Cognitive decline typical of frontotemporal lobar degeneration in transgenic mice expressing the 25-kDa C-terminal fragment of TDP-43

Abstract

Transactive response DNA-binding protein 43 (TDP-43) is the pathological signature protein in several neurodegenerative disorders, including the majority of frontotemporal lobar degeneration cases (FTLD-TDP), motor neuron disease, and amyotrophic lateral sclerosis. Pathological TDP-43 is mislocalized from its nuclear location to the cytoplasm, where it accumulates and is proteolytically cleaved to form C-terminal fragments. Although the 25-kDa C-terminal fragment of TDP-43 (TDP-25) accumulates in affected brain regions, its role in the disease pathogenesis remains elusive. To address this problem, we have generated a novel transgenic mouse that selectively expresses TDP-25 in neurons. We show that transgenic mice expressing TDP-25 develop cognitive deficits associated with the build-up of soluble TDP-25. These cognitive deficits are independent of TDP-43-positive inclusions and occur without overt neurodegeneration. Additionally, we show that the expression of TDP-25 is sufficient to alter the processing of endogenous full-length TDP-43. These studies represent the first in vivo demonstration of a pathological role for TDP-25 and strongly suggest that the onset of cognitive deficits in TDP-43 proteinopathies is independent of TDP-43 inclusions. These data provide a framework for understanding the molecular mechanisms underlying the onset of cognitive deficits in FTLD-TDP and other TDP-43 proteinopathies; thus, the TDP-25 transgenic mice represent a unique tool to reach this goal.

Figure 1

Generation of the TgTDP-25 mice. A: Schematic representation of the construct used to generate the TgTDP-25 mice. The entire mouse Thy1.2 genomic sequence is shown with exons depicted as boxes, and noncoding sequences as thin lines. TDP-25 (gray box) was cloned into exon 3 of the murine Thy1.2 gene. The expression cassette includes the Thy1.2 mRNA polyadenylation sequence. B–D: Representative Western blots and quantitative analyses of proteins extracted from the brains of 2-month-old TgTDP-25 mice (lines B, C, and F) and NonTg mice using the low-salt buffer. The arrow points to the transgene, and the arrowhead points to the endogenous, full-length TDP-43. E: Real-time PCR experiments indicated that the mRNA levels of the transgene did not change between 2 and 6 months of age. F: The graph shows a standard curve generated by real-time PCR by measuring the Ct values of known copy numbers of the transgene added to NonTg mouse DNA. The Ct values of the founders DNA was then plotted into the standard curve (dotted lines). G and H: Representative Western blots and quantitative analyses of proteins extracted from the indicated tissues. I and J: Representative Western blots and quantitative analyses of proteins extracted from different brain regions of TgTDP-25(B) mice (n = 4). Brn, brain; Cb, cerebellum; Cx, cortex; Hp, hippocampus; Hrt, heart; Kdn, kidneys; Lng, lungs; Lvr, liver; Msc, muscle; Ob, olfactory bulb, Sc, spinal cord; SP, spinal cord; Spl, spleen. Blots in panels B, G, and I were probed with an anti–TDP-43 polyclonal antibody from ProteinTech. β-Actin was used as a loading control, and quantifications of the Western blots were done by normalizing the protein of interest to β-actin. Data are presented as means ± SEM and analyzed by one-way analysis of variance.

Figure 2

The TgTDP-25 mice develop cognitive deficits. Behavioral testing was conducted in 2- and 6-month-old NonTg, TgTDP-25(B) and TgTDP-25(F) mice (n = 14/genotype/time point). A–D: The open-field activity test was conducted to measure spontaneous activity and anxiety. No statistically significant differences were found among the three groups (at any of the ages analyzed) in the distance covered during the exploration time (A) or the speed of exploration (B), indicating that gross motor function was intact in both lines of TgTDP-25 mice. Also, no differences among the groups were found when measuring the time spent in the periphery and center of the arena (C and D, respectively), indicating that the TgTDP-25 mice had no detectable anxiety defects. E: To measure motor coordination, we used the accelerating rotarod and found no statistically significant changes among the three groups analyzed. F: T-maze data show that at 2 months of age, working memory was similar among the three groups of mice. In contrast, 6-month-old TgTDP-25(B) mice performed significantly worse compared to NonTg and TgTDP-25(F) mice. G: Novel object recognition tests, a behavioral task highly dependent on the cortex, show that at 2 months of age all three groups of mice performed similarly to each other. At 6 months of age, however, both lines of TDP-25 transgenic mice were significantly impaired compared to NonTg mice as they spent less time exploring the new object compared to the NonTg mice. *P < 0.01, **P < 0.001. Data are presented as means ± SEM, and each time point was independently analyzed by one-way analysis of variance.

