Adult mice expressing a Braf Q241R mutation on an ICR/CD-1

OR I G INA L ART I C L E

Adult mice expressing a Braf Q241R mutation on anICR/CD-1 background exhibit a cardio-facio-cutaneoussyndrome phenotypeMitsuji Moriya1,†, Shin-ichi Inoue1,†, Sachiko Miyagawa-Tomita3,4,Yasumi Nakashima5, Daiju Oba1, Tetsuya Niihori1, Misato Hashi6,Hiroshi Ohnishi6, Shigeo Kure2, Yoichi Matsubara1,7 and Yoko Aoki1,*1Department of Medical Genetics and 2Department of Pediatrics, Tohoku University School of Medicine, Sendai,Japan, 3Department of Pediatric Cardiology and 4Division of Cardiovascular Development and Differentiation,Medical Research Institute, Tokyo Women’s Medical University, Tokyo, Japan, 5Department of Pediatrics, SeireiHamamatsuGeneral Hospital, Shizuoka, Japan, 6Department of Laboratory Sciences, GunmaUniversity GraduateSchool of Health Sciences, Gunma, Japan and 7National Research Institute for Child Health and Development,Tokyo, Japan

*To whom correspondence should be addressed at: Department of Medical Genetics, Tohoku University School of Medicine, 1-1 Seiryo-machi, Aoba-ku,Sendai 980-8574, Japan. Tel: +81-22-717-8139; Fax: +81-22-717-8142; Email: [email protected]

AbstractActivation of the RAS pathway has been implicated in oncogenesis and developmental disorders called RASopathies. Germlinemutations in BRAF have been identified in 50–75% of patients with cardio-facio-cutaneous (CFC) syndrome, which ischaracterized by congenital heart defects, distinctive facial features, short stature and ectodermal abnormalities. We recentlydemonstrated that mice expressing a Braf Q241R mutation, which corresponds to the most frequent BRAF mutation (Q257R) inCFC syndrome, on a C57BL/6J background are embryonic/neonatal lethal, with multiple congenital defects, preventing us fromanalyzing the phenotypic consequences after birth. Here, to further explore the pathogenesis of CFC syndrome, we backcrossedthese mice onto a BALB/c or ICR/CD-1 genetic background. On a mixed (BALB/c and C57BL/6J) background, all heterozygousBrafQ241R/+ mice died between birth and 24 weeks and exhibited growth retardation, sparse and ruffled fur, liver necrosis andatrial septal defects (ASDs). In contrast, 31% of the heterozygous BrafQ241R/+ ICR mice survived over 74 weeks. The survivingBrafQ241R/+ ICRmice exhibited growth retardation, sparse and ruffled fur, a hunchedappearance, craniofacial dysmorphism, longand/or dystrophic nails, extra digits and ovarian cysts. The BrafQ241R/+ ICR mice also showed learning deficits in the contextualfear-conditioning test. Echocardiography indicated the presence of pulmonary stenosis and ASDs in the BrafQ241R/+ ICR mice,which were confirmed by histological analysis. These data suggest that the heterozygous BrafQ241R/+ ICR mice show similarphenotypes as CFC syndrome after birth andwill be useful for elucidating the pathogenesis and potential therapeutic strategiesfor RASopathies.

†These authors contributed equally to this work.Received: September 16, 2015. Revised and Accepted: October 12, 2015

© The Author 2015. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]

Human Molecular Genetics, 2015, Vol. 24, No. 25 7349–7360

doi: 10.1093/hmg/ddv435Advance Access Publication Date: 15 October 2015Original Article

7349

Downloaded from https://academic.oup.com/hmg/article-abstract/24/25/7349/2384721by gueston 13 March 2018

http://www.oxfordjournals.org

IntroductionThe RAS/mitogen-activated protein kinase (MAPK) pathway playsimportant roles in cellular proliferation, differentiation, apoptosisand senescence (1). Germline mutations in components of theRAS/MAPK pathway cause partially overlapping developmentaldisorders called ‘RASopathies’ or ‘RAS/MAPK syndromes’ (2,3).These disorders include cardio-facio-cutaneous (CFC) syndrome,Noonan syndrome, Costello syndrome, Noonan syndrome withmultiple lentigines (previously known as LEOPARD syndrome),neurofibromatosis type 1, Legius syndrome (neurofibromatosis 1-like) and capillary malformation–arteriovenous malformationsyndrome (2–4). With the exception of Noonan syndrome withmultiple lentigines that results froma loss of function PTPN11mu-tation, these syndromes usually result from the activation of theRAS/MAPK pathway (2–4).

CFC syndrome is characterized by congenital heart defects, adistinctive facial appearance, short stature, moderate to severemental retardation, seizures and ectodermal abnormalities, suchas sparse, fragile hair and hyperkeratotic skin lesions (5,6). The car-diac defects observed in CFC syndrome are pulmonary valve sten-osis, hypertrophic cardiomyopathy, atrial septal defects (ASDs),ventricular septal defects (VSDs) and arrhythmia (5,6). Germlinemissense mutations in BRAF occur in 50–75% of CFC syndromepatients. Germline mutations in MAP2K1/2 (MEK1/2) and KRAShave been identified in∼25% and <2%of CFCpatients (7–9), respect-ively. The most common BRAF mutation in human cancers is thesubstitution of the residue valine at position 600 with glutamicacid (p.V600E), which has not been identified in CFC syndrome(2,10). In contrast, the most common mutation among CFC syn-drome patients with a BRAF mutation is an arginine for glutaminesubstitution at position 257 (p.Q257R), which increases the kinaseactivity of BRAF, similar to the BRAF V600E mutation (7,8,11). Thepresence ofmalignant tumors is rare in CFC syndrome; three indivi-duals with BRAF mutations have been reported to have developedacute lymphoblastic leukemia and non-Hodgkin’s lymphoma (5,12).

We have recently described heterozygous knock-in mice ex-pressing the Braf Q241R mutation on a C57BL/6J background(BrafQ241R/+ B6 mice); this corresponds to the BRAF Q257Rmutationin humans (13). The BrafQ241R/+ B6 mice exhibited embryonic/neonatal lethality. The BrafQ241R/+ B6 embryos and neonates hadmany characteristic features of patients with CFC syndrome, in-cluding skeletal abnormalities, lymphatic defects, cardiomegaly,pulmonary valve stenosis and VSDs. Prenatal treatment with aMEK inhibitor, PD0325901, or a histone 3 demethylase inhibitor,GSK-J4 or NCDM-32b, rescued the embryonic and postnatal lethal-ity in the BrafQ241R/+ B6 mice. Some CFC syndrome patients die ofhydrops fetalis, heart failure or chylothorax in the perinatal period(14–16), similar to the BrafQ241R/+ B6 mice. However, most CFC syn-drome patients survive to adulthood, showing a distinctive facialappearance, growth retardation, intellectual disability, heartdefects and skin abnormalities (9,12,15). To better understand themolecular pathogenesis of individuals with CFC syndrome fromthe neonatal period to adulthood, we analyzed knock-in miceexpressing a developmental-specific Braf Q241R mutation on aBALB/c or ICR genetic background.

