Activated Braf induces esophageal dilation and gastric epithelial hyperplasia in mice

Abstract

Germline mutations in BRAF are a major cause of cardio-facio-cutaneous (CFC) syndrome, which is characterized by heart defects, characteristic craniofacial dysmorphology and dermatologic abnormalities. Patients with CFC syndrome also commonly show gastrointestinal dysfunction, including feeding and swallowing difficulties and gastroesophageal reflux. We have previously found that knock-in mice expressing a Braf Q241R mutation exhibit CFC syndrome-related phenotypes, such as growth retardation, craniofacial dysmorphisms, congenital heart defects and learning deficits. However, it remains unclear whether Braf^Q241R/+ mice exhibit gastrointestinal dysfunction. Here, we report that Braf^Q241R/+ mice have neonatal feeding difficulties and esophageal dilation. The esophagus tissues from Braf^Q241R/+ mice displayed incomplete replacement of smooth muscle with skeletal muscle and decreased contraction. Furthermore, the Braf^Q241R/+ mice showed hyperkeratosis and a thickened muscle layer in the forestomach. Treatment with MEK inhibitors ameliorated the growth retardation, esophageal dilation, hyperkeratosis and thickened muscle layer in the forestomach in Braf^Q241R/+ mice. The esophageal dilation with aberrant skeletal-smooth muscle boundary in Braf^Q241R/+ mice were recovered after treatment with the histone H3K27 demethylase inhibitor GSK-J4. Our results provide clues to elucidate the pathogenesis and possible treatment of gastrointestinal dysfunction and failure to thrive in patients with CFC syndrome.

Introduction

The proto-oncogene BRAF is a serine/threonine protein kinase that plays major roles in the regulation of the phosphorylation and activation of MEK/ERK as RAS signal effectors. BRAF has three conserved domains: conserved region (CR) 1, CR2 and CR3. CR1 contains the RAS-binding domain and the cysteine-rich domain, which are involved in the membrane recruitment of RAF and the interaction with RAS; CR2 contains the 14-3-3 scaffolding protein binding sites; and CR3 contains the catalytic kinase domain (1). Somatic mutations in BRAF occur at a high frequency in numerous human cancers, particularly melanoma, papillary thyroid carcinoma and colorectal cancer (2). BRAF is ubiquitously expressed in murine organs at mid-gestation, and it is highly expressed in the fetal brain and adult cerebrum and testes (3,4). Braf knockout mice die between E10.5 and E12.5, owing to enlarged blood vessels, apoptotic death of differentiated endothelial cells and neuronal defects (5).

Cardio-facio-cutaneous (CFC) syndrome is a multiple congenital anomaly disorder, characterized by heart diseases, dysmorphic facial features, failure to thrive, intellectual disability, dermatologic abnormalities, seizures and gastrointestinal dysfunction (6,7). In 2006, our group and another group reported that CFC is an autosomal dominant genetic disorder caused by mutations in BRAF, MAP2K1, MAP2K2 and KRAS, all of which belong to the RAS/MAPK pathway, which regulates cell differentiation, proliferation, survival, apoptosis and senescence (8–10). The most common somatic mutation in BRAF is a substitution of the residue V600E in the CR3 domain, which accounts for 90% of BRAF mutations in cancers. In contrast, the most common mutation in CFC syndrome is the BRAF Q257R mutation in the cysteine-rich domain within the CR1. The BRAF mutations found in patients with CFC syndrome lead to the activation of the RAS/MAPK pathway as well as cancer-related BRAF mutations, such as BRAF V600E (8,11).

We have previously generated heterozygous knock-in mice on a C57BL/6 J genetic background (Braf^Q241R/+ B6) that express the Braf Q241R mutation, which corresponds to the human BRAF Q257R mutation. These mice show embryonic and neonatal lethality resulting from craniofacial abnormalities, edema, liver necrosis and congenital heart defects, including pulmonary valve stenosis and ventricular septal defects (12). Prenatal treatment with the MEK inhibitor PD0325901 or the histone H3K27 demethylase inhibitor GSK-J4 rescues the embryonic and neonatal lethality in Braf^Q241R/+ B6 mice. Furthermore, in a recent study, we found that the embryonic and neonatal lethality improved when the surviving Braf^Q241R/+ B6 mice treated with PD0325901 were crossed with closed-colony ICR/CD-1 mice. Approximately, 50% of the Braf^Q241R/+ mice on an ICR background (Braf^Q241R/+) died within 4 weeks after exhibiting failure to thrive. The surviving adult Braf^Q241R/+ mice showed typical CFC syndrome features, including cardiomegaly, pulmonary valve stenosis, atrial septal defects, craniofacial dysmorphism, mild growth retardation, learning deficits, hair abnormality, a hunched appearance and long and/or dystrophic nails (13). Furthermore, the Braf^Q241R/+ mice displayed extra digits and ovarian cysts, which have not been found in CFC syndrome. However, the primary cause of death in the Braf^Q241R/+ neonatal mice with failure to thrive remains unknown.

Feeding problems in patients with CFC syndrome are common causes of failure to thrive and poor growth in the neonatal period. Feeding problems include gastroesophageal reflux, sucking/swallowing difficulties, aspiration and oral aversion (6,14). Therefore, assisted feeding with a nasogastric or gastrostomy tube is commonly used in 40 to 50% of CFC individuals (6). The gastrointestinal dysfunction, including constipation and intestinal malrotation, also occurs in individuals with CFC syndrome. In the present study, we observed that Braf^Q241R/+ mice showed esophageal dilation and decreased motility due to an aberrant skeletal-smooth muscle boundary and gastrointestinal dysfunction. Furthermore, a pharmacological intervention with MEK inhibitors (PD0325901, MEK162, and AZD6244) or GSK-J4 reversed the digestive system abnormalities in the Braf^Q241R/+ mice.

Results

Braf^Q241R/+ mice develop esophageal dilation, hyperkeratosis, a thickened muscle layer and hyperproliferation of the muscularis mucosae in the forestomach

We have previously reported that approximately half of the Braf^Q241R/+ neonatal mice on an ICR background exhibit growth retardation and feeding problems, and die within 4 weeks. Thereafter, 23% (6 of 26) of the surviving Braf^Q241R/+ mice die in the adult stage from severe growth retardation and a decreased body weight (50% that of control mice) within 1 year (13). We therefore examined whether the feeding problems and severe growth retardation might be caused by esophageal and gastrointestinal abnormalities in the Braf^Q241R/+ mice. The stomachs of Braf^Q241R/+ mice with severe growth retardation were smaller than those of the Braf^+/+ mice (Fig. 1A). The Braf^Q241R/+ mice showed megaesophagus with digested food, malrotation of the stomach, dilation of the jejunum and cecum, and constipation with fecal impaction (Fig. 1A, Supplementary Material, Fig. S1A–D). Histological and immunohistochemical examinations revealed a cornified epithelium, an increased thickness of the muscle layer and hyperproliferation of the muscularis mucosae in the forestomach in the Braf^Q241R/+ mice (Fig. 1B); however, the mucosa in the glandular stomach was intact. In the Braf^Q241R/+ mice, the lower esophageal sphincter (LES) and proximal esophagus of the LES showed a cornified epithelium and decreased thickness of the muscle layer (Fig. 1C). Patients with achalasia and megaesophagus develop fibrosis of the LES during the later stages of the disease (15). Sirius red staining indicated no differences in fibrosis of the LES between the Braf^+/+ and Braf^Q241R/+ mice (data not shown).

