Neuronal-specific overexpression of a mutant valosin-containing protein associated with IBMPFD promotes aberrant ubiquitin and TDP-43 accumulation and cognitive dysfunction in transgenic mice - PubMed (original) (raw)

Neuronal-specific overexpression of a mutant valosin-containing protein associated with IBMPFD promotes aberrant ubiquitin and TDP-43 accumulation and cognitive dysfunction in transgenic mice

Carlos J Rodriguez-Ortiz et al. Am J Pathol. 2013 Aug.

Abstract

Mutations in valosin-containing protein (VCP) cause a rare, autosomal dominant disease called inclusion body myopathy associated with Paget disease of bone and frontotemporal dementia (IBMPFD). One-third of patients with IBMPFD develop frontotemporal dementia, characterized by an extensive neurodegeneration in the frontal and temporal lobes. Neuropathologic hallmarks include nuclear and cytosolic inclusions positive to ubiquitin and transactive response DNA-binding protein 43 (TDP-43) in neurons and glial activation in affected regions. However, the pathogenic mechanisms by which mutant VCP triggers neurodegeneration remain unknown. Herein, we generated a mouse model selectively overexpressing a human mutant VCP in neurons to study pathogenic mechanisms of mutant VCP-mediated neurodegeneration and cognitive impairment. The overexpression of VCPA232E mutation in forebrain regions produced significant progressive impairments of cognitive function, including deficits in spatial memory, object recognition, and fear conditioning. Although overexpressed or endogenous VCP did not seem to focally aggregate inside neurons, TDP-43 and ubiquitin accumulated with age in transgenic mouse brains. TDP-43 was also found to co-localize with stress granules in the cytosolic compartment. Together with the appearance of high-molecular-weight TDP-43 in cytosolic fractions, these findings demonstrate the mislocalization and accumulation of abnormal TDP-43 in the cytosol of transgenic mice, which likely lead to an increase in cellular stress and cognitive impairment. Taken together, these results highlight an important pathologic link between VCP and cognition.

Copyright © 2013 American Society for Investigative Pathology. Published by Elsevier Inc. All rights reserved.

PubMed Disclaimer

Figures

Figure 1

Figure 1

Generation of Thy-VCPA232E Tg mice. A: The Thy1.2-VCPA232E transgene construct shows the location of the sequence amplified by PCR genotyping (arrows; 0.5kb) and the probe used in Southern blot analysis (2.5 kb). B: Southern blot analysis of founder lines A, B, and C. Asterisks indicate the 4- and 5.5-kb bands in Thy1.2-VCPA232E Tg mice. Bands above 6 kb are undigested multiple copies of the transgene that were incorporated into genomic DNA. C: PCR genotyping shows amplification of 500 bp (asterisk) by the primer set shown as arrows in A. D: Immunoblot analysis of VCP in the brain. Increased steady-state levels of VCP at 98 kDa in the brain were detected in founder lines A, B, and C. Actin was used as loading control.

Figure 2

Figure 2

Overexpression of VCPA232E causes age-dependent cognitive impairment. Six- and 12-month-old mice were trained in hippocampal- and cortical-dependent behavioral tasks. A and B: The higher-expressing VCPA232E B line showed significant impairment in Morris water maze acquisition at 6 and 12 months of age, whereas the lower-expressing A line was not different from wild-type animals. C and D: Both 12-month-old Tg groups presented cognitive impairments when tested 24 hours later as measured by escape latency but not as measured by platform crosses. E: Both VCPA232E-expressing lines performed poorly in the object recognition task at 12 months of age but not at 6 months of age. †P < 0.05 from 50% (chance level, shown as a horizontal line). F: Only the higher-expressing B line showed impairments at 6 and 12 months of age when contextual fear conditioning was assessed (n = 10 to 12 per group). ∗P < 0.05, ∗∗P < 0.01. Data are given as means ± SEM.

Figure 3

Figure 3

Thy-VCPA232E mice display TDP-43 protein accumulation in the cytoplasm that co-localized with the stress granule marker TiA-1. A and B: Fluorescence microscopy analysis of 20-month-old mice revealed TiA-1 (red)–positive staining on wild-type (A) cortical sections but to a lesser degree than on VCPA232E (B). D: TiA-1 co-localized with TDP-43 (green) deposits in the cytoplasm of Tg brains but not NT (C). E and F: DAPI staining (blue) was used as nuclear marker. G and H: Merge images of NT (G) and Tg (H). I: Quantification of 12- and 20-month-old sections showed significant increases in TiA-1 levels on the latter age. Data are given as means ± SEM. Scale bars: 10 μm. n = 5 to 6 per group. ∗P < 0.05.

