Amyotrophic lateral sclerosis-related VAPB P56S mutation differentially affects the function and survival of corticospinal and spinal motor neurons

Abstract

The substitution of Proline with Serine at residue 56 (P56S) of vesicle-associated membrane protein-associated protein B (VAPB) has been linked to an atypical autosomal dominant form of familial amyotrophic lateral sclerosis 8 (ALS8). To investigate the pathogenic mechanism of P56S VAPB in ALS, we generated transgenic (Tg) mice that heterologously express human wild-type (WT) and P56S VAPB under the control of a pan-neuronal promoter Thy1.2. While WT VAPB Tg mice did not exhibit any overt motor behavioral phenotypes, P56S VAPB Tg mice developed progressive hyperactivities and other motor abnormalities. VAPB protein was accumulated as large punctate in the soma and proximal dendrites of both corticospinal motor neurons (CSMNs) and spinal motor neurons (SMNs) in P56S VAPB Tg mice. Concomitantly, a significant increase of endoplasmic reticulum stress and unfolded protein response and the resulting up-regulation of pro-apoptotic factor CCAAT/enhancer-binding protein homologous protein expression were observed in the CSMNs and SMNs of P56S VAPB Tg mice. However, only a progressive loss of CSMNs but not SMNs was found in P56S VAPB Tg mice. In SMNs, P56S VAPB promoted a rather selective translocation of VAPB protein onto the postsynaptic site of C-boutons that altered the morphology of C-boutons and impaired the spontaneous rhythmic discharges of SMNs. Therefore, these findings provide new pathophysiological mechanisms of P56S VAPB that differentially affect the function and survival of CSMNs and SMNs in ALS8.

Figure 1.

Generation of P56S VAPB Tg mice. (A) A schematic outline of the construct used to generate Tg mice. Human WT or P56S mutant VAPB cDNA followed by an internal ribosome entry site (IRES) and green fluorescent protein (GFP) expression cassette is cloned into the mouse Thy1.2 expression vector at the XhoI site. (B) qRT-PCR analysis for VAPB Tg expression in different mouse lines. Three and more mice were analyzed per genotype per line. WT B3 and P56S D3 VAPB lines have the highest and comparable expression level of Tg VAPB. (C) Western blots show the distribution of VAPB protein in the cerebral cortex (CX), spinal cord (SC), striatum (ST), hippocampus (HP) and cerebellum (CB) of 3-month-old nTg and littermate P56S-VAPB or WT-VAPB Tg mice. β-tubulin staining was used as the loading control. (D) Western blots show differential accumulation of WT and P56S VAPB protein in Triton X-100 soluble (TX-sol) and insoluble (TX-insol) fractions of brain extracts from different mouse lines. (E) Representative images show VAPB (red) expression in the CSMNs (top panels) and SMNs (bottom panels) of 3-month-old nTg, WT-VAPB and P56S-VAPB Tg mice. Five and more mice per genotype and more than 10 sections per mouse were examined. Arrows point to neurons with excessive aggregation of VAPB protein. Scale bar: 10 μm. (F) Representative images show ubiquitin (ubi, red) and CTIP2 (blue) staining in the CSMNs of 3-month-old nTg, WT-VAPB and P56S-VAPB Tg mice. Five and more mice per genotype and more than 10 sections per mouse were examined. Arrows point to neurons with excessive aggregation of ubiquitin. Scale bar: 20 μm. (G) Representative images show p62 (red) and CTIP2 (blue) staining in the CSMNs of 3-month-old nTg, WT-VAPB and P56S-VAPB Tg mice. Five and more mice per genotype and more than 10 sections per mouse were examined. Arrows point to neurons with excessive aggregation of p62. Scale bar: 20 μm. (H) Western blot analysis of p62 expression in the cerebral cortex of 3-month-old nTg, WT-VAPB and P56S-VAPB Tg mice. β-Actin was used as the loading control. Bar graph shows the quantification of p62 expression. Three mice were used for each genotype. Data are presented as mean ± SEM; ***P < 0.001. (I) Representative images show p62 (red) and VAPB (blue) staining in a CSMN of 3-month-old P56S-VAPB Tg mice. Five and more mice per genotype and more than 10 sections per mouse were examined. Scale bar: 10 μm.