Figure 1.

Esophageal and gastric phenotypes in Braf^Q241R/+ mice. (A) Representative esophageal and gastric appearance in 10-week-old mice, showing a markedly dilated esophagus in the Braf^Q241R/+ mice. (B) H&E-stained (left and middle panels) and α-SMA immunohistochemistry-stained (right panels) sections of the stomach tissues from Braf^+/+ and Braf^Q241R/+ mice at 10 weeks of age. Note the increased thickness of the muscular layer, the cornified layer and the hyperproliferation of the muscularis mucosae in the forestomach in the Braf^Q241R/+ mice. (C) H&E-stained longitudinal sections of 10-week-old mice, showing a thickened cornified layer in the Braf^Q241R/+ esophagus. (D) Western blotting analysis of BRAF in protein extracts from 8-week-old Braf^+/+ and Braf^Q241R/+ mice. GAPDH is shown as a loading control. (E, F) In situ hybridizations of Braf antisense (left panel) and sense (right panel) probes in esophagus (E) and stomach (F) sections from Braf^+/+ and Braf^Q241R/+ embryos at E18.5. (G) Sections of the esophagus and forestomach from 26-week-old Braf^+/+ were stained with antibodies against BRAF. (H) The esophagus width of the Braf^+/+ and Braf^Q241R/+ mice at 9–10, 22–27, 54–56 and 12–31 weeks of age. The data are shown as the mean ± SD [Braf^+/+ (n = 9–20) and Braf^Q241R/+ (n = 5–19)]. *P < 0.05; **P  < 0.01; ***P  < 0.001 versus Braf^+/+. (I) Longitudinal sections of the stomachs from the 10-, 26- and 54–56-week-old Braf^+/+ and Braf^Q241R/+ mice stained with H&E. Note the increased thickness of the muscular layer and cornified layer in the Braf^Q241R/+ forestomach at all stages.

BRAF is expressed at high levels in neuronal tissues and the testis in the adult stage (3,4). However, it is not known whether Braf is expressed in the esophagus and stomach in the embryonic and adult stages. A western blot analysis detected expression of the BRAF protein at high levels in the brain, lung, esophagus, stomach and testis tissues, and low levels were detected in the heart tissues of adult ICR mice (Fig. 1D). In situ hybridization of tissues from the ICR mice at E18.5 revealed signals that were widely distributed in the skin, nasal cavity, tooth germ, ganglion, pancreas, bladder, heart and neuronal tissues, including the cerebral cortex, hippocampus, striatum, thalamus, midbrain and spinal cord (Supplementary Material, Fig. S2). The expression of Braf was also detected both in the muscle layer and epithelium of the esophagus and stomach (Fig. 1E and F). In addition, immunohistochemistry staining confirmed that BRAF was expressed in the epithelium in the esophagus and forestomach of adult ICR mice (Fig. 1G). These results suggested that Braf is expressed in the esophagus and stomach at the embryonic and adult stages of ICR mice.

We investigated whether Braf^Q241R/+ mice without severe growth retardation exhibit megaesophagus, malrotation of the stomach, dilation of the jejunum and cecum, constipation with fecal impaction, and the hyperkeratotic phenotype in the LES and forestomach. The esophageal width was significantly larger in the Braf^Q241R/+ mice than in the Braf^+/+ mice (Fig. 1H, Supplementary Material, Table S1). In addition, the esophageal width in the Braf^Q241R/+ mice increased with age. In particular, the Braf^Q241R/+ mice with severe growth retardation had a prominent esophagus width (4.41 ± 0.36). The esophagus length in the Braf^Q241R/+ mice was longer than that in the Braf^+/+ mice, and the esophagus length to body length ratio was significantly higher (Supplementary Material, Table S1). Hyperkeratosis, an increased thickness of the muscle layer and hyperproliferation of the muscularis mucosae in the forestomach were observed in the Braf^Q241R/+ mice (Fig. 1I). Several mice-exhibiting megaesophagus showed hyperkeratosis of the LES (data not shown). Two of 12 and 3 of 10 Braf^Q241R/+ mice exhibited megastomach at 54 to 56 and 79 weeks of age, respectively (Supplementary Material, Fig. S1E). Other phenotypes, including dilation of the jejunum and cecum, constipation with fecal impaction and a reduced thickness of muscle layer in esophagus and LES, were not observed in the Braf^Q241R/+ mice without severe growth retardation. These data demonstrated that regardless of the degree of the growth retardation, the Braf^Q241R/+ mice showed a dilated esophagus, hyperkeratosis, an increased thickness of the muscle layer, and hyperproliferation of the muscularis mucosae in the forestomach.

Braf^Q241R/+ mice develop esophageal dilation, hyperkeratosis, a thickened muscle layer and hyperproliferation of the muscularis mucosae in the forestomach until postnatal day 14

To investigate when Braf^Q241R/+ mice develop esophageal dilation, hyperkeratosis, an increased thickness of the muscle layer, and hyperproliferation of the muscularis mucosae in the forestomach, we analysed Braf^+/+ and Braf^Q241R/+ mice from postnatal day (P) 0 to P20. Consistent with the decreased body weight and length in the Braf^Q241R/+ mice at P3, the stomachs in the Braf^Q241R/+ mice were smaller than those in the Braf^+/+ mice (Fig. 2A, Supplementary Material, Table S2). The Braf^Q241R/+ mice also had aerophagy (data not shown). The gross observations indicated dilated esophagus in the Braf^Q241R/+ mice at P7 (Fig. 2B). Furthermore, the morphometric evaluation revealed that the esophagus width in the Braf^Q241R/+ mice was significantly greater than that in the Braf^+/+ mice at P0, P7-P8 and P14-P16 (Fig. 2C, Supplementary Material, Table S2). Braf^Q241R/+ mice with a severe growth retardation and esophageal dilation at P7 displayed a thickened stratum corneum in the skin (Supplementary Material, Fig. S3). On the basis of the esophageal histology at P0-P7, the esophageal structures were similar between the Braf^+/+ and Braf^Q241R/+ mice. However, a detached cornified layer and food particles were observed in the esophagus in Braf^Q241R/+ mice at P10 and P14, and P20, respectively (Fig. 2D). At P3, the Braf^Q241R/+ mice developed a marked proliferation of the stratified squamous epithelium and muscularis mucosae in the forestomach (Fig. 2E, Supplementary Material, Fig. S4A). All Braf^Q241R/+ mice at P14 had a thickened muscle layer, although the increased thickness of the muscle layer was observed in 3 of 6 Braf^Q241R/+ mice at P10 (Fig. 2E, Supplementary Material, Fig. S4A). To examine whether the hyperkeratosis of the forestomach observed in the Braf^Q241R/+ mice might be caused by changes in cell proliferation, we performed immunostaining using Ki67 to assess cell proliferation. Ki67-positive-stained epithelial cells in the forestomach increased in Braf^Q241R/+ mice at P7 (Fig. 2F).

