Elevated expression of TDP-43 in the forebrain of mice is sufficient to cause neurological and pathological phenotypes mimicking FTLD-U

Abstract

TDP-43 is a multifunctional DNA/RNA-binding factor that has been implicated in the regulation of neuronal plasticity. TDP-43 has also been identified as the major constituent of the neuronal cytoplasmic inclusions (NCIs) that are characteristic of a range of neurodegenerative diseases, including the frontotemporal lobar degeneration with ubiquitin(+) inclusions (FTLD-U) and amyotrophic lateral sclerosis (ALS). We have generated a FTLD-U mouse model (CaMKII-TDP-43 Tg) in which TDP-43 is transgenically overexpressed in the forebrain resulting in phenotypic characteristics mimicking those of FTLD-U. In particular, the transgenic (Tg) mice exhibit impaired learning/memory, progressive motor dysfunction, and hippocampal atrophy. The cognitive and motor impairments are accompanied by reduced levels of the neuronal regulators phospho-extracellular signal-regulated kinase and phosphorylated cAMP response element-binding protein and increased levels of gliosis in the brains of the Tg mice. Moreover, cells with TDP-43(+), ubiquitin(+) NCIs and TDP-43-deleted nuclei appear in the Tg mouse brains in an age-dependent manner. Our data provide direct evidence that increased levels of TDP-43 protein in the forebrain is sufficient to lead to the formation of TDP-43(+), ubiquitin(+) NCIs and neurodegeneration. This FTLD-U mouse model should be valuable for the mechanistic analysis of the role of TDP-43 in the pathogenesis of FTLD-U and for the design of effective therapeutic approaches of the disease.

Figure 1.

Generation and characterization of CaMKII-TDP-43 Tg mice. (A) Physical map of the CaMKII-TDP-43 fragment for pronuclei injection. The orientation of transcription is indicated by the arrow. The positions of the short hybrid intron derived from an adenovirus splice donor, an immunoglobulin G splice acceptor, and the SV40 poly(A) addition sequence (pA) are indicated. The approximate locations of the Southern blotting and PCR probes are also indicated. Tg mice were identified by the presence of the 4.4-kb KpnI fragment on the Southern blot and the 523-bp PCR band on gel. The restriction sites on the map are as follows: K, KpnI; E, EcoRV; N, NotI; S, SfiI. (B) Genotyping of the Tg mice. The data from PCR (top) and Southern blotting (bottom) analysis of the tail DNAs are exemplified. NT represents the nontransgenic samples. (C) Western blotting of the protein extracts from the hippocampus, cortex, cerebellum, and spinal cord of the WT, NT, and Tg mice, respectively. (D) In situ hybridization patterns of TDP-43 transcripts in the brains of WT and TDP-43 Tg mice (Tg). Bars, 500 µm. (E) Immunostaining patterns of TDP-43 protein in the brains of WT and Tg mice. CA1, CA1 layer; CA3, CA3 layer; DG, dentate gyrus. Bars, 200 µm. Results in B–E are representative of three independent experiments.

Figure 2.

Behavioral performances of CaMKII-TDP-43 Tg mice. (A) Water maze tests of 2-mo-old WT and TDP-43 Tg mice. The learning/memory capabilities are expressed as the latencies exhibited in six consecutive sessions of the test. Results represent the mean ± SEM of three independent experiments (n = 20 mice/group). (B) Comparison of the cognitive functions of 2-mo-old WT and Tg mice in the fear-conditioning task. (C) Locomotor activity test of 2-mo-old WT and Tg mice. Results in B and C represent the mean ± SEM of three independent experiments and (n = 16 mice/group). (D) Abnormal limb-clasping of a 6-mo-old Tg mouse in comparison to a WT mouse when suspended by their tails. Results are representative of five independent experiments. (E) Rotarod tests of WT and Tg mice. The time until drop from the rotating rod (20 r.p.m.) are shown for mice at 2, 4, and 6 mo of age. (F) Performance of mice in 1-h novel object recognition tests. The results represent the mean ± SEM of three independent experiments (n = 10 mice/group). *, P < 0.05.

Figure 3.

