoma,

while KRAS mutations predominate in colorectal,

pancreatic, endometrial, lung, and cervical cancers,

and NRAS lesions are more commonly observed in

melanoma and liver carcinoma as well as in lymph-

oid and myeloid malignancies. In the present

study, we documented that, among the RAS paral-

ogs, NRAS is most commonly affected in embry-

onal RMS. Of note, all cancer-associated mutations

were observed to involve the glutamine residue at

codon 61, an unreported observation in these

tumors. We can confidently exclude a bias in sam-

pling since no apparent selection operated in

patient inclusion in this study, as well as any meth-

odological issue, because positive controls carrying

missense mutations affecting codons 12 and 13

were used in all DHPLC runs and bidirectional

sequencing was performed for each case exhibiting

a DHPLC variant profile. HRAS, KRAS, and NRASexon 1 direct sequencing for all mutation-negative

tumors was also performed, excluding occurrence

of lesions affecting those codons. One possible

explanation of this discrepancy could reside in the

observation that most of the previous reports did

not analyze the NRAS gene, suggesting a possible

underestimation of RAS lesions, and particularly of

those affecting codon 61. Predominant involve-

ment of mutations at codon 61 was previously

reported in other human cancers, such as multiple

myeloma (Kalakonda et al., 2001). These lesions

have previously been observed to confer growth

advantage and resistance to apoptosis to cancer

cells (Billadeau et al., 1997). Interestingly, the

same effect was not observed in cells expressing

an oncogenic codon 12 mutant, suggesting that the

higher than expected occurrence of RAS codon 61

mutations in embryonal RMS might reflect a cell

context-specific cancer-contributing effect of RAS

functional dysregulation. This observation is fur-

ther supported by the evidence of a nonrandom

occurrence of individual amino acid substitutions

involving cancer-associated NRAS mutations at

this codon. Similarly to what was observed in this

study, in thyroid cancers, the Q61R and Q61K

amino acid changes occur in the vast majority of

cases (COSMIC database, http://www.sanger.a-

c.uk/genetics/CGP/cosmic/). These findings sup-

port the idea that the functional consequences of

individual RAS mutants might contribute to trans-

formation in a cell context-specific fashion. In this

scenario, during the initial phases of RMS devel-

opment the NRASQ61K mutant might have a

higher potential to promote proliferation in skele-

tal muscle cell precursors, or impair myogenic

differentiation, and/or suppress differentiation-

associated apoptosis.

Structural analyses of HRAS indicate that glu-

tamine 61 plays a critical role in GTP hydrolysis,

and any substitution at this position is predicted

to impair catalysis (Der et al., 1986). In the pres-

ent series, we have identified the NRASQ61R and

the H/NRASQ61K mutant proteins in one and six

tumors, respectively. As elegantly demonstrated

by Donovan et al. (2002), both these amino acid

changes confer to HRAS a robust transforming

phenotype caused by a dramatic decrease in its

intrinsic GTPase activity. The variable transform-

ing potency of missense changes affecting codon

61, together with the evidence that mutant alleles

carrying the Q61R and the Q61L mutations were

only present in a fraction of cancer cells, suggest

that, among all possible mutations at codon 61,

only the glutamine-to-lysine change is likely to

have a crucial role in the process of tumor initia-

tion, representing an early event during RMS de-

velopment; any other mutations at the same

position might specifically contribute to clone

selection or disease progression. Consistent with

this hypothesis, both the embryonal RMS tumors

with a mutated RAS at codon 61 described by

Stratton et al. (1989) and Chen et al. (2006)

showed the Q61K amino acid substitution.

We also explored the possible involvement of

transducers participating in the RAS signaling

pathway in embryonal RMS tumorigenesis. While

PTPN11 mutations frequently occur in leukemia

and more rarely in solid tumors (Tartaglia and

Gelb, 2005), and BRAF defects represent the

most common genetic alteration in thyroid cancer

(Xing, 2005) and melanoma (Brose et al., 2002),

MEK1 and MEK2 were not found to be mutated

in human cancer thus far. While our data provide



further evidence confirming that BRAF and

PTPN11 defects do not play a key-role in embry-

onal RMS (Miao et al., 2004; Chen et al., 2006;

Martinelli et al., 2006) and exclude a major con-

tribution of MEK1 and MEK2 functional dysre-

gulation in these tumors, we cannot rule out the

possible involvement of other transducers in the

signaling pathway.