Figure 2.

Feeding problems, dilated esophagus, thickened cornified and muscle layer and hyperproliferation of the muscularis mucosa in the forestomach throughout postnatal development in Braf^Q241R/+ mice. (A) Representative esophageal and gastric appearance in 3-day-old mice, showing a small stomach and decreased stomach milk in the Braf^Q241R/+ mice. (B) Representative esophageal and gastric appearance in 7-day-old mice, showing a dilated esophagus in the Braf^Q241R/+ mice. (C) The esophagus width in the Braf^+/+ and Braf^Q241R/+ mice at P0, 3, 7–8 and 14–16. The data are shown as the mean ± SD [Braf^+/+ (n = 10–22) and Braf^Q241R/+ (n = 11–19)]. *P < 0.05; ***P < 0.001 versus Braf^+/+. (D) Longitudinal sections of the esophagus in Braf^+/+ and Braf^Q241R/+ mice at P7, 10, 14 and 20 stained with H&E. Note the detached cornified layer (P10 and 14) and food particles (P14) in the esophagus in the Braf^Q241R/+ mice. (E) Longitudinal sections of the stomach of mice at P0, 3, 7, 14 and 20 stained with H&E, showing the hyperproliferation of the muscularis mucosae in the Braf^Q241R/+ forestomach at P3, 7, 14 and 20, and the increased thickness of the muscular and cornified layer in the Braf^Q241R/+ forestomach at P14 and 20. (F) Ki67-stained longitudinal forestomach sections from Braf^+/+ and Braf^Q241R/+ mice at P7. The histogram shows the percentage of Ki-67-positive cells in the epithelium of forestomach. The data are shown as the mean ± SD [Braf^+/+ (n = 5–7) and Braf^Q241R/+ (n = 5–7)].

The esophagus in the Braf^Q241R/+ mice has an aberrant skeletal-smooth muscle boundary

The mammalian esophageal walls initially develop as smooth muscle. Subsequently, smooth muscle is replaced by skeletal muscle in a proximal-to-distal manner (16,17). In human adults, the upper third of the esophagus wall is composed of skeletal muscle cells, the middle third is composed of a mixture of smooth and skeletal muscle cells, and the lower third is composed of only smooth muscle cells. In contrast, the murine esophagus is composed of skeletal muscle cells, except for the LES, which is composed of smooth muscle cells (18). Several studies using knockout mice have suggested an association between a megaesophagus with delayed myogenic differentiation and dysplasia in the esophagus (17,19–22). To examine the myogenic differentiation in the esophagus in Braf^Q241R/+ mice, we performed immunostaining with markers of smooth muscle (α-smooth muscle actin; α-SMA) and skeletal muscle (sarcomeric actin; SA). All 10-week-old Braf^+/+ mice showed α-SMA expression in the LES and SA expression in the esophagus, whereas 6 of 8 Braf^Q241R/+ mice had an aberrant skeletal-smooth muscle boundary (Fig. 3A). The Braf^Q241R/+ mice exhibiting megaesophagus also showed α-SMA expression from the proximal to distal LES (data not shown). The smooth muscle cells in mouse esophagus are replaced by skeletal muscle cells at 14 days after birth (16,23). At P10, the skeletal-smooth muscle boundary in the LES was not clearly observed in either Braf^+/+ or Braf^Q241R/+ mice (Fig. 3B). At P14 and P20, the skeletal-smooth muscle boundary in the LES was clearly observed in the Braf^+/+ mice, whereas the Braf^Q241R/+ mice displayed delayed skeletal myogenesis. The Ki67 immunostaining revealed that Ki67-positive cells were observed in the transition zone, which is where smooth muscle cells and skeletal muscle cells are mixed, in both the Braf^+/+ mice at P10 and Braf^Q241R/+ mice at P14; however, Ki67-positive cells in the esophagus in the Braf^Q241R/+ mice were detected in the distal region of the LES (Fig. 3B and C). The decrease in Ki67-positive cells in the Braf^+/+ mice at P14 and P20 and the Braf^Q241R/+ mice at P20 resulted in the termination of the myogenic differentiation (Fig. 3C–E). At P0, the pharyngo esophagus was devoid of smooth muscle in Braf^+/+ mice, whereas α-SMA expression was observed in the Braf^Q241R/+ mice (Supplementary Material, Fig. S5A). Moreover, α-SMA expression was found in the esophagus on the thorax in Braf^Q241R/+ mice at P7, P10 and P14 but not in Braf^+/+ mice (Supplementary Material, Fig. S5B). Western blotting and quantitative reverse transcription-PCR analyses indicated comparable expression levels of α-SMA (Acta2), muscle-cell differentiation genes (Myod1, Myog) and skeletal muscle marker genes [SA (Acta1), Myh2a] between Braf^+/+ and Braf^Q241R/+ esophageal tissues in mice at the age of 10 weeks. The increased expression of Myh11 (smooth muscle cell marker), smooth muscle 22-kDa protein [SM22α (Tagln); smooth muscle cell marker] was also detected in the Braf^Q241R/+ mice (Fig. 3F and G). These data indicated that the replacement of smooth muscle cells with skeletal muscle cells in the esophagus is impaired in Braf^Q241R/+ mice at the neonatal stage.

Figure 3.