Electrophysiology study of the CaMKII-TDP-43 Tg mice. (A) Attenuated LTP in the hippocampus of 2-mo-old CaMKII-TDP-43 Tg mice. LTP was induced by strong tetanus stimulation in the stratum radiatum layer of CA1. Results represent the mean ± SEM of three independent experiments (n = 18 mice/group). (B) Recording of altered mEPSCs in the primary hippocampal culture of CaMKII-TDP-43 Tg mice. Whole-cell voltage clamp recordings of cultured hippocampal neurons (12–15 DIV) from the WT and TDP-43 Tg mice were performed. Representative traces of the mEPSCs recorded from the neurons are shown on top. The frequencies, amplitudes, decay τ, and rise τ of the mEPSCs are shown in the histograms. *, P < 0.05; ***, P < 0.001. The scale bars are 50 pA and 1 s, respectively. Results represent the mean ± SEM of three independent (n = 10 mice/group).

Figure 4.

Alterations of the levels of learning/memory-associated proteins in CaMKII-TDP-43 Tg mice. (A) Western blot analysis of different proteins in extracts prepared from isolated cerebral cortex and hippocampus of 2-mo-old WT and Tg mice (n = 2 mice/group). Results are representative of three independent experiments. Quantitative analysis of the Western blot data are shown in the lower portion of A. (B) Immunostaining analysis of GAD67 expression and measurement of GABA release. The primary hippocampal neuron cultures of the WT (left) and TDP-43 Tg (Tg, right) mice were double-stained with anti- GAD67 (green) and anti–TDP-43 (red). Bars, 50 µm. The statistical comparison of the relative GAD67 intensities of GAD67⁺ cells of the WT and Tg mice is shown in the left histogram below the confocal images. The data represent the mean ± SEM of three independent experiments (n = 20 mice/group). Shown in the lower right histogram is the statistical comparison of the GABA levels of the WT and Tg mice (n = 5; P < 0.05).

Figure 5.

Neurodegeneration of the brains of CaMKII-TDP-43 Tg mice. (A) Representative immunofluorescent images showing GFAP staining (green) in the hippocampus and cortex (CX) of 2-mo-old TDP-43 Tg and WT mice. Nuclei were labeled by DAPI (blue). CA1, CA1 layer; CA3, CA3 layer; DG, dentate gyrus. Bars, 100 µm. (B) Representative Western blotting patterns of the urea-soluble fractions of the brain extracts from the cortexes and hippocampi of the WT mice and Tg^+/+ mice at 2 mo and 6 mo, respectively. The arrow points to the unmodified form of TDP-43. Results in A–C are representative of five independent experiments.

Figure 6.

Immunofluorescence staining analysis of TDP-43 distribution in the neurons of the mouse brains. (A) The brain sections of the WT and Tg mice were costained with anti–TDP-43 (red), anti-NeuN (green), and DAPI (blue). TDP-43⁺ NCIs are indicated by arrows. One neuron each in the Tg^+/+ and WT samples (boxed) was magnified in the lower right corners (n = 5 mice/group). Bars, 20 µm. (B) Representative immunostaining pattern of a 6-mo-old Tg^+/+ mouse brain section exhibiting neuronal cells with TDP-43 (green)-containing NCIs that are also positive for the anti-ubiquitin (Ub) staining (red, arrows). High magnification photos of one of the cells with TDP-43⁺, Ub⁺ NCIs are shown in the lower right corners . Bars, 20 µm.

Figure 7.

Neuronal loss and apoptosis in the brains of the Tg mice. (A) Reduction of the brain mass of the 6-mo-old Tg mice. Whole brains were dissected and weighed. n = 5; P < 0.05. (B) Loss of neurons in the cortexes of the Tg mice. Coronal brain sections from 6-mo-old Tg and WT mice were immunostained with anti-NeuN. The average number of the neurons in the cortexes of the Tg mice was compared with that of the WT mice (right; n = 5 for each group and P < 0.05). Bars, 100 µm. (C) TUNEL assay of the brains of 6-mo-old Tg mice. Apoptotic nuclei are shown in green (arrows) and DAPI staining in blue. (D) Western blots of total brain extracts (cortexes + hippocampi) of 6-mo-old WT and Tg mice showing levels of total and active caspase-3 (n = 5 mice/group). Bars, 20 µm.