Sensorimotor and Neurocognitive Dysfunctions Parallel Early Telencephalic Neuropathology in Fucosidosis Mice

Abstract

Fucosidosis is a lysosomal storage disorder (LSD) caused by lysosomal α-L-fucosidase deficiency. Insufficient α-L-fucosidase activity triggers accumulation of undegraded, fucosylated glycoproteins and glycolipids in various tissues. The human phenotype is heterogeneous, but progressive motor and cognitive impairments represent the most characteristic symptoms. Recently, Fuca1-deficient mice were generated by gene targeting techniques, constituting a novel animal model for human fucosidosis. These mice display widespread LSD pathology, accumulation of secondary storage material and neuroinflammation throughout the brain, as well as progressive loss of Purkinje cells. Fuca1-deficient mice and control littermates were subjected to a battery of tests detailing different aspects of motor, emotional and cognitive function. At an early stage of disease, we observed reduced exploratory activity, sensorimotor disintegration as well as impaired spatial learning and fear memory. These early markers of neurological deterioration were related to the respective stage of neuropathology using molecular genetic and immunochemical procedures. Increased expression of the lysosomal marker Lamp1 and neuroinflammation markers was observed throughout the brain, but appeared more prominent in cerebral areas in comparison to cerebellum of Fuca1-deficient mice. This is consistent with impaired behaviors putatively related to early disruptions of motor and cognitive circuits particularly involving cerebral cortex, basal ganglia, and hippocampus. Thus, Fuca1-deficient mice represent a practical and promising fucosidosis model, which can be utilized for pathogenetic and therapeutic studies.

Keywords: behavior; fucosidosis; learning and memory; lysosomal storage disorder; motor function; mouse model; neuropathology.

Figure 1

Increased Lamp1 expression in the brain of Fuca1-deficient mice. (A) Immunofluorescence staining of brain sections from 3-month-old mice showed an increase in the lysosome-associated membrane protein 1 (Lamp1) in the cerebellum, hippocampus, motor cortex and striatum of Fuca1-deficient animals. Scale bars = 200 μm and 50 μm for magnifications of the hippocampus (CA1). Insets represent further magnifications of the boxed areas, clearly illustrating increased Lamp1 immunoreactivity in Fuca1-deficient mice. Nuclei were stained with DAPI. (B) Immunoblotting of tissue homogenates from 3-month-old mice revealed an increased amount in Lamp1 in the cerebrum but not in the cerebellum of Fuca1-deficient mice (mean ± SD, n = 3). Gapdh was used for normalization. Asterisk indicates significance of difference.

Figure 2

Increased expression of neuroinflammation markers in Fuca1-deficient mouse brains. (A) Quantitative PCR analysis of mRNA cerebrum and cerebellum revealed increased transcript levels for CD68, Iba1 and GFAP in 3-month-old Fuca1-deficient mice. (B) Tissue homogenates from cerebrum and cerebellum were separated by SDS-PAGE and blotted onto PVDF membrane. After blotting, the membrane was cut and the part of the membrane representing proteins larger than 25 kDa was analyzed for GFAP. An apparent increase in GFAP protein was found in the cerebrum but not in the cerebellum of 3-month-old Fuca1-deficient mice as shown by immunoblotting of tissue homogenates (mean ± SD, n = 3), but both comparisons did not reach significance. Gapdh was used for normalization. Immunofluorescence staining showed elevated amounts of CD68 (C) and GFAP (D) in the cerebellum, hippocampus, motor cortex and the dorsal striatum of 3-month-old Fuca1-deficient mice. The cell nuclei were stained with DAPI. Scale bars = 200 μm. Asterisk indicates significance of difference.

Figure 3

Reduced Plp1 and MBP mRNA expression but intact myelination in the brain of Fuca1-deficient mice. (A) Quantitative PCR analysis of mRNA from cerebrum and cerebellum revealed decreased levels of Plp1 and MBP in 3-month-old Fuca1-deficient mice. (B) Tissue homogenates from cerebrum and cerebellum were separated by SDS-PAGE and blotted onto PVDF membrane. After blotting, the membrane was cut and the part of the membrane representing proteins smaller than 25 kDa was analyzed for MBP. The amount of MBP was not significantly lower in the Fuca1 gene knock-out as shown by immunoblotting of cerebral and cerebellar tissue homogenates of 3-month-old mice (mean ± SD, n = 3). Gapdh was used for normalization. The Gapdh results shown here and in Figure 2B are identical, as the same blot was used for the analyses. (C) Myelination is unaffected in the Fuca1 gene knock-out as shown by immunofluorescence staining of MBP in brain sections of 3-month-old mice. The cell nuclei were stained with DAPI. Scale bars = 200 μm (cerebellum (left panel) and striatum) or 50 μm (cerebellum (right panel) and corpus callosum). Asterisk indicates significance of difference.

