Clinical characterization of Collagen XII-related disease caused by biallelic COL12A1 variants

Abstract

Objective: While there have been several reports of patients with dominantly acting COL12A1 variants, few cases of the more severe recessive Collagen XII-related disorders have previously been documented.

Methods: We present detailed clinical, immunocytochemical, and imaging data on eight additional patients from seven families with biallelic pathogenic variants in COL12A1.

Results: All patients presented with a consistent constellation of congenital onset clinical features: hypotonia, dysmorphic features, most notably gingival hypertrophy, prominent distal joint hyperlaxity, with co-occurring contractures of large joints, and variable muscle involvement, evident both clinically and on muscle imaging. Five patients presented with a severe congenital phenotype manifesting with profound weakness, significantly delayed or minimal attainment of motor milestones, respiratory insufficiency, and feeding difficulties. Three patients presented with mild-to-moderate muscle weakness and delayed milestones but were able to achieve independent ambulation. Patients were found to have biallelic loss-of-function COL12A1 variants, except for one family (p.I1393Ffs*11/p.A1110D). Consistent with the variable clinical spectrum, in vitro immunocytochemistry analysis in fibroblasts ranged from complete absence of Collagen XII expression in a patient with severe disease, to a mild reduction in a patient with milder disease.

Interpretation: Here we characterize the clinical presentation, muscle imaging, and dermal fibroblast immunostaining findings associated with biallelic variants in COL12A1, further establishing COL12A1 as a recessive myopathic Ehlers-Danlos syndrome (mEDS) gene, and expanding the clinical spectrum to include a milder EDS phenotype.

Figure 1

Common clinical and histological findings in patients with biallelic COL12A1 variants. (A) Wrist hyperlaxity; finger, and long finger flexor contractures (F1P1). (B) Hyperlaxity of the ankle (F5P6). (C) Prominent calcanei (F1P1). (D) Long finger flexor contractures (F2P2). (E) High‐arched palate with a deep midline groove and dental eruption cysts (F1P1). (F) Gingival hypertrophy (F5P5). (G) Narrow elongated face with myopathic facies and open mouth posture (F1P1). (H) Biopsy of the vastus lateralis from F7P8 at age 8 months revealed mild fiber size variation with both hypertrophied, rounded and polygonal atrophic fibers on hematoxylin and eosin (H&E) with (I) mild Type 1 fiber predominance and Type 1 hypotrophy on ATPase pH 4.6. (J) Biopsy of the vastus lateralis from F2P2 at age 7 years revealed very subtle variation in fiber size without clear predominance of Type 1 fibers on ATPase pH 4.3 (K).

Figure 2

Muscle imaging in patients with biallelic COL12A1 variants. (A) Muscle MRI, axial T1 at the thigh (top) and lower leg (bottom) for F1P1 (left, top) and F4P4 (middle, top), at age 15 months and 11 years, respectively. There was severe generalized muscle atrophy and fatty replacement in F1P1's proximal and distal lower extremity. Asymmetry and imaging artifact in the lower leg was due to contractures. F4P4's MRI demonstrated significant posterior thigh involvement with relative sparing of the semitendinosus. Preserved medial and anterior compartments of the thigh, with the exception of the rectus femoris with atrophy and fatty infiltrates noted around the central fascia. Muscle MRI in F5P5 (left, bottom) at age 6 years demonstrating normal appearance of muscles of the thigh and distal lower extremity. Muscle MRI of F7P8 (right, bottom) at age 7 years revealed relative sparing of the medial and anterior thigh compartments compared to the posterior thigh compartment, with selective atrophy of the rectus femoris, and to a lesser extent the vastus lateralis. Visualization of normal muscle imaging (far right). (B) Muscle ultrasound of F1P1 (age 15 months), F5P5 (age 6 years), and F5P6 (age 4 years) of the rectus femoris (RF) (a, e, and i), vastus lateralis (V. Lateralis) (b, f, and j), triceps (c, g, and k), and hamstring (d, h, and l). Imaging in P1 demonstrated diffuse muscle atrophy in the lower extremity (LE) more than upper extremity (UE) with dense granular increased echogenicity in the rectus femoris and vastus lateralis, and mixed pattern echogenicity in the triceps. Muscle ultrasound in F5P5 and F5P6 revealed mild–moderate granular/streaky echogenicity, most pronounced in the hamstrings and rectus femoris with no muscle atrophy.

Figure 3

Pedigrees and genetic findings in patients with biallelic COL12A1 variants. (A) Pedigrees of the families with biallelic COL12A1 variants. (B) Collagen XII α1 domains and structure with labeled recessive variants: COL12A1 recessive variants described in this paper (lower) and those reported previously (upper) were labeled on Collagen XII α1. The short isoform is depicted in bright colors; the additional domains in the long splice isoform are added as lighter symbols.

Figure 4

Collagen XII expression in cultured fibroblasts. (A) Immunofluorescence (IF) analysis of Collagen XII ECM formed in 5‐day culture, compared with normal control fibroblasts, F1P1 fibroblasts showed no detectable intracellular or extracellular immunoreactivity for Collagen XII; the signal of Collagen XII from F4P4 fibroblasts was detectable but significantly reduced while the signal from F5P5 is similar compared to normal control. (B) Immunoblotting of overnight conditioned medium of the fibroblast cultures with quantification. Compared to normal control fibroblasts, Collagen XII signal was barely detectable in F4P4 fibroblasts; was reduced in F5P5, but normal in both heterozygous unaffected carrier parents (P: p.I1393Ffs*11 / p.A1110D, F: p.A1110D, M: p.I1393Ffs*11) . P = patient; F = father; M = mother. Fibronectin was used as staining/loading control.