ResultsThe BrafQ241R/+ mice on a BALB/c background exhibitgrowth retardation, decreased survival, furabnormalities, liver necrosis and cardiac anomalies

To avoid strain-specific lethality in the knock-in mice expressingthe Braf Q241R mutation, we first crossed the surviving

PD0325901-treated BrafQ241R/+ B6 mice we had generated previ-ously (13) with the inbred BALB/c mouse strain. Notably, few ofthe BrafQ241R/+ mice on the BALB/c background (BrafQ241R/+ BALB/c mice) survived to adulthood. The male BrafQ241R/+ mice (threegenerations backcrossed; N3) were fertile, but the femaleBrafQ241R/+ mice were infertile. The BrafQ241R/+ mice displayed sig-nificant growth retardation and sparse and ruffled fur comparedwith the Braf+/+mice (Fig. 1A and B). Approximately 50 and 90% ofthe BrafQ241R/+ mice had died at postnatal day (P) 5 and weaning(P28), respectively, whereas the Braf+/+ and BrafQ241R/+ mice wereobtained at the expected Mendelian ratios at embryonic day (E)16.5 and P0 (Fig. 1C; Supplementary Material, Table S1). Histo-logical examinations revealed liver necrosis in 4 of 11 BrafQ241R/+

mice (36%) at P3 (data not shown). Most CFC syndrome patientshave heart defects, including pulmonary valve stenosis, hyper-trophic cardiomyopathy, ASDs, VSDs and arrhythmia, whichmay contribute to morbidity and mortality (5). At P5, theBrafQ241R/+ mice exhibited a significantly higher heart weight(HW) to body weight (BW) ratio, which is an index of cardiomeg-aly (17), compared with the Braf+/+ mice (Fig. 1D; SupplementaryMaterial, Table S2). Histologic analysis revealed a large secundumASD (ASDII) in five of six BrafQ241R/+ mice, but a small ASDII or pa-tent foramen ovale was also observed in one of six Braf+/+ mice(Supplementary Material, Fig. S1A). Cardiac valve defects and ap-parent abnormalities in other tissues were not observed in theBrafQ241R/+ mice (Supplementary Material, Fig. S1B). These resultssuggest that the BrafQ241R/+ BALB/c mice exhibit growth retard-ation, decreased survival, fur abnormalities, liver necrosis, an in-crease in the HW to BW ratio and ASDs.

The BrafQ241R/+ mice on an ICR background displaydecreased survival, growth retardation, a hunchedappearance, fur abnormalities, craniofacialdysmorphism, long and dystrophic nails, extra digits,ovarian cysts and learning deficits

Because all BrafQ241R/+ BALB/c mice died within 24 weeks, thesemice were not suitable for evaluating the pathogenesis of CFCsyndrome from the neonatal period to adulthood. We next back-crossed the surviving PD0325901-treated BrafQ241R/+ B6 mice forseven generations onto a closed-colony ICR mouse strain. Boththe male and female BrafQ241R/+ mice on the ICR background(BrafQ241R/+ ICR mice) were viable. All homozygous BrafQ241R/Q241R

mice died within 19 days (Supplementary Material, Table S3),whereas approximately half of the BrafQ241R/+ mice survived forover 4 weeks (Fig. 2A). Although the BWs of the surviving maleand female BrafQ241R/+ mice were normal from birth to 2 days,after 1 week, the weights of male and female BrafQ241R/+ micewere significantly lower than those of the Braf+/+ mice (Fig. 2B).Consistent with their reduced BW at P3, the stomach milk con-tents were decreased in all BrafQ241R/+ ICR mice compared withthe Braf+/+ mice, suggesting that the BrafQ241R/+ ICR mice have afeeding problem (data not shown). The BrafQ241R/+ mice had a dis-tinctive gross appearance, including a rounded head, sparse andruffled fur, with or without a hunched appearance such as ky-phosis (Fig. 2C; Supplementary Material, Fig. S2). To verify thecraniofacial dysmorphism, we performed X-ray computed tom-ography (CT) in 8-week-old Braf+/+ and BrafQ241R/+ mice. Theskull lengths of the BrafQ241R/+ mice were significantly shorterthan those of the Braf+/+ mice, consistent with their BW andsize (Fig. 2D). In contrast, the skull widths of the Braf+/+ andBrafQ241R/+ mice were comparable, resulting in a significantincrease in the width/length ratio (Fig. 2D).

7350 | Human Molecular Genetics, 2015, Vol. 24, No. 25


http://hmg.oxfordjournals.org/lookup/suppl/doi:10.1093/hmg/ddv435/-/DC1







It has been reported that a few patients with CFC syndromeexhibit fast-growing or dystrophic nails, clinodactyly, crowdedor overlapping toes and syndactly (18,19). Thirteen of 19 (68%)BrafQ241R/+mice had long and/or dystrophic nails (SupplementaryMaterial, Fig. S3A). Extra digits on the forelimbs and hindlimbwere also present in 14 (14/28; 50%) and 1 (1/28; 3.6%) BrafQ241R/+

mice, respectively (Supplementary Material, Fig. S3A–C),although polydactyly has not been reported in patients withCFC syndrome. In addition, at 10 weeks of age, 6 of 9 (67%) femaleBrafQ241R/+ mice developed a unilateral ovarian cyst filled withserous fluid, which has not been reported in CFC syndromepatients (Supplementary Material, Fig. S4A). At 26 weeks of age,

Figure 1. The BrafQ241R/+ BALB/cmice exhibit a phenotype of growth retardation, fur abnormalities and heart anomalies. (A) Representative gross appearance of the Braf+/+

and BrafQ241R/+ mice at P11. (B) The BWs of the Braf+/+ and BrafQ241R/+ mice were measured at the indicated ages (0, 2, 7, 14, 21 and 28 days). The data are shown as the

mean ± SD [Braf+/+ (n = 53) and BrafQ241R/+ (n = 6)]. *P < 0.0001 versus Braf+/+ (Student’s t-test). (C) Survival of the Braf+/+ and BrafQ241R/+ mice. Braf+/+, n = 31; BrafQ241R/+,

n = 53. *P < 0.0001 versus Braf+/+ (log-rank test). (D) Gross morphology of the hearts from the P5 Braf+/+ and BrafQ241R/+ mice. The histogram indicates the HW to BW ratio

of the P5 Braf+/+ and BrafQ241R/+ mice. The data are shown as the mean ± SD [Braf+/+ (n = 41) and BrafQ241R/+ (n = 30)]. *P < 0.001 versus Braf+/+ (Student’s t-test). Scale

bar = 500 μm. (E) Coronal sections of the hearts from the P5 Braf+/+ and BrafQ241R/+ mice stained with H&E. The right panels show the high magnification images. The

solid arrowheads indicate the ASDs. The scale bars in left panels = 500 μm, and those in right panels = 100 μm. RA, right atrium; LA, left atrium; RV; right ventricle; LV;

left ventricle.

Human Molecular Genetics, 2015, Vol. 24, No. 25 | 7351






6 of 6 (100%) female BrafQ241R/+ mice had unilateral or bilateralsevere ovarian cysts, suggesting an increase in ovarian cyst sizewith age (Supplementary Material, Fig. S4B). In agreement withthese observations, half of the female BrafQ241R/+ mice wereinfertile.