Delayed esophageal musculature in Braf^Q241R/+ mice. (A) Sections of the LES from 10-week-old Braf^+/+ and Braf^Q241R/+ mice were stained with antibodies against α-SMA or SA. (B–D) Sections of the LES from Braf^+/+ and Braf^Q241R/+ mice at P10, 14 and 20 were stained with antibodies against α-SMA, SA or Ki67. The right panel in Ki67 staining indicates higher magnification of the boxed region in C. The arrows indicate a skeletal-smooth muscle boundary to the esophagus (A-D). (E) Ki67-stained esophagus sections from Braf^+/+ and Braf^Q241R/+ mice at P10, P14 and P20. The histogram shows the percentage of Ki-67-positive cells in the skeletal-smooth muscle boundary to esophagus. The data are shown as the mean ± SD [Braf^+/+ (n = 5) and Braf^Q241R/+ (n = 5–6)]. (F) Esophageal protein extracts from 10-week-old Braf^+/+ and Braf^Q241R/+ mice were subjected to immunoblotting with the indicated antibodies. GAPDH is shown as a loading control. The data are shown as the mean ± SD [Braf^+/+ (n = 5) and Braf^Q241R/+ (n = 5)]. *P < 0.05 versus Braf^+/+. (G) The esophageal expression of Acta2, Myh11, Tagln, Myod1, Myog, Myh2a and Acta1 in 10-week-old Braf^+/+ and Braf^Q241R/+ mice was measured by quantitative reverse transcription-PCR. mRNA levels were normalized to those of Gapdh, and those in Braf^+/+ were set as 1. The data are shown as the mean ± SD [Braf^+/+ (n = 6) and Braf^Q241R/+ (n = 7)]. *P < 0.05 versus Braf^+/+.

Braf^Q241R/+ mice display decreased esophageal motility

Achalasia is characterized by an esophageal motility dysfunction with impaired relaxation of the LES and decreased peristalsis due to impaired inhibitory nitrergic neurons in the esophageal myenteric (Auerbach’s) plexus (24). To investigate whether Braf^Q241R/+ mice could develop achalasia, we performed immunostaining of the LES in Braf^+/+ and Braf^Q241R/+ mice by using antibodies against protein gene product 9.5 (PGP9.5; neuronal marker). There were no differences in the numbers of PGP9.5-positive cells in the Braf^+/+ and Braf^Q241R/+ mice [51.25 ± 28.7 (Braf^+/+, n = 4) versus 48.3 ± 32.5 (Braf^Q241R/+, n = 4) PGP9.5-positive cells/0.15-mm² fields, P = 0.97 (Mann–Whitney U-test)] (Fig. 4A). Braf^Q241R/+ mice with severe growth retardation and megaesophagus showed a loss of PGP9.5-positive cells in the esophagus but not in the LES (Supplementary Material, Fig. S6A). Esophageal Uchl1 (PGP9.5) mRNA levels in the Braf^Q241R/+ mice, but not PGP9.5 protein levels, were significantly higher than those in the Braf^+/+mice (Fig. 4B and C). Likewise, esophageal choline acetyltransferase (Chat1; presynaptic cholinergic marker) mRNA levels were significantly higher in the Braf^Q241R/+ mice than those in the Braf^+/+mice. The expression of the inhibitory neuron markers Nos1, acetylcholinesterase (Ache1), vasoactive intestinal polypeptide (Vip1), neurotrophic factors, Gdnf1 and Ngf were comparable in the Braf^+/+ and Braf^Q241R/+ mice (Fig. 4B). Western blotting analysis revealed that the levels of phosphorylated ERK protein were significantly lower in the Braf^Q241R/+ mice than in the Braf^+/+ mice [1.00 ± 0.46 (Braf^+/+) versus 0.45 ± 0.20 (Braf^Q241R/+), P = 0.03 (Mann–Whitney U-test)]. Consistent with the decreased phosphorylated ERK levels, the protein levels of kinase suppressor of RAS (KSR-1) were significantly higher in the Braf^Q241R/+ mice [1.00 ± 0.62 (Braf^+/+) versus 2.25 ± 0.42 (Braf^Q241R/+), P = 0.008 (Mann–Whitney U-test)] (Fig. 4C), thus suggesting that a negative feedback mechanism exists that normalizes constitutive ERK activation in the esophagus in Braf^Q241R/+ mice. There were no differences in the neuron markers, including nNOS, Tublinβ3 and c-kit, cell signaling factors, phosphorylated p38, AKT (Ser473) and AKT (Thr308) in the Braf^+/+ and Braf^Q241R/+ mice.

Figure 4.

Decreased esophageal motility without a loss of inhibitory neurons in Braf^Q241R/+ mice. (A) Longitudinal LES sections from 10-week-old Braf^+/+ and Braf^Q241R/+ mice were stained with an antibody against PGP9.5. (B) The esophageal expression of Uchl1, Nos1, Ache1, Vip1, Chat1, Gdnf1 and Ngf in 10-week-old Braf^+/+ and Braf^Q241R/+ mice was measured by quantitative reverse transcription-PCR. mRNA levels were normalized to those of Gapdh, and those in Braf^+/+ were set as 1. The data are shown as the mean ± SD [Braf^+/+ (n = 6) and Braf^Q241R/+ (n = 7)]. *P < 0.05 versus Braf^+/+. (C) Western blotting of esophagus tissues from 10-week-old Braf^+/+ and Braf^Q241R/+ mice. GAPDH is shown as a loading control. The arrowheads indicate the bands corresponding to each protein. (D, E) Measurement of the contraction force of the esophagus in 14- to 18-week-old Braf^+/+ and Braf^Q241R/+ mice in response to 200 μM carbachol (CCh) (D) or 60 mM KCl (E). The scale bars indicate tension (50 mg). (F) Quantification of the maximum contraction force of the esophagus in 14- to 18-week-old Braf^+/+ and Braf^Q241R/+ mice in response to 200 μM CCh and 60 mM KCl. The data are shown as the mean ± SD [Braf^+/+ (n = 9) and Braf^Q241R/+ (n = 10)]. *P  < 0.05 versus Braf^+/+.

Because the Braf^Q241R/+ mice developed esophageal dilation without a loss of neurons, as a result of impaired myogenic differentiation, we evaluated the esophageal motility function in the Braf^Q241R/+ mice at the age of 14–18 weeks. By exploring the nerve-independent and nerve-dependent contractions induced by a high concentration of KCl and carbachol (CCh), respectively, the contraction force of the LES was similar in the Braf^+/+ mice and Braf^Q241R/+ mice (Supplementary Material, Fig. S6B). In contrast, the esophageal tissues from the Braf^Q241R/+ mice showed significantly decreased contractile responses to CCh and slightly decreased responses to KCl (Fig. 4D–F).