Figure 4

Figure 4

Thy-VCPA232E mice show age-dependent cytosolic accumulation of polyubiquitinated proteins. A–H: VCPA232E cortical sections from 20-month-old mice showed increased levels of polyubiquitin proteins (red) (B) and cytoplasmic TDP-43 (green) accumulation (D) compared with NT mice (A and C). E and F: DAPI staining (blue) was used as nuclear marker. G and H: Merge images of NT (G) and Tg (H). I: Quantification of 12- and 20-month-old sections uncovered increased levels of polyubiquitinated proteins on the Thy-VCP brain. J: Immunoblot analysis showed age-dependent buildup of polyubiquitinated proteins. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as loading control. K: Polyubiquitinated proteins were accumulated in the cytosolic compartment of mutant VCP mice. Tubulin was used as loading control. L: Quantification of K. Scale bars: 10 μm. n = 5 to 6 per group. ∗P < 0.05, ∗∗P < 0.01. Data are given as means ± SEM.

Figure 5

Figure 5

A high-molecular-weight TDP-43 isoform accumulates in the cytoplasm of Thy-VCPA232E mice. A: Immunoblot analysis revealed a TDP-43 band of approximately 90 kDa only on mutant VCP mice at 6 and 12 months of age. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as loading control. B and C: Nuclear-cytosolic fractionation unveiled a Tg-specific TDP-43 high-molecular-weight (HMW) band in the cytosol compartment of 12- and 20-month-old animals. p84 and GAPDH were used as nuclear and cytoplasmic purity controls, respectively. D: A better resolution of the TDP-43 HMW band was obtained when samples were run in a 5% acrylamide gel. E and F: Quantification of the HMW TDP-43 band in the cytoplasmic fraction of animals aged 12 (E) and 20 (F) months (n = 6 to 7 per group). ∗P < 0.05, ∗∗P < 0.01. Arrowheads indicate HMW TDP-43 bands. Asterisks indicate full-length TDP-43 bands. Data are given as means ± SEM.

Figure 6

Figure 6

High-molecular-weight TDP-43 is present in different aggregation states. A and B: Sequential extraction of brain tissue showed high-molecular-weight TDP-43 (arrowheads) in detergent-soluble and detergent-insoluble fractions. C: High-molecular-weight ubiquitinated proteins were detected in LS and urea fractions (n = 6 per group).

Figure 7

Figure 7

High-molecular-weight TDP-43 interacts with VCP. A: Immunoprecipitation (IP) with a VCP antibody and immunoblot (IB) for TDP-43 uncovered an interaction between VCP and the Tg-specific TDP-43 high-molecular-weight isoform (arrowhead). B: High-molecular-weight TDP-43 is not affected by alkaline phosphatase treatment (arrowhead). C: Quantification of B (n = 4 per group). Asterisks indicate full-length TDP-43 bands. Data are given as means ± SEM.

Figure 8

Figure 8

SHSY-5Y neuroblastoma cells present TDP-43 cytosolic buildup when transfected with mutant VCP. Cells transfected with wild-type VCP showed TDP-43 (red)–positive staining mainly in the nucleus, with little to no cytoplasmic signal (A). Cells transfected with the common R155H (B) or severe A232E (C) VCP mutation also showed TDP-43 nuclear signal with small, but evident, TDP-43 deposits in the cytoplasm (arrowheads). D–F: Green fluorescence protein signal was used to identify transfected cells. G–I: DAPI staining (blue) was used as nuclear marker. J–L: Merge images of wild-type VCP, VCP(R155H), and VCP(A232E), respectively. M: TDP-43 granule size was larger in cells transfected with mutant VCP. Scale bars: 50 μm. ∗P < 0.05. Data are given as means ± SEM.

References

    1. Muller J.M., Meyer H.H., Ruhrberg C., Stamp G.W., Warren G., Shima D.T. The mouse p97 (CDC48) gene: genomic structure, definition of transcriptional regulatory sequences, gene expression, and characterization of a pseudogene. J Biol Chem. 1999;274:10154–10162. - PubMed
    1. Hirabayashi M., Inoue K., Tanaka K., Nakadate K., Ohsawa Y., Kamei Y., Popiel A.H., Sinohara A., Iwamatsu A., Kimura Y., Uchiyama Y., Hori S., Kakizuka A. VCP/p97 in abnormal protein aggregates, cytoplasmic vacuoles, and cell death, phenotypes relevant to neurodegeneration. Cell Death Differ. 2001;8:977–984. - PubMed
    1. Zalk R., Shoshan-Barmatz V. ATP-binding sites in brain p97/VCP (valosin-containing protein), a multifunctional AAA ATPase. Biochem J. 2003;374:473–480. - PMC - PubMed
    1. Wang Q., Song C., Li C.C. Hexamerization of p97-VCP is promoted by ATP binding to the D1 domain and required for ATPase and biological activities. Biochem Biophys Res Commun. 2003;300:253–260. - PubMed
    1. Wang Q., Song C., Li C.C. Molecular perspectives on p97-VCP: progress in understanding its structure and diverse biological functions. J Struct Biol. 2004;146:44–57. - PubMed

Publication types

MeSH terms

Substances

Supplementary concepts

Grants and funding

LinkOut - more resources