Figure 4

Fuca1-deficient mouse brains show normal levels of NeuN protein. (A) Immunofluorescence staining of brain sections from 3-month-old mice showed no alterations in NeuN protein in the cerebellum, hippocampus, motor cortex and dorsal striatum of Fuca1-deficient animals. Scale bars = 200 μm. (B) The amount of NeuN protein is unaffected in the Fuca1 gene knock-out as shown by immunoblotting of cerebral and cerebellar tissue homogenates from 3-month-old mice (mean ± SD, n = 3). Gapdh was used for normalization.

Figure 5

Fuca1-deficient mice exhibit accumulation of GM2 ganglioside throughout the brain. Representative immunofluorescence staining of brain sections from 3-month-old mice showed signs of GM2 ganglioside accumulation in all analyzed regions (cerebellum, hippocampus, cerebral cortex, striatum) of the Fuca1-deficient brain. Magnified images of boxed areas illustrate prominent secondary lipid storage. Scale bars = 200 μm. Nuclei were stained with DAPI.

Figure 6

Fuca1-deficient mice display impaired balance beam performance but no significant changes in open field behavior. (A,B) Balance beam performance in 3-month-old Fuca1-deficient mice (Fuca1 -/-, gray bars) and age-matched wildtype (WT) littermates (black bars). Fuca1-deficient mice were slower to traverse all square and round beam types in the balance beam test (A). They also made significantly more paw slips in comparison to WT littermates on all beams except SQ1 (B). (C,D) Open field behavior in 3-month-old Fuca1-deficient mice (Fuca1 -/-, gray bars) and age-matched WT littermates (WT, black bars). No significant changes were observed during open field exploration in 3-month-old Fuca1-deficient mice in comparison to WT animals, as illustrated by total path length (C) and the relative distance traveled in the center of the arena (D). All data are presented as mean ± standard error of the mean (SEM; both genotypes n = 15). Abbreviations: BB, balance beam; SQ, square; RO, round; OF, open field. Asterisk indicates significance of difference.

Figure 7

Fuca1-deficient mice show reduced exploratory activity but no changes in anxiety levels. (A–C) Elevated plus maze exploration in 3-month-old Fuca1-deficient mice (Fuca1 -/-, gray bars) and age-matched WT littermates (black bars). Total beam crossings tended to be decreased in Fuca1-deficient mice (A), but relative exploratory activity (B) and time (C) in open vs. closed arms was similar to WT littermates. (D) Y-maze exploration in 3-month-old Fuca1-deficient mice (Fuca1 -/-, gray bar) and age-matched WT littermates (WT, black bar). Fuca1-deficient mice were significantly less active in the Y-maze. The low amount of arm visits precluded assessment of spontaneous alternations for evaluation of working memory. All data are presented as mean ± SEM (both genotypes n = 15). Abbreviations: EPM, elevated plus maze. Asterisk indicates significance of difference.

Figure 8

Fuca1-deficient mice display impaired spatial learning during water maze acquisition. (A–F) Morris water acquisition in 3-month-old Fuca1-deficient mice (Fuca1 -/-, white circles) and age-matched WT littermates (black circles). Fuca1-deficient mice were significantly slower to locate the hidden platform during the first 3 days of acquisition (A, escape latency). Furthermore, they traveled more distance than WT mice before reaching the platform during days 2 and 3 (B, path length). However, these mice also showed reduced swimming velocity during days 1–4 (C) as well as a tendency for increased immobility (D). Adjusted escape latencies (E) and path lengths (F) showed residual defects in spatial learning of Fuca1-deficient mice after taking into account variability in swimming velocity (during days 2 and 2–3 respectively). All data are presented as mean ± SEM (both genotypes n = 15). Asterisk indicates significance of difference.

Figure 9

Fuca1-deficient mice display intact spatial but impaired fear memory (A–E). Probe trial performance in 3-month-old Fuca1-deficient mice (Fuca1 -/-, gray bars) and age-matched WT littermates (WT, black bars). Fuca1-deficient mice were slower to cross the virtual platform zone during the first probe trial (A) but also swam slower during this trial in comparison to WT littermates (B). Virtual escape latencies were adjusted taking into account variability in swimming velocity, showing no residual differences (C). Similar probe trial performance was further confirmed by other measures of spatial memory such as time spent in the target quadrant (D) and average distance to target (E). Importantly, only sufficiently mobile animals were included for probe trial analysis (see “Materials and Methods” section final n: WT = 12, Fuca1 -/- = 7). (F) Passive avoidance learning in 3-month-old Fuca1-deficient mice (Fuca1 -/-, gray bars) and age-matched WT littermates (black bars). Step-through latencies of Fuca1-deficient mice did not differ from WT mice during the training phase. During contextual memory assessment 24 h later, both genotypes showed an increase of step-through latencies, indicating significant retention. However, Fuca1-deficient mice showed significantly reduced latencies in comparison to WT mice. Notably, three Fuca1-deficient mice were excluded from final analysis as they did not enter the dark compartment during the training trial (final n: WT = 15, Fuca1 -/- = 12). All data are presented as mean ± SEM. Asterisk indicates significance of difference.