Somatic activation of AKT3 causes hemispheric developmental brain malformations

Abstract

Hemimegalencephaly (HMG) is a developmental brain disorder characterized by an enlarged, malformed cerebral hemisphere, typically causing epilepsy that requires surgical resection. We studied resected HMG tissue to test whether the condition might reflect somatic mutations affecting genes critical to brain development. We found that two out of eight HMG samples showed trisomy of chromosome 1q, which encompasses many genes, including AKT3, a gene known to regulate brain size. A third case showed a known activating mutation in AKT3 (c.49G→A, creating p.E17K) that was not present in the patient's blood cells. Remarkably, the E17K mutation in AKT3 is exactly paralogous to E17K mutations in AKT1 and AKT2 recently discovered in somatic overgrowth syndromes. We show that AKT3 is the most abundant AKT paralog in the brain during neurogenesis and that phosphorylated AKT is abundant in cortical progenitor cells. Our data suggest that somatic mutations limited to the brain could represent an important cause of complex neurogenetic disease.

Figure 1. MRIs of patients with hemimegalencephaly due to somatic mutations

(A, B) The first column shows an example of coronal T2-weighted and axial T2-weighted MRI images showing the brain of a normal 1-year-old. Note the symmetric size of the right and left hemispheres, labeled R and L to denote standard MRI convention. (C-J) Representative images from the brain MRIs of two patients with HMG, before and after surgical removal of the abnormal hemisphere, are shown. (C, D) HMG-1 has somatic trisomy of chromosome 1q. MRI before surgery showed left-sided hemispheric enlargement, abnormal cortical thickness and configuration, and enlarged left lateral ventricle in the coronal T2-weighted and axial T2-weighted images. The right hemisphere is smaller and appears normal. (E, F) Following left hemispherectomy surgery, there is cerebrospinal fluid (CSF) where the abnormal hemisphere had been, seen as bright signal in coronal and axial images taken at approximately the same plane as the pre-operative images. (G, H) HMG-3 has a somatic mosaic mutation in AKT3. Coronal T2-weighted and axial T2-weighted MRI images show right-sided hemispheric enlargement, abnormal cortical thickness and signal, abnormal white matter signal, and an enlarged lateral ventricle. (I, J) Following right hemispherectomy surgery, as in the previous case, CSF is visible as bright signal in place of the resected abnormal hemisphere.

Figure 2. Abnormal cortical development in hemimegalencephaly case HMG-1 with trisomy of chromosome 1q

(A) Low-power view (20x mag) of a gyrus from the cerebral cortex stained with haematoxylin and eosin (H&E) shows an abnormally contoured surface and variably thick cortical ribbon and molecular layer. (B) Analysis of subcortical white matter using Cresyl violet and Luxol Fast Blue (LFB) highlights numerous subcortical bands/islands of ectopic gray matter containing neurons and glia (*). (C) Immunohistochemical staining for phosphorylated neurofilament, SMI31, highlights scattered abnormal large neurons. (D) Rare small collections of neuroblast-like cells (microdysplasia) were present on H&E. (E) Immunohistochemical staining also demonstrated an abnormal number of proliferating Ki67 positive cells scattered throughout gray and white matter which had an atypical nuclear morphology (e). (A) 20x, (B) 200x (C, D, E) 600x.

Figure 3. Mosaic mutations in hemimegalencephaly: trisomy of chromosome 1q and an activating point mutation in AKT3

(A) Copy number for all of the chromosomes is shown for HMG-1; the estimated copy number for 1q is 2.41 (S.D. 0.12), consistent with mosaic trisomy 1q. Chromosome 1p as well as the other autosomes have normal copy number of 2, and chromosomes X and Y each show copy number of 1. (B) Copy number evaluation of Affymetrix 6.0 data shows the gain in copy number at chromosome 1q for HMG-1, with the x-axis representing nucleotide position along chromosome 1 and the y-axis denoting copy number. (C) Assuming a copy number of 2 for all regions in the DNA derived from leukocytes (white columns), the calculated copy number from the brain tissue (black columns) was 2.68 (S.D. 0.16) at 1q21.3, 2.76 (S.D. 0.20) at 1q31.1, and 2.73 (S.D. 0.13) at 1q42.2. (D) The AKT3 c.49G→A, p.E17K heterozygous mutation is present in the sequencing traces from brain-derived DNA (first row) and absent in the traces from leukocyte-derived DNA from HMG-3 (second row). The arrows point to AKT3 nucleotide position 49. Cloning results indicate that the mutation is present in 8/46 (17.4%) of the DNA reads from a brain tissue sample, suggesting that the mutation exists in the heterozygous state in 35% of the cells; traces from two clones are shown in the third and fourth rows, the trace in the third row showing the results of sequencing from a clone with the AKT3 c.49G→A mutation present (A) and the bottom row showing the results from a clone without the mutation but rather with the reference allele present (G).

Figure 4. Active Akt signaling in the developing cortex is enriched in apical progenitor cells and the cortical plate

(A) Immunohistochemistry of cortical sections at embryonic day (E)10.5 reveals Akt activity as assessed by pan-phospho(P)-Akt immunostaining (red) in the cortical plate and ventricular zone. (B-D) Higher magnification images of ventricular zone at E10.5 are shown. Overlay of P-Akt with P-Vimentin 4A4 (green) shows that dividing radial glial cells, which generate cortical pyramidal neurons and glial cells, are P-Akt positive. (E-H) At E14.5, dividing radial glial cells show a similar pattern of immunostaining for P-Akt and P-Vimentin 4A4. (I-K) High-magnification images of the areas delineated by white boxes in F-H demonstrate that P-Akt activity (marked by arrowheads) is not restricted to the P-Vimentin 4A4-positive-staining, M phase cells in the ventricular zone. The arrows indicate an example of a P-Akt-positive, P-Vimentin 4A4-negative cell. Nuclei are labeled with Hoechst. Scale bars, 50μm. MZ: marginal zone, CP: cortical plate, SP: subplate, IZ: intermediate zone, SVZ: subventricular zone, VZ: ventricular zone.

Figure 4. Active Akt signaling in the developing cortex is enriched in apical progenitor cells and the cortical plate

(A) Immunohistochemistry of cortical sections at embryonic day (E)10.5 reveals Akt activity as assessed by pan-phospho(P)-Akt immunostaining (red) in the cortical plate and ventricular zone. (B-D) Higher magnification images of ventricular zone at E10.5 are shown. Overlay of P-Akt with P-Vimentin 4A4 (green) shows that dividing radial glial cells, which generate cortical pyramidal neurons and glial cells, are P-Akt positive. (E-H) At E14.5, dividing radial glial cells show a similar pattern of immunostaining for P-Akt and P-Vimentin 4A4. (I-K) High-magnification images of the areas delineated by white boxes in F-H demonstrate that P-Akt activity (marked by arrowheads) is not restricted to the P-Vimentin 4A4-positive-staining, M phase cells in the ventricular zone. The arrows indicate an example of a P-Akt-positive, P-Vimentin 4A4-negative cell. Nuclei are labeled with Hoechst. Scale bars, 50μm. MZ: marginal zone, CP: cortical plate, SP: subplate, IZ: intermediate zone, SVZ: subventricular zone, VZ: ventricular zone.