Several studies have addressed the clinical rele-

vance of RAS mutations in tumors but, to our

knowledge, none of them analyzed the impact of

RAS gene defects in RMS. In this study, we

explored the clinical impact of somatic mutations

predicted to promote upregulation of RAS signal-

ing traffic by analyzing the possible association

between these lesions and clinical features, but

failed in identifying any significant association. A

trend, however, was observed for patients carrying

a RAS or a PTPN11 mutation to have a higher OS

probability (88% vs. 57%), although this difference

did not reach statistical significance possibly due

to the relatively small size of the study cohort. De-

spite RAS mutations have frequently been associ-

ated with aggressive tumor phenotypes and poor

outcome in thyroid cancer (Garcia-Rostan et al.,

2003), colorectal adenocarcinoma (Cerottini et al.,

1998), nonsmall-cell lung carcinoma (Nelson

et al., 1999), and acute myeloid leukemia (Kiyoi

et al., 1999), consistent with the present finding,

NRAS lesions were characterized by a favorable

outcome in comparison with BRAF mutations or

an uncharacterized genotype in patients with met-

astatic melanoma (Ugurel et al., 2007). The analy-

sis of a more numerous series is requested to

establish definitely whether these gene lesions

actually have a significant impact on embryonal

RMS pathogenesis and/or progression.

Notwithstanding the treatment of RMS has

improved in the last decade, patients with local-

ized high risk or metastatic RMS have still a very

poor prognosis with the current therapeutic strat-

egies. In these patients, novel and innovative

therapies must be attempted. The present and

available data support the idea that dysregulation

of RAS signaling plays a key role in RMS develop-

ment. A more detailed understanding of the mo-

lecular mechanisms and underlying gene defects is

required to guide future work directed at design-

ing novel target-specific therapeutic approaches.

ACKNOWLEDGMENTS

The authors are gratefully indebted to Dr.

Pierluigi Altavista (ENEA Research Center,

Rome, Italy) for his help in statistical analysis and

Serenella Venanzi (Istituto Superiore di Sanita,

Rome, Italy) for her excellent technical assistance.

REFERENCES

Barr FG, Chatten J, D’Cruz CM, Wilson AE, Nauta LE, NycumLM, Biegel JA, Worner RB. 1995. Molecular assays for chromo-somal translocations in the diagnosis of pediatric soft tissue sar-comas. J Am Med Assoc 273:553–557.

Billadeau D, Liu P, Jelinek D, Shah N, LeBien TW, Van Ness B.1997. Activating mutations in the N- and K-ras oncogenes dif-ferentially affect the growth properties of the IL-6-dependentmyeloma cell line ANBL6. Cancer Res 57:2268–2275.

Bos, JL. 1989. RAS oncogenes in human cancer: A review. CancerRes 49:4682–4689.

Brose MS, Volpe P, Feldman M, Kumar M, Rishi I, Gerrero R,Einhorn E, Herlyn M, Minna J, Nicholson A, Roth JA, AlbeldaSM, Davies H, Cox C, Brignell G, Stephens P, Futreal PA,Wooster R, Stratton MR, Weber BL. 2002. BRAF and RASmutations in human lung cancer and melanoma. Cancer Res62:6997–7000.

Carli M, Colombatti R, Oberlin O, Bisogno G, Treuner J, Kosciel-niak E, Tridello G, Garaventa A, Pinkerton R, Stevens M.2004. European intergroup studies (MMT4-89 and MMT4-91)on childhood metastatic rhabdomyosarcoma: Final results andanalysis of prognostic factors. J Clin Oncol 22:4787–4794.

Cerottini JP, Caplin S, Saraga E, Givel JC, Benhattar J. 1998.The type of K-ras mutation determines prognosis in colorectalcancer. Am J Surg 175:198–202.

Ciarapica R, Russo G, Verginelli F, Raimondi L, Donfrancesco A,Rota R, Giordano A. 2009. Deregulated expression of miR-26aand Ezh2 in rhabdomyosarcoma. Cell Cycle 8:172–175.

