The CMS19 disease model specifies a pivotal role for collagen XIII in bone homeostasis

Abstract

Mutations in the COL13A1 gene result in congenital myasthenic syndrome type 19 (CMS19), a disease of neuromuscular synapses and including various skeletal manifestations, particularly facial dysmorphisms. The phenotypic consequences in Col13a1 null mice (Col13a1^-/-) recapitulate the muscle findings of the CMS19 patients. Collagen XIII (ColXIII) is exists as two forms, a transmembrane protein and a soluble molecule. While the Col13a1^-/- mice have poorly formed neuromuscular junctions, the prevention of shedding of the ColXIII ectodomain in the Col13a1^tm/tm mice results in acetylcholine receptor clusters of increased size and complexity. In view of the bone abnormalities in CMS19, we here studied the tubular and calvarial bone morphology of the Col13a1^-/- mice. We discovered several craniofacial malformations, albeit less pronounced ones than in the human disease, and a reduction of cortical bone mass in aged mice. In the Col13a1^tm/tm mice, where ColXIII is synthesized but the ectodomain shedding is prevented due to a mutation in a protease recognition sequence, the cortical bone mass decreased as well with age and the cephalometric analyses revealed significant craniofacial abnormalities but no clear phenotypical pattern. To conclude, our data indicates an intrinsic role for ColXIII, particularly the soluble form, in the upkeep of bone with aging and suggests the possibility of previously undiscovered bone pathologies in patients with CMS19.

Figure 1

Single scaled mid-diaphyseal axial tomographic µCT images of male and female Col13a1^−/− mouse femurs and respective controls at the 4-, 12-, 25-, 35, and 72-week time points. Scale bars 1 mm.

Figure 2

Single-scaled mid-diaphyseal axial tomographic µCT images of male and female Col13a1^tm/tm mouse femurs and respective controls. From the 4- to 35-week time points, diaphyseal cortical bone morphology remained mostly unaltered in the Col13a1^tm/tm samples compared to the controls, whereas at 72 weeks a slight reduction in bone mass could be seen compared to the WT samples. Scale bars 1 mm.

Figure 3

Summary of morphometric parameters of diaphyseal cortical bone and distal femoral trabecular bone of male and female Col13a1^−/− mice and WT controls. Cortical area (Ct.Ar, A) and medullary area (Ma.Ar, B) were significantly reduced in the female Col13a1^−/− mice at the 72-week compared to the controls. There was no differences in cortical thickness (Ct.Th., C) between genotypes. Bone volume fraction (BV/TV, D) and trabecular number (Tb.N, E) were higher in 35-week-old Col13a1^−/− males compared to WT. There was no differences in trabucular thickness (Tb.Th., F) between genotypes. Trabecular separation (Tb.Sp, G) was higher in 72-week-old Col13a1^−/− females compared to WT. n(WT male): 3–6; n(Col13a1^−/− male): 4–7; n(WT female): 3–5; n(Col13a1^−/− female): 5–6; the whiskers represent min to max; *q < 0.05, **q < 0.01, ***q < 0.001 determined by two-way ANOVA and followed by the false discovery rate.

Figure 4

Summary of morphometric parameters of diaphyseal cortical bone and distal femoral trabecular bone of male and female Col13a1^tm/tm mice and WT controls. The cortical bone area (Ct.Ar, A) was significantly reduced in the female Col13a1^tm/tm mice at the 35- and 72-week and in males at the 72-week. The medullary area (Ma.Ar, B) was significantly lower in Col13a1^−/− mice at the age of 72-week compared to the controls The cortical thickness (Ct.Th, C) was reduced in the female Col13a1^tm/tm mice at 12-, 35- and 72-week-old mice. The trabecular bone was unaltered between Col13a1^tm/tm and WT samples (D–G). n(WT male): 4–6; n(Col13a1^tm/tm male): 4–6; n(WT female): 2–6; n(Col13a1^tm/tm female): 4–6; the whiskers represent min to max; *q < 0.05, **q < 0.01, ***q < 0.001 determined by two-way ANOVA and followed by the false discovery rate.

Figure 5

Single scaled axial tomographic µCT images of distal metaphyseal femoral cortical bone in female Col13a1^−/−, Col13a1^tm/tm and WT control mice at the age of 35 and 72 weeks. Whereas bone morphology was similar at the 35-week time point (A,B), cortical bone mass was clearly reduced in the Col13a1^−/− mice at the 72-week time point compared to the WT samples (C,D). The metaphyseal cortical bone mass was significantly reduced in the Col13a1^tm/tm mice at both time points compared to the WT controls (E–H). Scale bars 500 µm.

Figure 6

Morphometric parameters for HRCT analysis of metaphyseal cortical bone. The Col13a1^−/− female mice and respective WT controls (A–C). There were no statistically significant differences in the cortical area (Ct.Ar, A) or thickness (Ct.Th, B) or the medullary area (Ma.Ar, C) at the 35-week time point. At the 72-week time point, Ct.Ar and Ct.Th were significantly reduced in the Col13a1^−/− samples compared to the WT mice. The Col13a1^tm/tm female mice and respective WT controls (D–F). The cortical area (Ct.Ar, D) and thickness (Ct.Th, E) were significantly reduced in the Col13a1^tm/tm mice at the 35- and 72-week time points compared to the respective controls. The medullary area (Ma.Ar, F) was unaltered. n(WT female): 4–5; n(Col13a1^−/− female): 5; n(WT female): 3–6; n(Col13a1^tm/tm female): 4–6; the whiskers represent min to max; **q < 0.01 determined by two-way ANOVA and followed by the false discovery rate.

Figure 7

A 3D-reconstructed image of male and female mouse skulls, showing 25 anatomic landmarks used in the cephalometric analysis. Representative images of Col13a1^−/− and Col13a1^tm/tm female mice and respective WT controls at 4- and 35-week time points. For measurement data, see the Supplementary Tables S2 and S2, and Supplementary Figs. S7 and S8.