Multiple congenital malformations of Wolf-Hirschhorn syndrome are recapitulated in Fgfrl1 null mice

Abstract

Wolf-Hirschhorn syndrome (WHS) is caused by deletions in the short arm of chromosome 4 (4p) and occurs in about one per 20,000 births. Patients with WHS display a set of highly variable characteristics including craniofacial dysgenesis, mental retardation, speech problems, congenital heart defects, short stature and a variety of skeletal anomalies. Analysis of patients with 4p deletions has identified two WHS critical regions (WHSCRs); however, deletions targeting mouse WHSCRs do not recapitulate the classical WHS defects, and the genes contributing to WHS have not been conclusively established. Recently, the human FGFRL1 gene, encoding a putative fibroblast growth factor (FGF) decoy receptor, has been implicated in the craniofacial phenotype of a WHS patient. Here, we report that targeted deletion of the mouse Fgfrl1 gene recapitulates a broad array of WHS phenotypes, including abnormal craniofacial development, axial and appendicular skeletal anomalies, and congenital heart defects. Fgfrl1 null mutants also display a transient foetal anaemia and a fully penetrant diaphragm defect, causing prenatal and perinatal lethality. Together, these data support a wider role for Fgfrl1 in development, implicate FGFRL1 insufficiency in WHS, and provide a novel animal model to dissect the complex aetiology of this human disease.

Fig. 1.

Expression of Fgfrl1 at age E10.5. (A) Fgfrl1 expression is prominent in the brain, cranial placodes, pharyngeal arches and somites. (B) Cross section through the otic vesicle, first and second pharyngeal arches and nasal pit confirms Fgfrl1 expression in the cranial placodes, as well as in a subepithelial compartment of the pharyngeal arch (inset; arrowhead indicates the pharyngeal ectoderm without Fgfrl1 expression). (C) Fgfrl1 expression in the neural tube (arrowhead). (D–F) Fgfrl1 expression in the heart (D) and in the endocardial cushions (arrows) of the atrioventricular canal (E) and cardiac outflow tract (F). Abbreviations: av, atrioventricular canal; ey, eye; fb, forebrain; np, nasal pit; ot, outflow tract; ov, otic vesicle; pa, pharyngeal arch; so, somite. Bars, 1 mm (A,C); 0.5 mm (B,D–F).

Fig. 2.

Targeting of Fgfrl1 by homologous recombination to produce the Fgfrl1 null allele. (A) Schematic representation of the Fgfrl1 locus, targeting construct and Fgfrl1 null allele after Cre-mediated recombination of the targeted locus. Coding exons are depicted as open boxes and the 3′ and 5′ UTRs (untranscribed regions) are shown as black boxes. The FRT-flanked PGK neo^r cassette is depicted as a grey box. LoxP sites in the targeting construct flank exons 3–7 and the PGK neo^r cassette, which are removed after Cre-mediated deletion. Left arm and right arm external probes were used for screening embryonic stem (ES) cells to distinguish the endogenous allele from the targeted allele. Arrows indicate the position of the PCR primers (P1, P2, P3) used for genotyping. (B) Southern blot analysis using the left arm probe on ES cell genomic DNA after SpeI digest, indicating the wild-type (18.7 kb) and the Fgfrl1^flox allele (8.7 kb) in successfully targeted clones. (C) PCR analysis of embryos derived from matings between mice that are heterozygous for the Fgfrl1 null allele, showing the 245 bp wild-type band and the 310 bp band of the Fgfrl1 null allele. (D) Relative quantification of mRNA levels by RT-PCR for Fgfrl1, and the WHSCR genes Fgfr3, Whsc1 and Whsc2 in E12.5 embryos. Values are shown as fold induction in Fgfrl1 mutants relative to wild-type littermates following normalisation to Gapdh expression (mean±s.e.m.). ***P<0.01.

Fig. 3.

Comparison between Fgfrl1 wild-type (+/+) mice and homozygous mutant (−/−) mice. Wild-type (A) and Fgfrl1^−/− (B) embryos at age E14.5 showed no gross morphological differences. Wild-type (C) and Fgfrl1^−/− (D,E) embryos at age E16.5. At this stage, a subpopulation of Fgfrl1^−/− embryos (E) was severely affected, suffering prenatal lethality and showing developmental retardation. Mildly affected Fgfrl1^−/− embryos (D) display a short stature and a dome-shaped cranium. Wild-type (F) and severely affected Fgfrl1^−/− (G) embryos at age E16.5 within their yolk sac. Whereas the yolk sac of the wild-type embryo (F) shows a prominent vasculature, the yolk sac of the Fgfrl1^−/− embryo (G) lacks a clear vasculature and is devoid of blood. (H) Wild-type and mildly affected Fgfrl1^−/− embryos at age E18.5. The Fgfrl1^−/− embryos at this stage maintain the typical cranial and short stature phenotype. Comparison between the diaphragm from a wild-type (I) and Fgfrl1^−/− (J) embryo at age E18.5 shows that the lumbar (arrows) and sternal (arrowhead) muscle portions of the diaphragm were not present in Fgfrl1^−/−embryos and that the remaining costal portions of the diaphragm muscle were very thin in these embryos compared with in their wild-type littermates. The crural muscles (asterisk) in Fgfrl1^−/− mice were also smaller compared with control animals and confirm the overall defect in diaphragm development. Bars, 1 mm.