We next examined the behavioral performance of theBrafQ241R/+ mice in the open-field test, light–dark transition test,

Y-maze test and fear-conditioning test. Although the mutantmice showed a slight decrease in the total rearing number in theopen-field test [230.5 ± 29.4 (Braf+/+) versus 202.8 ± 34.9 (BrafQ241R/+),mean ± SD (n = 12), P = 0.046 (Student’s t-test)] and the ‘movingspeed’ in open-field test [11.2 ± 1.2 (Braf+/+) versus 10.1 ± 1.0 cm/s(BrafQ241R/+), (n = 12), P = 0.035 (Student’s t-test)], as well as in theY-maze test [12.0 ± 1.8 (Braf+/+) versus 10.4 ± 1.3 cm/s (BrafQ241R/+),

Figure 2. Decreased survival, growth retardation, fur abnormalities, a hunched appearance, craniofacial dysmorphism and learning deficits in BrafQ241R/+ ICR mice. (A)

Survival of the Braf+/+ and BrafQ241R/+ mice. Braf+/+, n = 28; BrafQ241R/+, n = 49. *P < 0.0001 versus Braf+/+ (log-rank test). (B) The BWs of the Braf+/+ and BrafQ241R/+ mice were

measured at the indicated ages. The upper and lower panels indicate the growth curves of the male and female mice, respectively. The data are shown as the

mean ± SD [male: Braf+/+ (n = 13) and BrafQ241R/+ (n = 10); female: Braf+/+ (n = 13) and BrafQ241R/+ (n = 8)]. *P < 0.01, **P < 0.001, ***P < 0.0001 versus Braf+/+ (Student’s t-test). (C)Representative gross appearance of the 26-week-old Braf+/+ and BrafQ241R/+ mice. The arrow indicates the hunched appearance. (D) Morphometric measurements of

the skulls from 8-week-old Braf+/+ and BrafQ241R/+ mice. The data are shown as the mean ± SD [Braf+/+ (n = 7) and BrafQ241R/+ (n = 11)]. **P < 0.001, ***P < 0.0001 versus Braf+/+

(Student’s t-test). (E) The Braf+/+ and BrafQ241R/+ mice were placed in a conditioning chamber, and three tone-shock pairs were delivered at the indicated points. A tone

was sounded for 30 s as the CS (bars) and a mild footshock was delivered for 2 s (arrows) at the end of the CS as the US (conditioning: top panel). Twenty-four hours

later, the mice were placed back into the same conditioning chamber and the animals’ freezing behavior was measured (context test: middle panel). Three to four hours

later, the mice were placed into a distinct chamber for 6 min (cued test: bottom panel). Three minutes after starting the cued test, the tone (CS) was sounded for 3 min

(bar). The freezing behavior was expressed as a percentage of the immobility time during each 1 min period. The data are the mean ± SD [Braf+/+ (n = 12) and BrafQ241R/+

(n = 11)], and the P-values for the comparison between the Braf+/+ and BrafQ241R/+ mice were determined by repeated-measures ANOVA.




(n = 12), P = 0.026 (Student’s t-test)], the overall behavior of theBrafQ241R/+ mice was essentially normal in the open-field test,light–dark transition test and Y-maze test (data not shown).Thus, despite the multiple abnormalities, including craniofacialdysmorphism and dystrophic nails, the locomotor activity, emo-tional or anxiety-like behavior and working memory were essen-tially normal in the BrafQ241R/+ mice. In contrast, the BrafQ241R/+

mice showed an obvious defect in the contextual fear-condition-ing test. In this experiment, an aversive mild footshock (uncondi-tioned stimulus; US) was paired with an auditory conditionedstimulus (CS) within a novel environment. Twenty-four hoursafter the training, the Braf+/+mice exhibited marked fear, freezingbehavior, in response to re-presentation of the context (contexttest), suggesting thatmice learned and rememberedanassociationbetween the context and US (Fig. 2E). The freezing behavior of theBrafQ241R/+ mice was significantly reduced compared with that ofthe Braf+/+ mice, suggesting a learning deficit in these mice(Fig. 2E). Neither the Braf+/+ nor BrafQ241R/+mice exhibited an appar-ent freezing response to a CS delivered in a different context (cuedtest), suggesting that mice of both genotypes failed to learn an as-sociation between the CS and US in the cued test (Fig. 2E). Wemayhave to optimize our experimental conditions of the cued test toanalyze themice on the ICRbackground,which exhibit very low le-vels of freezing in the fear-conditioning test compared with otherstrains, e.g. C57BL/6 (20).

The BrafQ241R/+ ICR mice exhibit ASDs and pulmonarystenosis

We next evaluated the heart structure and function in sevenBrafQ241R/+ adult mice as well as six Braf+/+ mice. Color-flow Dop-pler echocardiography showed visible blood flow from the left at-rium to right atrium in three of seven BrafQ241R/+ mice, suggesting

the presence of ASDs (Fig. 3A). The pulsed-wave Doppler wave-forms at the pulmonary valve in the Braf+/+ mice showed a lam-inar flow with normal outflow velocities (Fig. 3B), whereas, in allBrafQ241R/+ mice, the ejection speed was higher than in the Braf+/+

mice, suggesting that these BrafQ241R/+ mice have pulmonarystenosis (Fig. 3B). M-mode echocardiography to evaluate theheart function revealed that the left ventricular (LV) internalend-diastolic dimension (LVIDd), LV internal end-systolic dimen-sion (LVIDs), LV diastolic posterior wall thickness (LVPWTd) andend-diastolic interventricular septum (IVSd) in the Braf+/+ andBrafQ241R/+ mice were comparable (Fig. 3C; Supplementary Mater-ial, Table S4). Likewise, the LV fractional shortening (LVFS) andejection fraction (EF), an index of systolic function, in the Braf+/+

and BrafQ241R/+ mice were comparable (Fig. 3D). These data sug-gest that the BrafQ241R/+ mice have structural abnormalities, in-cluding ASDs and pulmonary stenosis, without obvious systolicand diastolic dysfunction.

Pulmonary valve stenosis was induced by thickenedvalve leaflets, without evidence of infundibular stenosisin the BrafQ241R/+ ICR embryos

To evaluate the development of the cardiac valves and heartstructure, we performed histologic examination of the heart inBrafQ241R/+ embryos at E16.5. Four of seven BrafQ241R/+ embryoshad thickened pulmonary valve leaflets compared with theBraf+/+ mice (Fig. 4A and B). Four of seven BrafQ241R/+ embryoshad a dramatic increase in the density of the trabeculae (hyper-trabeculation) (lower panel, Fig. 4A), althoughmild hypertrabecu-lation was observed in two of eight Braf+/+ embryos (data notshown). Measurement of the thickness of cardiac valves revealedthat the pulmonary and mitral valve leaflets were significantly

Figure 3. Pulmonary stenosis and ASDs in the BrafQ241R/+ ICR mice. Color-flow (A) and pulsed Doppler (B) echocardiograms from the parasternal long-axis view in the

BrafQ241R/+ mice. The arrow indicates an ASD in the BrafQ241R/+ mice. (B) Normal waveforms in the Braf+/+ mice and high-velocity waveforms during pulmonary stenosis

in the BrafQ241R/+mice. The histogram indicates the PApeak velocity of the Braf+/+ and BrafQ241R/+mice. LV, left ventricle; RV, right ventricle; LA, left atrium; RA, right atrium;

ASD, atrial septal defect; PA, pulmonary artery. (C) Representative M-mode echocardiograms from the 26- to 31-week-old Braf+/+ and BrafQ241R/+ mice. (D) The fractional

shortening (FS) and ejection fraction (EF) in the Braf+/+ and BrafQ241R/+ mice. The data are shown as the mean ± SD [Braf+/+ (n = 6) and BrafQ241R/+ (n = 7)].





thickened in the BrafQ241R/+ embryos compared with those in theBraf+/+ embryos (Fig. 4C). These results indicate that the BrafQ241R/+ embryos show hypertrabeculation and thickened pulmonaryandmitral valves at E16.5, whichmay lead to the pulmonarysten-osis observed by echocardiography in the BrafQ241R/+ mice at 26–31weeks of age.

