TDP-43 expression in mouse models of amyotrophic lateral sclerosis and spinal muscular atrophy

Abstract

Background: Redistribution of nuclear TAR DNA binding protein 43 (TDP-43) to the cytoplasm and ubiquitinated inclusions of spinal motor neurons and glial cells is characteristic of amyotrophic lateral sclerosis (ALS) pathology. Recent evidence suggests that TDP-43 pathology is common to sporadic ALS and familial ALS without SOD1 mutation, but not SOD1-related fALS cases. Furthermore, it remains unclear whether TDP-43 abnormalities occur in non-ALS forms of motor neuron disease. Here, we characterise TDP-43 localisation, expression levels and post-translational modifications in mouse models of ALS and spinal muscular atrophy (SMA).

Results: TDP-43 mislocalisation to ubiquitinated inclusions or cytoplasm was notably lacking in anterior horn cells from transgenic mutant SOD1G93A mice. In addition, abnormally phosphorylated or truncated TDP-43 species were not detected in fractionated ALS mouse spinal cord or brain. Despite partial colocalisation of TDP-43 with SMN, depletion of SMN- and coilin-positive Cajal bodies in motor neurons of affected SMA mice did not alter nuclear TDP-43 distribution, expression or biochemistry in spinal cords.

Conclusion: These results emphasise that TDP-43 pathology characteristic of human sporadic ALS is not a core component of the neurodegenerative mechanisms caused by SOD1 mutation or SMN deficiency in mouse models of ALS and SMA, respectively.

Figure 1

Spinal cord sections of transgenic SOD1^G93A mice (a-c, f) show anterior horn cell degeneration, hyaline aggregates (arrow) and vacuolation (a) not present in corresponding wild-type (d). Amorphous cytoplasmic aggregates are ubiquitinated and occasionally extend into proximal processes (b, c). TDP-43 remains located in the nucleus of anterior horn cells and shows no distinct aggregates (f). The appearances are essentially indistinguishable from wild-type (e) apart from possible slightly weaker TDP-43 signal in presumed reactive glia (f). Samples from transgenic Smn^-/-;SMN2;SMNΔ7 spinal cord (g-i) show residual motor neurons (g) without ubiquitin positivity (h) and preserved nuclear TDP-43 signal (i). Scale bar in all images = 20 μm.

Figure 2

(a) Wild-type mouse spinal cord sections stained for TDP-43 and SMN. Large alpha motor neurons show diffuse cytoplasmic SMN stain and punctate nuclear SMN, Cajal bodies. TDP-43 shows only nuclear and no cytoplasmic staining. (b) Representative merged TDP-43 and SMN stains for all genotypes. Note absence of Cajal bodies in the SMA mouse. The pattern of TDP-43 immunoreactivity is unaltered in SOD1^G93A and SMA mice. Scale bar = 20 μm.

Figure 3

Quantification of Cajal bodies per motor neuron nucleus for each genotype. Unpaired two-tailed t-test, p = 0.0004 (***) and p = 0.002 (**). Error bars represent standard deviation.

Figure 4

(a) Wild-type mouse spinal cord sections stained for coilin and SMN. Coilin co-localises with SMN in the motor neuron nucleus. (b) Coilin immunohistochemistry confirms reduction of Cajal bodies in the SMA mouse, but equal distribution between SOD1^G93A and WT animals. Scale bar = 20 μm.

Figure 5

Quantification of neuronal TDP-43 nuclear speckles in spinal cord sections. Approximately 0.9 μm thick confocal section through representative motor neuron nucleus (a) and automatic setting of brightness and contrast in Image J (b). No significant difference in TDP-43 speckle number is obvious between genotypes (c).

Figure 6

TDP-43 protein and transcript levels in CNS of transgenic SOD1 G93A mice compared with age matched controls (a) Immunoblot analysis of brain and spinal cord extracts from wild-type (WT) and transgenic mice (SOD1^G93A) at 4 months of age. Open arrow, putative dimeric hSOD1 species. (b) Phosphatase treatment of representative brain urea extracts. (c) Quantitative RT-PCR analysis of brain regions and spinal cords from mice.

Figure 7

TDP-43 protein and transcript levels in CNS of SMA mice (Smn^-/-;SMN2;SMNΔ7) and age-matched littermates (a) Immunoblot analysis of brain and spinal cord extracts from wild-type (WT) and affected (SMA) mice at 13 days of age. (b) Quantitative RT-PCR analysis of brain regions and spinal cords from mice.