Exome-wide rare variant analysis identifies TUBA4A mutations associated with familial ALS

Abstract

Exome sequencing is an effective strategy for identifying human disease genes. However, this methodology is difficult in late-onset diseases where limited availability of DNA from informative family members prohibits comprehensive segregation analysis. To overcome this limitation, we performed an exome-wide rare variant burden analysis of 363 index cases with familial ALS (FALS). The results revealed an excess of patient variants within TUBA4A, the gene encoding the Tubulin, Alpha 4A protein. Analysis of a further 272 FALS cases and 5,510 internal controls confirmed the overrepresentation as statistically significant and replicable. Functional analyses revealed that TUBA4A mutants destabilize the microtubule network, diminishing its repolymerization capability. These results further emphasize the role of cytoskeletal defects in ALS and demonstrate the power of gene-based rare variant analyses in situations where causal genes cannot be identified through traditional segregation analysis.

Figures

Figure 1. Rare Variant Analysis Identifies TUBA4A Mutations in FALS

(A) Manhattan plot displaying permutation-based corrected p values generated by a rare variant analysis of FALS. The dotted line represents a Pcorrected = 0.05. The points in yellow, red, and black for TUBA4A denote the p value for the discovery, replication, and joint analyses, respectively. (B) The evolutionary conservation of TUBA4A mutations is displayed. Mutated residues are shown in red. (C) The interactions of bovine TUBA1B and TUBB2B with the rat kinesin KIF5B (Protein Data Bank accession number 4ATX) are shown using the PyMOL Molecular Graphics System (v.1.5.0.5). All relevant residues are identical between bovine TUBA1B and human TUBA4A. Light blue, α-tubulin; green, β-tubulin; orange, kinesin; red, mutant or deleted residues identified in this study; spheres, residues involved in interprotein interactions. As shown, the C-terminal region interacts with both β-tubulin and the kinesin motor domain.

Figure 2. Mutant TUBA4A Alters Microtubule Polymerization

(A) PMNs transfected with HA-tagged TUBA4A constructs. (B) HEK293 cells expressing wild-type or mutant HA-TUBA4A were immunostained for ubiquitin. Ubiquitin-positive aggregates were visible only in the HA-TUBA4AW407X-transfected cells. (C) Nondenaturing (top) and SDS-gel electrophoresis (bottom) of HA-TUBA4A constructs after in vitro translation. (D) The intensity of the bands representing the α-/β-tubulin dimers was quantified. Bars represent mean and SEM (one-way ANOVA and Dunnett’s post hoc test, n = 6, *p < 0.05, ***p < 0.001). (E) Representative images of primary astroglial cells transfected with the HA-TUBA4A constructs. (F) The ratio between HA-TUBA4A protein distributions in filaments versus its overall levels was measured and distributed into four categories from mainly filamentous to mainly diffuse. The frequency of each phenotype was compared between the wild-type and mutant proteins (Kolmogorov-Smirnov test; n = 75–90 cells per condition from 3 experiments, **p < 0.01, ***p < 0.001). From left to right, bars represent the relative frequencies of the normal, mild, moderate, and severe phenotypes. Red, HA; green, ubiquitin; blue, DAPI. White boxes indicate the regions enlarged in the insets. Scale bars, 10 μm.

Figure 3. Mutant TUBA4A Alters Microtubule Dynamics

(A and C) Representative images of nocodazole-treated COS7 cells expressing wild-type and mutant TUBA4A (A, TUBA4AR320C; C, TUBA4AW407X). Cells were allowed to recover in nocodazole-free medium for 2.5, 5, 10, or 15 min prior to fixation. Red, HA; green, TUBA4A; blue, DAPI. White boxes indicate the regions enlarged in the insets. Scale bars, 10 μm. (B and D) The percentage of cells containing microtubules positive for the HA-tagged or the endogenous TUBA4A protein was quantified from more than 100 cells per condition from 3 independent experiments and compared between wild-type and mutant TUBA4A (B, TUBA4AR320C; D, TUBA4AW407X). Points represent mean and SEM (one-way ANOVA and Dunnett’s post hoc test; *p < 0.05, **p < 0.01, ***p < 0.001).

Figure 4. Mutant TUBA4A Destabilizes the Microtubule Network

(A) PMNs transfected with wild-type or mutant TUBA4A constructs were permeabilzed with 0.1% Triton X-100 for 5 min prior to fixation. (B and C) The fluorescence intensity of (B) HA-TUBA4A and (C) β-tubulin was quantified and compared. Bars represent mean and SEM (Kruskal-Wallis test, n = 23–34 cells per condition from 3 independent experiments, *p