The BrafQ241R/+ ICR mice do not develop cardiomyocytehypertrophy

In addition to the structural defects in the heart, including pul-monary valve stenosis and ASDs, patients with CFC syndromefrequently exhibit hypertrophic cardiomyopathy (5,6). Histologic

analysis showed that the heart size in the 26-week-old BrafQ241R/+

mice was comparable with that in the Braf+/+ mice (Fig. 5A). TheHW to BW ratio was significantly higher in the BrafQ241R/+ micethan in the Braf+/+ mice (Fig. 5B). However, there were no differ-ences in cardiomyocyte size (Fig. 5C) and number (data notshown) in the Braf+/+ and BrafQ241R/+ mice. The mRNA levels ofthe fetal cardiac genes Myh7 (mice) and MYH6 (human), whichare reactivated in the heart tissues of mice and humans withhypertrophic cardiomyopathy (21,22), were similar in the Braf+/+

and BrafQ241R/+mice (data not shown). Likewise, therewere no dif-ferences in themRNA levels of several hypertrophic cardiomyop-athy/heart failure markers, including Nppb and Atp2a2 (23,24)(data not shown). Sirius staining did not revealed fibrosis in the

Figure 4. Heart defects in the BrafQ241R/+ ICR embryos. H&E-stained sections of E16.5 embryonic Braf+/+ and BrafQ241R/+ hearts. (A) Serial sections from anterior to posterior.

The arrows and solid arrowheads indicate a dramatic increase in the density of the trabeculae and thickened valves, respectively. Scale bar = 500 μm. LV, left ventricle; RV,

right ventricle; LA, left atrium; RA, right atrium; PV, pulmonary valve; AV, aortic valve; TV, tricuspid valve; MV, mitral valve. (B) Representative image of the increased

thickness of the pulmonary valves in the E16.5 BrafQ241R/+ embryos. Scale bar = 100 μm. (C) The thicknesses of the cardiac valve leaflets were measured at their largest

diameter in serial sections of E16.5 Braf+/+ and BrafQ241R/+ embryos. The data are shown as the mean ± SD [Braf+/+ (n = 8) and BrafQ241R/+ (n = 7)]. *P < 0.05, **P < 0.0001

versus Braf+/+ (Student’s t-test).



Braf+/+ and BrafQ241R/+ mice (Fig. 5D). These observations sug-gested that the BrafQ241R/+ mice have a higher HW to BW ratiothan the Braf+/+ mice, which is not associated with hypertrophiccardiomyopathy.

To further explorewhether BrafQ241R/+micehave cardiomyocytedamage, we examined the ultrastructure of the cardiomyocytes of

the LV using transmission electron microscopy. Electron micros-copy analysis indicated a normal sarcomeric structure and sarco-plasmic reticulum in the Braf+/+ and BrafQ241R/+ mice (Fig. 5E).However, the mitochondria from two of four BrafQ241R/+ mice wereirregular in shape and heterogeneous in size (Fig. 5E and F). Half ofthe remaining mice had numerous granules in the mitochondria,

Figure 5. Cardiac phenotype of the BrafQ241R/+ ICR mice. (A) Transverse sections of the hearts from 26-week-old Braf+/+ and BrafQ241R/+ mice stained with H&E. Scale

bar = 1 mm. (B) The HW to BW ratios of the Braf+/+ and BrafQ241R/+ mice at 6, 12 and 23–31 weeks of age. The data are shown as the mean ± SD [Braf+/+ (n = 6–10) and

BrafQ241R/+ (n = 5–10)]. *P < 0.05, **P < 0.001, ***P < 0.0001 versus Braf+/+ (Student’s t-test). (C) Wheat germ agglutinin-stained transverse sections of LV cardiomyocytes in

26-week-old Braf+/+ and BrafQ241R/+ mice. The data are shown as the mean ± SD [Braf+/+ (n = 6) and BrafQ241R/+ (n = 7)]. Scale bar = 50 μm. (D) Sirius red-stained transverse

sections of the LV cardiomyocytes in 26-week-old Braf+/+ and BrafQ241R/+ mice. Scale bar = 20 μm. (E and F) Abnormal mitochondrial morphology in the BrafQ241R/+ ICR

mice. Electron micrographs of the LV cardiomyocytes in Braf+/+ (n = 3) and BrafQ241R/+ (n = 4) mice at from 22 to 25 weeks of age (E and F). The scale bars in the upper

panels = 5 μm, and those in the lower panels = 1 μm (E). (F) The electron micrograph shows numerous granules in the mitochondria of the BrafQ241R/+ mice. Scale

bar = 1 μm. (G) The protein extracts from 12-week-old Braf+/+ and BrafQ241R/+ hearts were subjected to western blotting with the indicated antibodies. The arrowheads

indicate the bands corresponding to each protein. The band intensity was quantified using ImageJ software and then normalized to GAPDH. The data are shown as

the mean ± SD [Braf+/+ (n = 6) and BrafQ241R/+ (n = 5)]. *P < 0.05 versus Braf+/+ (Student’s t-test).



an index of cell injury (Fig. 5F). To investigate whether the mito-chondrial function was altered in BrafQ241R/+mice, western blottinganalysiswasperformedusing cell extracts derived from the hearts.The levels of the MFN1 and OPA1 mitochondrial fusion proteins,but not MFN2, were significantly higher in the BrafQ241R/+ micethan in the Braf+/+ mice (Fig. 5G). In contrast, there were no differ-ences in the levels of the DLP1 and FIS1 mitochondrial fission pro-teins (Fig. 5G). These observations suggest that minor cell injurymay occur in the myocardium of BrafQ241R/+ mice, although thehistologic and mRNA expression analyses of the heart tissues didnot identify any changes, such as a hypertrophic reaction orfibrosis.

The AKT pathway is down-regulated in BrafQ241R/+

ICR embryos and adult hearts

Finally, to investigate whether RAS/MAPK signaling was alteredin the BrafQ241R/+ mice, we performed western blotting using celllysates of whole mouse embryos and adult hearts. The levels ofthe phosphorylated ERK and p38 proteins in the BrafQ241R/+

whole embryos and adult hearts were similar to those of theBraf+/+ samples (Fig. 6A and B). In contrast, the levels of phos-phorylated AKT (Ser473), which is not a direct target of BRAF,

were significantly lower in the BrafQ241R/+ whole embryos(Fig. 6A). The levels of phosphorylated AKT (Thr308) were signifi-cantly lower in both BrafQ241R/+ whole embryos and adult heartsthan in the Braf+/+ samples (Fig. 6A and B).

DiscussionIn the current study, we backcrossed Braf Q241R B6mice (13) intothe BALB/c (inbred) or ICR (outbred) genetic background. AllBrafQ241R/+ BALB/c mice died between birth and 24 weeks. In con-trast, 31% of the heterozygous BrafQ241R/+ ICR mice survived forover 74weeks. Both the BrafQ241R/+ BALB/c and ICRmice displayedthe characteristic features of CFC syndrome, including growth re-tardation, sparse and ruffled fur, an increase in the HW to BWratio and ASDs. In addition, the BrafQ241R/+ ICRmice had addition-al CFC syndrome-related phenotypes, such as a hunched appear-ance, craniofacial dysmorphism, long and/or dystrophic nails,learning deficits and pulmonary valve stenosis. In contrast,extra digits and ovarian cysts observed in the BrafQ241R/+ ICRmice have not been reported in CFC syndrome patients. We con-clude that themice expressing BrafQ241Rmutationwill help elu-cidate the pathologic features of patients with CFC syndromefrom birth to adulthood.

Figure 6. Decreased AKT signaling in BrafQ241R/+ ICR embryos andmouse hearts. (A and B) The protein extracts fromwhole embryos (at E13.5) and adult hearts (at 12 week

old) of the Braf+/+ and BrafQ241R/+ micewere subjected towestern blottingwith the indicated antibodies. The arrowheads indicate the bands corresponding to each protein.

The band intensity of each protein was quantified using ImageJ software, and then the phosphorylated proteins were normalized to the corresponding non-

phosphorylated proteins. The data are shown as the mean ± SD [Braf+/+ (n = 6) and BrafQ241R/+ (n = 5)]. *P < 0.05 versus Braf+/+ (Student’s t-test).