Figure 3

Increased soluble TDP-25 levels in 6-month-old TgTDP-25 mice. A: Representative Western blots of proteins extracted from the brains of 2- and 6-month-old transgenic and NonTg mice, in low- and high-salt buffers. Blots were probed with an anti–TDP-43 polyclonal antibody from ProteinTech. Two different exposure times are presented to show the less abundant, low molecular weight bands. B: Quantitative analyses of the blots from low-salt extraction show that the levels of full-length TDP-43 were similar between the TgTDP-25(B) and TgTDP-25(F) mice and did not change as a function of age. C: In contrast, we found the steady-state levels of TDP-25 (arrow) significantly increased as a function of age in both the TgTDP-25(B) and TgTDP-25(F) mice. D: Quantitative analyses of the blots from high-salt extracts show that the levels of full-length TDP-43 were similar between the TgTDP-25(B) and TgTDP-25(F) mice and did not change as a function of age. E: TDP-25 levels increased as a function of age in the high-salt fraction in both TgTDP-25 lines, suggesting a change in aggregation of this fragment as a function of age. n = 6/genotype/age. *P < 0.05, **P < 0.01. β-Actin was used as a loading control, and quantifications of the Western blots were done by normalizing the protein of interest to β-actin. Data are presented as means ± SEM and analyzed by two-way analysis of variance, with genotype and age as independent variables.

Figure 4

Absence of ubiquitin-positive inclusions and cell death in TgTDP-25 mice. A: Representative hippocampal sections from 6-month-old NonTg and TgTDP-25(B) mice stained with a phospho-specific TDP-43 antibody (n = 6/genotype). B: Cortical sections from 6-month-old TgTDP-25(B) mice were double stained with a C-terminal anti-TDP-43 antibody and with an anti-ubiquitin antibody. The overlay shows that at this age, the TgTDP-25(B) mice have no ubiquitin-positive TDP-43 inclusions. C: Representative cortical and hippocampal sections from 6-month-old NonTg and TgTDP-25(B) mice stained with Fluoro-Jade. D: Quantitative analysis of the Fluoro-Jade–positive cells, conducted as described in Materials and Methods, showed no statistically significant difference between the two groups of mice (n = 6/genotype). Data are presented as means ± SEM and analyzed by Student's t-test.

Figure 5

TDP-25 expression leads to altered processing of full-length endogenous TDP-43. A: Representative Western blots extracted from 2- and 6-month-old TgTDP-25 and NonTg mice. Different exposures are presented to show the less abundant, low molecular TDP-43 fragments. Proteins were extracted from the cytosolic and nuclear fraction, as depicted in the figure. TDP-43 blots were probed with an anti-TDP-43 polyclonal antibody from ProteinTech. β-Actin and LaminA were used as loading controls for the cytosolic and nuclear fractions, respectively. B: As expected, quantitation of the full-length TDP-43 band (panel A, top right blot, black arrow), shows that the levels of TDP-43 are significantly higher in the nucleus compared to the cytoplasm in both transgenic and NonTg mice. Additionally, the levels of TDP-43 were not different in the cytosolic or nuclear fractions between TgTDP-25(B) and NonTg mice. C: Quantitative analysis of the three bands of ∼39-kDa (panel A, top right blot, white arrow). Notably, these bands were not present in 2-month-old mice, irrespective of the genotype. In the nuclear fraction of 6-month-old mice, the levels of these fragments were significantly higher in NonTg mice compared to TgTDP-25(B) mice. In the TgTDP-25(B) mice, however, there was a striking increase of an ∼35-kDa band in the cytosolic fraction compared to NonTg mice, suggesting that in the TgTDP-25(B) mice, there is a possible redistribution of the ∼35-kDa bands from the nucleus (where they are more abundant in NonTg mice) to the cytosol. D: Quantitative analysis of the 25-kDa bands (panel A, bottom right blot, arrowhead) shows that the TDP-25 transgene is also present in the nuclear fraction, despite the lack of putative nuclear localization signals. *P < 0.05, **P < 0.01. β-Actin and LaminA were used as loading controls for cytosolic and nuclear fractions, respectively. Quantifications of the Western blots were done by normalizing the fragment of interest to β-actin of LaminA. Data are presented as means ± SEM and analyzed by two-way analysis of variance, with genotype and age as independent variables.