Chardin P, Yeramian P, Madaule P, Tavitian A. 1985. N-RASgene activation in the RD human rhabdomyosarcoma cell line.Int J Cancer 35:647–652.

Chen Y, Takita J, Hiwatari M, Igarashi T, Hanada R, Kikuchi A,Hongo T, Taki T, Ogasawara M, Shimada A, Hayashi Y. 2006.Mutations of the PTPN11 and RAS genes in rhabdomyosar-coma and pediatric hematological malignancies. Genes Chromo-somes Cancer 45:583–591.

Dee K, Freer M, Mei Y, Weyman CM. 2002. Apoptosis coincidentwith the differentiation of skeletal myoblasts is delayed by cas-pase 3 inhibition and abrogated by MEK-independent constitu-tive Ras signaling. Cell Death Differ 9:209–218.

Dentici ML, Sarkozy A, Pantaleoni F, Carta C, Lepri F, FereseR, Cordeddu V, Martinelli S, Briuglia S, Digilio MC, ZampinoG, Tartaglia M, Dallapiccola B. 2009. Spectrum of MEK1 andMEK2 gene mutations in cardio-facio-cutaneous syndrome andgenotype -phenotype correlations. Eur J Hum Genet 17:733–740.

Der CJ, Finkel T, Cooper GM. 1986. Biological and biochemicalproperties of human rasH genes mutated at codon 61. Cell44:167–176.

Donovan S, Shannon KM, Bollag G. 2002. GTPase activatingproteins: Critical regulators of intracellular signaling. BiochimBiophys Acta 1602:23–45.

Ferrari A, Bisogno G, Casanova M, Meazza C, Piva L, CecchettoG, Zanetti I, Pilz T, Mattke A, Treuner J, Carli M. 2002. Para-testicular rhabdomyosarcoma: Report from the Italian and Ger-man Cooperative Group. J Clin Oncol 20:449–455.

Garcia-Rostan G, Zhao H, Camp RL, Pollan M, Herrero A, Pardo J,Wu R, Carcangiu ML, Costa J, Tallini G. 2003. ras mutations areassociated with aggressive tumor phenotypes and poor prognosisin thyroid cancer. J Clin Oncol 21:3226–3235.

Gurney JG, Young JL, Jr., Roffers SD, Smith MA, Bunin GR.1999. Soft tissue sarcomas. In: Ries LAG, Smith MA, GurneyJG, Linet M, Tamra T, Young JL, Bunin GR, editors. CancerIncidence and Survival among Children and Adolescents:United States SEER Program 1975–1995, SEER Program 99-4649. Bethesda: National Cancer Institute, pp. 111–124.

Herrmann C. 2003. Ras-effector interactions: After one decade.Curr Opin Struct Biol 13:122–129.

Hof P, Pluskey S, Dhe-Paganon S, Shoelson SE. 1998. Crystalstructure of the tyrosine phosphatase SHP-2. Cell 92:441–450.



Kalakonda N, Rothwell DG, Scarffe JH, Norton JD. 2001. Detec-tion of N-Ras codon 61 mutations in subpopulations of tumorcells in multiple myeloma at presentation. Blood 98:1555–1560.

Kiyoi H, Naoe T, Nakano Y, Yokota S, Minami S, Miyawaki S,Asou N, Kuriyama K, Jinnai I, Shimazaki C, Akiyama H, Saito K,Oh H, Motoji T, Omoto E, Saito H, Ohno R, Ueda R. 1999. Prog-nostic implication of FLT3 and N-RAS gene mutations in acutemyeloid leukemia. Blood 93:3074–3080.

Konieczny SF, Drobes BL, Menke SL, Taparowsky EJ. 1989. In-hibition of myogenic differentiation by the H-ras oncogene isassociated with the down regulation of the MyoD1 gene. Onco-gene 4:473–481.

Kratz CP, Steinemann D, Niemeyer CM, Schlegelberger B,Koscielniak E, Kontny U, Zenker M. 2007. Uniparental disomyat chromosome 11p15.5 followed by HRAS mutations in embry-onal rhabdomyosarcoma: Lessons from Costello syndrome. HumMol Genet 16:374–379.

Langenau DM, Keefe MD, Storer NY, Guyon JR, Kutok JL, LeX, Goessling W, Neuberg DS, Kunkel LM, Zon LI. 2007.Effects of RAS on the genesis of embryonal rhabdomyosarcoma.Genes Dev. 21:1382–1395.