Two knock-in mice expressing Braf mutations have been re-ported: mice expressing the BRAF L597V mutation (BRAF L597Vmice) and mice expressing a hypomorphic V600E allele (BRAFV600E mice) (25,26). The BRAF L597V mutation has been identi-fied in three patients with Noonan syndrome/CFC syndrome(11,27,28), whereas the V600E mutation has been identified insomatic cancers, but not in CFC syndrome. These mice havebeen shown to exhibit growth retardation, a reduced life spanand craniofacial dysmorphism (25,26), which were observed inour BrafQ241R/+ ICR mice. BRAF V600E mice have been reported toexhibit seizures, one of the CFC syndrome phenotypes. In con-trast, our BrafQ241R/+ ICR mice showed fur abnormalities, ahunched appearance (such as kyphosis) and long and/or dys-trophic nails, which has not been reported in other mouse mod-els. Most patients with CFC syndrome show neurologicabnormalities, including intellectual disability and a develop-mental delay with speech. Indeed, we have found, for the firsttime, that the BrafQ241R/+ ICR mice display learning deficits.Three individuals with BRAF mutations have been reported tohave developed acute lymphoblastic leukemia and non-Hodg-kin’s lymphoma (5,12). Lung adenoma and melanocytic hyper-plasia have been observed in BRAF V600E mice, and benigntumors, including skin papillomas and intestinal polyps, havebeen observed in BRAF L597V mice (25,26). Our survivingBrafQ241R/+ ICRmice had not developed obvious benign andmalig-nant tumors as of 1 year of age (data not shown), which is consist-ent with the fact that malignant tumors are rare in CFCsyndrome. Collectively, these results suggest that our BrafQ241R/+

ICR mice will be useful for exploring the pathogenesis of the furabnormalities, skeletal abnormalities, long and/or dystrophicnails and learning deficits in CFC syndrome.

Heart defects are one of the most common features observedin individuals with CFC syndrome. Valvular pulmonary stenosis,hypertrophic cardiomyopathy, ASDs, VSDs and arrhythmia havebeen observed in 50, 38, 28, 11 and 20% of patients with CFC syn-drome, respectively (6,12). Indeed, echocardiography of the 26- to31-week-old BrafQ241R/+ ICR mice suggested pulmonary stenosisand ASDs. Histologic analysis of the heart showed that theBrafQ241R/+ ICR embryos had thickened pulmonary and mitralvalve leaflets and hypertrabeculation, confirming the finding ofpulmonary stenosis observed by echocardiography. ASDs werefrequently observed in the neonatal BrafQ241R/+ BALB/c mice, sug-gesting that the Braf Q241R allele contributes to the developmentof congenital heart disease. In contrast, the surviving BrafQ241R/+

ICR mice did not develop hypertrophic cardiomyopathy, even at1 year of age (data not shown), although an increase in the heartto BW ratio was observed in the BrafQ241R/+ ICR mice. BRAF L597Vand BRAF V600E mice have been found to exhibit cardiomegalyand hypertrophic cardiomyopathy (25,26). Congenital heart de-fects, such as pulmonary valve stenosis, ASDs and VSDs, havenot been reported in these mice (25,26). These results suggestthat the Q241R mutation, which is frequently identified in CFCsyndrome, affects the development of the heart structure andvalve leaflets. However, it is possible that the differences in theheart defects in these Braf mutant mice depend on multiple fac-tors, such as the genetic background, environmental conditionsand the type of Braf mutation.

The genetic background influences the phenotype of genetic-ally engineered mice. In mouse models of RASopathies, Arakiet al. reported that approximately half of the knock-in mice (ona mixed 129S4/SvJae × C57BL/6J background) expressing thePTPN11 D61G mutation, which is associated with Noonan syn-drome, survived for >3 weeks (29). In contrast, almost all mice(∼97%) were dead within 3 weeks when these PTPN11 D61G

mice were crossed with C57BL/6J mice. Furthermore, these miceon a BALB/c background were equally as viable as those on a129S4/SvJae × C57BL/6J background. Urosevic et al. reported thatsurvival was increased when knock-in mice expressing a hypo-morphic BRAF V600E allele (on a mixed 129Sv/J × C57BL/6 back-ground) were backcrossed onto an ICR background, but notonto a C57BL/6 background (26). In our mouse model of CFC syn-drome, the BALB/c or ICR genetic background ameliorated theembryonic lethality associated with the Braf Q241R B6 mice, des-pite the differences in the survival period after birth. However,the BrafQ241R/+ B6 mice exhibited embryonic/neonatal lethality,with edema, heart defects, lymphatic defects and craniofacial ab-normalities (13). These findings demonstrate that mice on the129, BALB/c or ICR genetic backgrounds are more likely to havea survival advantage than those on a C57BL/6 background inmouse models of RASopathies, suggesting that there are strain-specific modifiers.

It remains unknown why phosphorylated AKT (Ser473)and/or AKT (Thr308) expression were reduced in the BrafQ241R/+

ICR embryos and adult hearts. Interestingly, we have previouslyidentified these phenomena in BrafQ241R/+ B6 embryo hearts(13), although they have not been found in other knock-inmouse models for RASopathies, including Noonan syndrome,Costello syndrome and Noonan syndrome with multiple lentigi-nes. Constitutive MEK/ERK signaling activation in the heart isassociated with cardiac hypertrophy (30). Likewise, myocardial-specific transgenic mice expressing AKT have been shown todevelop increased cardiomyocyte size and HW to BW ratios(31). In the present study, the BrafQ241R/+ ICR mice exhibited anincrease in the HW to BW ratio, whereas the expression of phos-phorylated AKT (Thr308), but not phosphorylated ERK, wasdecreased. These results suggest that the decreased expressionof phosphorylated AKT (Thr308) could represent a negative feed-back mechanism for preventing hypertrophic cardiomyopathy.

Recent studies have demonstrated that treatment with theMEK inhibitors PD184352 and PD0325901 or with a fibroblastgrowth factor receptor 1 (FGFR1) inhibitor, SU-5402, amelioratesthe cell motility defects in zebrafish expressing CFC mutationsduring early embryogenesis (32,33). We have reported that pre-natal treatment with PD0325901 or with a histone 3 demethylaseinhibitor, GSK-J4 or NCDM-32b, rescued the embryonic and post-natal lethality in BrafQ241R/+ B6 mice (13). Moreover, treatmentwith both PD0325901 and GSK-J4 further rescued the embryoniclethality compared with the BrafQ241R/+ B6 mice treated withPD0325901 or GSK-J4 alone (13). In the present study, we usedthe ICR/CD-1 outbredmouse strain. Because this strain of outbredmice has genetic heterozygosity comparablewith the differencesin humans, these mice are commonly used for pharmacologicstudies (34–36). Thus, our new mouse model, the BrafQ241R/+ ICRmice, will be useful not only for elucidating the pathophysiologyof the clinical features observed in CFC patients but also forevaluating the therapeutic effects of drugs and compounds onCFC syndrome.

Materials and MethodsMice

The generation of BrafQ241RB6micehas been previously described(13). C57BL/6J, BALB/c and ICRmicewere purchased from CharlesRiver Laboratories Japan (Yokohama, Japan). The mice weremaintained under a 12-h light/12-h dark cycle and fed a conven-tional laboratory diet, Labo MR Stock (Nihon Nosan Kogyo, Yoko-hama, Japan). Generation of BrafQ241R mice on a BALB/c or ICR



background: the surviving PD0325901-treated BrafQ241R B6 micewere generated as previously described (13) and backcrossedonto BALB/c mice for three generations and onto ICR mice formore than seven generations. The animal experiments wereapproved by the Animal Care and Use Committee of TohokuUniversity.