Fig. 4.

Skeletal anomalies in Fgfrl1^−/− embryos. (A) Comparison between the whole skeletons of wild-type (+/+) and homozygous Fgfrl1 mutant (−/−) embryos at E18.5, with ossified areas in red and cartilage in blue, revealed a general hypoplasia of all skeletal elements in Fgfrl1 mutant embryos, including shortened axial and appendicular skeletons, malformation of the vertebrae (arrowhead), shortening of the limb girdles and a reduction in the size of the ribcage. (B) Ventral view of the skull, showing midfacial and mandibular hypoplasia and an anterior-shifted foramen magnum in Fgfrl1^−/− mice. (C) Dorsal view of the skull, revealing a delay in suture closure between frontal and parietal bones in Fgfrl1^−/− mice. The frontal and parietal bones in Fgfrl1^−/− mice are also thin compared with those in wild-type littermates. Comparison between cross sections of wild-type (D) and homozygous mutant (E) heads shows enlargement of the brain ventricles and brain overgrowth (arrow) in Fgfrl1^−/− mice. (F,G) Ossification of the cervical vertebrae (arrowhead) is delayed in homozygous mutants. (H,I) Fgfrl1^−/− mice display a shortened sternum with abnormal ossification that is most prominent in the manubrium and the xiphoid process. The sternum was also bending inward in the mutant background, contributing to the reduced thoracic diameter. (J,K) Fgfrl1^−/− mice show incomplete ossification of the hyoid bone and hypoplasia of permanent laryngeal cartilage elements. Abbreviations: bo, basioccipital; bs, basisphenoid; cc, cricoid cartilage; fm, foramen magnum; fr, frontal; hb, hyoid bone; ma, manubrium; pa, parietal; sp, sphenoid; tc, thyroid cartilage; tr, trachea; xp, xiphoid process. Bars, 1 mm.

Fig. 5.

Congenital heart defects in Fgfrl1^−/− embryos. At E14.5, a comparison between wild-type (A–C) and Fgfrl1^−/− (D–F) hearts reveals ventricular septation defects [compare (A) with (D)], and thickening of the atrioventricular [compare (B) with (E)] and semilunar [compare (C) with (F)] valves (arrowheads) in Fgfrl1^−/−embryos. (G–L) The Fgfrl1^−/− mutant phenotypes persist at E18.5: ventricular septation defects [compare (G) with (J)], and thickening of the atrioventricular [compare (H) with (K)] and semilunar [compare (I) with (L)] valves (arrowheads). Bars, 0.5 mm.

Fig. 6.

Comparison of placental development between wild-type (+/+) and homozygous Fgfrl1 mutant (−/−) embryos. (A,B) Histological analysis of placentas at E14.5 reveals no gross morphological differences between wild-type and Fgfrl1^−/− embryos. (C) Placentas of severely affected Fgfrl1^−/− embryos at E16.5 were pale compared with those of wild-type littermates. At E16.5, mildly affected Fgfrl1^−/− embryos have a normal placenta [compare (D) with (E)], whereas the structure of the labyrinth in placentas of severely affected Fgfrl1^−/− embryos is highly disorganised (F). Abbreviations: de, decidua; la, labyrinth; sp, spongiotrophoblast. Bars, 1 mm.

Fig. 7.

Foetal anaemia in Fgfrl1^−/− embryos. (A–C) The peripheral blood in E16.5 Fgfrl1^−/− embryos shows a 27% decrease in red blood cell count (A), a 37% decrease in the haemoglobin level (B) and 23% less haematocrit (C) compared with wild-type littermates. Data are presented as mean±s.e.m. (***P<0.01). (D) Dot plots of the typical CD71 and Ter119 staining pattern on 10,000 viable cells that were dispersed from wild-type (WT) and Fgfrl1^−/− (KO) whole foetal livers at E14.5 and E16.5. The axes indicate the relative logarithmic fluorescence units for Ter119-PE-Cy7 (x-axis) and CD71-PE (y-axis). The cells in regions R1 to R5 represent erythroblasts at different developmental stages (R1 contains the least, and R5 the most, differentiated cells) and are defined by their characteristic CD71 and Ter119 staining pattern. These regions can be classified as follows: primitive progenitor cells and proerythroblasts (R1), proerythroblasts and early basophilic erythroblasts (R2), early and late basophilic erythroblasts (R3), chromatophilic and orthochromatophilic erythroblasts (R4), and late orthochromatophilic erythroblasts and reticulocytes (R5) (Zhang et al., 2003). The numbers shown in the boxed areas indicate the frequency of cells in each region, as a percentage of total liver cells. Asterisks indicate the regions (R3, R4, R5) with significantly different values between Fgfrl1^−/− and wild-type embryos at E14.5 (P<0.01).

Fig. 8.

Comparison between WHSCRs in different mammalian species. Alignment between the human and orangutan (primates), mouse and rat (rodents), and horse and cow (ungulates) orthologous genes, which in humans are located at 4p16.3, reveals a high degree of conservation in chromosomal organisation between primates and ungulates, but shows that rodents have a very different arrangement. The chromosomal region that encompasses a part of 4p16.3 in humans has been broken up in rodents, separating Fgfrl1 from the WHSCRs over rather distant locations on the same chromosome (~75 Mb in the mouse and ~81 Mb in the rat). (Sequences have been derived from the Ensembl Genome Browser.)