Cystic cerebellar dysplasia and biallelic LAMA1 mutations: a lamininopathy associated with tics, obsessive compulsive traits and myopia due to cell adhesion and migration defects

Abstract

Background: Laminins are heterotrimeric complexes, consisting of α, β and γ subunits that form a major component of basement membranes and extracellular matrix. Laminin complexes have different, but often overlapping, distributions and functions.

Methods: Under our clinical protocol, NCT00068224, we have performed extensive clinical and neuropsychiatric phenotyping, neuroimaging and molecular analysis in patients with laminin α1 (LAMA1)-associated lamininopathy. We investigated the consequence of mutations in LAMA1 using patient-derived fibroblasts and neuronal cells derived from neuronal stem cells.

Results: In this paper we describe individuals with biallelic mutations in LAMA1, all of whom had the cerebellar dysplasia, myopia and retinal dystrophy, in addition to obsessive compulsive traits, tics and anxiety. Patient-derived fibroblasts have impaired adhesion, reduced migration, abnormal morphology and increased apoptosis due to impaired activation of Cdc42, a member of the Rho family of GTPases that is involved in cytoskeletal dynamics. LAMA1 knockdown in human neuronal cells also showed abnormal morphology and filopodia formation, supporting the importance of LAMA1 in neuronal migration, and marking these cells potentially useful tools for disease modelling and therapeutic target discovery.

Conclusion: This paper broadens the phenotypes associated with LAMA1 mutations. We demonstrate that LAMA1 deficiency can lead to alteration in cytoskeletal dynamics, which may invariably lead to alteration in dendrite growth and axonal formation. Estimation of disease prevalence based on population studies in LAMA1 reveals a prevalence of 1-20 in 1 000 000.

Trial registration number: NCT00068224.

Keywords: Clinical genetics; Genetics; Movement disorders (other than Parkinsons); Myopia; Neurosciences.

Conflict of interest statement

Competing interests None declared.

Published by the BMJ Publishing Group Limited. For permission to use (where not already granted under a licence) please go to http://www.bmj.com/company/products-services/rights-and-licensing/

Figures

Figure 1

Clinical and neuroimaging features of patients with LAMA1 mutations. (A) Clinical features of patients showing distinct ophthalmologic characteristics. Patients 1 (A), 2 (B) and 3 (C) had unremarkable facial features. Note residual strabismus of patient 1 after corrective surgery. (D) Composite colour fundus photography of patient 1’s right eye demonstrating a dysplastic optic nerve with nasalised retinal vasculature (arrow) and diffuse chorioretinal atrophy involving the macula and midperiphery. (E) Composite fundus autofluorescence of the same eye, showing a ‘cobblestone’ pattern (arrow) indicating chorioretinal atrophy. (F) Composite colour fundus photograph of patient 2, the older sister of patient 1, showing moderate chorioretinal atrophy (arrow). (G) Colour fundus photo of patient 3 showing normal-appearing posterior pole and peripheral lattice degeneration (arrow), despite only mild myopia. (H–S) Brain MRI demonstrates cerebellar dysplasia associated with cysts. Brain MRI findings of patient 1 (H–J), patient 2 (K–M), patient 3 (N–P) and a healthy control (Q–S) are shown. (H) Patient 1 at 7 months of age has multiple cortical/subcortical cysts in the cerebellar vermis (arrow), dysplastic cerebellar vermis and enlarged fourth ventricle (arrowhead) with an abnormal rectangle-like shape (midsagittal-T1). The midbrain is mildly elongated, and the pons is mildly reduced in size. (I) Subcortical cysts (arrows) and dysplasia are also seen in the cerebellar hemispheres (axial-T1). (J) At 10 years 7 months, dysplasia in cerebellar vermis and hemispheres with multiple cortical and subcortical cysts (arrow) remain visible. (K) Patient 2 has hypoplasia and dysplasia of the cerebellar vermis with cortical and subcortical cysts, enlarged fourth ventricle (asterisk), elongated midbrain and pons appearing small (midsagittal-T1). (L) Cerebellar hemispheres are dysplastic and with multiple cortical and subcortical cysts in patient 2 (axial-T2). (M) Cortical and subcortical cysts within the left cerebellar hemisphere of patient 2 (parasagittal-T1) are shown. Images (K–M) were taken at age 6 years and 10 months. (N) Patient 3 has a normal sized but severely dysplastic vermis (white dot) with lack of separation of the vermian lobules (midsagittal-T1). Axial (O) and coronal (P) T2-weighted images of patient 3 show a markedly dysplastic appearance of the cerebellar hemispheres and the vermis with a diffusely abnormal configuration of the cerebellar folia and sulci (black arrows). Multiple small cysts are noted in the cerebellar hemispheres bilaterally (white arrows). Images (N–P) were taken at 4 years 3 months. (Q) A healthy 12-year-old control with normal anatomy of the posterior fossa structures (midsagittal-T1) and a normal triangle-shaped fourth ventricle is shown. Normal morphology of the cerebellum with an ‘onion-like’ foliation/fissuration pattern and white matter arborisation in the cerebellar hemispheres seen in axial (R) and coronal (S) T2 weighted images.

