Mek1Y130C mice recapitulate aspects of human cardio-facio-cutaneous syndrome

Abstract

The RAS/MAPK signaling pathway is one of the most investigated pathways, owing to its established role in numerous cellular processes and implication in cancer. Germline mutations in genes encoding members of the RAS/MAPK pathway also cause severe developmental syndromes collectively known as RASopathies. These syndromes share overlapping characteristics, including craniofacial dysmorphology, cardiac malformations, cutaneous abnormalities and developmental delay. Cardio-facio-cutaneous syndrome (CFC) is a rare RASopathy associated with mutations in BRAF, KRAS, MEK1 (MAP2K1) and MEK2 (MAP2K2). MEK1 and MEK2 mutations are found in ∼25% of the CFC patients and the MEK1^Y130C substitution is the most common one. However, little is known about the origins and mechanisms responsible for the development of CFC. To our knowledge, no mouse model carrying RASopathy-linked Mek1 or Mek2 gene mutations has been reported. To investigate the molecular and developmental consequences of the Mek1^Y130C mutation, we generated a mouse line carrying this mutation. Analysis of mice from a Mek1 allelic series revealed that the Mek1^Y130C allele expresses both wild-type and Y130C mutant forms of MEK1. However, despite reduced levels of MEK1 protein and the lower abundance of MEK1 Y130C protein than wild type, Mek1^Y130C mutants showed increased ERK (MAPK) protein activation in response to growth factors, supporting a role for MEK1 Y130C in hyperactivation of the RAS/MAPK pathway, leading to CFC. Mek1^Y130C mutant mice exhibited pulmonary artery stenosis, cranial dysmorphia and neurological anomalies, including increased numbers of GFAP⁺ astrocytes and Olig2⁺ oligodendrocytes in regions of the cerebral cortex. These data indicate that the Mek1^Y130C mutation recapitulates major aspects of CFC, providing a new animal model to investigate the physiopathology of this RASopathy. This article has an associated First Person interview with the first author of the paper.

Keywords: Cardio-facio-cutaneous syndrome; MEK1 Y130C mutation; Mouse model; Neurological defects; Pulmonary artery stenosis; RAS/MAPK pathway.

Fig. 1.

Generation of the Mek1^Y130C allele. (A) Mek1 gene-targeting strategy for the generation of the Y130C point mutation. Exons are represented as white boxes. The targeting vector contains a PGKneo selection cassette flanked by loxP sites (black triangles), a PGK-DTA selection cassette (gray box), an A to G substitution in exon 3 (asterisk), and a new XbaI site (in bold) inserted into the third intron to facilitate the identification of the targeted allele by Southern blot analysis. Insertion of the targeting vector by homologous recombination will generate the Mek1^Y130C-neo-targeted allele. Cre-mediated recombination of the Mek1^Y130C-neo allele will generate the Mek1^Y130C allele. Location of the 5′, internal and 3′ probes used for Southern analyses is indicated as gray boxes (a,b,c, respectively). (B) Schematic representation of DNA fragments obtained after EcoRI, XbaI and NheI digestions detected by Southern analyses with the 5′, 3′ and internal probes, respectively. (C) Southern blot analyses for ES cell screening (left panel; XbaI digestion with probe c), germline transmission (right panel; EcoRI digestion with probe a) and Cre-mediated deletion (lower panel; NheI digestion with probe b). The position of the different alleles is indicated on the side of the gels. (D) Sequence analysis of amplified exon 3 fragment from Mek1^+/+ (left panel) and Mek1^+/Y130C (right panel) specimens, confirming the presence of the A to G transition mutation, leading to the substitution of a tyrosine for a cysteine.

Fig. 2.

The Mek1^Y130C allele contains a partial duplication of the Mek1 gene. (A) Schematic of the endogenous Mek1 gene, the Mek1^Y130C allele and the Mek1 null allele with the expected DNA fragments after StuI digestion and detection with probe b. (B) Southern blot analysis of tail DNA from a litter obtained following Mek1^+/−×Mek1^+/Y130C breeding. DNA was digested with StuI, blotted and hybridized with probe b. Mice positive for the Mek1^Y130C allele always carried a wt allele even when a null allele was detected. The wt allele showed a stronger signal equivalent to the one of wt mice, suggesting Mek1 gene duplication in the Mek1^Y130C allele. (C) Whole-genome sequencing alignment reveals a 61.5 kb Mek1 duplication starting 6.5 kb upstream of the Mek1 transcription start site and ending in the fifth intron. Only one break point was observed between the fifth intron in 5′ and Mek1 promoter sequences in 3′, suggesting intragenic duplication occurring by unequal crossing over.

Fig. 3.