Figure 2.

P56S VAPB Tg mice develop abnormal motor phenotypes. (A) The body weights of male nTg (n = 15), WT VAPB Tg (n = 12) and P56S VAPB Tg (n = 12) mice are measured at 2, 6, 12, 15 and 18 months of age. Data are presented as mean ± SEM; *P < 0.05. (B and C) Open-field tests show the ambulatory (B) and rearing activities (C) of nTg and VAPB Tg mice at different age (n ≥ 10 per genotype per time point). Data are presented as mean ± SEM; ***P < 0.001. (D) Longitudinal rotarod tests reveal the latency to fall of nTg and VAPB Tg mice from the accelerating rotatory rod at different age (n ≥ 10 per genotype per time point). Data are presented as mean ± SEM; *P < 0.05, **P < 0.01. (E and F) In gait analysis, the average stride length and time of nTg (n = 15), WT VAPB Tg mice (n = 12) and P56S VAPB Tg mice (n = 12) tested at 2, 6, 12, 15 and 18 months of age. Data are presented as mean ± SEM; ***P < 0.001.

Figure 3.

P56S VAPB Tg mice develop progressive loss of CSMNs. (A) Unbiased stereological estimation of the number of Nissl-stained layer V large pyramidal neurons in the motor cortex of nTg, WT and P56S VAPB Tg mice at 3, 10 and 18 months of age (n ≥ 3 per genotype per time point). Data are presented as mean ± SEM; *P < 0.05, **P < 0.01, ***P < 0.001. (B) Unbiased stereological estimation of the number of Nissl-stained SMNs in the lumber spinal cord of nTg and P56S VAPB Tg mice (n ≥ 3 per genotype per time point). Data are presented as mean ± SEM. (C) Representative images show CTIP2 staining in the motor cortex of 18-month-old nTg and P56S VAPB Tg mice. Three and more mice per genotype and more than 10 sections per mouse were examined. Scale bar: 500 μm. (D) Unbiased stereological estimation of the number of CTIP2-positive neurons in the layer V motor cortex of nTg and P56S VAPB Tg mice at 3, 12 and 18 months of age (n ≥ 3 per genotype per time point). Data are presented as mean ± SEM; **P < 0.01, ***P < 0.001. (E) Unbiased stereological estimation of the number of CTIP2-positive neurons in layer V posterior cortex of nTg and P56S VAPB Tg mice (n ≥ 3 per genotype per time point). Data are presented as mean ± SEM. (F) Representative images show GFP signals (green) in the motor cortex and thoracic spinal cord of 12-month-old WT and P56S VAPB Tg mice. The sections were counterstained with Topro3 (blue). The layer V motor cortex was marked by ‘-V-’. The CST was pointed by an arrow. Three and more mice per genotype and more than three sections per mouse were examined. Scale bars: 500 μm (top panel), 200 μm (bottom panel).

Figure 4.

P56S VAPB inclusion causes ER stress in the CSMNs of P56S VAPB Tg mice. (A) Representative images show BiP (red) and VAPB (blue) staining in the CSMNs of 3-month-old nTg, WT-VAPB and P56S-VAPB Tg mice. Five and more mice per genotype and more than 10 sections per mouse were examined. Arrowheads point to neurons with excessive accumulation of BiP and VAPB. Asterisks point to neurons with less VAPB accumulation. Scale bar: 20 μm. (B) Representative images show GFP (green), PDI (red) and VAPB (blue) staining in the CSMNs of 3-month-old P56S-VAPB Tg mice. Five and more mice per genotype and more than 10 sections per mouse were examined. Arrowheads point to neurons with excessive accumulation of PDI and VAPB. Scale bar: 20 μm.

Figure 5.