Cell adhesion- and epithelium development-associated gene expression is increased in the forestomach in Braf^Q241R/+ mice

We then examined the neuronal cells and the motility function in the forestomach in Braf^Q241R/+ mice. In agreement with the increased thickness of the muscle layer in the forestomach in the Braf^Q241R/+ mice at P10 and P14, the Braf^Q241R/+ mice exhibited a marked increase in the numbers of S100 [At P10: 803.1 ± 854.1 (Braf^+/+, n = 6) versus 2134.6 ± 1164.2 (Braf^Q241R/+, n = 6) S100-positive area (μm²)/muscle layer (mm), P = 0.04 (Mann–Whitney U-test); At P14: 1008.1 ± 825.2 (Braf^+/+, n = 5) versus 3541.9 ± 735.6 (Braf^Q241R/+, n = 5), P = 0.008 (Mann–Whitney U-test)] (Fig. 5A) or PGP9.5-positive neural cells (data not shown), as compared with Braf^+/+ mice. Similar results were obtained in mice at 10 weeks of age [477.7 ± 300.91 (Braf^+/+, n = 4) versus 1972.2 ± 947.4 (Braf^Q241R/+, n = 4) PGP9.5-positive area (μm²)/muscle layer (mm), P = 0.03 (Mann–Whitney U-test)] (Fig. 5B). Although the neural cells increased in the forestomach in the Braf^Q241R/+ mice, the analysis of the contractile response in the forestomach to CCh revealed no differences between the Braf^+/+ and Braf^Q241R/+ mice (data not shown). To determine whether the neuronal gene expression and cell signaling pathways, including ERK, p38 and Akt, were altered in the Braf^Q241R/+ mice, western blotting was performed using cell extracts from the forestomach. The protein levels of PGP9.5, nNOS, Tublinβ3, c-Kit, phosphorylated ERK, p38 and AKT were comparable between the Braf^+/+ and Braf^Q241R/+ mice (Supplementary Material, Fig. S7).

Figure 5.

Increased expression of cell adhesion- and epithelium development-associated genes in the forestomach in Braf^Q241R/+ mice. (A) Longitudinal forestomach sections from Braf^+/+ and Braf^Q241R/+ mice at P 1, 7, 10 and 14 were stained with an antibody against S100. (B) Longitudinal forestomach sections from 10-week-old Braf^+/+ and Braf^Q241R/+ were stained with an antibody against PGP9.5. (C–E) The expression of the cell and biological adhesion-associated genes (C), epithelium-associated genes (D) and keratin genes (E) in the forestomach in 3-day-old Braf^+/+ and Braf^Q241R/+ mice, as measured by quantitative reverse transcription-PCR. mRNA levels were normalized to those of β-actin, and those in Braf^+/+ are set to 1. The data are shown as the mean ± SD [Braf^+/+ (n = 7) and Braf^Q241R/+ (n = 7)]. *P  < 0.05; **P < 0.01; ***P  < 0.001 versus Braf^+/+. (F) H&E-stained forestomach sections from 10-week-old Braf^+/+ and Braf^Q241R/+ mice, showing the disorganized basal layer in the forestomach in Braf^Q241R/+ mice. (G) Forestomach sections from 10-week-old Braf^+/+ and Braf^Q241R/+ mice were stained with an antibody against keratin 5.

To investigate the mechanism of the hyperkeratosis, thickened muscle layer and hyperproliferation of the muscularis mucosae in the forestomach of the Braf^Q241R/+ mice, a microarray analysis was performed using forestomach tissues from the Braf^+/+ and Braf^Q241R/+ mice at P3. Using a 1.5-fold expression difference as the cut-off, we found 115 up-regulated genes and 38 down-regulated genes. The up-regulated genes were analysed by using the database for annotation, visualization and integrated discovery (DAVID), which is a functional annotation tool. The top three terms in biological process were cell adhesion, biological adhesion and epithelium development, and the top three terms in cellular components were desmosome, extracellular region and integral to plasma membrane (Supplementary Material, Table S3). RT-PCR revealed that the mRNA levels of Cdh13, Cgref1 and Gpnmb, which were selected as genes categorized in cell adhesion and biological adhesion of biological process, were significantly increased in Braf^Q241R/+ mice compared with Braf^+/+ mice (Fig. 5C). In the associated genes in epithelium development, the expression of Etv4, Krt14 (basal cytokeratin) and Sprr1b (a squamous differentiation marker) was also significantly higher in Braf^Q241R/+ mice than in Braf^+/+ mice (Fig. 5D). Consistently with the hyperkeratosis phenotype and the increased Krt14 expression in the Braf^Q241R/+ mice, the Krt5 (paired with Krt14) mRNA levels, but not the KRT5 protein levels, were significantly higher in the Braf^Q241R/+ mice than in the Braf^+/+ mice (Fig. 5E, Supplementary Material, Fig. S7A). The expression of suprabasal cytokeratin, Krt13, was also significantly higher in the Braf^Q241R/+ mice than in the Braf^+/+ mice. In addition, the histological analysis and immunohistochemistry using the anti-keratin 5 antibody revealed that the forestomach in the Braf^Q241R/+ mice had a disorganized basal layer and increased keratin 5 expression (Fig. 5F and G).

MEK inhibitors and GSK-J4 rescued the gastrointestinal phenotypes in the Braf^Q241R/+ mice

We have previously reported that treatment with the MEK inhibitor PD0325901, or the histone demethylase inhibitors GSK-J4 or NCDM-32 b, prevents the embryonic and postnatal lethality in Braf^Q241R/+ mice on a C57BL/6 J background (12). These results suggest that the inhibition of ERK activation and alteration of histone modification may be useful for effective treatment of patients with CFC syndrome. We therefore investigated the effects of MEK inhibitors (i.e., PD0325901, MEK162, AZD-6244 and trametinib), histone demethylase inhibitors (i.e., GSK-J4 and NCDM-32 b) and other inhibitors (i.e., SAHA, rapamycin and FK506) on the gastrointestinal abnormalities in Braf^Q241R/+ mice. A 13-day treatment with PD0325901, but not GSK-J4, in nursing female mice resulted in an increase in body weight and body length in 14-day-old Braf^Q241R/+ mice. The body weight and length of the MEK162-treated Braf^Q241R/+ mice were significantly greater than those of the DMSO-treated Braf^Q241R/+ mice and were comparable to those of the MEK162-treated Braf^+/+ mice (Fig. 6A, Supplementary Material, Table S4). Because the trametinib, rapamycin and FK506 treatments caused a severe weight loss in the Braf^+/+ and Braf^Q241R/+ mice, we did not evaluate these compounds in this study (Supplementary Material, Table S4). The esophagus width of the MEK162 and GSK-J4-treated Braf^Q241R/+ mice, but not the PD0325901-treated Braf^Q241R/+ mice, was significantly decreased compared with that of DMSO-treated Braf^Q241R/+ mice, although there were significant differences between the drug-treated Braf^+/+ and Braf^Q241R/+ mice (Fig. 6A, Supplementary Material, Table S4). The gross observations showed that the stomach size and milk content in the MEK162-treated Braf^Q241R/+ mice were comparable with those in the MEK162-treated Braf^+/+ mice (data not shown). The myogenic differentiation in the esophagus in 3 of 5 GSK-J4-treated Braf^Q241R/+ mice was similar to that in the DMSO-treated Braf^+/+ mice (Fig. 6B). Furthermore, the thickened muscle layer in the forestomach was improved in 10 of 14 PD0325901-treated Braf^Q241R/+ mice, all MEK162-treated Braf^Q241R/+ mice and 6 of 8 AZD6244-treated Braf^Q241R/+ mice (Fig. 6C). Hyperkeratosis and hyperproliferation of the muscularis mucosae in the forestomach were not observed in 7 of 14 PD0325901-treated Braf^Q241R/+ mice and 6 of 8 MEK162-treated Braf^Q241R/+ mice (Fig. 6C). These results indicated that the MEK inhibitors, including PD0325901, MEK162 and AZD6244, or the histone H3K27 demethylase inhibitor GSK-J4, partly improve growth retardation and the esophageal and gastric phenotypes in Braf^Q241R/+ mice.