MEK1 Y130C expression levels. (A) Schematic representation of Mek1 mRNA with the eleven exons. Blue and red boxes represent the translated region (amino acids 1 to 393) and the kinase domain, respectively. The position of the Y130C point mutation is indicated, as well as that of primers C and D used for qRT-PCR. (B) Mek1 mRNA expression levels were assessed by qRT-PCR analysis on RNA isolated from Mek1^+/+ (n=4), Mek1^+/− (n=4) and Mek1^Y130C/− (n=6) kidneys. (C) MEK1 protein levels were evaluated by western blot analysis of total protein extracts from Mek1^+/+ (n=3), Mek1^+/− (n=3) and Mek1^Y130C/− (n=3) kidneys. Vinculin was used as a loading control. Quantification showed a significant diminution of Mek1 mRNA and MEK1 protein levels in Mek1^+/− and Mek1^Y130C/− mutants compared to Mek1^+/+ specimens. Values are reported as mean±s.e.m. (D) qRT-PCR products obtained from Mek1^+/+ and Mek1^Y130C/− samples analyzed in B were sequenced. Sequence from Mek1^Y130C/− samples showed equal representation of A and G at the mutation site, indicating the presence of transcript encoding the Y130C mutation. (E) Relative wt MEK1 and MEK1 Y130C protein levels were quantified by PRM-targeted mass spectrometry using the CNSPYIVGFYGAFYSDGE wt and CNSPYIVGFCGAFYSDGE Y130C peptides (mean±s.d.; n=2 biological replicates).

Fig. 4.

RAS/MAPK pathway hyperactivation in MEFs carrying Mek1^Y130C allele. (A-C) Mek1^+/+-, Mek1^+/Y130C- and Mek1^Y130C/−-established MEF lines were stimulated with 20% FBS (in A, Mek1^Y130C/− MEFs were not tested), 2 ng/ml of FGF2 (B) or 2 ng/ml of EGF (C), and phosphorylation of ERK1/2 was assessed by quantitative immunoblotting along with total vinculin as a loading control. (D) Three Mek1^+/+, Mek1^Y130C/− and Mek1^Y130C/Y130C primary MEF cultures were treated with 2 ng/ml EGF and phosphorylation of ERK1/2 was assessed. Values are reported as mean±s.e.m. in arbitrary units (n=3).

Fig. 5.

Mek1^Y130C mutant mice present cardiac and cranial anomalies. (A) E13.5 control (Mek1^+/+ and Mek^+/−) and Mek1^+/Y130C, Mek1^Y130C/− and Mek1^Y130C/Y130C mutant embryos did not present overt morphological anomalies. (B) H&E staining of transverse sections from pulmonary arteries of E13.5 control (Mek1^+/+ and Mek1^+/−) and Mek1^+/Y130C, Mek1^Y130C/− and Mek1^Y130C/Y130C mutant embryos. Pulmonary stenosis was observed in all specimens carrying a Mek1^Y130C allele (arrowhead). e, esophagus; lb, left bronchia; lpa, left pulmonary artery (arrow); rb, right bronchia; rpa, right pulmonary artery. (C) Measurement of arterial lumen area confirmed the pulmonary stenosis for the left and right arteries in Mek1^+/Y130C, Mek1^Y130C/− and Mek1^Y130C/Y130C mutants. (D) Immunostaining for phospho-ERK (pERK) was performed on sections of pulmonary arteries from E13.5 Mek1^+/+, Mek1^+/− and Mek1^Y130C/− embryos. No difference was observed. (E) Morphometric characteristics of mice skulls. Length and width of the skull as well as inner canthal width were measured in cm on Alcian Blue/Alizarin Red-stained skulls. Mek1^Y130C/− mice presented facial dysmorphia with reduced skull width and increased inner canthal width. Mek1^Y130C/Y130C mutants presented reduced skull length. *P<0.05; **P<0.01; ***P<0.005. Scale bars: 25 µm (A); 100 µm (D); 200 µm (B).

Fig. 6.

Increased number of GFAP⁺ cells in cortical and hippocampal sections of Mek1^Y130C/Y130C mutants. (A) Representative coronal brain sections stained for GFAP. Red dashed line boxes denote an area of sensory cortex shown in C-F; yellow dashed line boxes denote an area of hippocampal CA1 presented in G-J. (B) Quantification of relative density of GFAP⁺ cell counts is shown as a relative number of positive cells/mm² (n=3). (C-F) Mek1^Y130C/Y130C animals exhibited an increased number of GFAP⁺ astrocytes in the sensory cortex in comparison to wt animals. (G-J) Analysis of GFAP-labeled astrocytes in hippocampal CA1 revealed a modest but significant increase in the relative density of cells/mm². *P<0.05; **P<0.01. Scale bars: 50 µm (G); 100 µm (C).

Fig. 7.

Mek1^Y130C/Y130C cortices exhibit increased density of Olig2⁺ cells. (A) Representative coronal brain sections stained for the oligodendrocyte transcription factor (Olig2). Red boxes denote an area of sensory cortex presented in C-F. (B) Quantification of relative densities of Olig2⁺ cells in adult sensory cortex (n=3). (C-F) The relative density of Olig2⁺ cells was assessed in radial columns of sensory cortex. Note the increased density of Olig2⁺ cells in the mutant sensory cortex (E,F) relative to controls (C,D). **P<0.01. Scale bars: 100 µm.