Genotyping

The genomic DNA was isolated from the tail tissue using theDNeasy Blood and Tissue Kit (for embryos; QIAGEN, Hilden, Ger-many) or the Maxwell 16 Mouse Tail DNA Purification Kit (foradult mice; Promega, Madison, WI). The Braf+/+, BrafQ241R/+ andBrafQ241R/Q241R mice were genotyped by polymerase chain reac-tion using KOD FX Neo (TOYOBO, Osaka, Japan) with the primers5′ GTGTTGTTCTGCCCATACTTACTGC 3′ and 5′ GTGACTTAATGTACAGCATGGATCA 3′.

X-ray CT scan

Themouse skulls were scanned using CT scans (Latheta LCT200;Hitachi Aloka Medical Ltd, Tokyo, Japan) and visualized with theVGStudio Max 2.0 software (Volume Graphics GmbH, Heidelberg,Germany). The craniofacial measurements weremade accordingto The Jackson Laboratory standard protocols and procedures(http://craniofacial.jax.org/standard_protocols.html).

Behavioral tests

The behavioral tests were performed with male littermates. Theopen-field test, light–dark transition test, Y-maze test, and con-textual and cued fear-conditioning tests were performed withequipment from O’Hara & Co. Ltd (Tokyo, Japan), essentially aspreviously described, with slight modifications (37,38). In thesetests, the digital images of the mice were recorded with a CCDcamera and were analyzed with ImageJ-OFC, ImageJ-LD1, TimeYM1 or Time FZ1 software (O’Hara & Co. Ltd). ImageJ-OFC and Im-ageJ-LD1 aremodified software products based on the ImageJ pro-gram (http://rsb.info.nih.gov/ij). The sequence of the behavioraltests described above was started with Braf+/+ and BrafQ241R/+ ICRmice at 7–10 weeks of age, and took 2–5 weeks to complete. Forthe open-field test, the individual mice were placed at the cornerof the open-field apparatus (50 × 50 × 40 cm) illuminated at 150lux, and their behavior was recorded over a 20 min test periodfrom the obtained digital images. The distance traveled, verticalactivity (rearing, measured by counting the number of photobeaminterruptions) and the time spent in the central area were calcu-lated by analyzing the images using the ImageJ-OFC software.For the light–dark transition test, the mice were allowed to movefreely between the two equal-size chambers (21 × 21 × 25 cm),which were separated by a door. One chamber was illuminatedat 390 lux, and the other chamber was dark (10 lux). At the startof the test, the mice were placed into the dark chamber, andtheir behavior was monitored for 10 min. The first latency toenter the light chamber, total number of transitions, time spentin each chamber and distance traveled were calculated by analyz-ing the images using the ImageJ-LD1 software. For the Y-mazetest, the spontaneous alternation behavior was examined in anacrylic Y-shaped maze. This maze consists of three identicalarms (length: 40 cm; high: 12 cm; width at bottom: 3 cm; widthat top: 12 cm) converging at the center of a triangular area toform a symmetrical Y shape (120° of angular deviation fromeach other). A mouse was placed at the end of one arm and wasallowed to freely explore the maze for 5 min. The sequence and

total number of arm entries were recorded. The percent alterna-tion is the number of triads containing entries into all threearms divided by the maximum possible number of alternations(the total number of arms entered minus 2) × 100. The data wereanalyzed automatically using the Time YM1 software. For thefear-conditioning test, the individual mice were placed into theconditioning chamber (10.5 × 10.5 × 10.5 cm, 200 lux, inside asound-proof box),whichhad transparent acrylicwalls anda stain-less steel grid floor, and allowed to freely explore the apparatus for88 s. An auditory CS (60 dB, 10 kHz) was sounded as the CS for 30 sthrougha speakeron the ceiling of the sound-proof box.At the endof the tone presentation, a scrambled electrical footshock (0.3 mA,2 s) was delivered from the metal grid floor as the US. Two moretone-shock pairs were given at 88 s intervals (conditioning ses-sion). Twenty-four hours after the conditioning, the mice wereplaced back into the conditioning chamber for 5 min without theCS and US presentations (context test). At 3–5 h after the contexttext, the mice were placed into a distinct chamber with opaquewhite walls for 6 min (cued test). Three minutes after startingthe test, the tone (CS) was sounded for 3 min. The images werecaptured at a rate of one frame per second. For each pair of succes-sive frames, the area (pixels) within which themousemoved wasmeasured. When the area was below a certain threshold (100 pix-els), the mouse was judged to be ‘immobile’. The optimal thresh-old (number of pixels) for the definition of immobility wasdetermined by adjustment based on human observation. The im-mobilitywas automatically determined from the images using theTime FZ1 software and expressed as a percentage of the immobil-ity time in each 1 min period.

Western blotting

Whole mouse embryos and the LV myocardia of adult mice werelysed in lysis buffer [(10 m Tris–HCl, pH 8.0 and 1% sodiumdode-cyl sulfate (SDS)] containing phosphatase and protease inhibitorcocktails (P5726 and P8340, respectively; Sigma-Aldrich, St. Louis,MO). These lysates were centrifuged at 20 400g for 30 min at 15°C,and the protein concentration was determined by the Bradfordmethod using the Bio-Rad Protein Assay (Bio-Rad Laboratories,Hercules, CA), with bovine serum albumin as the standard. Thelysates were subjected to SDS–polyacrylamide gel electrophoresis[4–20% Criterion TGX Precast Gels (Bio-Rad Laboratories)] andtransferred to nitrocellulose membranes (Trans-Blot Turbo trans-fer pack; Bio-Rad Laboratories) using the Trans-Blot Turbo Trans-fer System (Bio-Rad Laboratories). Antibodies specific for thefollowing proteins were used (with catalog numbers in paren-theses): ERK1/2 (9102), phospho-ERK1/2 (9101), p38 (9212), phos-pho-p38 (4511), AKT (9272), phospho-AKT (on Ser473; 9018),phospho-AKT (on Thr308; 2965) and GAPDH (2118) from Cell Sig-naling (Danvers, MA); MFN1 (sc-50330) and FIS1 (sc-98900) fromSanta Cruz Biotechnology (Santa Cruz, CA, USA); DLP1 (611112)and OPA1 (612606) from BD Transduction Laboratories (San Jose,CA, USA); andMFN2 (M6319) from Sigma-Aldrich. All of themem-branes were visualized using the Western Lightning ECL-Plus Kit(PerkinElmer,Waltham,MA). The band intensitieswere quantifiedusing ImageJ software. The levels of the phosphorylated proteinswere normalized to the non-phosphorylated proteins.

Histology and immunohistochemistry

The embryonic hearts were perfusedwith phosphate-buffered sa-line and 10% neutral-buffered formalin from the placenta. Theembryonic, neonatal (at P5) and adult hearts were fixed in 10%neutral-buffered formalin, and the fixed hearts were embedded



http://craniofacial.jax.org/standard_protocols.html






http://rsb.info.nih.gov/ij






in paraffin. The embedded tissues were sectioned at 6 (embryonichearts), 8 (neonatal hearts) or 3 μm (adult hearts). The sectionswere stained with hematoxylin and eosin (H&E) or picrosirius red.

To determine the cross-sectional areas of the cardiomyocytes,the sections were stained with fluorescein isothiocyanate-labeledwheat germ agglutinin (FITC-WGA, L4895; Sigma-Aldrich) for 1 hat room temperature. The slides were mounted with the ProLongGold antifade reagent with 4′, 6-diamidino-2-phenylindole. Thecardiomyocytes with nuclei were quantitated using ImageJsoftware.