Figure 6.

MEK inhibitors and a histone demethylase inhibitor improve the esophageal and gastric phenotypes in Braf^Q241R/+ mice. (A–C) Nursing female mice were injected intraperitoneally daily with PD0325901 [5 mg/kg body weight (BW)], MEK162 (10 mg/kg BW), AZD-6244 (5 mg/kg BW), GSK-J4 (10 mg/kg BW) or dimethyl sulfoxide (DMSO) for 13 days, and the pups were dissected at P14. The body weight, body length and esophagus width in Braf^+/+ and Braf^Q241R/+ mice at P14. The data are shown as the mean ± SD [DMSO; Braf^+/+ (n = 19) and Braf^Q241R/+ (n = 13). PD0325901; Braf^+/+ (n = 18) and Braf^Q241R/+ (n = 17). MEK162; Braf^+/+ (n = 18) and Braf^Q241R/+ (n = 16). GSK-J4; Braf^+/+ (n = 13) and Braf^Q241R/+ (n = 23)]. *P < 0.05; **P < 0.01; ***P < 0.001, Tukey–Kramer test (DMSO versus PD0325901, MEK162 or GSK-J4). NS, not significant. (B) Longitudinal forestomach sections from Braf^Q241R/+ mice were stained with an antibody against α-SMA. (C) H&E-stained forestomach sections from Braf^Q241R/+ mice.

Discussion

Individuals with RASopathies, such as CFC syndrome, Costello syndrome and Noonan syndrome, have gastrointestinal abnormalities, including feeding and swallowing difficulties, in the neonatal period (6,25–27). Furthermore, nasogastric or gastrostomy tube feedings are required in patients with CFC syndrome and Costello syndrome (6). A patient with neurofibromatosis type 1 has been reported to have an achalasia-like disorder of the esophagus (28). Thus, the feeding problem is common in individuals with RASopathies. However, the feeding problems, esophageal dilations and gastrointestinal dysfunctions have not been reported in the following mouse models of RASopathies: knock-in mouse model expressing Braf L597V or V600E mutation (CFC syndrome model), Hras mutation (Costello model) or Ptpn11, Raf1 or Kras mutations (Noonan syndrome model) (29–34). These mutations all lead to the activation of the RAS/MAPK pathway. In this study, we provide the evidence that knock-in mice expressing a gain-of-function Braf Q241R mutation show esophageal dilation, decreased esophageal motility and impaired myogenic differentiation. Notably, the Braf^Q241R/+ mice on a mixed BALB/c x C57BL/6 background showed severe growth retardation and esophageal dilation with high penetrance, as compared with Braf^Q241R/+ mice on an ICR background (Supplementary Material, Table S5), thus suggesting that these phenotypes may be genetic background-independent and associated with the Braf Q241R mutation. The Braf^Q241R/+ mice also displayed hyperkeratosis, a thickened muscle layer and hyperproliferation of the muscularis mucosae in the forestomach. Furthermore, we found that treatment with the MEK inhibitors PD0325901, MEK162 and AZD6244, or a histone H3K27 demethylase inhibitor GSK-J4, partly improved the growth retardation, esophageal dilation and gastrointestinal dysfunction observed in Braf^Q241R/+ mice. Thus, mice expressing the Braf Q241R mutation provide a useful model for clarifying the pathogenesis of esophageal dilation and gastrointestinal dysfunction and testing treatments for RASopathies.

A dysregulation in myogenic differentiation in the esophageal wall has been associated with esophageal dilation and esophageal dysmotility. Protein kinase C alpha (PKCα), frizzled 4 and paired box protein (Pax) 7 knockout mice have been reported to develop megaesophagus with an aberrant skeletal-smooth muscle boundary (19,21,22). Mice lacking a multifunctional cell surface receptor Cdo and collagen 19a1 have also been shown to have impaired myogenic differentiation in the esophagus, thus resulting in the development of megaesophagus and an abnormal esophageal motility (17,20). Our study demonstrated that Braf^Q241R/+ mice show an esophageal dilation and a decreased esophageal contractile response to CCh after impaired myogenic differentiation. The RAS/MAPK pathway plays a crucial role in the myogenic differentiation of smooth muscle to skeletal muscle (35). The overexpression of Raf1 or MEK1 negatively regulates myogenic differentiation in 10T1/2 cells, thus inhibiting skeletal muscle cell differentiation through a myocyte enhancer factor 2 or MyoD dependent mechanism (36,37). DA-Raf1, a splicing variant of Araf and a dominant negative antagonist of the RAS/MAPK pathway, has also been shown to act as a positive regulator of skeletal muscle cell differentiation (38). In this study, our data indicated that treatment with the MEK inhibitors PD0325901, MEK162, AZD-6244 and trametinib had no effect on the myogenic differentiation in Braf^Q241R/+ mice. In contrast, mice treated with the histone H3K27 demethylase inhibitor GSK-J4 showed recovery from the delayed and impaired myogenic differentiation. An increased expression of a histone H3K27 methylase, Ezh2, has been reported to inhibit skeletal muscle differentiation through an SRF- and MyoD-dependent mechanism (39). These data suggest that the modulation of histone modifications may be important for the myogenic differentiation in the esophagus wall.

Dysregulation of the RAS/MAPK pathway is associated with esophageal dilation or achalasia in mice. Old mice lacking the Ras association domain family 1 isoform A (Rassf1a) have been found to have megaesophagus with a loss of neural cells (40). Sprouty 2 is a negative regulator of the RAS/MAPK signaling pathway, and Sprouty 2 knockout mice have been reported to show neuronal hyperplasia and achalasia due to ERK activation (41). Furthermore, lsc is the murine homologue of human p115 RhoGEF, and Isc-null mice have been found to exhibit progressive dilation and impaired peristalsis in the esophagus (42). Remarkably, all these models have common defects in neuronal development. In contrast, the Braf^Q241R/+ mice had esophageal dilation and decreased esophageal motility without any abnormalities in neuronal development. Thus, our mouse models provide a different tool to investigate the effects of the RAS/MAPK pathway on esophageal development and motility.