Figure 2

LAMA1 null mutations in patients with lamininopathy. (A) Pedigrees of the two families including chromatograms of their identified LAMA1 variants. (B) Quantitative real-time PCR results for LAMA1 mRNA expression in fibroblast from patient 2 and patient 3 compared with control using five different probes spanning LAMA1 cDNA. Values are percentage expression of LAMA1 in patient cells compared with control cells, normalised to ACTB (error bars represent S.D., n=4 from three independent experiments; **p<0.01, Mann–Whitney test); under the histogram, localisation of the different probes on LAMA1 transcript (NM_005559.3) is shown. (c) Immunoblot of fibroblast lysates of patient 2, patient 3 and control, showing absence of LAMA1 in the patients. Loading was controlled by vinculin (VCL).

Figure 3

LAMA1-deficient cells exhibit impaired adhesion and migration. (A) Equal numbers of control and patient 2 fibroblasts were plated on wells coated with either bovine serum albumin (BSA), collagen-I (Col-I) or Laminin-111 and allowed to attach. OD values represent amounts of DNA in cells that remain attached to the wells after 90 min of incubation. Bar graph gives mean and SD of five replicates from three independent experiments; *p

Figure 4

LAMA1 deficiency disrupts cell architecture for focal adhesion in fibroblasts. Equal numbers of cells from control and patient 2 fibroblasts were plated on non-coated or laminin-111-coated coverslips. Laminin-111 is a heterotrimer complex composed of LAMA1, LAMB1 and LAMC1. (A) Cells were imaged after staining with phalloidin (red) that labels actin filaments, and DAPI (blue) that labels the nuclei. On uncoated coverslips, the control cells show intact, highly organised cytoskeletal organisation, and numerous filopodia. In cells of patient 2 and patient 3, there is less complexity in cytoskeletal network, and filopodia structures are remarkably reduced or absent, indicating the importance of LAMA1 in the focal adhesion complex. On laminin-111-coated coverslips, control cells appear to spread more than on non-coated coverslips, while the filopodia length is increased. Cells of patient 2 and patient 3 show increased number of filopodia on laminin-coated coverslips. Scale bar represents 20 μm. (B) On scanning electron microscopy, control cells on uncoated slides appear thoroughly spread and flat with well-formed filopodia structures at the edges, as expected in normal control fibroblasts. Patient 2 and patient 3 cells on uncoated slides generally appear ‘lifted’, creating a three-dimensional appearance. The number of filopodia in patient 2 cells are markedly lower as compared with control cells; this number normalised to almost similar to control cells when patient 2 and patient 3 cells were grown on laminin-coated slides. In addition, microvilli protrusions, which are presumably unattached filopodia or lamellipodia, are seen on the surface of the cells. Scale bars in (B) represent 50 μm in B1–2 and B4–5, 10 μm in B3, B6, B7 and B9 and 25 μm in B8. (C) GTPase activation assay of Cdc42, Rac1 and RhoA was performed in control and patient 2 fibroblasts. Results show a significant decrease of the activated form (GTP-bound) of Cdc42 in patient 2 and patient 3 fibroblasts compared with control; no difference was observed for Rac1 and RhoA (error bars represent SD, n=6; *p

Figure 5

Effects of LAMA1 knockdown in neuronal cells (A–C); LAMA1 knockdown in neuronal cells reduced the number of filopodia. Non-targeting (control) siRNA and siRNA to LAMA1 (LAMA1-siRNA) were introduced to neuronal cells differentiated from Neural Stem Cells (NSC). Immunohistochemistry with phalloidin (A) showed that LAMA1-siRNA reduced the number of neurites and branching, in addition to reduction of filopodia (red, phalloidin; blue, DAPI stain for nucleus). Scanning electron microscopy (SEM) (B,C) confirmed that the number of filopodia is reduced when LAMA1 is knocked down (LAMA1-siRNA) as compared with control. Filopodia were partially recovered when wild-type hLAMA1 was reintroduced to cells (LAMA1-siRNA+Rescue). Magnified images of (B) are shown in (C). Scale bars represent 100 μm in (A), 10 μm in (B) and 2 μm in (C). (D) LAMA1 deficiency increases susceptibility to apoptosis. To quantify apoptosis, caspase-3/7 activity was measured at 12 and 30 h after plating control (white bars), patient 2 (light grey bars) and patient 3 cells, with or without the addition of the known apoptosis inducer staurosporine. Results show an increased susceptibility of patient’s cells to apoptosis. Bars represent SD for three replicates of three experiments; *p<0.05; **p<0.01 (Mann–Whitney test).