Echocardiography

Transthoracic echocardiography was performed on miceanesthetized with 4.0–1.0% isoflurane (150–200 ml/min) with aNemio 35 ultrasound system (Toshiba) provided with a 12 MHzimaging transducer. The hemodynamics and LV systolic functionwere analyzed as previously described (39). An index of systolicfunction was defined as the percentage of LVFS with echocardio-graphic measurements of LVIDd and LVIDs using the followingstandard equation: % LVFS = [(LVIDd–LVIDs)/LVIDd] × 100. Pul-monary valve disease (stenosis) or septal defects were assessedby the Doppler examination, as previously described (39).

Electron microscopy

The left ventriclemyocardia of the adultmicewere dissected andfixed overnight with 2% paraformaldehyde and 2% in 0.1 M phos-phate buffer (PB) (pH 7.4) at 4°C. The fixed samples were thenwashed three times with 0.1 M PB for 30 min each, postfixedwith 2% osmium tetroxide in 0.1 M PB for 2 h at 4°C and dehy-drated in graded ethanol solutions. The samples were placed ina 70:30mixture of propylene oxide and Quetol-812 resin (NisshinEM, Tokyo, Japan) for 1 h, transferred to a fresh 100% resin andthen polymerized at 60°C for 48 h. Ultra-thin sections (70 nm)were cut using an ultramicrotome (Ultracut UCT; Leica, Vienna,Austria) and stainedwith 2% uranyl acetate and a lead stain solu-tion (Sigma-Aldrich). The stained sections were observed with atransmission electron microscope (JEM-1400Plus; JEOL, Tokyo,Japan).

Statistical analysis

All statistical analyses were performed using Prism software(ver. 6.01; GraphPad Software, Inc., San Diego, CA). The datawere analyzed using Student’s t-test for unpaired samples andthe χ2 test for the differences between the observed and expecteddistributions. The mouse survival was calculated using theKaplan–Meier method and analyzed using the log-rank test.The differences were considered significant at a P-value of < 0.05.

Supplementary MaterialSupplementary Material is available at HMG online.

AcknowledgementsWewish to thank Riyo Takahashi, Kumi Kato andYoko Tateda fortechnical assistance and Fumiko Date, Hiroaki Nagao, KenjiYoshiwara and Shingo Takahara for technical assistance andfor discussions of the experimental data. We also acknowledgethe support of the Biomedical Research Core of Tohoku Univer-sity Graduate School of Medicine.

Conflict of Interest statement. None declared.

FundingThis work was supported by the Funding Program for the NextGeneration of World-Leading Researchers (NEXT Program) fromthe Ministry of Education, Culture, Sports, Science and Technol-ogy of Japan to Y.A. (LS004), by the Practical Research Project forRare/Intractable Diseases from the Japan Agency for Medical Re-search and development, AMED to Y.A., by the Japan Society forthe Promotion of Science (JSPS) KAKENHI Grant Number26293241 to Y.A., and by JSPS KAKENHI Grant Numbers25860839 and 15K19598 to S.I.

References1. Malumbres, M. and Barbacid, M. (2003) RAS oncogenes: the

first 30 years. Nat. Rev. Cancer, 3, 459–465.2. Aoki, Y., Niihori, T., Narumi, Y., Kure, S. and Matsubara, Y.

(2008) The RAS/MAPK syndromes: novel roles of the RASpathway in human genetic disorders. Hum. Mutat., 29, 992–1006.

3. Tidyman, W.E. and Rauen, K.A. (2009) The RASopathies: de-velopmental syndromes of Ras/MAPK pathway dysregula-tion. Curr. Opin. Genet. Dev., 19, 230–236.

4. Rauen, K.A. (2013) The RASopathies.Annu. Rev. Genomics Hum.Genet., 14, 355–369.

5. Pierpont, M.E., Magoulas, P.L., Adi, S., Kavamura, M.I., Neri, G.,Noonan, J., Pierpont, E.I., Reinker, K., Roberts, A.E., Shankar, S.et al. (2014) Cardio-facio-cutaneous syndrome: clinical fea-tures, diagnosis, and management guidelines. Pediatrics,134, e1149–e1162.

6. Allanson, J.E., Anneren, G., Aoki, Y., Armour, C.M., Bondeson,M.L., Cave, H., Gripp, K.W., Kerr, B., Nystrom, A.M., Sol-Church, K. et al. (2011) Cardio-facio-cutaneous syndrome:does genotype predict phenotype? Am. J. Med. Genet. CSemin. Med. Genet., 157C, 129–135.

7. Niihori, T., Aoki, Y., Narumi, Y., Neri, G., Cave, H., Verloes, A.,Okamoto, N., Hennekam, R.C., Gillessen-Kaesbach, G., Wiec-zorek, D. et al. (2006) Germline KRAS and BRAF mutations incardio-facio-cutaneous syndrome. Nat. Genet., 38, 294–296.

8. Rodriguez-Viciana, P., Tetsu, O., Tidyman, W.E., Estep, A.L.,Conger, B.A., Cruz, M.S., McCormick, F. and Rauen, K.A.(2006) Germline mutations in genes within the MAPK path-way cause cardio-facio-cutaneous syndrome. Science, 311,1287–1290.

9. Narumi, Y., Aoki, Y., Niihori, T., Neri, G., Cave, H., Verloes, A.,Nava, C., Kavamura, M.I., Okamoto, N., Kurosawa, K. et al.(2007) Molecular and clinical characterization of cardio-facio-cutaneous (CFC) syndrome: overlapping clinical mani-festations with Costello syndrome. Am. J. Med. Genet. A.,143A, 799–807.

10. Davies, H., Bignell, G.R., Cox, C., Stephens, P., Edkins, S., Clegg,S., Teague, J., Woffendin, H., Garnett, M.J., Bottomley, W. et al.(2002) Mutations of the BRAF gene in human cancer. Nature,417, 949–954.

11. Sarkozy, A., Carta, C., Moretti, S., Zampino, G., Digilio, M.C.,Pantaleoni, F., Scioletti, A.P., Esposito, G., Cordeddu, V.,Lepri, F. et al. (2009) Germline BRAF mutations in Noonan,LEOPARD, and cardiofaciocutaneous syndromes: moleculardiversity and associated phenotypic spectrum. Hum. Mutat.,30, 695–702.

12. Abe, Y., Aoki, Y., Kuriyama, S., Kawame, H., Okamoto, N., Kur-osawa, K., Ohashi, H., Mizuno, S., Ogata, T., Kure, S. et al. (2012)Prevalence and clinical features of Costello syndrome andcardio-facio-cutaneous syndrome in Japan: findings from a




nationwide epidemiological survey. Am. J. Med. Genet. A.,158A, 1083–1094.

13. Inoue, S.,Moriya,M.,Watanabe,Y.,Miyagawa-Tomita, S.,Niihori,T., Oba, D., Ono, M., Kure, S., Ogura, T., Matsubara, Y. et al. (2014)New BRAF knockin mice provide a pathogenetic mechanism ofdevelopmental defects and a therapeutic approach in cardio-facio-cutaneous syndrome. Hum. Mol. Genet., 23, 6553–6566.

14. Wong Ramsey, K.N., Loichinger, M.H., Slavin, T.P., Kuo, S. andSeaver, L.H. (2014) The perinatal presentation of cardiofacio-cutaneous syndrome. Am. J. Med. Genet. A., 164, 2036–2042.

15. Reynolds, J.F., Neri, G., Herrmann, J.P., Blumberg, B., Coldwell,J.G., Miles, P.V. and Opitz, J.M. (1986) Newmultiple congenitalanomalies/mental retardation syndrome with cardio-facio-cutaneous involvement—the CFC syndrome. Am. J. Med.Genet., 25, 413–427.

16. Witters, I., Denayer, E., Brems, H., Fryns, J.P. and Legius, E.(2008) The cardiofaciocutaneous syndrome: prenatalfindingsin two patients. Prenat. Diagn., 28, 53–55.