Lassar AB, Thayer MJ, Overell RW, Weintraub H. 1989. Trans-formation by activated ras or fos prevents myogenesis by inhibi-ting expression of MyoD1. Cell 58:659–667.

Lollini PL, De Giovanni C, Landuzzi L, Nicoletti G, Scotlandi K,Nanni P. 1991. Reduced metastatic ability of in vitro differenti-ated human rhabdomyosarcoma cells. Invasion Metastasis 11:116–124.

Martinelli S, Carta C, Flex E, Binni F, Lucci Cordisco E, MorettiS, Puxeddu E, Tonacchera M, Pinchera A, McDowell HP,Dominaci C, Rosolen A, Di Rocco C, Riccardi R, Celli P, Pic-ardo M, Genuardi M, Grammatico P, Sorcini M, Tartaglia M.2006. Activating PTPN11 mutations play a minor role in pediat-ric and adult solid tumors. Cancer Genet Cytogenet 166:124–129.

Maurer HM, Beltangady M, Gehan EA, Crist W, Hammond D,Hays DM, Heyn R, Lawrence W, Newton W, Ortega J, RagabAH, Raney B, Ruymann FB, Soule E, Tefft M, Webber B,Eharam M, Vietti TJ. 1988. The Intergroup RhabdomyosarcomaStudy I: A final report. Cancer 61:209–220.

McDowell HP. 2003. Update on childhood rhabdomyosarcoma.Arch Dis Child 88:354–357.

McDowell HP, Meco D, Riccardi A, Tanno B, Berardi AC,Raschella G, Riccardi R, Dominici C. 2007. Imatinib mesylatepotentiates topotecan antitumour activity in rhabdomyosarcomapreclinical models. Int J Cancer 120:1141–1149.

Miao J, Kusafuka T, Fukuzawa M. 2004. Hotspot mutations ofBRAF gene are not associated with pediatric solid neoplasms.Oncol Rep 12:1269–1272.

Mitin N, Rossman KL, Der CJ. 2005. Signaling interplay in Rassuperfamily function. Curr Biol 15:563–574.

Nelson HH, Christiani DC, Wiencke JK, Mark EJ, Wain JC, Kel-sey KT. 1999. Implications and prognostic value of K-ras muta-tion for early-stage lung cancer in women. J Natl Cancer Inst91:2032–2038.

Ohali A, Avigad S, Naumov I, Goshen Y, Ash S, Yaniv I. 2008.Different telomere maintenance mechanisms in alveolar andembryonal rhabdomyosarcoma. Genes Chromosomes Cancer47:965–970.

Parham DM, Ellison DA. 2006. Rhabdomyosarcomas in adultsand children: An update. Arch Pathol Lab Med 130:1454–1465.

Patton RB, Horn RC, Jr. 1962. Rhabdomyosarcoma: Clinical andpathological features and comparison with human fetal andembryonal skeletal muscle. Surgery 52:572–584.

Reed N, Gutmann DH. 2001. Tumorigenesis in neurofibromato-sis: New insights and potential therapies. Trends Mol Med7:157–162.

Sarkozy A, Carta C, Moretti S, Zampino G, Digilio MC,Pantaleoni F, Scioletti AP, Esposito G, Cordeddu V, Lepri F,Petrangeli V, Dentici ML, Mancini GM, Selicorni A, Rossi C,Mazzanti L, Marino B, Ferrero GB, Silengo MC, Memo L,Stanzial F, Faravelli F, Stuppia L, Puxeddu E, Gelb BD, Dalla-

piccola B, Tartaglia M. 2009. Germline BRAF mutations inNoonan, LEOPARD, and cardiofaciocutaneous syndromes:Molecular diversity and associated phenotypic spectrum. HumMutat 30:695–702.

Stevens M, Rey A, Bouvet N, Ellershaw C, Flamant F, HabrandJL, Marsden HB, Martelli H, Sanchez de Toledo J, Spicer RD,Spooner D, Terrier-Lacombe MJ, van Unnik A, Oberlin O.2005. Treatment of nonmetastatic rhabdomyosarcoma in child-hood and adolescence: Third study of the International Societyof Paediatric Oncology–SIOP Malignant Mesenchymal Tumor89. J Clin Oncol 23:2618–2628.