P56S VAPB inclusion leads to up-regulation of CHOP-mediated UPR in the CSMNs of P56S VAPB Tg mice. (A) Representative images show CHOP (red) and VAPB (blue) staining in the CSMNs of 3-month-old nTg and P56S-VAPB Tg mice. Five and more mice per genotype and more than 10 sections per mouse were examined. Arrowheads point to neurons with excessive accumulation of CHOP in the nucleus. Scale bar: 15 μm. (B) Representative images show CHOP (red) and CTIP2 (blue) staining in the CSMNs of 3-month-old nTg and P56S-VAPB Tg mice. Five and more mice per genotype and more than 10 sections per mouse were examined. Arrowheads point to neurons with excessive accumulation of CHOP in the nucleus, where the intensity of CTIP2 staining is decreased. Scale bar: 30 μm. (C, D and F) Western blots show ATF4 (C), CHOP (D) and CTIP2 (F) expression in the nuclear extract of cerebral cortices from 3-month-old nTg, WT-VAPB and P56S-VAPB Tg mice. Anti-TATA-binding protein (TBP) was used as the loading control. Bar graphs show the quantification of ATF4, CHOP and CTIP2 expression. Three mice were used for each genotype. Data are presented as mean ± SEM; *P < 0.05, **P < 0.01. (E) Western blot analysis of Bcl-2 expression in the cerebral cortex of 3-month-old nTg, WT-VAPB and P56S-VAPB Tg mice. β-actin was used as the loading control. Bar graph shows the quantification of Bcl-2 expression. Three mice were used for each genotype. Data are presented as mean ± SEM; **P < 0.01.

Figure 6.

Abnormal translocation of P56S VAPB onto the postsynaptic site of C-boutons alters the morphology of C-boutons in SMNs. (A) Representative images of VAPB (red) and synaptophysin (blue) staining in the SMNs of 3-month-old P56S VAPB Tg mice. Five and more mice per genotype and more than 10 sections per mouse were examined. Scale bar: 10 μm. (B) Representative image shows VAPB located at the postsynaptic membrane of C-boutons in SMNs of 3-month-old P56S VAPB Tg mice. Three mice more than 10 sections per mouse were examined. Scale bar: 100 nm. (C) Representative images show CHAT (green) and VAPB (red) staining in the SMNs of 3-month-old nTg, WT and P56S VAPB Tg mice. Three and more mice per genotype and more than 10 sections per mouse were examined. Arrow points to a C-bouton with CHAT and VAPB staining juxtaposing each other. Scale bar: 10 μm. (D) Representative image shows CHAT staining after 3D deconvolution treatment. Three and more mice per genotype and more than 10 sections per mouse were examined. Scale bar: 10 μm. (E and F) Bar graphs show the average volume (E) and length (F) of C-boutons in the SMNs of 3-month-old P56S VAPB Tg mice. Three mice were used for each genotype. 86, 83 and 92 C-bountons were analyzed for 3-month-old nTg, WT and P56S VAPB Tg mice, respectively. Data are presented as mean ± SEM; ***P < 0.001.

Figure 7.

Intensity of C-bouton-mediated long-lasting discharges in VRs was reduced in isolated spinal cords from P56S VAPB Tg mice. (A) Representative traces show the discharges recorded activity from the L5 VR in isolated spinal cords from P10 nTg and P56S VAPB Tg pups. Seven and more mice per genotype were examined. The discharges were evoked by stimulating the hemisected cord near the central canal to activate the neurons giving rise to C-boutons. (B) Line graph depicts the integrated, averaged and normalized intensity of discharges from L5 VR in P10–P11 nTg and P56S VAPB Tg pups. Data are presented as mean ± SEM. At stimulus intensities of 10–20 μA *P < 0.05 and at 40–500 µA **P < 0.01 (n = 7, nTg versus P56S VAPB Tg). (C) Atropine blocked the long-lasing discharges in the L5 VR evoked by repetitive stimulation of the medial spinal cord from P10 to P11 nTg and P56S VAPB Tg pups. Lines are fitted to the means of the integrated and averaged discharges recorded from the L5 VR (n = 4 experiments). (D and E) Representative traces (D) and quantification (E) of the intensity of the integrated, averaged discharge in L2 VR in the disinhibited spinal cord from P10 nTg and P56S VAPB Tg pups. Data are presented as mean ± SEM; ***P < 0.001 (n = 3 experiments).