The Braf^Q241R/+ knock-in mouse exhibited a cornified epithelium and a thickened muscle layer in the forestomach in which Braf expression was observed. Microarray and quantitative real-time PCR analysis revealed an increased expression of Cdh13, Cgref1 and Gpnmb, genes associated with cell adhesion. In agreement with the cornified epithelium in the forestomach, we also detected an increase in genes associated with cell proliferation (Etv4), epithelium development and barrier function in stratified squamous epithelial (Sprr1b), and basal layer keratin formation in the cornified epithelium (Krt5 and Krt14), suggesting the effects of the Braf Q241R mutation and/or excess epithelial cells in the forestomach. The activation of the RAS/MAPK pathway is associated with the proliferation of stratified squamous epithelial (43,44). Conditional Raf1 transgenic mice with a keratin 14 promoter have been shown to develop epidermal hyperplasia and hyperkeratosis in the skin (45). Notably, PD0325901, MEK162 and AZD6244 treatments ameliorated the cornified epithelium and thickened muscle layer in the forestomach, thus suggesting that these phenotypes in Braf^Q241R/+ mice are caused by activation of the MEK/ERK pathway.

It remains unknown why phosphorylated ERK expression was not changed in the esophagus and forestomach tissues from the Braf^Q241R/+ mice, although the Braf^Q241R/+ mice had esophageal dilation and gastrointestinal dysfunction. We hypothesized that the pERK expression depends on BRAF protein level. Thus, we investigated the protein levels of pERK in several tissues, including brain, lung, kidney, esophagus and forestomach, of the Braf^+/+ and Braf^Q241R/+ mice at P7 after treatment with DMSO or MEK162. Consistent with the higher BRAF expression (Fig. 1D), the protein levels of pERK were higher in brain and lung in the Braf^Q241R/+ mice than those in the Braf^+/+ mice (Supplementary Material, Fig. S8). In contrast, there were no differences in the pERK levels in kidney between the Braf^+/+ and Braf^Q241R/+ mice. The pERK levels were slightly higher in lysates of esophagus and forestomach in the Braf^Q241R/+ mice than in the Braf^+/+ mice (Supplementary Material, Fig. S9). These results suggest that ERK activation could depend on the level of BRAF expression in tissues. After MEK162 treatment, decreased pERK levels were observed in lung and kidney in the Braf^+/+ or Braf^Q241R/+ mice (Supplementary Material, Fig. S8). pERK levels were not changed in brain, esophagus or forestomach tissues from the Braf^+/+ and Braf^Q241R/+ mice in spite of MEK162 treatment (Supplementary Material, Figs. S8 and S9). We are not able to explain the reason why abnormalities of esophagus and forestomach were partially rescued by MEK inhibitors at this point. It is possible that cell-specific activation of pERK could be important for the phenotype, which was not detected in western blotting using whole tissue lysates. Alternatively, subtle decrease of pERK with MEK inhibitor, which might not be detected in western blotting, could be effective to improve the phenotypes. The third possibility is that negative/positive feedback could regulate the ERK phosphorylation so that the level of pERK keeps constant in affected tissues.

Here, we report that Braf^Q241R/+ mice display esophageal dilation and dysmotility and gastrointestinal dysfunction. Our recent study has shown that treatment with MEK inhibitors and/or histone demethylase inhibitors leads to the amelioration of several phenotypes in Braf^Q241R/+ mice on a C57BL/6 J background, including the hypertrophy of the cardiac valves, edema, mandibular hypoplasia and liver necrosis (12). In addition, in the present study, we found that treatment with MEK inhibitors or a histone H3K27 demethylase inhibitor partly rescued the growth retardation, esophageal dilation and gastrointestinal abnormalities in Braf^Q241R/+ mice. Our data suggested that treatment with MEK inhibitors and histone H3K27 demethylase inhibitors may be useful for the treatment of RASopathies with digestive system abnormalities.

Materials and Methods

Mice

Heterozygous knock-in mice on an ICR background expressing Q241R mutation, Braf^Q241R ICR, have been previously reported (12,13). To maintain the Braf^Q241R ICR strain, we purchased female ICR mice from Charles River Laboratories Japan (Yokohama, Japan) and mated them with male Braf^Q241R ICR mice. The mice were housed under a 12-h light/12-h dark cycle, fed Labo MR Stock (Nihon Nosan Kogyo, Yokohama, Japan) and were given water ad libitum. The animal experiments were approved by the Animal Care and Use Committee of Tohoku University.

Genotyping

DNA was extracted from the tail tissue using a Maxwell 16 Mouse Tail DNA Purification Kit (Promega, Madison, WI). The genotyping of the Braf^+/+ and Braf^Q241R/+ mice was carried out via PCR using KOD FX Neo (TOYOBO, Osaka, Japan) with the primers 5’-GTGTTGTTCTGCCCATACTTACTGC-3’ and 5’- GTGACTTAATGTACAGCATGGATCA-3’.

Western blotting

Dissected brain, lungs, heart, liver, esophagus, stomach, colon, kidneys, testis and spleen tissues were homogenized in lysis buffer (10 mM Tris-HCl, pH 8.0 and 1% SDS) containing phosphatase and protease inhibitor cocktails (P5726 and P8340, respectively; Sigma-Aldrich, St. Louis, MO). These lysates were centrifuged at 20, 400 x g for 30 min at 15 °C, and the protein concentration was determined using a Bio-Rad Protein Assay (Bio-Rad Laboratories, Hercules, CA) with BSA as the standard. Protein lysates were separated by 4–20% SDS-polyacrylamide gel electrophoresis [Criterion TGX Precast Gels (Bio-Rad Laboratories)] and transferred to nitrocellulose membranes (Trans-Blot Turbo transfer pack; Bio-Rad Laboratories) using a Trans-Blot Turbo Transfer System (Bio-Rad Laboratories). After being blocked with 5% non-fat milk for 1 h at room temperature, the membranes were incubated with the antibodies to the following proteins (with catalog numbers in parentheses): ERK1/2 (9102), phospho-ERK1/2 (9101), p38 (9212), phospho-p38 (4511), AKT (9272), phospho-AKT (on Ser473; 9018), phospho-AKT (on Thr308; 2965), nNOS (4236), c-Kit (3074) and GAPDH (2118) from Cell Signaling (Danvers, MA); BRAF (sc-166) from Santa Cruz Biotechnology (Santa Cruz, CA, USA); KSR-1 (611576) from BD Transduction Laboratories (San Jose, CA, USA); α-SMA (M0851) and SA (M0874) from DAKO (Glostrup, Denmark); SM22α (ab14106) from Abcam (Cambridge, UK); PGP9.5 (AB1761-I) from Millipore; and Tublin β3 (801201) and Keratin 5 (905501) from BioLegend (San Diego, CA, USA). The signals were visualized using a Western Lightning ECL-Plus Kit (PerkinElmer, Waltham, MA). The band intensities were quantified using ImageJ software (http://rsbweb.nih.gov/ij/). The phosphorylated proteins were normalized to the non-phosphorylated proteins.