Subramanian S, Lui WO, Lee CH, Espinosa I, Nielsen TO,Heinrich MC, Corless CL, Fire AZ, van de Rijn M. 2008.MicroRNA expression signature of human sarcomas. Oncogene27:2015–2026.

Stratton MR, Fisher C, Gusterson BA, Cooper CS. 1989. Detec-tion of point mutations in N-ras and K-ras genes of humanembryonal rhabdomyosarcomas using oligonucleotide probesand the polymerase chain reaction. Cancer Res 49:6324–6327.

Tartaglia M, Gelb BD. 2005. Germ-line and somatic PTPN11mutations in human disease. Eur J Med Genet 48:81–96.

Tartaglia M, Mehler EL, Goldberg R, Zampino G, Brunner HG,Kremer H, van der Burgt I, Crosby AH, Ion A, Jeffery S,Kalidas K, Patton MA, Kucherlapati RS, Gelb BD. 2001. Muta-tions in PTPN11, encoding the protein tyrosine phosphataseSHP-2, cause Noonan syndrome. Nat Genet 29:465–468.

Tartaglia M, Kalidas K, Shaw A, Song X, Musat DL, van derBurgt I, Brunner HG, Bertola D, Crosby A, Ion A, KucherlapatiRS, Jeffery S, Patton MA, Gelb BD. 2002. PTPN11 mutationsin Noonan syndrome: Molecular spectrum, genotype-phenotypecorrelation, and phenotypic heterogeneity. Am J Hum Genet70:1555–1563.

Tartaglia M, Niemeyer CM, Fragale A, Song X, Buechner J, JungA, Hahlen K, Hasle H, Licht JD, Gelb DB. 2003. SomaticPTPN11 mutations in juvenile myelomonocytic leukemia, mye-lodysplastic syndromes and acute myeloid leukemia. Nat Genet34:148–150.

Tartaglia M, Martinelli S, Cazzaniga G, Cordeddu V, Iavarone I,Spinelli M, Palmi C, Carta C, Pession A, Arico M, Masera G,Basso G, Sorcini M, Biondi A. 2004. Genetic evidence for line-age- and differentiation stage-related contribution of somaticPTPN11 mutations to leukemogenesis in childhood acute leu-kemia. Blood 104:307–313.

Tartaglia M, Martinelli S, Stella L, Bocchinfuso G, Flex E, Cor-deddu V, Zampino G, van der Burgt I, Palleschi A, PetrucciTC, Sorcini M, Schoch C, Foa R, Emanuel PD, Gelb BD.2006. Diversity and functional consequences of germline andsomatic PTPN11 mutations in human disease. Am J Hum Genet78:279–290.

Trahey M, McCormick F. 1987. A cytoplasmic protein stimulatesnormal N-ras p21 GTPase, but does not affect oncogenicmutants. Science 238:542–545.

Tsumura H, Yoshida T, Saito H, Imanaka-Yoshida K, Suzuki N.2006. Cooperation of oncogenic K-ras and p53 deficiency inpleomorphic rhabdomyosarcoma development in adult mice.Oncogene 25:7673–7679.

Ugurel S, Thirumaran RK, Bloethner S, Gast A, Sucker A, Muel-ler-Berghaus J, Rittgen W, Hemminki K, Becker JC, Kumar R,Schadendorf D. 2007. B-RAF and N-RAS mutations are pre-served during short time in vitro propagation and differentiallyimpact prognosis. PLoS ONE 2:e236.

Wilke W, Maillet M, Robinson R. 1993. H-ras-1 point mutationsin soft tissue sarcomas. Mod Pathol 6:129–132.

Xia SJ, Pressey JG, Barr FG. 2002. Molecular pathogenesis ofrhabdomyosarcoma. Cancer Biol Ther 1:97–104.

Xing M. 2005. BRAF mutation in thyroid cancer. EndocrinolRelat Cancer 12:245–262.

Yoo J, Robinson RA, Lee JY. 1999. H-ras and K-ras gene muta-tions in primary human soft tissue sarcoma: Concomitant muta-tions of the ras genes. Mod Pathol 12:775–780.



ras signaling dysregulation in human embryonal rhabdomyosarcoma

Documents