17. Muth, J.N., Bodi, I., Lewis, W., Varadi, G. and Schwartz, A.(2001) A Ca2+-dependent transgenic model of cardiac hyper-trophy a role for protein kinase Cα. Circulation, 103, 140–147.

18. Armour, C.M. and Allanson, J.E. (2008) Further delineation ofcardio-facio-cutaneous syndrome: clinical features of 38 in-dividuals with proven mutations. J. Med. Genet., 45, 249–254.

19. Rauen, K.A., Pagon, R.A., Adam, M.P., Ardinger, H.H., Wallace,S.E., Amemiya, A., Bean, L.J.H., Bird, T.D., Dolan, C.R., Fong, C.T., Smith, R.J.H. and Stephens, K. (1993) (eds), In GeneReviews(R). University of Washington, Seattle, Seattle, WA.

20. Adams, B., Fitch, T., Chaney, S. and Gerlai, R. (2002) Altered per-formance characteristics in cognitive tasks: comparison of thealbino ICR andCD1mouse strains.Behav Brain Res, 133, 351–361.

21. Lowes, B.D., Minobe, W., Abraham, W.T., Rizeq, M.N., Bohl-meyer, T.J., Quaife, R.A., Roden, R.L., Dutcher, D.L., Robertson,A.D., Voelkel, N.F. et al. (1997) Changes in gene expression inthe intact human heart. Downregulation of alpha-myosinheavy chain in hypertrophied, failing ventricular myocar-dium. J. Clin. Invest., 100, 2315–2324.

22. Marin, T.M., Keith, K., Davies, B., Conner, D.A., Guha, P., Ka-laitzidis, D., Wu, X., Lauriol, J., Wang, B., Bauer, M. et al.(2011) Rapamycin reverses hypertrophic cardiomyopathy ina mouse model of LEOPARD syndrome-associated PTPN11mutation. J. Clin. Invest., 121, 1026–1043.

23. Schoenfeld, J.R., Vasser, M., Jhurani, P., Ng, P., Hunter, J.J.,Ross, J. Jr., Chien, K.R. and Lowe, D.G. (1998) Distinct molecu-lar phenotypes in murine cardiac muscle development,growth, and hypertrophy. J. Mol. Cell Cardiol., 30, 2269–2280.

24. Nakagawa, O., Ogawa, Y., Itoh, H., Suga, S., Komatsu, Y., Kishi-moto, I., Nishino, K., Yoshimasa, T. andNakao, K. (1995) Rapidtranscriptional activation and early mRNA turnover of brainnatriuretic peptide in cardiocyte hypertrophy. Evidence forbrain natriuretic peptide as an “emergency” cardiac hormoneagainst ventricular overload. J. Clin. Invest., 96, 1280–1287.

25. Andreadi, C., Cheung, L.K., Giblett, S., Patel, B., Jin, H., Mercer,K., Kamata, T., Lee, P., Williams, A., McMahon, M. et al. (2012)The intermediate-activity (L597V)BRAFmutant acts as an epi-static modifier of oncogenic RAS by enhancing signalingthrough the RAF/MEK/ERK pathway. Genes Dev., 26, 1945–1958.

26. Urosevic, J., Sauzeau, V., Soto-Montenegro, M.L., Reig, S.,Desco, M., Wright, E.M., Canamero, M., Mulero, F., Ortega, S.,

Bustelo, X.R. et al. (2011) Constitutive activation of B-Raf in themouse germ line provides a model for human cardio-facio-cutaneous syndrome. Proc. Natl. Acad. Sci. USA, 108,5015–5020.

27. Pierpont, E.I., Pierpont, M.E., Mendelsohn, N.J., Roberts, A.E.,Tworog-Dube, E., Rauen, K.A. and Seidenberg, M.S. (2010) Ef-fects of germline mutations in the Ras/MAPK signaling path-way on adaptive behavior: cardiofaciocutaneous syndromeand Noonan syndrome. Am. J. Med. Genet. A., 152A, 591–600.

28. Stevenson, D.A., Schwarz, E.L., Carey, J.C., Viskochil, D.H.,Hanson, H., Bauer, S., Weng, H.Y., Greene, T., Reinker, K.,Swensen, J. et al. (2011) Bone resorption in syndromes of theRas/MAPK pathway. Clin. Genet., 80, 566–573.

29. Araki, T., Chan, G., Newbigging, S., Morikawa, L., Bronson, R.T.and Neel, B.G. (2009) Noonan syndrome cardiac defects arecaused by PTPN11 acting in endocardium to enhance endo-cardial-mesenchymal transformation. Proc. Natl. Acad. Sci.USA, 106, 4736–4741.

30. Bueno, O.F., DeWindt, L.J., Tymitz, K.M.,Witt, S.A., Kimball, T.R., Klevitsky, R., Hewett, T.E., Jones, S.P., Lefer, D.J., Peng, C.F.et al. (2000) The MEK1-ERK1/2 signaling pathway promotescompensated cardiac hypertrophy in transgenic mice.EMBO J., 19, 6341–6350.

31. Shioi, T., McMullen, J.R., Kang, P.M., Douglas, P.S., Obata, T.,Franke, T.F., Cantley, L.C. and Izumo, S. (2002) Akt/protein ki-nase B promotes organ growth in transgenic mice. Mol. Cell.Biol., 22, 2799–2809.

32. Anastasaki, C., Estep, A.L., Marais, R., Rauen, K.A. and Patton,E.E. (2009) Kinase-activating and kinase-impaired cardio-facio-cutaneous syndrome alleles have activity during zebra-fish development and are sensitive to small molecule inhibi-tors. Hum. Mol. Genet., 18, 2543–2554.

33. Anastasaki, C., Rauen, K.A. and Patton, E.E. (2012) Continuallow-level MEK inhibition ameliorates cardio-facio-cutaneousphenotypes in zebrafish. Dis. Model Mech., 5, 546–552.

34. Fujitani, T., Hojo, M., Inomata, A., Ogata, A., Hirose, A., Nishi-mura, T. and Nakae, D. (2014) Teratogenicity of asbestos inmice. J. Toxicol. Sci., 39, 363–370.

35. Cappon, G.D., Bowman, C.J., Campion, S.N., Chmielewski, G.,Hurtt, M.E., Finch, G.L. and Lewis, E.M. (2012) Developmentaltoxicity study of lersivirine inmice. Birth Defects Res. B Dev. Re-prod. Toxicol., 95, 225–230.

36. Elzeinova, F., Novakova, V., Buckiova, D., Kubatova, A. and Pe-knicova, J. (2008) Effect of low dose of vinclozolin on repro-ductive tract development and sperm parameters in CD1outbred mice. Reprod. Toxicol., 26, 231–238.

37. Miyakawa, T., Yamada, M., Duttaroy, A. and Wess, J. (2001)Hyperactivity and intact hippocampus-dependent learningin mice lacking the M1 muscarinic acetylcholine receptor.J Neurosci, 21, 5239–5250.

38. Takeuchi, H., Iba,M., Inoue,H., Higuchi,M., Takao, K., Tsukita, K.,Karatsu, Y., Iwamoto, Y., Miyakawa, T., Suhara, T. et al. (2011)P301Smutant human tau transgenicmicemanifest early symp-toms of human tauopathies with dementia and altered sensori-motor gating. PLoS One, 6, e21050.

39. Kokubo, H., Miyagawa-Tomita, S., Nakashima, Y., Kume, T.,Yoshizumi, M., Nakanishi, T. and Saga, Y. (2013) Hesr2 knock-outmice develop aortic valve diseasewith advancing age.Ar-terioscler. Thromb. Vasc. Biol., 33, e84–e92.



Documents

Adult mice expressing a Braf Q241R mutation on an ICR/CD-1