New murine Niemann-Pick type C models bearing a pseudoexon-generating mutation recapitulate the main neurobehavioural and molecular features of the disease

Abstract

Niemann-Pick disease type C (NPC) is a rare neurovisceral disease caused mainly by mutations in the NPC1 gene. This autosomal recessive lysosomal disorder is characterised by the defective lysosomal secretion of cholesterol and sphingolipids. No effective therapy exists for the disease. We previously described a deep intronic point mutation (c.1554-1009 G > A) in NPC1 that generated a pseudoexon, which could be corrected at the cellular level using antisense oligonucleotides. Here, we describe the generation of two mouse models bearing this mutation, one in homozygosity and the other in compound heterozygosity with the c.1920delG mutation. Both the homozygotes for the c.1554-1009 G > A mutation and the compound heterozygotes recapitulated the hallmarks of NPC. Lipid analysis revealed accumulation of cholesterol in the liver and sphingolipids in the brain, with both types of transgenic mice displaying tremor and ataxia at 7-8 weeks of age. Behavioural tests showed motor impairment, hyperactivity, reduced anxiety-like behaviour and impaired learning and memory performances, features consistent with those reported previously in NPC animal models and human patients. These mutant mice, the first NPC models bearing a pseudoexon-generating mutation, could be suitable for assessing the efficacy of specific splicing-targeted therapeutic strategies against NPC.

Figure 1. Splicing pattern generated by the c.1554-1009 G > A (Imagine) mutation.

(A) Scheme of part of the Npc1 Imagine allele, between exons 9 and 11, containing the human intron 9 with the mutation and the pseudoexon generated by the DNA change. (B) WT and alternative transcripts detected by RT-PCR (using the primers indicated by arrows) in the brain cortex (CX) and liver (LV) of WT and I/I mice.

Figure 2

Western blot analysis of NPC1 protein expression in the (A) liver and (B) brain cortex isolated from WT and Npc1^{imagine/imagine} mice. Tubulin was used as the loading control.

Figure 3

Brain lipid storage in (A, black) I/I and (B, black) I/P mice compared to (WT, white) wild-type mice. Data are expressed as mean ± SD. *p < 0.05, **p < 0.01 and ***p < 0.001. Cer, ceramide; SM, sphingomyelin; dhCer, dihydroceramide; dhSM, dihydrosphingomyelin; GlcCer, glucosylceramide; LacCer, lactosylceramide.

Figure 4

(A) Total hepatic cholesterol. Stacked bar chart showing quantification of esterified and unesterified cholesterol in WT and I/I or I/P mice. Data are shown as mean ± SD. **p < 0.01. (B) Histological analysis of liver tissue. Foamy macrophages in hepatocytes of I/I mice are indicated by black arrows. Neutrophil infiltration is marked with a black circle. Scale bar, 100 μm.

Figure 5. Immunohistochemical analysis of calbindin-D28k in the cerebellum.

Coronal cerebellar sections were obtained from (A–C) 8-week-old wild-type mice and (D–F) mutant mice. Purkinje cells were immunolabelled (arrows) using an antibody against calbindin-D28K. In mutant mice, these cells showed soma reduction (arrows) (D–F) compared to WT animals (A–C), as well as alterations in the dendritic arborisation pattern (arrow heads) (D–F vs A–C). Abbreviations: ML, molecular layer; PC, Purkinje cells; GL, granular layer; WM, white matter. Scale bar, 100 μm.

Figure 6. Changes in body weight in I/I and I/P mice compared to WT.

***p < 0.001.

Figure 7. Circadian activity.

Distance travelled (in cm) every 30 minutes for 24 hours by (A) I/I and (B) I/P animals compared to WT. Grey dashed box represents the active phase of the circadian cycle. Data are expressed as mean ± SEM. Genotype effect: *p < 0.05; **p < 0.01. Age effect: ^#p < 0.05; ^##p < 0.01.

Figure 8

Novel object recognition test in (A–D) I/I and (E–H) I/P mice. (A and E) Habituation session. Spontaneous alternation (expressed as the number of alternations divided by the total number of entries minus 2) and the total number of entries into the arms of the Y maze. Data are expressed as mean ± SEM. (B and F) Time of exploration during the familiarisation session. (C and G) Time of exploration during the recognition session. (D and H) Discrimination index was calculated by subtracting the time spent exploring the familiar object from that spent exploring the novel object and then dividing the sum by the total exploration time, before multiplying it by 100 to obtain a percentage.

Figure 9

Elevated plus maze test for (A–D) I/I and (E–H) I/P mice. (A and E) Stacked bar chart showing the percentage of time spent in open or closed arms or in the centre of the maze. (B and F) Ratios of the time spent in the open arms to that spent in the closed arms. (C and G) Number of entries into the arms. (D and H) Stacked bar charts showing the distance travelled (cm) in open or closed arms or in the centre of the maze. Data are expressed as mean ± SEM. *p < 0.05; **p < 0.01.

Figure 10

Rotarod motor coordination test for (A–C) I/I and (D–F) I/P mice. (A and D) Training session. Scores indicate the number of times the mice fell off the apparatus at 4 rpm. (B and E) Latency to fall off at different velocities. (C and F) Latency to fall off at increasing velocity from 4 to 40 rpm in 1 minute in the acceleration session. Data are expressed as mean ± SEM. ***p < 0.001.

Figure 11

Hot plate test in (A) I/I and (B) I/P mice compared to WT. Latency to lick front paws and to jump was recorded. Data are expressed as mean ± SEM. *p < 0.05; **p < 0.01.

Figure 12. Analysis of paw prints.

(A) Schematic of the footprints analysed in I/I mice compared to WT. The mean distance between steps and strides (cm) in (B) I/I and (E) I/P mice. Number of steps recorded in 20 cm of runway in (C) I/I and (F) I/P mice. Paw angles measured as degrees in (D) I/I and (G) I/P mice. Data are expressed as mean ± SEM. *p < 0.05; ***p < 0.001.