In situ hybridization

Mouse embryos at E18.5 were fixed with G-fix (Genostaff, Tokyo, Japan) at room temperature and decalcified with G-Chelate Mild (Genostaff). The decalcified embryos were then embedded in paraffin on CT-Pro20 (Genostaff) using G-Nox (Genostaff) and sectioned at 8 μm. To generate digoxigenin-labeled anti-sense and sense RNA probes, a 908 bp DNA fragment of mouse Braf cDNA (nucleotide position 2530-3437) was prepared from mouse 1^st Strand cDNA (Genostaff). The digoxigenin-labeled RNA probes were synthesized with in vitro transcription using DIG RNA Labeling Mix (Roche Diagnostics, Mannheim, Germany). Hybridization was carried out using an ISH Reagent Kit (Genostaff). The tissue sections were deparaffinized with G-Nox and rehydrated using an ethanol series and PBS. The sections were fixed with 10% formalin in PBS and treated with 4 μg/ml of Proteinase K (Wako Pure Chemicals, Osaka, Japan) in PBS for 10 min at 37 °C. The sections were then re-fixed with 10% formalin in PBS for 10 min at room temperature and placed in 0.2 N HCl for 10 min at room temperature. Hybridization was performed with RNA probes at concentrations of 300 ng/ml in G-Hybo (Genostaff) for 16 h at 60 °C. Staining was performed with an NBT/BCIP solution (Sigma-Aldrich) overnight. Sections were counterstained with Kernechtrot stain solution (Muto Pure Chemicals, Tokyo, Japan) and mounted with G-Mount (Genostaff).

Histology and immunohistochemistry

The tissues were fixed in 10% neutral buffered formalin and embedded in paraffin. Embedded tissues were sectioned at 3 μm and stained with hematoxylin and eosin.

For the immunohistochemistry, the tissue sections were deparaffinized with xylene and rehydrated using an ethanol series and PBS. Immunohistochemistry was performed with the Histofine simple stain kit (Nichirei Bio Sciences, Tokyo, Japan) and the antibodies to the following proteins (with catalog numbers in parentheses): BRAF (9433) from Cell Signaling; α-SMA (M0851) and SA (M0874) from DAKO; PGP9.5 (AB1761-I) from Millipore; Ki67 (418071) from Nichirei Bio Sciences; and Keratin 5 (905501) from BioLegend. The signals were visualized by using a DAB Substrate Kit (Nichirei Bio Sciences). The nuclei were stained with hematoxylin.

Quantitative reverse transcription-PCR and microarray analysis

Total RNA from the esophagus or forestomach was extracted using TRIzol reagent (Invitrogen, Carlsland, CA) and purified with an RNeasy Mini Kit (Qiagen, Hilden, Germany). RNA was reverse transcribed into cDNA using a High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA). The PCR reactions were performed using FastStart Universal Probe Master (ROX) (Applied Biosystems) with StepOnePlus (Applied Biosystems). The amplification primers and hydrolysis probes were designed using Universal ProbeLibrary Assay Design Center (https://qpcr.probefinder.com/roche3.html).

The gene expression analyses of the forestomach tissues were performed with SurePrint G3 Mouse GE 8x60K v2 Microarrays (Agilent Technologies, Santa Clara, CA, USA) with a Low Input Quick Amp Labeling Kit One-Color (Agilent Technologies), a Gene Expression Hybridization Kit (Agilent Technologies) and a Gene Expression Wash Pack (Agilent Technologies) according to the manufacturer’s protocol. The data were analysed with Agilent Feature Extraction Software, 11.5.1.1.

Measurement of esophageal contraction

Esophageal strips were isolated under a microscope after the mice were anesthetized with isoflurane, and were placed in Krebs solution (118.4 mM NaCl, 4.7 mM KCl, 2.5 mM CaCl₂, 1.2 mM MgSO₄, 1.2 mM KH₂PO₄, 25 mM NaHCO₃ and 11.7 mM glucose). The esophageal strips were cut into 5 mm sections and mounted vertically in a Magnus apparatus (Panlab, Barcelona, Spain) filled with Krebs solution at 37 °C and continuously bubbled. Responses to the esophageal contraction were measured under 200 μM CCh or 60 mM KCl conditions. The esophageal contraction was recorded with an isometric force transducer (ADInstruments, Sydney, Australia) and analysed with a Power Lab system (ADInstruments).

Mouse treatment

For the postnatal drug treatments, PD0325901 (Sigma-Aldrich), MEK162 (Active Biochem, Maplewood, NJ, USA), AZD-6244 (Selleckchem, Houston, TX), trametinib (Cayman Chemical), NCDM-32 b (Wako Pure Chemicals), GSK-J4 (Cayman Chemical), SAHA (Cayman Chemical), rapamycin (LC Laboratories, Woburn, MA) and FK506 (Sigma-Aldrich) were prepared using dimethyl sulfoxide. The prepared drugs or vehicles were i.p. injected into nursing female mice daily, beginning on P1 and continuing until P13.

Statistical analysis

All statistical analyses were performed using Prism software (ver. 6.01; GraphPad Software, Inc., San Diego, CA). The data analyses were performed with Student’s t-test, Welch’s t-test or Mann–Whitney U-test for comparisons between two groups or a one-way analysis of variance followed by the Tukey-Kramer test for comparisons between multiple experimental groups. The differences were considered significant at a P-value < 0.05.

Supplementary Material

Supplementary Material is available at HMG online.

Acknowledgements

We wish to thank Riyo Takahashi, Kumi Kato, Yoko Tateda, Aya Inoue and Yuya Miyata for their technical assistance. We also acknowledge the support of the Biomedical Research Core of Tohoku University Graduate School of Medicine.

Conflict of Interest statement. None declared.

Funding

The Funding Program for the Next Generation of World-Leading Researchers (NEXT Program) from the Ministry of Education, Culture, Sports, Science and Technology of Japan to Y.A. (LS004), the Grants-in-Aid from the Practical Research Project for Rare/Intractable Diseases from the Japan Agency for Medical Research and Development, AMED to Y.A. (16eK0109021h0003), and the Japan Society for the Promotion of Science (JSPS) KAKENHI Grant Number 26293241 and 16K15522 to Y.A. and JSPS KAKENHI Grant Number 15K19598 to S.-I.I.